(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Berkeley And Malebranche"

> DO 

64455 



BERKELEY 
AND 

MALEBRANCHE 




I 






55? 



y 

CO 



1 



OXFORD 
UNIVERSITY PRESS 

AMEN HOUSE E.C. 4 

London Edinburgh Glasgow 
New York Toronto Melbourne 
Capetown Bombay Calcutta 

Madras Shanghai 
HUMPHREY MILFORD 

PUBLISHER TO THE 
UNIVERSITY 



PRINTED IN GREAT BRITAIN 



PREFACE 

OUT of a close examination of the Commonplace Book 
sprang my desire to reconstruct the genesis of the 
Berkeleian philosophy, by following up the clues contained 
in the two famous note-books. Reading where Berkeley 
read, using sometimes the volumes he used, I studied the 
authorities he mentions. Malebranche was to me, at the 
outset, simply one of a large number of such authorities. 
The extent of Berkeley's debt to the Recherche de la VMte 
came to me as an unanticipated discovery. The following 
pages contain my attempt at detailed verification and show 
surmise passing into assurance. 

As befits a study in origins, the book gives an exposition 
of its thesis on the basis of empirical fact. I have avoided 
the a priori of history, and have taken no account of logical 
relationship. 

My thesis is somewhat complex. I have tried to show that 
Berkeley, when he was preparing to .write his early books, 
made a thorough study of the Recherche > that" the Berkeleian 
philosophy still bears the specific impress of that study, and, 
in particular, that the conception of seeing all things in God 
is at the back of the Berkeleian idea. The questions of 
literary dependence and of doctrinal affinity are closely knit, 
and I have not found it practicable to treat them separately. 
The case rests on evidence drawn from several sources, and 
the evidence, naturally, is not all on the same level of impor- 
tance or cogency. A few parallels and resemblances have 
been included, partly because they might be missed, if not 
mentioned, and partly because, though they are not proof, 
they may lend support. Some readers will find, I hope, that 



vi PREFACE 

my main contention is already familiar; to some, I doubt 
not, it will come as a novelty; all, perhaps, may be glad to 
have the evidence for the connexion presented with a 
fullness not hitherto attempted. 

This is primarily a book about Berkeley. My task obliged 
me to sketch the features of his metaphysic, but has not 
required me to deal at equal length with Malebranche's 
system. I have attempted to prove affinity, but not identity. 
There are elements in the philosophy of Malebranche that 
were repugnant to Berkeley and are alien to Berkeleianism. 
While I have not defined my attitude to either system, I have 
treated Berkeley's doctrine sympathetically, at times to the 
extent of illustrating or even defending a tenet of his. I am 
not without hope that the emphasis here placed upon the 
Malebranchian factor may help to correct one-sided inter- 
pretations which portray Berkeley as a subjective idealist or 
a solipsist, and to check the habit of treating this distinctive 
philosophy as a mere link between Locke and Hume. 

I have to thalit Professor H. S. Macran and Mr. F. La T. 
Godfrey, my colleagues, and Mr. T. E. Jessop, Ferens Pro- 
fessor of Philosophy in the University College of Hull, for 
reading the proof-sheets. Their suggestions and criticisms 
have been of great service to me. My debt to books is not so 
easy to determine. My first aim, in some measure carried 
out, has been to read Berkeley himself and all the books he is 
known to have read before 1710; but I have not neglected 
recent works on that account, and in particular I recognize 
my indebtedness to Dr. G. A. Johnston's edition of the 
Commonplace Book. I have had for some months past the 
advantage of using Professor Jessop 's bibliography, before 
it went to press, and I have drawn from it not only facts, but 



PREFACE vii 

also encouragement. Many, besides professed Berkeleiaiis, 
will read with satisfaction its plain record, down the 
decades, of Berkeley's gathering influence upon European 
and American thought. 

A. A. L. 

Trinity College, Dublin. 



CONTENTS 

FRONTISPIECE. Page 118 of the Commonplace Book 

LIST OF ABBREVIATIONS xi 

CHRONOLOGICAL LIST OF BERKELEY'S PHILO- 
SOPHICAL WORKS xii 

INTRODUCTION i 

New light on Berkeley's sources. The Commonplace Book 

Berkeley's debt to Locke and Malebranche 

His originality 

Contemporary and subsequent views as to his sources 

I. THE RULES OF METHOD 13 

The Arithmetica and Miscellanea Mathematica 

The 'Angelic Rules' 

Malebranche's Method 

The Commonplace Book, an exercise in Method 

Principles and Faculties 

II. THEORIES OF VISION 25 
The genesis of the New Theory 

The Theory of Vision in relation to the Principles 

Sketch of Berkeley's theory 

The Molyneux Problem 

Heterogeneity of sight and touch 

Inverted retinal image 

Molyneux 's Dioptrics 

Malebranche on Vision 

Detailed parallels between the Theory of Vision and the Search 

III. THE BACKGROUND OF IMMATERIALISM 47 
Matter, a recent problem 

Raphson, Cheyne, and Newton on Matter 
Theologians, King and Clarke, on Matter 
Bayle's influence on Berkeley 

IV. PROBLEMATICAL IDEALISM AND ESSE PERCIPI 56 

Inception of the 'Immaterial Hypothesis* 
Malebranche's Doctrine of Matter 
Primary and Secondary Qualities 
Modifications of the soul 
The 'conversion' to immaterialism 

V. SEEING ALL THINGS IN GOD 70 
Berkeley's term 'idea* 

Locke on the nature of ideas 
Idea in the Commonplace Book 
Interpretation of Principles, section i 
Malebranche's doctrine of ideas 



x CONTENTS 

The five possibilities 

Real ideas 

Ideas of the imagination 

Malebranche on the imagination 

Occasionalism 

VI. THE PRINCIPLES OF KNOWLEDGE 92 

Genesis of Berkeley's epistemology 
Knowledge without ideas 
Knowledge of ideas 
Knowledge of the Self 
Knowledge of other finite spirits 
Knowledge of God 

VII. ABSTRACT IDEAS 126 
True and false abstraction 

The absurd triangle 

Locke on abstract ideas 

Development of his doctrine 

Real and nominal essences 

Abstract ideas in the Commonplace Book 

Malebranche on abstraction 

VIII. THE ALLEGED WITHDRAWAL 148 

Abstract ideas in the Theory of Vision 

In the Principles The Introduction and the Draft Intro- 
duction 

Consequences of rejecting abstract ideas 
In the Three Dialogues 
In the De Motu 
In the Alciphron 

The changes in the third edition 

In the Analyst and Defence of Free-Thinking in Mathematics 
In Siris 
The solidarity of Berkeley's philosophy 

APPENDIX I. BERKELEY'S COMMONPLACE BOOK 178 
Its Purpose 
The marginal signs 
Interpretation of the symbols 
The Fellowship Examination and the date 
Its structure 
Orthography of 'idea* 
Conclusions as to date 
Classification of entries 
Textual corrections 

APPENDIX II. ^PHE PROBLEM OF BERKELEY'S VISITS 
TO MALEBRANCHE 208 

INDEX 211 



LIST OF ABBREVIATIONS 

C.P.B. Berkeley's Commonplace Book. The number, if given, 
is that of the entry as numbered in G. A. Johnston's edition. 

T.V. Berkeley's An Essay towards a New Theory of Vision. 

T.V.V. Berkeley's The Theory of Vision .... Vindicated. 

Prin. or Principles. Berkeley's A Treatise concerning the Prin- 
ciples of Human Knowledge. 

Int. or Intro. Berkeley's Introduction to the Principles. 

Hylas. Berkeley's Three Dialogues between Hylas and Philonous. 

Ale. Berkeley's Alciphron or the Minute Philosopher. 

Search. De la Recherche de la Vdrite, by Pere Malebranche 
(N.B. A. Collier, 1713, uses this abbreviation in his Clavis 
Universalis) . 

L.L. Life and Letters of George Berkeley, by A. C. Fraser (1871). 

Works. The Works of George Berkeley ', by A. C. Fraser (1901). 

The Three Dialogues, unlike the majority of Berkeley's works, 
has no numbered sections. So in references to it I give the page 
number in Fraser's (1901) edition of the Works, vol. i. 
Quotations from Malebranche 's Recherche de la Verite are taken 
from the English translation by T. Taylor, ist edition, 1694, 
Oxford and London. 



CHRONOLOGICAL LIST OF BERKELEY'S 
PHILOSOPHICAL WORKS 



Date of writing 
or publication 

c. 1706 
c. 1707-8 

1707 



1709 

Nov. -Dec. 1708 
May 1710 
May 1713 
1721 
1725 

1732 



1733 
1734 



1734 
1735 

1735 
1744-7 

1752 
1752 



Abridged titles 

Of Infinites 

Commonplace Book 

Arithmetica and Miscellanea Mathe- 
matica 

Theory of Vision (first edition, and 
second edition revised) 

Draft introduction to the Principles 

Principles (first edition) 

Three Dialogues (Hylas) (first edition) 

De Motu 

Three Dialogues (second edition, un- 
changed) 

Alciphron (first edition, and second 
edition revised) with Theory of 
Vision (further revised, appended 
to both editions) 

Theory of Vision Vindicated 

Principles (second edition, revised) 
with Three Dialogues (third edition, 
revised) appended 

Analyst 

Defence of Free- Thinking in Mathe- 
matics 

Reasons for not replying 

Siris (at least six editions in 1744, 
another in 1747) 

Alciphron (third edition, further re- 
vised) 

De Motu (reprinted in Miscellany) 



INTRODUCTION 

A STUDY in the origins of Berkeley's philosophy is emin- 
ently practicable. The reasons are plain. In the first 
place, Berkeley completed the main lines of his philosophy 
in very early manhood, while the creative impulse was 
strong within him and the external formative influences 
comparatively few. Again, we possess in the Commonplace 
Book an eyewitness of his book-making. Here, open to our 
inspection, are the two note-books in which the young 
author entered systematically the main heads of the philo- 
sophy he was struggling to shape and express ; note-books 
in which he records his silent arguments with himself and 
with his authorities. Thus we know what authorities he 
consulted and can estimate fairly accurately the degree of 
respect he paid to each. The reliability of this source of 
information is unquestioned. The relation between the 
Commonplace Book and the Theory of Vision and the Prin- 
ciples is very close. The two publications could hardly be in 
their present shape but for the work done in these two note- 
books, and certainly the note-books would be to us a sheer 
mystery if we had not the publications in which they issued. 
There is scarcely a point in his published philosophy not 
dealt with in his Commonplace Book, except knowledge of 
other spirits. We might infer that during the period 1707-8 
Berkeley could not have consulted an author or book at all 
extensively without disclosing the fact to us. There is little 
need to say, as in some studies of origins, 'he must have 
used this source'; we can say 'he did use this source', and 
can often point to chapter and verse. 

Further, we are fortunately placed for this study in that 
we possess a group of early subsidiary writings of his, 
which at certain points throw light on the genesis of his 

4046 



2 INTRODUCTION 

thought. These are, in their probable time-order, the 
Description of the Cave of Dunmore, the 'Of Infinites', 
the Arithmetica and Miscellanea Mathematica, the draft 
Introduction to the Principles, the draft of the Principles 
sect. 85-145, and the letters to Percival. 

Berkeley has been so long in the public eye that it would 
be natural to assume that we already know all there is to 
know about his sources. But new documents bring new 
knowledge. The Commonplace Book is still, in one sense, a 
new document. First published by Fraser in 1871, not till 
G. A. Johnston's edition of 1930 did it become available in 
a form proper for systematic study. Even yet perhaps, in 
spite of the studies of Lorenz, Aaron, and others, the higher 
criticism of the document is not complete. The Hermathena 
article, reprinted as an appendix to this book, is an attempt 
at a comprehensive solution of the chief problems. Its 
conclusions about the purpose, structure, and date of the 
Commonplace Book have been here relied upon by the 
author. It is hoped that they may prove fairly trustworthy. 
As my study in the origins of Berkeley's thought sprang 
directly from a critical examination of the Commonplace 
Book, I may perhaps be allowed to assemble here the 
main results of that examination, referring the reader to 
the Appendix for the detailed evidence. These are as 
follows : 

(1) The Commonplace Book is not a collection of casual 
'jottings', but is a definite composition, undertaken 
by Berkeley as a preparatory study in the argument 
for immaterialism. It is therefore to be regarded as 
a preliminary stage in the composition of the New 
Theory of Vision and the Principles, and to be viewed 
in intimate connexion with these two works. 

(2) Johnston's text, on the whole, gives the entries in 
their proper sequence ; but the entries, numbered by 



INTRODUCTION 3 

him 903-45, should be restored to their manuscript 
position after 395. l 

(3) The Commonplace Book proper was not begun before 
December 7, 1706, and was finished before Novem- 
ber 1708. There are grounds for holding that it was 
not begun till after June 9, 1707, and was completed 
on August 28, 1708. 

(4) The apparatus of marginal signs and the change in 
the orthography of the word 'idea' can throw some 
light upon the development of Berkeley's thought 
and the making of his books. 

A. J. Balfour opens his essay on Berkeley's life and letters, 
with the words, 'It is as a descendant in the true line of 
succession from Locke to the modern schools of thought 
. . . that he is and that he'deserves to be chiefly remembered'. 
Fraser says, 'Berkeley is the immediate successor of Locke 
and . . . was educated by collision with the Essay on Human 
Understanding 9 . 2 G. A. Johnston is emphatic to the same 
effect: 'From Locke only did he really derive anything of 
the first importance. The original impulse and direction 
of his philosophy came from Locke, and from Locke also 
the great Gemeingut of ideas which makes the continuity 
between them as remarkable as their differences.' 3 

1 Dr. Dawes Hicks suggests taking entry No. 70 as the first. 
I have examined the manuscript ad hoc and cannot see in it any 
justification for the suggestion. Sixty-nine and seventy are more 
or less continuous in subject-matter. Dr. Dawes Hicks does not 
press the suggestion or build on it. If he did, I think he would 
find it raises more problems than it settles. His difficulty is 
solved, I think, by the consideration that Berkeley was, at this 
stage, taking immaterialism hypothetically, and perhaps arguing 
with future readers. See G. Dawes Hicks, Berkeley, p. 27. 

2 Works, vol. i, (i 901), p. xi. The same judgement occurs else- 
where in Fraser 's writings. 

3 G. A. Johnston, The Development of Berkeley's Philosophy, 
p. 32. cf. pp. 67-73 an d 370 for the influence of Malebranche. 



4 INTRODUCTION 

My purpose is to try to recall opinion to an earlier and, 
I believe, a truer view. With the Fraser- Johnston judge- 
ment about Berkeley's sources firmly fixed in my mind, 
when engaged upon a close study of the Commonplace Book, 
I happened on the following entry (274), 'From Malbranch 
Locke, and my first arguings it can't be prov'd that extension 
is not in matter.' Pondering the implications of that entry, 
I studied the other references to and mentions of Male- 
branche in the Commonplace Book. Gradually I became 
convinced that Berkeley studied the Recherche de la Verite 
fully and thoroughly. Seeing reason to think that he read it 
in Taylor's translation, 1 1 made a close study of that trans- 
lation and found the debt to be greater than I had antici- 
pated. Returning to Berkeley's works with the details of 
the Search fresh in my mind I thought that the familiar 
words and thoughts of the Principles took on a new mean- 
ing, and the Berkeleian system a new firmness and depth 
and solidity. Hence this book. 

A student who reads Berkeley on the look out for Male- 

1 T. Taylor, M.A., of Magd. Coll. in Oxford, 1694. Berkeley 
probably used the second edition, of 1700, which contains the 
short tract on Light and Colours 'never before printed in any 
language*, perhaps referred to in C.P.B. 932. A copy is in T.C.D. 
Lib., with 1701 in ink on cover. My reasons for thinking Taylor's 
translation used are many. I need not give them seriatim, as they 
come out in the body of my work. There are several verbal echoes 
of Taylor in the C.P.B., notably the term 'illustration' in 812 for 
Eclair cissement , Malebranche's technical term for excursus. Com- 
mentators have missed this meaning of 'illustration', which is 
found e.g. in Keill's Exam. Burners Theory, 1698, p. 8, 'The 
oftner I read his (Malebranche's) long Illustration on this point*. 
For 'outness* in Taylor's translation of the 'illustration* men- 
tioned in C.P.B. 812, v. infra, p. 46. Accordingly I quote uni- 
formly from Taylor's translation of the Recherche and use his 
title for the book, the Search, the title familiar to Berkeley and 
to his English contemporaries. 



INTRODUCTION 5 

branche, and Malebranche on the look out for Berkeley, is 
in serious danger of 'reading in' what is not there. I have 
been alive to the danger, and my consequent attempt to be 
'objective' may be my apology for the excessive documenta- 
tion of the following pages. 

My primary aim, then, is to establish on internal evidence 
the position that Berkeley used the Search extensively, and 
that Malebranche was to him not merely a Cartesian (Male- 
branche often distinguishes himself from the Cartesians), 
but at any rate for a time, indeed for the critical time, a 
master-mind, one of his two primary sources, comparable 
in importance to Locke and, in some ways, of higher 
importance. 

What difference can the establishment of so narrow and 
so literary a thesis make ? What does it matter to philo- 
sophers to-day what Berkeley's sources were? I shall try 
to suggest in the course of my argument, especially in 
Chapters V and VI, a full answer to those questions. Here 
I will only say that if Bergson's remark 1 is just, that every 
philosophy of recent times has had to take its start by 
reckoning with the contentions of Berkeley, then the source 
of Berkeleianism matters to all philosophers to-day; for 
on the source the spirit of the philosophy will in large 
measure depend. Fraser wrote (and many moderns would 
endorse his words) : 'Berkeley would be less at home in the 
"divine vision" of Malebranche than among the "ideas" of 
Locke'. 2 I should like, I venture to confess, to see that 
judgement reversed. I believe that Fraser here is wrong 
in point of fact, and that this and similar judgements of his 
have led many to misconceive the spirit of the philosophy. 
Neglect of the Malebranche factor in Berkeleianism takes 
the heart out of it, reducing its solid reality to flimsy dream, 

1 Quoted with approval by Dawes Hicks, op. cit. p. 285. 

2 Works, vol. i, p. 217. 



6 INTRODUCTION 

negating Berkeley's claim to stand in the Tantheon of god- 
like figures'. Recent studies, notably Dr. Dawes Hicks's 
Berkeley, have shown that Berkeley is no mere link between 
Locke and Hume; but Berkeley's right to the rank of 
Leader of Philosophy will be open to challenge until the 
lengthening tradition of his sole dependence upon Locke 
is once and for all broken. 

I have not attempted to give a complete presentation 
here of Berkeleianism. It would be unnecessary to do so 
while Eraser's edition of the Works and the careful and 
critical studies in the system by Johnston and Dawes Hicks 
are accessible. But I have tried so to present my primary 
argument that the Theory of Vision and the Principles will 
stand out sharply, in their main positive features, before 
the reader's eye. I have scarcely noticed Berkeley's impor- 
tant answers to objections and arguments in support. The 
book may thus serve as an introduction to Berkeleianism. 
At the same time advanced students will find in it several 
critical studies in disputed points of Berkeleian exegesis. 

I have not even attempted here to survey the whole field 
of Berkeley's sources. My study then is not open to the 
charge of underestimating Locke's influence upon Berkeley. 
I have not tried to give an estimate of it. Others have per- 
formed that task. To neglect the Lockian element in 
Berkeley would be to fly in the face of the facts, 1 and I have 
no wish to go from one extreme to another. The balanced 
view surely would be that Malebranche and Locke both 
exerted a potent influence upon the young Irishman in his 
plastic days, and that the two influences were so hetero- 
geneous that a comparison of weight and value is futile. 
Many reflective scholars can remember some schoolmaster 
of theirs to whom they are indebted for years of solid 

1 Within two years of its publication (1690) Locke's Essay was 
on the course at Trinity College Dublin. 



INTRODUCTION 7 

grounding. But they remember too a personality who 
crossed their path perhaps after school and even after 
college days, and who was to them an intellectual stimulus 
of a higher order. Contact may have been brief, but the 
influence deep and strong. It was so with Berkeley, I 
submit. Locke taught him, but Malebranche inspired him. 

My limited objective should exempt me also from the 
charge of impugning Berkeley's originality. I would de- 
precate that charge. Originality, it should be unnecessary 
to say nowadays, does not stand in thinking the thing all 
out for oneself de novo. The 'original' philosopher is no 
recluse out of touch with his times, but the philosopher who 
makes creative, or at least novel, use of the material fur- 
nished by the representative thought of his day. Berkeley 
did so, but it is beside my purpose to show it. Suffice it 
here to say that Berkeley was original in all his ways. He 
nearly hanged himself in his college days, as an experi- 
ment, no doubt, in trains of ideas. He was original in wish- 
ing to give America the episcopate and make her spiritually 
independent. He was original in his views on Irish govern- 
ment and economics. He was original in his therapy. 
'Neminem transcripsi : nullius scrinia expilavi', he wrote in 
the Preface of his first publication. It is not to be thought 
that such a man would slavishly copy or re-edit the system 
of Malebranche or of Locke. 

If he found in Malebranche the hint of the heterogeneity 
of sight and touch, it was of himself to see the connexion 
with the Molyneux problem, and find there a clue to the 
nature of things. Thousands read Malebranche without 
dreaming of immaterialism. Norris and Collier, contem- 
poraries of Berkeley, steeped themselves in Malebranche 's 
philosophy, but originated little. The divine spark was not 
in them. Berkeley's rejection of abstract ideas, his attempt 
to turn ideas into things, his account of the sensible world, 



8 INTRODUCTION 

be they right or wrong, were strokes of genius. And he 
knew his own originality. 'So far as authority is of any 
weight with me', he wrote in his draft Introduction to the 
Principles. 1 Twenty-five years later he writes: 'The only 
advantage I pretend to is that I have always thought and 
judged for myself.' 2 The entry in the Commonplace Book 
which gave rise to this study in origins brackets his own 
'first arguings 5 with his two other authorities, Malebranche 
and Locke. 

This study is as far then from detraction as from adulation. 
It is simply an attempt to elicit and weigh the facts about 
Berkeley's use of the Search. Some may think that the thesis, 
however intended, would, if established, discredit Berkeley 
by proving him guilty of suppressio veri, if not of suggestio 
falsi. Neither charge can, I think, be sustained ; nor does 
my thesis lend colour to either of them. Berkeley's words 
about Malebranche's system and his silences will be dis- 
cussed later. Here I will only say that Berkeley nowhere 
denies his indebtedness to the Oratorian himself. An author 
is not bound to declare his sources, even if he knows them, 
and perhaps he ought not to do so, if, as in Berkeley's case, it 
is clearly against his interest to do so. After all, Berkeley in 
November 1713, within a week of his first arrival in Paris, 
secured an introduction to Father Malebranche and made 
arrangements to visit him. 3 If, when he went, (it may have 
been to say 'thank you',) he stayed to dispute, perhaps the 
fault was not all on the side of the younger man. 

1 Works, vol. iii, p. 363 ; cf. C.P.B. 464. 

2 Def. of Free-thinking in Math. 19. 

3 Rand, Berkeley and Percival, p. 129; Fraser, L.L., p. 67; 
v. Appendix ii, p. 208, for discussion. Locke's own debt to Male- 
branche was considerable ; yet the Essay contains no recognition, 
and his posthumously published Examination treats the Search 
with scant respect. Essay, bk. iv, c. 19 seems to have Malebranche 
in view. See Ep. to Reader and Fam. Letters, p. 42. 



INTRODUCTION 9 

A few representative views upon Berkeley's sources, in 
addition to those already quoted, may be collected here. 
Clarke and Whiston, the leaders of English thought of that 
day, who read the Principles a few months after its publi- 
cation rank its author with Malebranche and Norris. 1 
Bolingbroke writes to Swift, regretting that he had missed 
seeing Berkeley, 'I would not by any means lose the oppor- 
tunity of knowing a man who can espouse in good earnest 
the system of Father Malebranche' (July 24, 1725, L.L., 
p. 118). The Ada Eruditorum (1727, pp. 380-3), after 
reviewing the Three Dialogues, says: 'De origine, quicquid 
Autor dissimulet, sic sentimus, ex Cartesii, Malebranchii et 
Spinosae philosophiarum mixtura prognatum hoc XvpLKov 
Orjpiov*. Baxter's Enquiry (1733, vol. ii. 2) says little of 
Berkeley's sources, but suggests that Berkeley may have 
learned immaterialism from Descartes. Dr. Clayton's 
Essay on Spirit (1750, v. L.L., p. 324 note) opens with the 
statement that Spinoza's opinion 'with some few alterations 
hath been embraced and cultivated by P. Malbranche and 
Bishop Berkeley'. Reid (Inquiry c. i, 5) writes : 'Berkeley's 
arguments are founded upon the principles which were 
formerly laid down by Descartes, Malebranche and Locke', 
and (Int. Powers, Essay II, c. 10) : 'Berkeley's system follows 
from Mr Locke's by very obvious consequence.' On the 
whole Reid affiliates Berkeley to Locke, while Hamilton, 
according to Graham (Idealism, p. 124), finds 'the whole of 
Berkeley's philosophy wrapped up in Malebranche V. 
Beattie (On Truth, 1770, pt. ii. 2. i) says : 'The substance, or 
at least the foundation, of Berkeley's argument against the 
existence of matter may be found in Locke's Essay 2 and in 

1 Rand, op. cit., p. 87, with Berkeley's reply, pp. 88-9; cf. 
Fraser, L.L., pp. 45-6. 

2 The author of the Remarks, p. 41 in the 1776 ed. of the 
Principles, categorically denies this statement as regards Locke. 

4046 c 



io INTRODUCTION 

the Principia of Des Cartes.' Dr. C. R. Teape, in his preface 
to Berkeleian Philosophy (1870) quotes one of the authors 
of Essays and Reviews to the effect that Berkeley 'was not 
a closet-thinker, like his master Malebranche'. To come 
to more recent times, we have already quoted Fraser and 
Johnston. Dr. Dawes Hicks on the whole follows the 
English tradition, saying 'Undoubtedly the teachers to 
whom Berkeley was chiefly indebted were Descartes and 
Locke, above all Locke; and it was from a careful and cri- 
tical study of their systems that his new principle emerged/ 1 
From this welter of opinion certain things emerge. 
Berkeley, in his lifetime, was regarded as a disciple of Male- 
branche. Subsequently he came to be regarded as a Lockian. 
The new opinion was a natural growth. In the course of 
time British acquaintance with Malebranche sank, and the 
fame of Berkeley rose. National sentiment adopted him as 
the English philosopher in succession to Locke. It may 
therefore be in place here to mention the danger of over- 
estimating the degree to which the young Berkeley was 
anglicized. There are two national sentiments to be con- 
sidered, and to hold the balance is not easy. To speak of 
him, without qualification, as an English philosopher cannot 
be right. Leslie Stephen's statement 2 'Berkeley always con- 
sidered himself an Englishman', is misleading, if not mis- 
taken. Berkeley was born and bred in Ireland. His educa- 
tion was entirely Irish. He speaks of himself as an Irishman 
several times in the Commonplace Book. Newton to him was 
'a philosopher of a neighbouring nation'. 3 As with many 

1 Dawes Hicks, Berkeley, p. 70. He adds, however, when deal- 
ing with Berkeley's relation to contemporary philosophers (p. 
229), *I cannot help thinking that Berkeley's indebtedness to the 
writings of Malebranche is apt to be underrated.' 

2 Diet. Nat. Biog. art. *G. Berkeley'. 

3 Principles, no, ist ed. only. 



INTRODUCTION n 

of English descent settled in Ireland, brought up among 
kindly men of another faith and race, his sentiments were 
necessarily mixed and his loyalties divided. But credit must 
go where it is due. Berkeley's system in so far as it forms 
part of the heritage of international philosophy was com- 
plete before he set foot in England, and in England he 
wrote little or no philosophy. 1 

All modern students of Berkeleianism owe a great debt to 
A. C. Fraser, and I do not belittle that debt in giving a reason 
why Eraser's judgement about the origin of Berkeley's 
thought must be received with caution. The fact is that 
Fraser misconceived the state of education in Ireland at the 
time, and therefore came to his subject with a slight preju- 
dice against the education of the young Fellow of Trinity 
College. Holding erroneously that at that time 'Ireland, like 
Scotland, was in a state of provincial barbarism', 2 he came 
to the mistaken judgement (repeated since by other writers) 
implied in his words, 'It does not seem that his scholarship 
or philosophical learning was extensive.' This was the time 
of Swift and King, 3 writers of European reputation, of the 
Molyneuxs, Browne, Gilbert, Palliser, and the scores of 
cultured men who founded the Dublin Society. Berkeley, 
while still too young to take the degree of Master of Arts, 
read Latin, Greek, French, and some Hebrew, had pub- 
lished a work on mathematics in Latin, was versed in con- 
temporary works on theology, philosophy, the higher 
mathematics, and physics, and within two years of pro- 
ceeding M.A. he had written books that are, after two 

1 The De Motu was written in France, the Alciphron in Rhode 
Island, the Theory of Vision Vindicated and possibly the Analyst 
in England, the remainder in Ireland. 

2 L.L., p. 62, v. also Works, vol. i, p. 351, and vol. i, p. 4. 

3 Swift, like Berkeley, was a product of Kilkenny College and 
Trinity College. King was educated at Dungannon Royal School 
and Trinity College. 



12 INTRODUCTION 

centuries, still considered good literature as well as pro- 
found philosophy. 

Then, as now, there was direct commerce in ideas be- 
tween Dublin and the Continent. It was only natural that 
a man of Berkeley's breadth of vision should read and 
respect the great works then being produced in England, 
and should yet take an independent line in their regard, just 
because he was in immediate contact with continental 
thought. Berkeley read Leibniz as early as 1706. Leibniz 
knew of Berkeley's speculations. 1 Two draft Latin letters 
by Berkeley to Leclerc are extant, and the Theory of Vision 
was reviewed at length in the Bibliotheque Choisie, Amster- 
dam, in 1711. Further, we know that Malebranche was 
valued in Berkeley's immediate circle. The Search is placed 
only second to Locke's Essay in the Preface of Molyneux's 
Dioptricks. 2 There is, however, no need to concern our- 
selves with the probabilities of the case. It is certain on the 
evidence of the Commonplace Book that Berkeley used the 
Search extensively. Accordingly this study, basing itself on 
clear external evidence, proceeds to investigate the internal 
evidence of connexion, and thereby to provide a measure, 
it is hoped, of the degree of importance to be attached to 
that connexion. 

1 Leibniz to Des Bosses, March 1715, ed. Erdmann, pt. i, 
p. 726. 

2 v. also Locke's letter to Molyneux of March 28, 1693, Fam. 
Letters, p. 42. In the T.C.D. copy of Malebranche's De Inqui- 
renda Veritate from the Palliser gallery there is a statement penned 
on the fly-leaf under Palliser's name, to the effect that in the 
opinion of Edwards, Stillingfleet, and Burnit 'Malebranche de- 
serves not that mighty esteem he has obtained with some*. 



I 

THE RULES OF METHOD 

FROM his earliest days of authorship Berkeley was con- 
scious of method and critical of method (C.P.B. 405, 
447). The purpose of this chapter is to show that in his 
method of setting and solving problems, in his actual writ- 
ing, and in the spirit of his approach to philosophical ques- 
tions he was directly influenced by what Malebranche had 
said on methodology. Book vi of Malebranche 's Search is 
devoted to method. Berkeley, as we shall show, gives in 
his De Ludo Algebraico a reference to the central feature 
of Book vi and speaks of it appreciatively. So there is 
a solid foundation in fact for our attempt to trace out 
in detail the connexion between the methods of the two 
thinkers. 

Malebranche in Book vi is attempting a Novum Organum. 
In his own words he wants 'to give the mind all the perfec- 
tion it can naturally attain to'. He wants to increase the 
mind's capacity and to perfect its application and attention. 
In Book vi, Part i, he investigates the bases of geometry, 
arithmetic, and algebra. These, he argues, are the disci- 
plines proper to form the philosophic mind. Geometry is 
the proper discipline for the imagination ; for it gives, in 
simple lines, illustrations of forces and of movements, and 
so deepens the mind's attention. Arithmetic makes no ap- 
peal to the imagination, since it does not work in the con- 
crete. Arithmetic uses abstract symbols, and thus develops 
the range and capacity of the mind. Algebra is a higher 
development of the arithmetical method. Algebra is the 
model science for methodology ; for in virtue of its powers 
of abridgement, it is intellect's short-hand, opening up 
illimitable expanses of new territory for the mind. 



i 4 THE RULES OF METHOD 

In Book vi, Part 2, Malebranche deals with the actual 
conduct of the mind in its search for truth. He elaborates 
a code of rules, modelled largely upon Descartes 's rules; 
he illustrates the working of his rules, and makes them an 
occasion for an outline study of the logic of Aristotle and 
the physics of Descartes. Before giving a detailed account 
of these rules we must trace Berkeley's connexion with 
them. 

The methodology of Malebranche was, in some measure, 
common property to all adherents of the 'New Learning*. 
A good deal of it is implicit in Locke's Essay and explicit in 
Locke's Of the Conduct of the Understanding. Berkeley 
might have learned it from Locke or from other writers. 
In point of fact, at the time his attention was called to 
method, he came definitely and directly under the influence 
of the Oratorian. Berkeley's Arithmetica and Miscellanea 
Mathematica furnish the evidence as regards method. The 
Commonplace Book supports it, and so, to a lesser extent, 
do his main publications of the period. 

The Arithmetica was published early in 1707. The im- 
minent fellowship examination, no doubt, occasioned its 
publication then. But the work had been on the stocks for 
two or three years, and was, perhaps, designed originally 
as a study of and an exercise in method. Berkeley intends 
to smooth the path of students of elementary mathematics 
by reducing to principle the network of rules. He writes in 
the Preface : 'Nempe id mihi imprimis propositum fuerat 
ut numeros tractandi leges ex ipsis principiis, proprii exer- 
citii et recreationis causa, deducerem. Quod et deinceps 
horis subsecivis prosecutus sum.' At the conclusion of 
the preface we find a curious phrase, 'Monstravi porro ad 
quern collimaverim scopum.' Berkeley is here obeying a 
general direction given by Malebranche, and the phrase 
must surely be an echo of Taylor's quaint translation of 



THE RULES OF METHOD 15 

that direction 'Still having before your eyes the scope 1 
you aim at'. 

The climax of the publication is the De Ludo Algebraico, 
a piece which, as he says, 'miscuit utile dulci'. From his 
game of algebra he passes to an extravagant but quite 
serious commendation of the 'mirifica algebrae vis'. He 
marshals his authorities an impressive list Halley, 
Temple, Bacon, Descartes, Malebranche, Locke, quoting 
from most of them. Locke comes in for very high praise; 
but the thoughts and, not infrequently, the words are 
those of Malebranche. Berkeley calls algebra 'universae 
matheseos nucleus et clavis'. Malebranche had called it 'an 
universal science . . . the key of all other'. 2 Berkeley adds 
that some call algebra 'the foundation of all the sciences' 
the very term which Malebranche (bk. vi. i, 5) applies to 
arithmetic and algebra. It is 'difficult to assign the limits of 
algebra', says Berkeley; and the unlimited range of know- 
ledge opened up to the mind by the algebraic method is the 
theme of several passages in the Search (ii. 2. 3, iii. i. 3, iv. 
u,vi.i.5). 

This adulation of algebra has no intrinsic importance ; it 
was only a passing phase. I refer to it partly as evidence 
of Berkeley's close reading of the Search (n.b. Berkeley's 
'alibique passim'), and partly because it introduces the 
following important passage: 'Regulae quidem quas hie in 
quaestionum solutione observandas tradit lib. 6 part. 2 cap. 
i. quaeque tarn sunt eximiae, ut meliores angelum non 
fuisse daturum credat auctor quidam ingeniosus; illae, 
inquam, regulae angelicae ex algebra desumi videntur.' 3 

1 Bk. vi, 2. i. Neither the French nor the Latin translation 
(terme, terminus) suggests the word 'scope*. The phrase recurs 
in the De Motu, 38. 

2 Search bk. iv. 1 1 ; cf . Collier, Clavis Univer sails. 

3 Works, vol. iv, p. 61. 



i6 THE RULES OF METHOD 

Berkeley here states the facts correctly ; for Malebranche 
himself (vi. i. 5, vi. 2. i) speaks of these rules as originating 
in algebra, and he mentions the fifth and sixth as especially 
algebraic. 

I must consider in detail these 'angelic rules' ; for they 
express the spirit of his inquiry, and must have influenced 
the interrogations which the young querist put to the uni- 
verse. The code of rules is founded upon that of Descartes. 
Malebranche advances them as a continuation of the mathe- 
matical disciplines, as an alternative to the logic of the 
schools, and therefore as enabling the mind to secure all 
the perfections it is capable of. He claims that the rules are 
simple, coherent, and practical. The code forms in fact a 
system of empiricist method. 1 

The principle upon which they all depend is * Truth with 
evidence*. That principle was common to all the empiri- 
cists of the day, and was the spear-head of their attack upon 
obscurantism, dogmatism, and authority. The general 
rule which results, is expressed in two propositions: (i)* 
We ought only to reason upon such things whereof we have 
clear and distinct ideas. (2) We must begin with the most 
easy and simple subjects and insist long upon them before 
we undertake the inquiry into such as are more composed 
and difficult. Malebranche devotes five chapters to these 
two parts of his general rule. He then treats the six deri- 
vative rules for resolving questions, styling the truthseeker 
the 'Querist'. Now Berkeley was a querist all his life, 2 and 
these six rules for systematic querying were clearly and 
consciously before him in all his early work. 

1 Leclerc, Logica (used by Berkeley, v. C.P.B. 361), pt. iii, c. 4, 
on Method (1692), follows these rules fairly closely; v. his note 
on c. v. 

2 v. conclusion of the Analyst, of the Defence of Free-thinking, 
and Querist. 



THE RULES OF METHOD 17 

The first rule is : * We must very distinctly conceive the 
state of the question proposed to be resolved : that is, have 
ideas of the terms so distinct as that we may compare them 
together, and discover the relations which we look for.' 1 
The querist must put his difficulty in question form. Scores 
of the Commonplace Book entries are questions ; all the more 
striking and concrete ones at their first appearance are intro- 
duced by Qu : or Quaere. To put the problem as a query is 
one way of clarifying the subject under discussion. The 
query tends to get rid of ambiguity and verbal equivocation, 
makes for clear and distinct ideas, and releases the issue 
from the tyranny of words and jargon. The dialectic of 
question and answer is the characteristic method of the 
Commonplace Book. Berkeley's diligent pursuit of clarity 
and of distinct ideas is illustrated by the following entries 
(and there are many others): 25, 27, 30, 226, 367, 545, 
549, 585, 597, 640, 650. Malebranche points out that some 
queries are not in words, and he regards experiments as 
queries addressed to the author of nature. Berkeley was 
fond of such practical querying. He experimented in colour 
composition, and he conducted psychological experiments 
to find out the effect on the mind of such experiences as 
hanging, silence, and wearing inverting glasses. Further he 
followed up in thought experimental conceits such as the 
blind man made to see, the dumb man, and the solitary 
man. 

The second rule is : 'We must try by an essay of thought 
to discover one or several intermediate ideas, that may be 
a means or common measure to know the relations that 
are betwixt those things.' Two entries in the Commonplace 
Book (444, 709) show him perhaps trying to put this rule 
into practice. But as his views on abstract ideas clarified, 

1 Malebranche lists the rules in bk. vi, 2, i , and discusses them 
seriatim in chapters 7-9. 

4046 D 



i8 THE RULES OF METHOD 

he abandoned the method of intermediate ideas, along with 

the doctrine of certainty and demonstration (741). 

The third and fourth rules prescribe respectively isola- 
tion of subject-matter and analysis. Berkeley has clearly 
cultivated the habit of keeping to the point and of excluding 
irrelevant issues. His outline of immaterialism (Prin. i- 
33) is a model of precision. In dealing with objections and 
consequences he is forced to go over the same ground more 
than once, but even where redundant he is always relevant. 
His formed habit of analysis is seen in the construction of 
his books. The Theory of Vision and the Principles fall into 
clearly marked divisions, sections and sub-sections. The 
marginal signs in the Commonplace Book are part of the 
machinery of analysis. To this practice of analysis we may 
trace his clear and simple views of things, his perspicuity, 
and his limpid style. His forte is to see distinctions, con- 
cealed by custom and by language. 1 

The fifth and sixth rules are described by Malebranche 
as particularly algebraic in spirit. The fifth rule lays down 
that we must Abridge ideas and dispose them in the imagi- 
nation or write them upon paper, that they may no longer 
clog and fill up the capacity of the mind'. This rule is for 
very intricate questions, such as require great extent of 
mind. The sixth rule prescribes that when 'the ideas of all 
the things that necessarily require examination are clear, 
familiar, abridged and disposed and ranged in good order 
in the imagination or written upon paper, they are all to 
be compared by the rules of complication, one with the 
other alternately, either by the view of the mind alone, 
or by the motion of the imagination, attended with the 
view of the mind, or by the calculation of the pen joined 
to the attention of the mind and imagination'. Clearly 
Malebranche has in his mind's eye, as model, the setting 

1 For instances of analysis see C.P.B. 237, 323, 396, 752, 865. 



THE RULES OF METHOD 19 

out of some involved mathematical problem in simple 
algebraic notation. 

The Commonplace Book is, I think, a result of and, indeed, 
an embodiment of this pair of rules. 'We have got the 
Algebra of pure intelligences', he says (C.P.B. 926). He 
means, no doubt, that what is real to man is symbolic nota- 
tion to the angelic intelligence (cf . T. V. 155). But we may 
legitimately adapt his phrase and say that the Commonplace 
Book is the algebra of a good human intelligence. In its 
entries mind, imagination, and pen join to fix as with precise 
notation the fluid features of his thought. At first sight it is 
a purposeless compilation, random jottings for Berkeley ; 
for us a fund of quotations. On careful examination it 
proves to be a definite composition, an algebraic abridge- 
ment of the material of his books. The brevity and point 
of the earlier entries is noticeable. Berkeley throughout its 
pages aims at clear, distinct, and coherent ideas. We might 
style the Commonplace Book a 'Recherche de la Verite' con- 
ducted under the auspices of the 'angelic rules' of method. 
Berkeley began it shortly after he was engaged upon the 
final pages of the De Ludo Algebraico 1 where these rules 
are commended and where his language is redolent of the 
Search. Malebranche tells his readers how to 'improve the 
attention and extension of the mind' and make it 'more 
enlightened, sagacious, and piercing' (bk. vi, 2. i. cf. De 
Lud. Alg. 'Ingenium sagax, intellectum capacem, judicium 
acre'). The only way, he says, to enlarge the mind's 'capa- 
city' is to reduce to scale, i.e. abridge the ideas put into it. 
This abridgement is to be followed by repeated comparison 
of ideas. If ever there was a task needing a great extent and 
capacity of the mind, it was that of the young seeker after 
truth, when in the opening pages of the Commonplace Book, 

1 Probably prepared for the press early in 1707 (v. Appendix, 
p. 204). Locke's Opera Posthuma, there quoted, appeared in 1706. 



20 THE RULES OF METHOD 

he abridges his ideas and outlines the 'immaterial hypothe- 
sis'. On a small canvas (1-27) he crowds eternity, time, 
space, primary and secondary qualities, the real world, 
infinite divisibility with the attendant mathematical prob- 
lems, the nature of the soul, and the psychology of vision. 
'Capacity' must have been his prayer. His consciousness 
of the duty of comparing ideas appears repeatedly in the 
Commonplace Book, e.g. 47, 51, 309, 873, and notably in 
the 'demonstration' by comparison (<)i8b-C)22) which ori- 
ginally concluded the first note-book. 

Malebranche represents the search after truth as an intelli- 
gent adventure. 'We should always know whither we are 
going,' he says. Berkeley as he fills his note-books knows 
whither he is going. He is not jotting down ideas on isolated 
topics. All his themes are connected, and he is struggling 
for clear vision and clear expression of that connexion. His 
writing is for him a great adventure. There lies before him 
'much terra incognita to be travel'd over and discovered. 
A vast field for invention' (561). And Malebranche takes 
care to promise a reward for persistency with varied effort. 
He undertakes that the methodical seeker will attain by 
intuition an 'infallible Principle' which he can use in the 
furtherance of knowledge. He writes : 'We must above all 
take care not to satisfy ourselves with some glimpse or likely- 
hood ; but begin anew so often the comparisons that are 
conducible to discover the truths enquired after, as that we 
may not withhold our assent to it, without feeling the secret 
lashes and reproofs of our internal master that answers our 
questions, that is, the application of our minds, and the 
desires of our heart. Then will that truth serve as an infal- 
lible Principle, to proceed in the acquisition of sciences' 
(bk. vi. 2. i) . One could hardly have a better summary of 
the method and achievement of Berkeley's first (B) note- 
book, which culminates in his irrefragable intuitive prin- 



THE RULES OF METHOD 21 

ciple formally stated in 923-4 (cf. 287) and demonstrated 
in form twice in the previous entries (903-22). l Berkeley 
thus attains the reward that Malebranche had promised. 
He has the deep satisfaction of the explorer who achieves 
his aim and sees his method vindicated. As from a Pisgah 
peak, Berkeley sees all of a sudden, par simple vue, what he 
had come to look for. After long analysis and wearisome 
study of intricate queries, he frees himself from the lets 
and hindrances of jargon and conventional thought. Then 
an idea can be like nothing but an idea ; abstract general 
ideas are ghosts, and infinitesimals are nothings. The mist 
of matter thus melts away, and the Promised Land of being 
flames in the morning light of young percipience. 

It is clear then that Berkeley not only praised these 'rules 
angelic' of Malebranche, but practised them. Perhaps they 
did not leave a deep specific impress upon his work in other 
respects, but it can hardly be doubted that the algebra that 
is in them and over them stayed with him, and showed in 
his doctrine of sense symbolism. 2 He soon gave up his early 
hope of finding in algebra the key of universal knowledge, 
but the relation of the algebraic sign to the quantity signified 
became to him of decisive importance. Algebra seems to 
have cured him of the geometrical method, to which he was 
at one time devoted. Demonstration was the fashion of the 
hour. Defoe in his Consolidator ( 1 705) has some chaff about 
the moon-blind understanding, 'furnisht with Hocoscopes', 
Microscopes, Tellescopes, Caeliscopes, Money-scopes, and 
the D . . .L and all of glasses' making its way by the aid of 
'Physics, Politics, Ethics, Astronomy, Mathematics, and 
such sort of bewildering things, with vast difficulty to a little 

1 See App. p. 198 and, if possible, consult the manuscript. 
Editions cannot convey the full significance of these entries. 

2 See T. V. 140. Int. Prin. 19. cf. Works, vol. iii, p. 373, 
Ale. vii. 12-3. 



22 THE RULES OF METHOD 

Minute-spot, call'd Demonstration.' 1 Berkeley began the 
Commonplace Book as a demonstrator. He will indeed 
demonstrate all his doctrines (592). Certain entries are 
marked as axioms (369, 371 ; cf. 517, 532, 792, 793). At the 
end of the first note-book he wrote a double demonstration 
of the New Principle more geometrico (903-24). This speci- 
men is an interesting relic. It consists of two distinct lines 
of proof. The subordinate principles are listed, abridged, 
and numbered. Inferences are drawn in form. The first 
proof ignores comparison, the second is built on the fact 
and act of comparison. The two proofs meet in the same 
conclusion, * Nothing like an idea can be in an unperceiving 
thing* (918, 922). 

Towards the end of the Commonplace Book (870), he 
writes : 'I must not pretend to promise much of demonstra- 
tion. I must cancell all passages that look like that sort of 
pride, that raising of expectation in my readers.' We must 
connect this change of method with the fact that in the 
meantime one of the pivotal principles of the demonstration, 
'all significant words stand for ideas' (904), had been com- 
pletely upset. He writes, 'Some words there are w ch do 
not stand for ideas' (671). He develops this thought and 
makes an important principle of it in connexion with abstract 
ideas in his introduction to the Principles ( 19, 20). He 
there says that names are 'for the most part used as letters 
are in Algebra, in which ... it is not requisite that in every 
step each letter suggest to your thoughts that particular 
quantity it was appointed to stand for.' Thus the algebraic 
method is the high-road to symbolism; for the algebraic 
sign is negligible in itself, but it can suggest an indef- 
initely wide area of reality. Vision is to him, we might 
say, the algebraic sense, on account of 'the vast extent, 
number, and variety of objects that are at once, with so 
1 Consolidator, pp. 58 sqq. 



THE RULES OF METHOD 23 

much ease and quickness, and pleasure, suggested by it* 
(T.V. 148). 

Malebranche does not give separate treatment to algebra, 
for he regarded it as a higher extension of arithmetic. 
Berkeley appears to have followed his example. In the 
Principles, ( 121-2) we find no separate study of algebra; 
but what he says of arithmetic holds of the sister science. 
He attacks the view that arithmetic is a theoretical science 
leading to a knowledge of eternal verities. Arithmetic is a 
practical science. The primary purpose of numbering is 
'ease of memory and help of computation'. So arithmetic 
begins by abridging. Strokes, points, roman figures, and 
arabic figures are successive developments of the primary 
impulse, the figure '5' being simply an arbitrary abridge- 
ment of 'IIIII'. Hence 'the great use of the Indian figures 
above the Roman shows arithmetic to be about signs not 
ideas' (C.P.B. 815). Then the subject-matter of arith- 
metic, and a fortiori of algebra, is neither abstract ideas nor 
ideas different from the numbers themselves, but signs deri- 
ving from particular things and meaning particulars, but by 
the nature of signification achieving concrete generality. 

Passing from Malebranche 's formal rules of method, we 
may note in passing that Berkeley in his Introduction to the 
Principles, 1-5, seems to have Malebranche in view. He 
contrasts his own conception of the task of philosophy with 
that of 'the wisest men who have thought our ignorance 
incurable', and his phrase 'those lets and difficulties, which 
stay and embarrass the mind in its search after truth' all but 
names the author of the Recherche de la Veritd. These 'lets 
and difficulties' form the subject-matter of the first five books 
of the Search. The words 'darkness and intricacy in the 
objects' echo Malebranche 's phrase the 'dark, intricate and 
confused' effects of Nature (bk. vi. 2. 4). The 'natural defect 
in the understanding' is a favourite topic with Malebranche, 



24 THE RULES OF METHOD 

who often describes in that way the effect of the Fall and 
of the Union with the body. 

The contrast between Principles and Faculties, which 
dominates this section, derives perhaps from the same 
source. The self-partiality which puts the blame on our 
faculties is scornfully referred to in the Commonplace Book 
as 'The Excuse 1 (300, 363, 364; cf. 760). Berkeley met this 
'excuse* in Locke and in the 'mathematicians' ; he met it, 
much emphasized, in Malebranche 's trenchant indictment 
of the whole natural man, his sense, imagination, unaided 
intellect, passions, and will. Malebranche qua monk 
was acutely conscious of the defects in our faculties ; qua 
scientist he realized the importance of correct principles. 
His study of method culminates in the long and close review 
of principles which occupies the greater part of book vi, 
part 2. Here he contrasts the true principles of the Cartesian 
philosophy with the false principles of the Aristotelian. He 
shows that in certain points Descartes himself has been false 
to his own principles, and that in the case of cohesion 
Descartes's account must be corrected by the True Prin- 
ciples of physics, viz. 'the simple Principles of Extension, 
Figure, Motion'. Berkeley's repeatedprotestagainst'learned 
dust' and his appeal to common sense are well known. In a 
similar vein Malebranche attacks 'the Incomprehensible 
Principles of the falsely learned'. 

Instruction in method formed the least part of Berkeley's 
debt to Malebranche ; for the method did not make the doc- 
trine. But the facts given above are important as showing 
how seriously and systematically Berkeley took the Search. 
He was not content to know the more spectacular parts of 
Malebranche's system. His contemporaries were prepared 
to discuss and criticize the doctrine of seeing all things in 
God. But Berkeley went beneath the surface ; he dug down 
and saw the foundations. 



II 

THEORIES OF VISION 

BERKELEY'S first philosophical publication, An Essay 
towards a New Theory of Vision, appeared in 1709, prob- 
ably not later than May. 1 His second philosophical work, 
A Treatise concerning the Principles of Human Knowledge, 
was 'in the press' by March i , 1710, and appeared in May. 2 
A good deal of the preparatory work of both books was done 
concurrently. The Introduction to the Principles was in an 
advanced stage by November 1708, i.e. nearly six months 
before the issue of the Theory of Vision. It seems strange 
that a young author coming before the world with a daring 
metaphysic should have weighted his studies with a psycho- 
logical investigation of great moment, requiring close and 
technical knowledge. Accordingly the relation between the 
two books calls for fuller treatment than it has yet received. 
Fraser twice speaks of the Theory of Vision as a 'tentative 
juvenile essay'. He recognizes that Berkeley was an im- 
materialist when he wrote it. He complains of its 'studied 
reticence', and says that it 'was meant to prepare the way for 
the exposition and defence of the new theory of the material 
world'. Dr. Dawes Hicks prefers another explanation. He 
thinks that Berkeley may have written a good part of it 
before 1707, 'whilst he was still entertaining the belief that 
the objects of touch exist independently of being perceived, 
and that he considered it beside his purpose in finishing the 
volume to disturb that view, seeing that he was about to do 

1 In the dedication Berkeley speaks of Percival who came to 
Dublin in Oct. or Nov. 1708 as his friend of 'these few months*; 
and in appendix to 2nd ed. he says that the operation of June 29, 
1709, occurred 'soon after the first edition*. 

2 Rand, op. cit., pp. 73, 86. 
4046 E 



26 THEORIES OF VISION 

so in another treatise which was then on the eve of publica- 
tion'. 1 There is truth in both accounts; but neither seems 
adequate. The Theory of Vision is more than the thin end 
of the wedge. It is not merely ancillary; yet the Principles 
has not superseded it. All the evidence indicates that the 
Principles was conceived first, and the Theory of Vision 
second. Berkeley was a master of tactics, and he would 
not have disdained to make converts by an economy of 
truth, but we cannot think that a work so sustained and 
comparatively independent was mainly meant to 'insin- 
uate' his metaphysic. 

My view in brief is this: Berkeley's thoughts turned to- 
wards a systematic study of vision because, on the received 
theory, the facts of vision were an insuperable objection to 
immaterialism. His study of Malebranche provided a clue 
to a new theory which removed the gravest objections. He 
published it separately because there was then a demand for 
a theory of vision. 

Let us take the last point first. Why should a tract on 
vision by an unknown and very junior Fellow of Trinity 
College reach a second edition in one year, be answered by 
the famous Archbishop King, and be reviewed almost at 
once on the Continent ? The Principles of Human Knowledge 
made no such stir. Clearly there was a demand for works on 
vision. The demand had existed for some time. In 1682 
Dr. William Briggs presented to the Royal Society a small 
physiological work entitled A New Theory of Vision. Bar- 
rows *s Optical Lectures were well known to the learned, and 
Barrow (T.V. 29) called for a new theory. Molyneux's 
Dioptricks was published in 1692.* The eighteenth century 
opened with an outburst of visionism. Newton's Optics was 
published in English in 1704. Defoe 3 says jestingly, 'A 

1 Dawes Hicks, Berkeley, pp. 39, 40. 

2 See below p. 38, note. 3 1705, Consolidator, p. 57. 



THEORIES OF VISION 27 

generation have risen up, who to solve the difficulties of 
supernatural systems, imagine a mighty vast Something, 
who has no form but what represents him to them as one 
Great Eye. This infinite Optick they imagine to be Natura 
Naturans . . . the soul of man therefore, in the opinion of 

these naturallists is one vast Optick Power From hence 

they resolve all Beings to Eyes.' 

This general interest was largely practical. Spectacles, 
telescopes, and microscopes, had focused attention on 
vision, but the theoretical implications were studied too. 
Arthur Collier, who was working in the same field as 
Berkeley about January lyoS, 1 after outlining the Aristote- 
lian theory of vision says (p. 39) : 'From the old I proceed to 
the hypothesis of vision which is a part of the new philo- 
sophy. Everyone, I suppose, has heard of the doctrine of 
seeing the Divine Ideas, or (as Mr Malebranche expresses 
it) seeing all things in God/ It is noteworthy that Collier's 
Part I is related to his Part II much as Berkeley's Theory 
of Vision is related to his Principles. In Part I he proves that 
the visible world is not external, and in Part II he proves 
that there is no invisible world external. 

Now just as Collier knew quite well when he began to 
write Part I what the conclusion of Part II would be, so 
Berkeley, when he began (probably in the spring of 1708) 
to draft the Theory of Vision, was familiar with the thesis 
of the Principles. But in publishing the Theory of Vision 
separately without an explicit denial of matter, he was just 
taking one thing at a time. He had written a good book 
which his public would read. It was self-contained, and 
yet was a natural stepping-stone to his metaphysic. 

There is no antagonism between the two books. They 
are complementary. The Theory of Vision is not super- 
seded by the Principles. In the Principles ( 44) Berkeley 
1 See R. Benson's Collier, p. 18. 



28 THEORIES OF VISION 

refers to the earlier work for 'fuller information'. He re- 
published the Theory of Vision in 1732, and issued a vindi- 
cation and explanation of it in the following year. 1 In 
America he wrote of 'the design and connexion* of his 
philosophical works, and desired his disciples there to 
read them in the order in which he published them, taking 
care to send them copies of the 'Principles, the Theory, and 
the Dialogues'. 2 The psychology of the theory is indepen- 
dent of the metaphysic to which it leads. In both books he 
asserts the internality of ideas of sight or the ideality of 
objects seen. In the earlier book he allows the reader to 
assume the externality of ideas of touch, or the reality of 
objects touched, not as an economy of truth, but because it 
was foreign to his immediate purpose to carry the argument 
further. 

On the eve of the publication (March i, 1710) of the 
Principles Berkeley speaks of the two books together. He 
tells Percival that the Principles will vindicate the usefulness 
of the Theory of Vision and will make its argument 'appear 
subservient to the ends of morality and religion*. But 
Percival was not a philosopher, and could not appreciate 
the real connexion. We must look rather to Berkeley's 
published statements. We have first, in the dedication to 
Percival, the vague statement, 'my thoughts concerning 
Vision have led me into some notions so far out of the 
common road . . .'. The words may not refer to immaterial- 
ism; if they do, they mean that his study of the Molyneux 
problem, &c., helped to establish his new creed. They 
could hardly mean that writing his book on vision made him 
an immaterialist. The time-factor, if no other reason, bars 

1 The Theory of Vision Vindicated not only gives a masterly 
summary of the earlier work, but adds several new arguments, 

e.g. 46, 5i,5S-7. 

2 Letter to Johnson, March 24, 1730, Works, vol. ii, p. 20. 



THEORIES OF VISION 29 

that explanation. The Preface to the Principles mentions a 
'long and scrupulous inquiry'. Again, more precisely, 'The 
opinion of matter I have entertained some years', he writes 
to Percival on September 6, 1710, and he speaks of the time 
'long since . . . when the conceit was warm in my imagina- 
tion'. Further the Commonplace Book makes it certain, I 
think, that he was an immaterialist before he began the 
Theory of Vision. That essay is structurally dependent 
upon the Commonplace Book, and could scarcely have been 
begun, was certainly not far advanced, until at least the first 
ndte-book was filled with entries. The earliest possible 
date 1 for the opening entries is, I hold, December 1706, and 
certainly Berkeley held the immaterialist hypothesis then. 
In the first two pages of the Commonplace Book we see, to 
adapt Reid's simile, the young Samson with his arms around 
the two pillars of the house, time and space. 'Time a sensa- 
tion: therefore onely in ye mind' (13). 'Extension a 
sensation, therefore not without the mind' (18). The entries 
read like the death-knell of a world. Yet the slight change 
of rhythm 'onely in' to 'not without' is perhaps signifi- 
cant. Space that is seen is more external than time that is 
heard. Visible space is the stronger pillar of the two. Sam- 
son stops to think. 'In the immaterial hypothesis, the wall 
is white, fire hot, &c' (19 ; cf . Search, vi. 2. 2.). There's the 
rub. I see the wall and see it white. I feel the fire, and feel 
it hot. This surely was the 'difficulty' of which he speaks 
in the Principles (43). In this difficulty he turns to Male- 
branche's rules of method. He outlines the 'immaterial 
hypothesis' in its main bearings (C.P.B. 20-6), and then 
addresses himself at once to the study of vision. In the 
entries 27, 28 he sets himself in query form the main prob- 
lem of the Theory of Vision. From that on up to the last 
quarter of the first note-book topics of vision predominate, 
1 See Appendix, p. 200. 



3 o THEORIES OF VISION 

though the wider issues are never dropped. About the time 
he made the entries 283-7, he seems to have satisfied himself 
that Seeing' is no objection to immaterialism. The 'im- 
material hypothesis' thereupon or shortly after passes into 
the Principle. The remaining entries in that and the other 
note-book are about the wider issues, though an odd entry 
with a Theory of Vision marginal sign is found in the later 
pages (88 1 is the latest). 

Thus the Commonplace Book confirms the explicit state- 
ment of the Principles ( 43) about the origin of the Theory of 
Vision, 'that we should in truth see external space, and bodies 
actually existing in it, some nearer, others farther off, seems 
to carry with it some opposition to what hath been said of 
their existing nowhere without the mind. The considera- 
tion of this difficulty it was that gave birth to my Essay 
towards a New Theory of Vision , which was published not 
long since.' That passage makes it clear that when he began 
to shape the argument for immaterialism he was met at the 
outset by an obstacle that seemed to bar further progress 
we see the outside world ; we see the world outside. There- 
fore there is a world and it is outside. He turned aside to 
deal with that difficulty. His New Theory of Vision removed 
the barrier. So Berkeley's essay on vision, though presented 
as an independent work, and certainly possessing an 
independent value, is essentially and by origin an integral 
p'art of his new philosophy. 1 The influence of Malebranche 
upon this part of the Berkeleian philosophy is therefore an 
index to the degree of his influence upon the whole system. 

Sketch of Berkeley's Theory of Vision 

The developed and characteristic part of Berkeley's 
theory is his account of the object seen. Before discussing 
that part, we shall piece together his account of the subject 
1 See quotation from Collier, p. 27 above. 



THEORIES OF VISION 31 

seeing. For Berkeley, seeing is a compound act. It is more 
than the act of the eye, qua eye. It is shadowed by the 
judgement and by incipient movements of other organs, 
especially of touch. In his words, the 'turn of the eyes is 
attended with a sensation' ( 16), and he finds it acknow- 
ledged, as regards remote distance, that the estimate is an 
act of judgement rather than of sense ( 3). His insistence 
upon the suddenness and swiftness of the act is a recogni- 
tion of this complexity, and an attempt to introduce an 
apperceptive unity. As regards terms, he often uses in- 
differently seeing, sensing, perceiving and judging; but he 
can be precise when he wishes (e.g. 'The eye, or, to speak 
truly, the mind, perceiving . . .' 36). He knows well the dis- 
tinction between 'the first act of vision' ( 106) and its con- 
comitants, between sight and perceiving by sight ( 99). 
Pure vision is simple apprehension of colour ( 43, 156, 
158). It is the postulate of the Molyneux problem ( 106), 
and is the sole type of sensible act open to 'an intelligence 
or unbodied spirit, which is supposed to see perfectly . . . 
but to have no sense of touch' ( 153). Berkeley uses with 
effect certain terms, which imply the complexity of the 
process, but subordinate the notion of act. Of these the 
most characteristic is 'suggestion' ( 16 and passim), which 
covers both the stimulus and the mind's semi-conscious 
response (cf. 'unperceived transit' 145, and the vague 
terms 'attend' and 'experience', which are frequent). 

He carefully characterizes one part of the subjective pro- 
cess. The elements that blend, be they the joint actions of 
different senses, be they memory and sense, or mind and 
sense, do so by virtue of custom and not of necessity. 'On 
the demonstration of this point the whole theory depends* 
(T.V. y App.). There is no necessary connexion. There is no 
'natural geometry' in the act of seeing. Seeing distance or 
magnitude is not like a demonstration, in which the figure 



32 THEORIES OF VISION 

being given the properties flow by necessity. It is a mistake 
to confuse vision with optics. Optical angles, whether made 
by the axes of vision at the object or by diverging rays on 
the pupil, are not really there in rerum natura. So judge- 
ments of distance cannot, in fact, flow from them, as is gen- 
erally supposed, by iron necessity. Nor can what is outside 
be like what is within the mind. 'I saw shame ... So I see 
. . . distance.' This entry (240) is the only one in the Com- 
monplace Book marked for insertion in all the three original 
parts of the essay, and to this comparison he returns repeat- 
edly (T.V. 9, 23, 65 ; cf. 143-5). Thus Berkeley sub- 
stitutes arbitrary connexion of unlike elements for * natural 
geometry'. Colours of the countenance are not like passions ; 
yet in seeing red there, by custom I learn to see shame 
there. Such is the proper comparison for 'how we see*. 

We come now to Berkeley's treatment of the object seen. 
Here, too, he divides and conquers. He divides what I see 
from what I touch. He distinguishes what I see immed- 
iately from what I see mediately. He connects the mediate 
object with the object touched or tangible, and thus we have 
a dual object correlated with a dual activity of the subject. 
My eye sees the immediate object; my mind sees, i.e. I 
infer or perceive, the mediate object. What is presented 
to my eye is visible, purely and altogether visible ; yet it 
suggests what is not visible, and what may, in most sen- 
sations, be regarded as the tangible. 

That visible objects differ in kind from tangible is the 
key position of the new theory of vision. At this stage he is 
not concerned to probe the metaphysical implications of the 
distinction, which he says 'cannot be too often inculcated 
in treating of vision' (T.V. 91). The ideist philosophy was 
conveniently ambiguous, and by calling both objects 'ideas' 
he leaves the ground open for the ultimate argument of the 
Principles. Most of his contemporaries would admit the 



THEORIES OF VISION 33 

propriety of calling the datum of sight an 'idea', and the 
Lockians would admit that if it is strained to call 'what I 
touch' an idea, at least there is an idea of 'what I touch'. 
Locke had taught that there are ideas common to both 
senses, that, in fact, we see and touch the same thing. 
Berkeley joins issue with him, asserting what I see is one 
object and what I touch is another object. While both 
objects may be called and are 'ideas', visible ideas are stated 
by Berkeley to be in the mind, while the reader is allowed 
to assume that tangible ideas are external. The design of 
the essay is expressed in i, and the first part of the design 
has caught the commentator's eye rather too exclusively. But 
from 119 and 127 we see that in Berkeley's opinion the 
core of what is new in the new theory is this heterogeneity. 1 
The supposed fact that we do not see and touch the same 
object had been stamped on Berkeley's imagination by the 
Molyneux problem. He reproduces the problem in Locke's 
words together with Locke's solution in T.V. 132. The 
Molyneux problem was of capital importance to Berkeley 
when he was designing the essay. About fifteen entries in 
the Commonplace Book raise questions about the blind man 
made to see, and the earliest entries (27, 28, 32), dealing 
with the heterogeneity of sight and touch, raise the question 
in connexion with the Molyneux problem. Molyneux and 
Locke both said 'no', the blind man in the problem would 
not know which was the cube and which the square. Berke- 
ley's private opinion was that the blind man would not 
recognize either of them to be even bodies (C.P.B. 32), and 
that the question would be to him 'downright bantering and 
unintelligible' (T.V. 135). But he makes capital for his 
main thesis by saying that the solution given by Locke and 

1 So also Theory of Vision Vindicated, 41, 42. 'This stating 
of the matter placeth it on a new foot, and in a different light 
from all preceding theories/ 

4046 p 



34 THEORIES OF VISION 

Molyneux is wrong unless the space of sight and the space 

of touch are specifically distinct (T.V. I33). 1 

The main and comprehensive thesis of the essay, 'so 
remote from, and contrary to the received notions and 
settled opinion of mankind', is formally laid down in the 
following proposition : 'the extension, figures, and motions 
perceived by sight are specifically distinct from the ideas of 
touch, called by the same names ; nor is there any such thing 
as one idea, or kind of idea, common to both senses' (T.V. 
127). This proposition can be gathered from 'several 
places' of the essay, but in the sections immediately follow- 
ing Berkeley gives a 'demonstration' of its truth by three 
separate arguments, adding the solution of the Molyneux 
problem in 'farther confirmation of our tenet'. 

The New Theory, then, starts from the 'heterogeneity' 
and works round to it again through a study of distance, 
magnitude, and situation. That this is the correct analysis 
of the essay is established by the passage above quoted, by 
the Commonplace Book, and by The Theory of Vision Vin- 
dicated, where he calls the heterogeneity 'this main part and 
pillar' of the theory. 2 

We shall now briefly review the three steps by which he 
leads his readers to his own starting-point. 3 The section on 
'Distance' ( 2-51) may be summarized thus. It is physic- 
ally impossible for distance itself to be seen ( 2), as an 
object is seen. Distance is not perceived by means of lines 
and angles ( 13-15). Another type of medium must be 

1 Since the above was written I came across the following 
statement: 'His philosophy can alone be truly known, when seen 
germinating from the question of Molyneux' (Teape, Berkeleian 
Philosophy (1870), p. 3). For the original problem see Locke, 
Fam. Letters, p. 37. 

2 T.V.V. 41 ; v. also 15, 32, and passim. 

3 N.B. In The Theory of Vision Vindicated the order of these 
steps is reversed. 



THEORIES OF VISION 35 

sought. Distance is suggested by the 'turn of the eye' and 
by confused appearance and by other contributing circum- 
stances ( 28). This explains the Harrovian case', i.e. the 
problem which Dr. Barrow felt so acutely that he called in 
question the principles of optics, and demanded a new 
theory of vision ( 29-40). From this account of distance 
it follows, in Berkeley's view, that not colour alone, the 
proper and immediate object of sight, but also extension, 
figure and motion, are at no distance from the mind, but as 
near as pain ( 41-4). He then introduces his readers to 
the principle that was his own starting-point, and explains 
the illusion of visual distance by our refusal to admit the 
heterogeneity of the objects seen and touched, called by 
the same name (45-51). 

Berkeley's second step is to do for magnitude what the 
first step did for distance. The section on 'Magnitude' 
( 52-87) follows the course of the first section. We do 
not perceive magnitude by means of lines and angles. Real 
invariable magnitude is perceived principally by three 
means: (i) the everchanging visible magnitude; (2) con- 
fusion or distinctness of the visible appearance; (3) its 
vigorousness or faintness. There are as well some second- 
ary factors, e.g. the disposition of the eye, and the number, 
quality, and nature of intermediate objects. There is no 
necessary connexion between these features and tangible 
size. Great size means simply that the object contains' a 
large number of points or minima, and minima may be 
either tangible or visible. There is no reason other than 
custom why confusion and vigorousness should suggest 
smaller size and fewer minima, while distinctness and 
faintness do the reverse. We see magnitude in no other 
way than 'we see shame or anger in the looks of a man' 
( 65). The apparent size of the horizontal moon is then 
considered and explained on these principles. The section 



36 THEORIES OF VISION 

ends with a brief discussion of the minimum visibile and 
microscopic sight. The chief new point in this section is 
that the connexion between the visible and tangible, while 
arbitrary in the sense that it might have been otherwise, is not 
capricious ; for it conduces to the preservation of the body. 

The third section ( 88-120) has the broad title Situation, 
but it is entirely devoted to the narrow problem of the 
inverted retinal image. Berkeley may have taken the title 
from Descartes 's section on Situs (Dioptrics, c. vi.,9). As 
we shall see below, Malebranche provided the model for 
the sections on 'Distance* and 'Magnitude', but he has 
nothing corresponding to this third section. The section 
jyas an afterthought with Berkeley. From the marginal 
signs in the Commonplace Book we see that the three sections 
originally planned were on distance, magnitude, and hetero- 
geneity ; for the corresponding entries are numbered 1,2,3, 
respectively. There are, however, 7 or 8 entries on the in- 
verted retinal image, and these and these only are marked 3 3 . 1 

This 'one mighty difficulty' (T.V. 88) became to him 
'the principal point in the whole optic theory' ( T. V.V. 52). 
For its solution is clear proof that the tangible object is not 
the proper object of sight. He first considers Molyneux's 
explanation, using Molyneux's own diagram. This ex- 
planation, 'allowed by all men as satisfactory', is in effect 
that of Descartes, and involves the absurd geometria innata 
of the blind man with crossed sticks, and the other assump- 
tions of external visible distance. His own explanation turns 
on his sharp distinction between visible and tangible ideas 
(T.V. 97 sqq.). By 'erect man out there' we mean a 
tangible man with tangible feet on the tangible earth. The 
rays of light from the 'erect man out there' form an inverted 
image on the tangible retina. Taught by experience, when 
we want to see distinctly the objects imaged on the higher or 
1 See Appendix, p. 185 and frontispiece. 



THEORIES OF VISION 37 

lower parts of the retina, we turn the eyes down or up res- 
pectively. Hence, though we do not see 'up' and 'down', 
any more than we see outness, yet we say we see a man's 
head up and his heels down, because we see the head and 
heels distinctly by turning our eyes up and down respectively. 

The remainder of the essay turns on Berkeley's doctrine 
of abstract ideas ( 122-60). This section too was an 
addition to the original plan. 1 He was half-way through the 
Commonplace Book before he came on the question of 
abstraction. There are several parallels between this sec- 
tion and the Introduction to the Principles ; so we may 
presume that Berkeley added it in the late autumn of 1708, 
when, we know for certain, he was working at that Intro- 
duction. It is a highly important section, for it directly 
incorporates the new theory of vision in Berkeley's wider 
philosophy. An abstract idea of space, a mere product of 
fancy, is at the back of our belief that we see and touch the 
same object. The supposed idea common to two senses is 
nothing more or less than this abstraction, which gains in 
dignity and acquires apparent substance by being regarded 
as the subject-matter of geometry. Thus Berkeley brings 
the root-cause of men's mistakes about vision under the 
most inclusive of all the erroneous principles that, in his 
opinion, trouble knowledge. 

The heterogeneity of visible and tangible ideas is the 
'main part and pillar* of the New Theory. What I see is not 
what I touch, nor like it. This result is negative. It explains 
our errors, but is not positive knowledge. Berkeley is not 
content to leave the matter there. He discusses the question 
'How visible extension and figures come to be called by the 
same name with tangible extension and figures if they are 
not of the same kind with them.' (T.V. 139). His answer 2 

1 Abstraction is not treated in the Theory of Vision Vindicated. 

2 Based, I think, on Search , bk. ii, pt. 2, c. 3, see below, p. 46. 



38 THEORIES OF VISION 

leads to a positive expression of his theory. Visible words 
are symbols. So too are visible ideas. The series square is 
not a bit like a tangible square, yet to Englishmen that 
group of letters suggests a tangible square. Now a visible 
square, also unlike a tangible square, has parts that in some 
measure correspond to the parts of the other. So the visible 
square is a sign the world over, not conventional like the 
letters, not Variable and of human institution'. Thus he 
reaches his highest and widest conclusion, that 'the proper 
objects of vision constitute the Universal Language of 
Nature' (T.V. 147). Berkeley's reference to this passage 
in the Principles ( 44; also 66 and 108) shows that he 
regarded this truth as the net outcome of the essay, and in 
the Principles and in the Alciphron (dial, iv) he expands it 
into a main proof of the existence of God, and, we might 
almost say, into a solution of the enigma of the universe. 

Praise has been given to Berkeley's theory of vision, even 
by those who reject the metaphysic on which it rests. It is 
a fine piece of constructive reasoning, involving a wide 
sweep of the philosophic mind, as well as exact knowledge 
of optical science. A young man in his twenty- fourth year 
wrote it. So an inquiry into the influences that moulded 
the work is of unusual interest. We know that Berkeley 
used Barrow's Optical Lectures, Descartes's Dioptrics, and 
Newton's Optics. But his two outstanding authorities in 
this field were Molyneux and Malebranche. 

William Molyneux' s Treatise of 'Dioptricks* (Dioptrica Nova) 

This book was so important for Berkeley that some ac- 
count of it is required. It was published apparently in 
1692, with a dedication to the Royal Society. 1 It is written 

1 The above date seems correct, though others have given it 
as 1690 or 1691. The dedication is dated 1690. The T.C.D. 
Library copy, ex dono Authoris, is inscribed: 'Almae Matri 



THEORIES OF VISION 39 

from the standpoint of the New Learning. It claims to be 
the first book on the subject ever published in English. The 
Dedication speaks highly of Locke and of the Recherche de 
la Verite. Berkeley relied on Molyneux 's book for the tech- 
nique of his subject, and actually reproduces some of the 
diagrams contained in it. The opening of the Theory of 
Vision is verbally indebted to pp. H3-I4. 1 Molyneux treats 
of the physics of light and of plain vision with the naked 
eye. He deals with the structure and function of the eye, 
explaining the crystalline, retina, &c. He passes on to con- 
sider distinct, confused, clear and faint, near and distant 
vision. He discusses optic angles, apparent and real dis- 
tance, apparent and real magnitude. He concludes with a 
section on glasses, concave and convex, the speculum, the 
microscope, and telescope. The book is only once men- 
tioned in the Commonplace Book (203) ; but there can be no 
doubt that many other entries dealing with the technical 
side of optics are based upon it. The Theory of Vision 
( 29, 40, 89) treats it as one of his primary authorities. 
Now Molyneux, like Barrow and Newton, is eloquently 
silent at the point of contact between optics and philosophy. 
For instance, of the retinal image Molyneux says : * Which 
representation is there perceived by the sensitive soul (what- 
ever it be), the manner of whose actions and passions He 
only knows who created and preserves it' (p. 104). Or 
again, of the inverted retinal image Molyneux writes : 'But 
this quaery seems to encroach too nigh the enquiry into the 
manner of the Visive faculties' perception* (p. 105 ; cf. p. 
289). He recognizes that visual perception lies outside his 
province, adding, ' 'Tis not properly the eye that sees'. Thus 
Molyneux's treatise points on to a metaphysic of vision. 

Academiae Dubliniensi humillime offert Alumnus gratissimus 
Gulielmus Molyneux.' 
1 Cf. Works, vol. i, p. 108. 



40 THEORIES OF VISION 

Malebranche on Vision 

Malebranche 's influence on Berkeley extended beyond 
the technique to the philosophy of the subject, and was 
therefore profound. It is curious that while obscure writers 
are named in the Theory of Vision, Malebranche receives 
neither acknowledgement nor mention. Possibly Berkeley 
was so conscious of the points of difference that he did not 
realize his debt. More probably his silence was prudence. 
In some quarters 'Malebranche' spelled enthusiasm, and 
enthusiasm was literally a sin. It was safe for Collier, settled 
in his Wiltshire rectory, to connect his system with that of 
Malebranche. It was another matter for the ambitious 
young Irish Protestant to do so. Only those who know the 
conditions can appreciate the point. Berkeley was absolutely 
dependent for promotion on the favour of Dublin Castle. A 
Roman Catholic monk who wrote bitterly of 'heretics', 1 
and who called the English 'those wretched people, those 
children of this world', attacking the English Crown, the 
Church, and the State, 2 would not be the most profitable 
patron for Berkeley's first important venture in authorship. 
It seems certain that Berkeley deliberately avoided men- 
tioning Malebranche. For instance, in connexion with the 
moon problem a recognition of the care that Malebranche 
devoted to that problem would have been easy and natural. 
His mention in 75 of Gassendi, Descartes, and Hobbes is 
due, no doubt, to the fact that Molyneux's paper before the 
Royal Society, written 'to rouse philosophers up to enquire 
anew after this surprising Phenomenon', exposes the errors 
of those three thinkers. But when he goes on to review the 
very next paper in the Transactions, that by Wallis, one 
would have expected a mention of Malebranche among 
the 'others' who had anticipated Wallis's explanation. It 

1 Search, bk. ii, 2, c. 4. 2 Search, bk. ii, 3, c. 2. 



THEORIES OF VISION 41 

is worth noting that Gassendi's 'false principle' referred 
to in the Appendix to the Theory of Vision, namely the 
alteration in the pupil, is noted in the Commonplace Book 
(264) in connexion with Malebranche. 

There is no doubt whatever that Berkeley studied the 
Search while he was preparing to write the Theory of Vision. 
In the Commonplace Book some fourteen entries name Male- 
branche ; as many more refer to his views ; two entries (264, 
266) actually include the reference to bk. i, ch. 6, where 
Malebranche begins his study of vision. 

Malebranche gives his general theory of sense perception 
in bk. i, chs. 10-14. ^ s analysis of the act of perceptual 
judgement is close to that of the Berkeleian theory. What 
looks single, proves on analysis to be dual. The one world 
of perception turns into two worlds, the world within, and 
the world without (hors de nous). The act of natural judge- 
ment is two acts of sense, combined in 'compound sensa- 
tion'. The object sensed is dual, the mediate object and the 
immediate. Finally existence itself is in a manner bisected, 
and the theory is outlined in the following words : 'There 
are two sorts of Beings ; Beings which our soul immediately 
sees and others which she knows only by the mediation of 
the former. When, for instance, I perceive the sun arising 
I first perceive that which I immediately see, and because 
my perception of the former is only occasioned by some- 
thing without me ... I judge the former sun which is in my 
soul, to be without me and to exist.' 1 Vision, whether the 
term mean the act of seeing or the object seen, is, for Ber- 
keley, compound sensation. 2 I suggest that Malebranche, 

1 Search, bk. i, c. 14. For the illustration cf. C.P.B. 898. 

2 Berkeley does not use the phrase 'compound sensation* ; but 
his 'sudden judgement of sense* ( 77) is close to 'judgement of 
the senses' which Malebranche uses as a synonym for 'compound 
sensation', bk. i, c. 14. 

4046 G 



42 THEORIES OF VISION 

to that extent, supplied the outline of the New Theory, 
and that Berkeley filled in the outline by his original con- 
tributions, namely, the distinction between the data of 
sight and of touch in the complex percept, and the expres- 
sion of the whole in terms of idea. 

As regards motion Malebranche is close to the full 
Berkeleian distinction. He distinguishes (bk. i. 8) two sorts 
of motion, the one being visible, the other not. He argues 
that we do not know the true quantity of motion, adding, 
'This argument is only a corollary of that which I have said 
of extension.' Berkeley reproduces the thought and per- 
haps the words of this passage, dismissing motion similarly 
in an offhand way. 'Now that visible motion is not of the 
same sort with tangible motion seems to need no further 
proof, it being an evident corollary from what we have 
shown concerning . . . extension' (T.F. 137). 

The Cartesian division of the primary qualities, exten- 
sion, figure, and motion, without reference to solidity and 
the other Lockian variants, appears at most of the turning- 
points of Berkeley's argument. It is also the framework of 
Malebranche 's section on vision in Book i. Chapter 6 is 
devoted to extension, i.e. size or magnitude. Chapter 7 con- 
siders figure. Chapters 8 and 9 treat nominally of motion, 
<but in effect of distance. Similarly Berkeley gives a serial 
'treatment of the primary qualities, to some extent following 
a line of his own, but perhaps influenced by the Search. Like 
Malebranche he gives pride of place to distance, allowing 
it to oust motion. To figure Malebranche concedes only a 
brief notice, declaring that figure is 'not a thing of an abso- 
lute kind, but its nature consists in the relation which is 
between the parts which terminate some space'. Similarly 
Berkeley twice (T.V. 105, 124) declares that 'figure is the 
termination of magnitude', and he treats it along with 
magnitude. Berkeley's section on situation was, I have 



THEORIES OF VISION 43 

shown above, an afterthought, due no doubt to an indepen- 
dent study of the retinal problem (p. 36). 

The comparison of the sense-datum to language, the 
arbitrary connexion between sign and thing signified, and 
the contrast between man-made connexions and the uni- 
versal connexions established by the Will of God, these 
Berkeleian principles may all be found in the Search. 
Occasionalism taught that there is no necessary connexion 
between changes in the body and the accompanying changes 
in the soul. Berkeley did not accept occasionalism, but this 
tenet, in the form of faith in the real operation of the Will 
of God, became an essential of his theory. Malebranche 
points out that what we call * heaven' the Greeks call 'oura- 
nos', the Hebrews 'shamajim'. The connexion between the 
word and the thing is arbitrary and man-made. But, he 
notes, sensations like colour, taste, &c., have no such 
variable character. No words of mine can explain to a 
man what heat is or colour. If he does not know and I 
want to tell him, I must speak to him in the universal 
language that God has established. I must, that is, impress 
his organs of sense. *I must bring him to the fire, and 
show him a piece of painting/ 1 

It would seem that in the plan of the book and in the main 
lines of the argument, the Theory of Vision is indebted to 
the Search. To trace the dependence further would be 
tedious and unnecessary here; but I have collected in an 
appendix to this chapter some of the more striking parallels. 

'Dependence' seems scarcely the word in the case of so 
independent a thinker. Berkeley was no copyist. His 
authorities were his sources ; they helped to mould but did 
not make his thought. My aim is to show that the way to the 
heart of Berkeleianism lies through Malebranche. If the 
facts bear me out, a fair presentation of them cannot detrac t 
1 Bk. i. 13; see also bk. ii. 2. 3 and elsewhere. 



44 THEORIES OF VISION 

from Berkeley's achievement. His achievement is not in 
question. It stands above detraction. They detract who 
would make his views the dreams of an egoist out of touch 
with his day. His Irish originality and genius stand out all 
the more sharply against the French background. He used 
Malebranche so much and so fruitfully, and yet reached 
conclusions so different. Malebranche's confessed aim in 
analysing the 'most comprehensive' of the senses is at one 
blow to shake man's natural faith in all his powers. His sys- 
tem is built upon a conscious and radical distrust of the 
evidence of our eyesight. It was a stroke of genius on 
Berkeley's part to study afresh under Malebranche's tuition, 
the 'most comprehensive' of the senses, and to find its evi- 
dence, misunderstood and misrepresented indeed, but in its 
true interpretation reliable, reassuring, and uplifting. 

APPENDIX 
Parallels between the < Theory of Vision' and the 'Search* 

N.B. S. = Search 

Commonplace Book: the following entries refer to Male- 
branche on vision: 264, 266, 278, 296, and probably 60, 83, 
126, 127, 441. 

The term 'comprehensive' of sight is common to Male- 
branche, Locke, and Berkeley; cf. Descartes 's latissime patens 
\Dioptrics, c. i). 

The welfare of the body determines our visual apprehension, 
especially in the case of magnitude. S. bk. i, c. 6 and 20, T.V. 
59 > 85. 'The principal thing I would have remembered' 
(Malebranche). 

Glasses as proof of not seeing extension: S. i. 6, C.P.B. 63. 

The animate atomies which indicate eternal creation: .1.6, 
C.P.B. 60; cf. letter to Percival in Rand, op. cit., pp. 83, 84. 

The varying length of a six-foot rule: S. i. 6, T.V. 61; 
cf. C.P.B. 87-9. 



THEORIES OF VISION 45 

'If we had eyes after the manner of microscopes': S. i. 6; 
cf. 'Were our eyes turned into the nature of microscopes', 
T.V. 86; cf. Locke, Essay, ii. 23, 12. 

Right lines not verifiable: S. i. 7, C.P.B. 126. The cube: 
S. i. yandiii. 2. 2, C.P.B. 83. 

The steeple, interjacent objects, fields, houses, &c., seen 
behind a wall: S. i. 7, T.V. 70, 73, 76, 77. 

Blind man with two sticks (optical angle): T.V. 42, S. i. 9, 
and Descartes, Dioptrics, c. vi and diagrams. 

Natural Geometry: S. i. 9, T.V. 19 and App. 

Both give great prominence to the apparent size of the hori- 
zontal moon. Malebranche refers to it under 'Figure', 'Dis- 
tance' and 'Judgement'. Though Berkeley omits to mention 
Malebranche, there are two indications of special study of this 
source: (i) The moon problem leads M. to say (S. i. 7) that 
optics only instructs us 'how to put fallacies on our eyes' and 
he speaks of 'this cheat'. B. in the same connexion asks (T.V. 
74), 'What is it can put this cheat on the understanding ?' 
(2) B. says (T.V. 70, 74) that the visible magnitude of the 
moon 'remains the same or is rather lesser'. The inconsistent 
phrase puzzles Fraser, who asks, 'Why lesser?' The answer 
may be that M. says so. M., denying that atmospheric refrac- 
tion is the explanation, says that the retinal image of the moon 
is smaller as refracted, because astronomers have proved that 
the moon's diameter grows greater as she climbs (S. i. 9). 

Malebranche in S. i. 9 on 'Distance' distinguishes six 
'mediums' of judging distance. Berkeley uses the word 'med- 
iums' ( T.V. 22). B. refers to all the mediums, accepting 
some, rejecting others. The mediums are: (i) The optical 
angle, or the optical angle with the disposition of the eye; cf. 
T. V. 19 and 42. (2) The disposition of the eye accompanying 
the angle made by the rays converging on the retina. (3) The 
retinal image. (4 and 5) The force of the object's action upon 
the eye and the distinctness and clearness, the faintness and 
confusion of the image. (6) Interjacent objects. 

B. illustrates the relativity of sensation by the case of an 
Englishman meeting a foreigner who used the same words in 



46 THEORIES OF VISION 

the contrary sense (TV. 32). So Malebranche (S. i. 13) 
speaks of the foreigner, who, on a winter's day praised 'cold' 
water, thinking 'cold' was the word for hot. 

B., contrasting geometry with algebra, speaks of the 'extra- 
ordinary clearness and evidence of geometry', and assigns as 
the reason, 'the very ideas themselves' being copied out and 
exposed to view upon paper (T.V. 150); so also S. vi. i. 3, 
'By drawing lines upon paper geometricians draw, as I may 
say, answerable ideas upon their mind.' 

'All men have the idea of a square upon sight of a square ; 
because that connexion is natural; but it may be very well 
doubted whether all men have the idea of a square when they 
hear the word 'square' pronounced ; because that connexion is 
altogether arbitrary' (S. ii. 2. 3). So B. (T.V. 152; cf. 140) 
says: 'There is indeed this difference betwixt the signification 
of tangible figures by visible figures and of ideas by words 
that whereas the latter is variable and uncertain, depending 
altogether on the arbitrary appointment of men, the former is 
fixed and immutably the same in all times and places. A visible 
square, for instance, suggests to the mind the same tangible 
figure in Europe that it doth in America.' 

'Outness.' Where did Berkeley find that 'barbarous but ex- 
pressive term ? ' x The Oxford Dictionary gives no instance before 
Berkeley. I have only met it in one other place, namely, in the 
famous 'Illustration' appended to the Search on the difficulty 
of proving the existence of bodies. B. refers to this 'Illustration' 
more than once in the C.P.B., e.g. 697 and 830. The passage 
is: 'Is it not evident that there are outnesses (des dehors) and 
remotenesses ... in the intelligible world which is the im- 
mediate object of our mind?' N.B. There are many other 
indications, amounting to proof that Berkeley read Male- 
branche in Taylor's translation, e.g. C.P.B. 812: 'Malbranch 
in his Illustration'. Taylor invariably gives 'Illustration' for 
the technical term 'ficlaircissement'. 

1 Thomas Brown, Philosophy of the Human Mind, 1820, vol. i, 
p. 491. 



Ill 

THE BACKGROUND OF IMMATERIALISM 

PROFESSOR DAWES HICKS and Professor Aaron have stated 
that immaterialism was 'in the air' when Berkeley wrote, 
That statement is correct, if it means that immaterialism 
is strongly suggested to us by Cartesian ideism, but not, I 
think, if it be taken to mean that there were Berkeleians 
before Berkeley. The reception accorded to Berkeley's 
Principles 1 shows that immaterialism was not in the air of 
London. Collier too had no hearing. Thinkers were not 
waiting for immaterialism, nor ready for it. As we study the 
books Berkeley read, we shall see that his contemporaries 
differed about the nature of matter, but were ready to turn 
and rend any one who denied its existence. Materialism 
was much 'in the air' that Berkeley breathed. 

Religion was the driving-force of his immaterialism 
'Matter once allowed, I defy any man to prove that God is 
not matter' (C.P.B. 634; cf. 308). But Berkeley did not 
isolate his religion. Philosophy, mathematics, and science 
combined with religion to thrust upon his notice the prob- 
lem of matter. Mathematicians needed matter, for they 
demonstrated its infinite divisibility, the postulate of the 
new theory of infinitesimals. Physics assumed matter, for 
'attraction', the 'great mechanical principle now in vogue* 
(Prin. 103), required something to attract. The microscope 
was exploring the microcosm of matter, and had confirmed 
the doctrine of the infinite divisibility. A world lies hid in 
a mite. In a seed there lies a tree itself seeding yet another 
tree (Search, i. 6). The telescope had advanced the frontiers 
of knowledge and of matter. Lastly, under the influence of 
Spinoza's 'infinite attribute', theologians were beginning to 
1 V. Percival to B. August 26, 1710 (Rand, op. cit., p. 80). 



48 THE BACKGROUND OF IMMATERIALISM 

regard extension or real space as an attribute of Deity, and 
Clarke had based on matter a demonstration of the essential 
attributes. 

Matter had only recently become a problem. Norris 1 
writes: 'Suarez proves that first matter is not so a pur a 
potentia but that it has some entative actuality belonging to 
it, that is, that tho' it be in pura potentia receptiva as to any 
formal act, yet it is not in pura potentia objectiva as to reality 
of being, but is a real something (however incomplete) and 
actually extra nihil.' This elusive ghost, Pura Potentia extra 
Nihil, could never be a danger to religion, nor to anything 
except clear thinking. Indeed, the ghost was much in de- 
mand to support certain dogmas of the medieval church. 
Transubstantiation had shielded matter down the centuries, 
as matter was the ground of transubstantiation. 2 

The scholastic tradition died hard. Sir Kenelme Digby, 
writing about the middle of the seventeenth century, devotes 
400 pages of his Two Treatises to the physics of bodies, in 
preparation for his lofty address to his soul. The type of his 
physics can be gathered from his definition of colour, 'the 
term of the diaphanous body', the definition being 'con- 
firmed by Aristotle's authority, reason, and experience* 
(c. 29). Though clearly he believed in matter, he scarcely 
mentions it. It was of no interest to him. He says in effect 
(p. 426), 'It is unnecessary to study matter after bodies'. 

Half a century of the method of doubt changed all that. 
Cartesian metaphysic raised the status of matter and focused 
attention upon its nature. The speculative examined the 
evidence for its existence. Physicists studied its content 
and when, with the discrediting of Descartes' vortices, the 

1 Theory of the Ideal World, 1701-4, pt. i, p. 78. 

2 A. Collier, Clavis Univ., pt ii, c. 9, assembles several defini- 
tions and descriptions of matter, and in his Conclusion connects 
matter and transubstantiation. 



THE BACKGROUND OF IMMATERIALISM 49 
force of cohesion was transferred from the thing to the 
ether, the solidity of matter vanished, leaving behind it the 
clear and distinct idea of extension. In England Hobbes's 
materialism had taken deep root, and Locke's material 
substance was solid extension, divisible and mobile, which 
might even think. Subtle French matter and solid English 
matter, both were fair targets for the young Irishman. 

A brief account of Raphson will illustrate the mathe- 
matician's attitude to matter. Joseph Raphson, F.R.S., in 
1697* published his vigorous essay De Spatio Realiseu Ente 
Infinite, annexed to a mathematical treatise. Berkeley's 
first mention of it is in his essay Of Infinites (1706-7). He 
criticizes Raphson for speaking of the infinitesimal as 'quasi 
extensa'. In the Commonplace Book Berkeley twice names 
Raphson (308, 839; cf. Prin. 117). In both these entries 
Berkeley has a serious charge to make. Raphson, although 
a critic of Spinoza, followed him in so interpreting the 
divine immensity as, in effect, to make God extended, 
demonstrating (ch. v) that space is, inter alia^ 'actus purus, 
incorporeum, immutabile, aeternum, omni-continens, 
omni-penetrans, attributum (viz. immensitas) primae 
causae'. Each of these phrases constitutes an infringement 
of the divine prerogative. Raphson opposes Inexten- 
sionism, which is the doctrine of those who hold that 
there is no extension in addition to matter. He maintains 
that space is real and is distinct from matter, claiming 
antiquity and the authority of Henry More for this view. 
Berkeley long remembered Raphson. Writing to the Rev- 
erend S. Johnson on March 24, 1730, he speaks of mathe- 
maticians who have attributed extension to God, 'one of 
whom in a treatise De Spatio Reali pretends to find out 
fifteen of the incommunicable attributes of God in space'. 2 

We pass on to a tougher type of 'materialism' that of Dr. 

1 Fraser, vol. ii, p. 19 n., gives 1706 2 v. c. v especially. 

4046 



50 THE BACKGROUND OF IMMATERIALISM 
Cheyne. He also was a Fellow of the Royal Society. In 1705 
he published his Philosophical Principles of Natural Religion. 
Like Raphson, Keill, Barrow, and almost all the 'mathema- 
ticians', he held the infinite divisibility of matter, and gives a 
full geometrical demonstration of it (c. iv. 7). His words on 
infinitesimals are quoted and criticized by Berkeley in his Of 
Infinites. Three entries in the Commonplace Book name 
Cheyne (388 duplicates 93 1 , 458). His book must have im- 
pressed Berkeley. Itsstyleisclearandvigorous. Itistheistic. 
It strongly maintains Providence and the Divine Conserva- 
tion . Cheyne expounds the mechanics of bodies , celestial and 
terrestrial, the anatomy of the human body, the philosophy 
of the organism, of sensation and colour. He disagrees with 
Raphson as to extension, saying, 'Time and Space are no real 
things, nor complete substances' (c. iv. 8). His is a robust 
doctrine of matter. He writes that gravitation is 'a principle 
annex'd to matter by the Creator' (c. i, 27), and again: 
'The existence of matter is a plain demonstration of the 
existence of a Deity. I believe nobody doubts that there 
now exists a quantity of solid mass, out of which the celestial 
and terrestrial bodies were form'd ; and tho' perhaps in our 
most solid bodies there be more Pores than Parts, or more 
vacuity than solidity, yet there is still sufficient not to permit 
,us to doubt of the existence of matter' (c. iii, 3). 

The Newtonian principles are too well known to need 
notice here. Berkeley refers to them repeatedly in the 
Commonplace Book, and one of his early queries is 'How to 
reconcile Newton's 2 sorts of motion with my doctrine ?' (30). 
He discusses the Principia with special reference to abso- 
lute motion, in the Principles ( 110-15). In his letter to 
Johnson (Works, vol. ii, p. 19), he says : 'Sir Isaac Newton 
supposeth an absolute Space, different from relative, and 
consequent thereto ; absolute Motion different from relative 
motion; and with all other mathematicians he supposeth 



THE BACKGROUND OF IMMATERIALISM 51 

the infinite divisibility of the finite parts of this absolute 
Space; he also supposeth material bodies to drift therein/ 
Since Berkeley's immaterialism is as much an assertion of a 
spiritual God as of an ideal world, itisof importance to glance 
at the view of matter held by the theologians whose works he 
read. King's De Origine Mali 1 (1702) opens with a statement 
about sense-perception and sensible qualities in Lockian 
terms. He then contrasts the variable qualities and 'that which 
continues under all these changes'. When greater changes 
happen to a piece of wax, its essence and appellation change. 
Still, under all mutations it is always extended, capable of 
motion and rest, and has always some parts separable and 
mutually exclusive. The substance which carries along with 
it these qualities is called matter. When a portion of matter is 
removed another succeeds into its place. Place or space is dis- 
tinct from matter because the same space receives successively 
different bodies (cf.C.P.B. 137). King is a tough 'materialist', 
but he decides definitely against extensionism. No doubt 
with Spinoza and Raphson in view, he denies the necessary 
existence of space. He says that we attribute necessary exis- 
tence to space as the vulgar attribute existence to secondary 
qualities. Again, he says, * Whether there be any such thing 
as space or no, whether its extension be distinguished from 
the extension of body or not : be it nothing at all, be it mere 
privation of contact, as some are pleased to term it : be it mere 
possibility or capacity of existing . . . yet still it is an indolent 
thing, it neither acts nor is in the least acted upon. ... It 
cannot therefore be the cause of matter, or impress motion on 
it. There must then necessarily be another Cause of matter 
and motion'. 2 This passage well illustrates the theologians' 
interest in matter and the points at issue as regards space. 

1 The references to this work in the C.P.B. 143-6, 158-62 have 
hitherto escaped notice. See App. p. 187. 

2 King, De Grig. Mai. c. i, ii. 18, trans. 1731. 



52 THE BACKGROUND OF IMMATERIALISM 

Clarke's Demonstration of the Being and Attributes of God 1 
defines the contemporary English attitude to the Cartesian 
doctrine of matter. He says that the known properties of 
matter are dependence, finitude, divisibility, passivity, and 
unintelligence. It has therefore no active powers. He says 
that the Cartesians are reduced to the absurdity of making 
matter a necessarily existing being ; and he maintains that 
Spinoza holds that the material world is the only self- 
existing, or necessarily existing being. Matter is not a neces- 
sary being. Answering the objection, derived from Hobbes 
and tenderly treated by Locke, that perception might arise 
out of matter, as colours and sounds do from figures and 
motion, he says that there is nothing in the bodies that is like 
the secondary qualities, but that the secondary qualities are 
thoughts or modifications of the mind. He charges Hobbes 
and his followers with holding an * ambiguous and confused 
use of the word matter', and says that thinking and willing 
cannot possibly be qualities or affections of matter, unless 
we will confound the ideas of things and mean by matter not 
what the word commonly is used to signify, a solid substance 
capable of division, figure, and motion, but substance in 
general or the unknown substance, capable of powers or 
properties entirely different from these. Clarke thus is defi- 
nitely opposed to both the Cartesian and the Hobbesian 
conceptions of matter. In effect he says 'matter need not 
have been, but it is there, and is extremely important to us 
for purposes of life and thought 5 . He holds that matter is 
not required for demonstrating the being of God, but is 
required for demonstrating the attributes of God. He says 
for instance (3rd ed., p. 55), that the intelligence of the self- 
existent being 'cannot be demonstrated a priori; but aposte- 
riori almost everything in the world demonstrates to us this 
great truth'. 

1 Boyle Lectures preached 1704-5, published 1705-6. 



THE BACKGROUND OF IMMATERIALISM 53 
From this review of typical contemporary thought, it is 
clear that the view that Berkeley's immaterialism expressed 
'what was in the air at the time* is paradoxical. Any one can 
throw a stone at matter to-day. It was not so then. Matter, 
as well as being, as it always is, an object of the popular 
imagination, was then a mathematical object, a scientific 
object, a metaphysical object, and a theological necessity. 
Berkeley approached the subject of matter from all these 
angles, and had read most of the relevant contemporary 
literature. 1 Against this popular, perplexed, but, in his 
view, idle entity he brought to bear all the engines of his 
resourceful mind. 

A review of the background of Berkeley's immaterialism 
would be incomplete without a reference to the Cartesians. 
Malebranche is dealt with fully in the next chapter. Here 
something must be said of Bayle. I suspect that Bayle 
exerted considerable influence upon Berkeley, but I cannot 
prove it. A copy of the Dictionary was sold at the auction 
of the Berkeleys' library. A duplicated entry in the Common- 
place Book (373, 420) couples Bayle with Malebranche. The 
remaining evidence is subjective. In his early published 
works Berkeley never mentions the Dictionary? and there 
are obvious reasons for his silence. Yet he could hardly 
have left that famous work unconsulted, and the following 
articles seem to me to have left their mark upon the Pnw 7 
ciples: 'Anaxagoras', 'Epicurus', 'Leucippus', 'Rodon', 
'Rorarius', 'Zabarella', and, more especially, Tyrrho' and 
'Zeno'. The main topics concerned are the infinite divisi- 
bility, the nature of extension, and the sensible qualities. 

1 Barrow's lectures on geometry, mechanics, and optics, and 
Keill's Introductio ad veram Physicam, 1702, are referred to re- 
peatedly in the Commonplace Book. Their point of view is that 
of the mathematicians above reviewed. 

2 Bayle is mentioned in Theory of Vision Vindicated, 6. 



54 THE BACKGROUND OF IMMATERIALISM 
Who, if not Bayle, taught Berkeley to reject so uncom- 
promisingly and with such repeated emphasis the infinite 
divisibility 1 ? Keill,Cheyne, Barrow, Raphson, Newton, and 
Malebranche all the great names supported it. Locke, 
following Descartes, speaks of the difficulties whether 'we 
grant or deny it* (Essay, ii. 23. 31 ; cf. ii. 17. 12). Berkeley 
in his early paper Of Infinites cites Locke's distinction 
between infinity and infinite as decisive. But Locke's criti- 
cism of 'the idea of a body infinitely little' is very guarded, 
and in great contrast to the conviction expressed by Berke- 
ley and Bayle. Bayle, speaking of Anaxagoras and the 
Homoeomeriae, says: 'He cannot get out of this difficulty 
but by the divisibility of matter in infinitum, which is to 
imitate a man, who, to avoid the thrust of a sword throws 
himself headlong into a precipice.' 

Bayle seems to have reserved the full vigour of his amazing 
pen for his article on * Zeno' . Here we have an extraordinarily 
clear analysis of deep and abstract subjects. Point after 
point in it would have come home to Berkeley. Bayle's 
problem here is the relation between the infinite divisibility 
and extension. The entry in the Commonplace Book (26) 
which rounds off the opening sketch of the immaterial hypo- 
thesis, relates the infinite divisibility and external extension 
exactly in Bayle's logical way. Bayle says : 'The divisibility 
in infinitum is an hypothesis embraced by Aristotle and 

almost all the professors of philosophy You ought to lay 

aside your disjunctive syllogism and make use of this hypo- 
thetical one. If extension existed, it would be composed 
of mathematical points or physical points or of parts divi- 
sible in infinitum. But it is not composed of mathematical 
points or physical points or of parts divisible in infinitum. 
Therefore it does not exist.' 

1 Some fifteen entries in the C.P.B, name it. Scores refer to it. 
See also Prin. 123-32. 



THE BACKGROUND OF IMMATERIALISM 55 
Then after two formal disproofs of the divisibility, he 
connects the issue with sensible qualities, bridging the gulf 
between primary and secondary in Berkeley's manner (v. 
also sub art. Tyrrho'): 'all the ways of suspension which 
destroy the reality of corporeal qualities overthrow the real- 
ity of extension. The modern philosophers . . . have so well 
apprehended the foundation of the epochs with relation to 
sounds, odours, heat, cold . . . colours, &c. that they teach 
that all these qualities are perceptions of our mind and do 
not exist in the objects of our senses. Why should we not 
say the same of extension ? If a being void of colour, yet 
appears to us under a colour determined as to its species, 
figure, and situation, why cannot a being without any exten- 
sion be visible to us under an appearance of determinate 
extension?' Bayle then refers to those geometrical prob- 
lems, which students of the Commonplace Book know to have 
been much in Berkeley's mind. These are the sides of a 
square equal to the diagonal, and the equality of concentric 
circles. Bayle says that these do not prove that matter is 
divisible in infinitum^ but only make it appear 'that exten- 
sion doth not exist any where but in our minds'. He quotes 
from Malebranche's 'Illustration on Bodies', and adds: 'I 
was obliged to prove that there are stronger objections than 
those of Malebranche.' Bayle concludes the 'Zeno' article 
with a severe criticism of the principles of mathematics, an4 
of the doctrine of infinitesimals. Sir William Hamilton 1 
pointed out that Bayle anticipated Berkeley as regards the 
assimilation of primary and secondary qualities. Ought not 
we who have the Commonplace Book, and can watch Berke- 
ley's thought in the making, to go farther than Hamilton, 
andseeinBayle one of Berkeley's important sources, perhaps 
ranking next in importance to Malebranche and Locke ? 
1 Reid, Inquiry, c. vi, 6. n. 



IV 

PROBLEMATICAL IDEALISM AND ESSE 
PERCIPI 

IF Berkeley's immaterialism had a source other than his 
own genius, there is, I think, no clear evidence as to what 
that source was. There is, however, abundant evidence that 
for the argument supporting immaterialism, expounding it, 
developing it, Berkeley was greatly indebted to the Cartesian 
idealism, and more especially to Malebranche's Search. 1 

Immaterialism, as expounded in the Principles, was a new 
thing. It was Berkeley's invention. He was * the founder of 
a sect called the Immaterialists'. 2 No study of sources is 
likely to upset his claim to that title. 

It is very extraordinary, we think, that a young man 
should have invented a philosophy that still grips European 
and American thought . But youth has these great thoughts . 
Berkeley had remarkable natural endowments. He could 
look within and without, up and down with equal facility 
and interest. He received a broad-based liberal education 
that encouraged expression and that did not condemn him 
prematurely to middle-aged caution. His words on the 
education of the ancients apply to his own education. Like 
theirs, his mind 'seems to have been more exercised and 
less burdened, than in later ages' (Stris, 298). Further, 
Berkeley believed in God with heart and mind. In his meta- 

1 Berkeley seems to have confined his reading of Malebranche 
to the Search. Possibly there are echoes in his writings of the 
Meditations Chrtiiennes, e.g. *Tu pnetres les Cieux, tu perces 
les abimes'; cf. Intr. Prin^ 24. But the same thought occurs in 
the Preface of the Search. So also Nature and Grace gives most 
of the accidental evils mentioned in Prin. 151, but no doubt they 
were stock subjects. 

2 Swift to Lord Carteret, September 3, 1724. L.L. p. 102. 



ESSE PERCIPI 57 

physic he took God seriously. For God's sake he wanted 
to prove immaterialism, and perhaps that wish was father 
to the thought. He speaks once as if immaterialism had 
been an intuition, disclosed to him, long debated, and finally 
made public. Stung by the charge of insincerity, he writes 
to Percival on September 6, 1710, about his three-months- 
old book : 'God is my witness that I was and do still remain 
entirely persuaded of the non-existence of matter and the 
other tenets published along with it. ... Nothing less thai} 
a full conviction not only of the truth of my notions but also 
of their usefulness . . . could have engaged me to make them 
public. I may add that the opinion of matter I have enter- 
tained some years ; if therefore a motive of vanity could have 
induced me to obtrude falsehoods on the world, I had long 
since done it when the conceit was warm in my imagination 
and not have staid to examine and revise it both with my 
own judgment and that of my ingenious friends.' 

He was aged 25 when he wrote the words, so that the 'long 
since* is relative to the train of his thronging ideas ; but the 
'some years' is more precise. We may be certain then that 
the immaterialist conceit warmed in his imagination not 
later than 1706-7, perhaps earlier. Berkeley graduated in 
1704. Then he stayed on in college, waiting no doubt for 
a fellowship to fall vacant, and reading the course for the 
fellowship examination .* Some time during that three years' 
post-graduate reading, the great thought came to him. Cer- 
tainly by the end of 1 706 he knew well Malebranche 's Search . 
If he drew his inspiration thence, and if, as seems probable, 
he began the Commonplace Book soon after winning a fellow- 
ship in June 1707, we understand how he was able to begin 
it with a full-fledged statement (C.P.B. 1-26) of the 'im- 
material hypothesis' in all its main bearings. 

1 Cf. 'aliis studiis occupato', De Ludo Alg., and v. Appendix, 
p. 189. 

4046 r 



5 8 PROBLEMATICAL IDEALISM 

The Commonplace Book is certainly part of the 'revising 
it with my own judgment', of which he wrote to Percival. 
When he had filled 30 or 40 of its pages, the hypothesis 
became proof, and the negation, immaterialism, took posi- 
tive shape as the New Principle. Malebranche figures in 
these pages, 1 and it will therefore be in place here to give an 
account of his doctrine of matter. 

Malebranche 's Doctrine of Matter 

As a headline for his discussion of the secondary qualities, 
Malebranche writes (i. 10): 'We often see things that have 
no existence, nor ever had, and it ought not to be concluded 
that a thing is actually without us, from our seeing it without 
us.' Things have 'a real existence . . . though it be a very 
hard thing to prove it'. He speaks to the same effect in bk. 
vi. 2. 6. As an elucidation of these two passages he appends 
his well-known 'Illustration' (falaircissement, i.e. excursus), 
entitled 'that 'tis very difficult to prove the Existence of 
Bodies ; what we ought to esteem of the proofs which are 
brought of their existence'. We are on firm ground in mak- 
ing this Illustration our starting-point. For the Common- 
place Book refers to it several times (697, 698, 812, 813, 830, 
and perhaps 274). Malebranche does not distinguish here 
between body and matter. He passes from the one to the 
other as if they were synonyms. He is discussing matter as 
embodied and body as material. The discussion is carried 
on in full view of the distinctions already established by him 
between immediate and mediate object, between voir and 
regarder. Malebranche does not doubt, in fact he here 
affirms, the m-existence of what we see (voir)ihe visual 
datum. He does doubt, is not far from denying the out- 
existence of what we behold (regarder). 

1 Especially in those represented by entries 260-300, where 
Berkeley's sense of discovery is at its height. 



AND ESSE PERCIPI 59 

The Illustration opens with a resume of Malebranche 's 
general teaching about the double union. Knowledge aris- 
ing from the soul's union with the body is not comparable 
to knowledge arising from her union with the Word of God. 
Circumambient bodies have not in them to inform us of 
their existence. As has been already proved by him, 'The 
testimony of the senses is never exactly true, but commonly 
every way false.' Here the senses do not report 'for' or 
'against'. 'What reason is there from the reports of our . . . 
senses to conclude there are actually bodies without us and 
that they are like those we see, I mean, those which are the 
immediate object of our soul ?' It is then idle, he thinks, to 
say 'We see external bodies', for in point of fact we do not. 
External bodies, if such there be, are not and cannot be the 
immediate object of our minds. The same argument holds 
of the body. The body, which is united to the soul, 'cannot 
give light to reason*. Therefore Malebranche appeals to 
another court. Since things cannot speak, and the senses 
do not speak (nor would be decisive if they spoke), only 
God can prove to us the existence of matter. Has He done 
so P 1 His answer is 'Yes' and 'No', inclining to 'No'. Male- 
branche is trying to reconcile conflicting interests. He has 
in view different types of knowledge, and different sources 
of knowledge. So the issue becomes perplexed and his 
answer indeterminate. To such an extent is the issue per- 
plexed that Malebranche here advances from the pros and 
cons of the materialist hypothesis to the pros and cons 
of the immaterialist hypothesis. The Illustration which 
begins with the difficulty of proving the existence of matter 

1 Malebranche's theory of the union of man's intellect with 
the Word of God enables him to pass easily from theological 
terms to rationalist. So the same question appears in the alterna- 
tive form : Only reason can prove to us the existence of matter. 
Does reason prove it ? 



60 PROBLEMATICAL IDEALISM 

ends with the possibility of proving the non-existence of 
matter. 

Descartes, he says, has gone as far as bare reason can go 
towards proving the existence of bodies. But he has not 
proved their existence. God speaks by evidence and by 
faith. Faith assures us of the existence of matter. But this 
may mean no more than that material bodies exist for faith, 
are, in fact, part of the setting of the Christian Faith. It 
does not prove that they would exist if not believed in. 
There can be no cogency, for faith, in a proposition about 
a world of imfaith. Revelation assures us that 'prophets, 
apostles, sacred writ, and miracles' exist ; but, he holds, the 
appearance of these things would have the same effect as 
the reality. 

Malebranche passes then from faith to evidence. Here 
reason is inconclusive. 'We are not invincibly carried to 
believe there is anything existing besides God and our own 
mind/ We know in general that God is no deceiver. But 
has He in this particular told us by reason that matter exists ? 
In effect Malebranche answers in the negative. He grants 
to Descartes that God has given us a 'strong inclination' or 
'natural propension' to believe in matter. But, he holds, 
God does not compel us to believe it. If we believe it, we 
believe it, not God in us. Besides, the 'natural propension' 
works by the sensible impression, and the fallibility of the 
senses has been proved up to the hilt. If they are wrong in 
the one deliverance, they may be wrong in the other. 

In his final summing up, Malebranche 's hesitation is 
very marked. He puts revelation on one side, and proceeds 
to reason thus. We have naturally a strong inclination to 
believe there are external bodies. Therefore we have more 
reason to hold the existence of matter than its non-existence. 
This argument is clearly based on a consideration of prob- 
abilities, and falls far short of demonstration. If it were 



AND ESSE PERCIPI 61 

not for scripture, the question would be unresolved. Male- 
branche the monk gives a decision ; Malebranche the thinker 
leaves it undecided. He says, ' We cannot deny the existence 
of bodies, through a principle of religion'. The reference is 
probably to transubstantiation ; it supports Hamilton's con- 
tention that only fidelity to the Catholic doctrine stood 
between Malebranche and Berkeley's views on matter. 1 

Berkeley writes (C.P.B. 697): * Scripture and possibility 
are the onely proofs with Malbranch. Add to these what he 
calls a great propension to think so.' Malebranche himself 
had exposed the 'propension'. Berkeley searches Scripture 
and finds no matter there, and he had only ridicule for the 
poor 'bare possibility' (Principles, 75). Malebranche 
was a 'patron of matter'; he could not therefore teach 
immaterialism. But there can be little doubt that he set 
forward on his way thither the travelling man. 'From Mai- 
branch, Locke, & my first arguings it can't be prov'd that 
extension is not in matter' (C.P.B. 274). He had tried 
Malebranche and Locke. He had under their guidance 
analysed the sensible qualities. He had steeped himself in 
the relativity arguments. He failed to find there a basis for 
immaterialism. So he sought and found another founda- 
tion. But before dealing with his intuitive approach to im- 
materialism, we must discuss the argument from the sen- 
sible qualities in which Malebranche was his chief guide. 

Primary and Secondary Qualities 

Berkeley wanted no half-measures. He was not content 
to deny the matter of metaphysicians. He did so, but he 
went farther and denied the matter of the 'man in the 
street'. He came down to empirical facts, and challenged 
the qualities of things. We suppose, rightly or wrongly, 

1 Sir W. Hamilton, Note P, on Malebranche 's theory, appended 
to Reid's Works. 



62 PROBLEMATICAL IDEALISM 

that there is in the things about us something that enables 
them to stand on their own feet. This is their material 
quality. It guarantees, we think, material reality apart from 
mind. Of material qualities we make material things. Of 
material things we build the material world. Material qua- 
lity is therefore the foundation of the whole structure of 
reality external to the mind. So in attacking such qualities, 
Berkeley was striking a blow at the heart of materialism. 

That an examination of the sensible qualities gave direc- 
tion to his immaterialism, we see from the marginal signs 
of the Commonplace Book . In Berkeley's own index we find 
the two abbreviations T* for 'primary and secondary 
qualities' and 'M' for 'matter*. There are some 36 T' 
entries. Of these 8 mention Malebranche or the Cartesians, 
and another 10 or 12 contain references to the Search. 
Three 'P' entries mention Locke and possibly 4 others refer 
to him. Several 'M J entries also refer to the Search. About 
half of the T' entries are also marked 'M'. So probably his 
first intention was to treat the qualities by themselves ; but 
afterwards he decided to treat matter and its qualities 
together. 

Locke taught that qualities in bodies fall under three 
heads (Essay, ii, c. 8) : (i) Primary qualities, viz. bulk, figure, 
number, situation, and motion or rest of their solid parts. 
These are inseparable and original. (2) Secondary qualities 
immediately perceivable, viz. colour, taste, &c. These are 
powers to produce different ideas in us, and are usually 
called sensible qualities. (3) Secondary qualities mediately 
perceivable, e.g. the sun's power to make wax white. These 
operate on other bodies and make them able to produce 
varied ideas in us. They are usually called 'powers'. 1 

1 King had given great vogue to the doctrine in Dublin, not 
only in his De Origine Mali, but also in his famous sermon on 
* Predestination* (May 15, 1709). He says: 'Light and colours are 



AND ESSE PERCIPI 63 

Malebranche does not use the terms primary and second- 
ary, but he recognizes corresponding categories. The 
former category, dealt with in Book i, cc. 6-9, comprises 
extension, figure, and motion. These are real qualities, 
though our judgements concerning them are mostly false. 
They are extrinsical, without us, and 'wholly independent 
on our mind/ (i, c. 10). Save for solidity, the list agrees 
with Locke's primary qualities. The other class consists 
of the sensible qualities proper. They correspond roughly 
to Locke's secondary qualities. As studied seriatim in Book 
i, cc. 10-13, they are pleasure, pain, heat, cold, light, colour, 
sound, touch, &C. 1 The order is of importance in Male- 
branche 's argument. These qualities are not external to us, 
but are modifications of the soul. 

Malebranche tries to keep these two categories of qualities 
distinct. The former qualities as apprehended are ideas, 
the latter are 'sentiments'. Malebranche is, by his own 
admission, loose in his use of the term idea (see Illust. on 
bk. i, 3). But whenever he is speaking precisely he draws 
a sharp line between the idea which represents that which 
is outside us and the sensations which represent to us what 
we find within us, and are therefore called technically 
'modifications of the souP. The distinction is essential to 
his system. In order that our calm clear ideas may be in 
God, and be seen there, they must be cut clean away from 
the disturbing obscure modes of mind which arise from 
union with the body, and which cannot without impiety 
be attributed to God. The secondary qualities then, for 
Malebranche, are modifications of the human soul caused 
by the general Will of God on the occasion of motions of 
the animal spirits, sense-fibres, and brain. 

nothing but effects ... no such things at all in nature but only 
in our minds.' 

1 For other lists slightly varying see bk. iii. i . i . and bk. iii. 2. 5. 



64 PROBLEMATICAL IDEALISM 

Bayle pointed out that 'sauce for the goose is sauce for the 
gander'. The argument that internalizes the secondary 
internalizes the primary. 1 Locke, 2 apparently blind to the 
same difficulty in his own system, shows that on Male- 
branche's principles the colour of the marigold and the 
number and figure of its leaves must be on the same footing 
in respect of mind, divine and human. Berkeley, whether 
he read it in Bayle or Locke or thought it out for himself, 
made the assimilation of primary and secondary qualities 
the spear-head of his attack. In the opening statement of 
the immaterial hypothesis (C.P.B. 20) he writes: Trimary 
ideas prov'd not to exist in matter after the same manner 
yt secondary ones are prov'd not to exist therein/ His 
resolve to equalize the status remains unaltered through- 
out the Commonplace Book. 3 He must have noticed the 
difference of opinion as to which qualities were primary 
and which were not, as well as the arbitrary character of 
the distinction. But his main argument is simply that in 
point of fact colour is always extended. 

In the Principles we find this argument at the outset. In 
9~ T 3 h e develops the thesis that primary and secondary 
qualities are in point of fact inseparable. That is his own 
line of proof. In 1 4 he glances at an auxiliary line of proof. 
Primary qualities can be shown not to be in matter 'after the 
same manner as modern philosophers prove certain sensible 
qualities to have no existence in matter'. His two illustra- 
tions of this 'manner' show that he has Locke (ii. 8.21) and 
Malebranche (i. 13) in view. Berkeley goes on to show his 
dissatisfaction with this mode of arguing. In effect he says 
that arguments based on the variability of primary qualities, 
and on the relativity of sensation, carry you only part of the 

1 So Hamilton, Reid, Inquiry, c. vi. n. 2 Exam. Maleb., 40-1 . 
3 v. 123, 330, 380, 452, 813, 894, 938 for extension and colour; 
cf. 231, 704, 844, for the generalized thesis. 



AND ESSE PERCIPI 65 

way. At best they form arguments ad hominem, and they 
represent little advance upon Descartes 's problematic 
idealism^ They prove that we do not know the absolute 
magnitude and distance of objects ; they do not prove that 
there is no absolute magnitude nor distance. With these 
unsatisfactory lines of argument he contrasts 'the arguments 
fore-going', i.e. those of 9-13 which are his own, and are 
based upon intuitive perception of the meaning of the exist- 
ence ( 3) of sensible things. 

The assimilation of primary and secondary qualities 
might well support scepticism, as with Bayle, or indeed 
materialism. That it became a leading argument for im- 
materialism should be traced probably to its conjunction 
with the doctrine of modifications of the soul. Primary 
qualities are where secondary qualities are. But where is 
that? Malebranche might have supplied the answer. 
Secondary qualities are in the soul, because they are modi- 
fications of the soul. 

The term 'modification* involved a break with scholasti- 
cism, as Bayle shows (sub 'Spinoza'). It cut across the old 
doctrine of substance and accidents. While Locke left the 
secondary qualities dependent on the mind, Malebranche, 
calling them 'modifications', takes them into the mind and 
makes them part of it. Light and colours then are not only 
relative to mind, but mental, as are love, joy, and hatred. 
They are so near us, he holds, that we cannot cause them 
nor comprehend them. In the attempt to grasp them, we 
extrude and externalize them. Yet our clear and distinct 
idea of extension excludes them. They must be, he argues, 
in the mind, because there is nowhere else for them to be. 
The most serious consequence of this doctrine, patent to 
the true Cartesians and to Berkeley, was its reaction upon 
our conception of the mind. Descartes's 'first knowledge' 
disappears. The idea of the soul must go (Search, i. 12, 13). 

4046 v 



66 PROBLEMATICAL IDEALISM 

We simply do not know what we are . We have no knowledge 
of the fabric of the mind, and cannot know what modifica- 
tions it can endure. Our partial self-knowledge is no more 
than conscience or sentiment interieur. Cartesians resented 
this perversion of the master's doctrine, and ridiculed Male- 
branche's 'rain-bow soul'. It may be noted that Clarke, who 
in general opposed Cartesianism, accepted the view that the 
secondary qualities are 'thoughts or modifications of the 
mind' (Dem., 3rd ed., p. 59). 

Did Berkeley accept it ? He was profoundly influenced 
by it. It was the background for his assimilation of the 
primary and secondary qualities. Both are alike in being 'not 
without the mind'. Further, the method of the first Hylas 
dialogue follows the lines of the Search (i. 12). Both argu- 
ments take the qualities seriatim and internalize them all. 
Malebranche treats pain and pleasure first. They are 'strong 
and lively sensations'. They modify the soul so sensibly 
that, for instance, we never locate the pain in the needle 
(cf. C.P.B. 441). Heat and cold also are vigorous and 
readily become pain. Unfelt they would be nothing; yet, 
with an effort, at times we locate them in the object. But 
light and colours, qua sensations, are faint. So we divest 
our own soul, where they properly belong, 'to cloathe and 
beautify the objects that are without us'. 

Berkeley takes the qualities in the same order ; he uses the 
same terms, strong, lively, vigorous, faint, of ideas of sense 
( 3> 33)> an< ^ ne adopts in part ( 27) Malebranche 's 
argument that having no idea of the soul we do not know 
what modifications she is capable of. Once at least, (T.V. 
94) he uses the technical phrase 'modifications of the soul', 
coupling it vaguely with 'thoughts, desires, and passions'; 
and he approaches Malebranche 's doctrine in the Theory 
of Vision 41, where, summing up his account of distance, 
he states that the objects of sight are 'thoughts or sensations 



AND ESSE PERCIPI 67 

each whereof is as near to him as the perceptions of pain or 
pleasure or the most inward passions of his soul'. 

Influence is one thing, doctrine another. On the whole it 
would not be correct to say that Berkeley accepted Male- 
branche's doctrine of modifications. This is an important 
point for interpreters of Berkeleianism, involving the issue 
closely and acutely argued by Dr. Dawes Hicks 1 as to the 
way of mode versus way of idea. I quite agree with Dr. 
Dawes Hicks that Berkeley 'in his more guarded moments' 
held firmly the 'duality of subject and object', but I am not 
convinced that instances of less guarded moments are many 
or serious. There were definite obstacles in the way of his 
acceptance of Malebranche's 'modifications'. The chief 
obstacle was his distinction between active spirit and passive 
idea. Light and colours, tastes and sounds cannot, for 
Berkeley, be modifications of mind, part of the fabric of the 
mind; for they are passive and not active. They are idea- 
things, not spirit-things. They are objects that are not sub- 
jects though necessarily related to a subject. 

Although he did not accept the implications of the 
Modification' doctrine, yet from it Berkeley would get the 
support he needed for his assimilation of the primary and 
secondary qualities, thereby laying a strong foundation for 
his argument for immaterialism. If a ray of light, composed 
of globules, almost a mathematical line, is in the truth of 
things an intimate part of the mind, then there can be no 
absurdity in stating that the very elements of the world we 
see are not without the mind. Where lines of light are, there 
colour must be. Therefore length and breadth and depth 
along with light and colour are not without the mind. 

The term 'conversion' has been used to describe the 
beginnings of Berkeley's immaterialism. The term is not 

1 Berkeley, pp. 1 10 sqq. Note the explicit 'not generated from 
within by the mind itself* (Prin. 90). 



68 PROBLEMATICAL IDEALISM 

inappropriate, provided we remember that here, as in most 
conversions, the sudden and the gradual were blended. In 
Berkeley's conversion we may distinguish three crises. 
First came the initial intuition or inspiration. All we know 
of it is what he says in his letter to Percival quoted above 
(p. 57). We may presume that the 'conceit' came to him in 
1705-6, tracing it to the religious motive at work in his 
mind, critical from childhood and now aghast at the rising 
tide of materialism. He turned it over in his mind and dis- 
cussed it with his friends. Then came the second crisis, 
probably in the summer of 1707. The 'conceit' now takes 
written form as 'the immaterial hypothesis'. The young 
Lockian is thinking of turning Malebranchian. In accor- 
dance with Malebranche 's rules of method, he compiles 
the Commonplace Book, testing the hypothesis by the facts. 
After pondering the mathematical problems involved, and 
making a carefulstudy of vision mainlyunder Malebranche 's 
tuition, and scrutinizing the material qualities in close con- 
nexion with Malebranche 's doctrine of modifications of the 
soul, he met the third wave, perhaps early in 1708. At the 
third crisis the 'convert' found full salvation. The hypo- 
thesis that matter is nothing is verified, and takes positive 
shape in the affirmation of the New Principle. At this point 
Berkeley consciously transcended his sources, went beyond 
Locke, beyond Malebranche, beyond his own 'first argu- 
ings', and launched out into the uncharted sea of essepercipi. 
The Commonplace Book supplies the evidence. Towards 
the end of the first note-book we find entries touched with 
the discoverer's joy. We see (274) the young thinker at the 
end of his tether, and his first method bankrupt. Locke, 
Malebranche, and his own 'first arguings' have brought 
him far, but have failed him. So he turns to a new method, 
and in the entry 287 he records his Heureka. From this on, 
he subordinates the argument from sense relativity, and 



AND ESSE PERCIPI 69 

substitutes that from the clear and distinct idea of percep- 
tion or existence. The terms become interchangeable. The 
hypothesis of matter is not needed, for spiritual substance 
explains all the facts. His parallel study of vision reached 
a head about the same time (277, 283, 286). The inverted 
retinal image points to the same conclusion. For if the retinal 
image were a photograph of something outside the mind, 
erect vision by inverted image would be impossible. At the 
same time too (272-3) he seems to have satisfied himself as 
to one of the most urgent of the mathematical problems in 
the case, the incommensurability of the diagonal and side. 
'The diagonal is commensurable with the side', and accord- 
ingly a proof of the infinite divisibility of matter falls to the 
ground. So on all the main counts Berkeley concludes that 
there can be no 'Outside the mind'. 'Me percipit . . . horror' 
exclaimed Lucretius, 1 when he saw the secrets of existence. 
Just so, his great antithesis, Berkeley, is staggered by the very 
simplicity of his intuitive solution. He writes: 'I wonder 
not at my sagacity in discovering the obvious tho' amazing 
truth. I rather wonder at my stupid inadvertency in not find- 
ing it out before, 'tis no witchcraft to see we know nothing 
but our thoughts or what these think' (287, prima manu). 2 

1 De Rerum Natura, iii. 28-9: 'His ibi me rebus quaedam 
divina voluptas percipit adque horror . . .' 

2 Berkeley's denial of matter is not to be limited in scope, nor his 
belief in real things impugned. The unflinching letter of Sept. '6, 
1710 (Rand, p. 81) shows that he had not a particle of doubt of the 
existence of sensible things created by God, yet had not a shred of 
belief in any sort of matter. It is instructive to compare his 'entirely 
persuaded of the non-existence of matter ' with his more guarded, 
'non-existence of what philosophers call material substance* 
(Aug. 7, 1713, Rand, p. 123). Yet the two phrases differ only in 
prudence, not in meaning. For the second Hylas dialogue (1713) 
lists seven types of exploded matter, and concludes that matter is 
proved impossible 'in every particular sense that either you or any 
one else understands it in*. Works, vol. i, pp. 437-40. 



V 
SEEING ALL THINGS IN GOD 

BERKELEY'S denial of matter is not to be divorced from his 
affirmation of ideas. The denial catches attention ; but 
the affirmation holds it. The present-day and permanent 
attraction of Berkeleianism lies not so much in its challeng- 
ing negation as in its bold assertion as to the nature of the 
object perceived and known. This assertion is implicit in 
Berkeley's use of the term idea. So it is important to watch 
the usage in the making, to see the new coin struck, and to 
take note of the mint from which it issued. 

Berkeley thought out his doctrine of idea gradually during 
the years 1707-8. A marked change in his style of writing 
the word 'idea' occurred in that period. 1 The change is so 
pronounced that it is probably significant. The introduc- 
tion of the capital letter seems to show that Berkeley meant 
to give the word a new meaning. In any case the term Idea 
at the end of the Commonplace Book had richer connotation 
than the term idea at the beginning. In the interval Berkeley 
had learned to idealize the thing and to spiritualize the idea. 

These two lessons could be learned from the Search, but 
not from Locke's Essay. From Locke Berkeley would learn 
to think and speak and try to know in terms of idea. But 
Locke could not teach any deeper lesson about ideas, for he 
nowhere gives a critical study of the nature of idea. He keeps 
clear of material ideas and of such views of ideation as Hook 
had laid before the Royal Society. 2 But as to the stuff of 
which they are made, their habitat, and the relation between 

1 See Appendix, p. 194. 

2 'The ideas of sight, he thinks, are formed of a kind of matter 
resembling the Bononian stone or some kind of phosphorus.' 
Reid, IntelL Powers, Essay II, c. 9. 



SEEING ALL THINGS IN GOD 71 

my ideas and yours Locke is virtually silent. In the opening 
of his Examination of Malebranche's Opinion he speaks of 
his 'unaffected ignorance' as to the nature and manner of 
ideas in our understanding. In the Essay he takes ideas for 
granted, as elements in thought recognized by all. There 
is no knowledge without them, there is real knowledge by 
their means. Each man finds the simple ideas in his mind 
and manipulates them, framing, compounding, abstracting, 
and comparing them. Locke in fact gives a psychology of 
idea and a metaphysic by means thereof, but he attempts no 
metaphysic of the idea itself. 1 

Clearly a young Lockian would have much to learn about 
ideas and much to unlearn. A study of the nature of idea 
was an essential part of Berkeley's struggle for consistent 
expression. The famous Malebranche-Arnauld contro- 
versy, a brilliant discussion of the status of idea, must have 
been known to him. 2 He would have found a clear outline 
of the points there at issue in Leclerc's little manuals of the 
new Logic which we know he used (C.P.B. 361). But in the 
Search itself he would have found all he needed. For Book 
iii, Part 2, the section in which the theory of seeing all things 
in God is expounded, has the title 'Of the Nature of Ideas', 
and justifies its title. Chapters 1-6 coupled with the 
* Illustrations' constitute a searching and instructive analysis 
of idea which could not fail to further in Berkeley's alert 
mind the twofold process of idealizing the object and 
spiritualizing the idea. 

The Commonplace Book throws considerable light upon 
this process. In the first note-book the word 'idea' occurs 

1 In the Essay, bk. ii, c. i, 1-5 he glances at their source 
and nature, assuming their origin in things and their representa- 
tive function. See also J. Sergeant, Solid Philosophy , p. 22. 

2 Leclerc, Pneum., Sect. I, c. v, De Idearum Natura, *ut uter 
vicerit haud facile constet.' 



72 SEEING ALL THINGS IN GOD 

often in the Lockian sense, but only one or two entries (e.g. 
392) towards the end disclose any reflection upon the nature 
of the idea. It is otherwise in the second note-book. Here 
ideism is a recurrent theme. The relations of idea to idea, 
of idea to word, of idea to thing, and of idea to soul are dis- 
cussed over and over again, andlooked atfrommany angles. 1 
During the months 2 occupied in the filling of that note- 
book Berkeley consciously discarded several Lockian doc- 
trines. He gave up the distinction between adequate and 
inadequate ideas. He considered abstract ideas and abstract 
particular ideas and abstract general ideas. He reached his 
revolutionary doctrine of abstraction. Nearly as momen- 
tous, yet commonly ignored, was his tacit abandonment of 
the Lockian doctrine of simple and complex ideas. This 
distinction, fundamental for Locke, is prominent at the 
opening of the Commonplace Book, 3 but towards the end, 
and in the published works it has practically disappeared 
from sight. 

There are certain entries, viz. 492, 576, 591, 615, 665, 
which seem to me to decide the interpretation of Principles, 
i, an important and recently disputed paragraph. 
Dr. Dawes Hicks (op. cit., p. 109) writes : 'At the beginning 
of the Principles, he starts by accepting, somewhat preci- 
pitately, Locke's threefold division, according to which 
there are simple ideas of sense, simple ideas of reflection, 
and complex ideas', and he adds in a footnote to 'preci- 
pitately': 'Because later he is all the while contending that 
of the mind and its operations there are no "ideas". I do 
not think that Dr. G. A. Johnston's interpretation of this 

1 v. 424, 471, 479-8o, 486, 492, 510, 527-8, 575-6, 591, 615, 
653, 665, 670, 696, 770, 788, 819, 855, 884. 

2 No doubt, the spring and summer of 1708. 

3 The entries are marked +, which I take to be the obelus; 
v. Appendix, p. 186, 



SEEING ALL THINGS IN GOD 73 

passage is tenable/ Here I side with Dr. Johnston. The 
missing word must be * objects* not 'ideas'. In his formal 
opening 'survey' of the objects of human knowledge Berke- 
ley could not, so it seems to me, have accepted Locke's ideas 
of reflection ; nor, in view of the entries 1 cited above, could 
it be said that haste had betrayed him into accepting them. 
Berkeley had deliberated upon this opening sentence. It 
shows every sign of careful wording. Possibly, so as not to 
estrange readers at the outset, he meant it to suggest the 
well-known Lockian distinction. He may be insinuating 
hi point of view ; he may be so stating the Lockian divi- 
sions that in the stating they are quietly transformed. That 
may be so. Yet no tactics of his could alter the fact that he 
had irretrievably made up his mind that the perception of 
white is not different from white. 2 

What, then, does Berkeley in the first sentence of the 
Principles mean by 'or else such as are perceived by attending 
to the passions and operations of the mind' ? He is surveying 
'the objects of human knowledge' in their tripartite division. 
So grammar requires, as Dr. Johnston points out, that such 
should qualify 'objects', and not 'ideas'. What objects he 
has in view seems to me quite clear. But Dr. Johnston, 3 I 
venture to think, has missed giving an adequate exegesis of 
the passage, because he looks almost entirely to Locke for the 
origins of Berkeley's thought. The clue to the passage is, 

1 e.g. 'not to call the operations of the mind ideas* (492). < Mem. 
To begin the ist Book not with mention of sensation and reflec- 
tion* (576). N.B. The word printed 'sensation' in the second half 
of 576, i.e. 'instead of sensation', is almost illegible. It cannot, 
in my opinion, be 'sensation*; it might be 'reflection', but is pro- 
bably 'those*. 

2 v. C.P.B. 591, correct reading, 'Wherein I pray you does 
the perception of white differ from white.' N.B. Berkeley does 
not equate the act of perceiving and the perception. 

3 Development , pp. 143-6. 

4046 L 



74 SEEING ALL THINGS IN GOD 

I think, Malebranche's statement 'there are things we know 
without ideas'. These 'things' for both Malebranche and 
Berkeley are God, the self, and other spirits. It would be 
impossible, surely, for Berkeley to exclude these from his 
opening 'survey of the objects of human knowledge'. To 
pass them over would have been virtual scepticism. This 
class of objects is set aside during the first half of the 
Principles, but it is resumed in 86, and comes in for detailed 
treatment from 135 on, when he has 'dispatched what we 
intended to say concerning the knowledge of ideas'. 

The opening sentence of the Theory of Vision states the 
design of the book. Just so the opening sentence of the 
Principles, rightly understood, outlines its scope and con- 
tents. The principles of knowledge are concerned with 
objects of knowledge, and those objects are, for Berkeley, 
ideas of sense, imagination, and memory, on the one hand ; 
on the other, the objects which are not ideas but spirits, 
namely, the self, other finite spirits, and Deity. If this be 
the correct interpretation, the question must be faced : does 
the phrase 'such as are perceived by attending to the 
passions and operations of the mind' adequately describe 
that hemisphere of knowledge, called 'knowledge of spirits' ? 
I think it does. It is not a full description, but it is not 
positivelyincorrect,andthereforeitissufficientforBerkeley's 
immediate purpose. The self is thus known. Knowledge 
of God is based in 29 on attention to the contrast between 
'passions' and 'operations'. Knowledge of other minds is 
mediated by the other modes of knowing. So consider- 
ing that Berkeley's argument is new, and that knowledge 
of ideas was his main concern in Part I, the description 
is adequate for the purposes of his opening survey of the 
field. 1 

1 Eraser misquotes the passage in his footnote 'ideas perceived 
by attending etc.*, Works, vol. i, p. 272. Browne, Procedure, 



SEEING ALL THINGS IN GOD 75 

Malebranche's Doctrine of Ideas 

Berkeley would meet Malebran die's doctrine of ideas in 
two forms : in the Search itself and in Locke's Examination 
of P. Malebranche's Opinion. This latter work, posthu- 
mously published, appeared in 1706. Berkeley speaks highly 
of the little volume that contains it, in his De LudoAlgebraico. 
But it is far from likely that Berkeley owed to it his know- 
ledge of the famous theory. It is certain that he went to the 
fountain-head and studied the nature of ideas in the Search 
itself. It must be remembered that for Locke all our ideas, 
if not man-made, at least arise in the process of man's know- 
ing. For Berkeley the major part of our ideas are not human 
in origin or in nature. Idea is the characteristic word of 
Locke's Essay. Idea is the characteristic word of Berkeley's 
Principles. But in the interval the word has been trans- 
formed. In Locke's Essay 'idea' speaks of sober English 
common sense, of sunlight, and the highway of life. In the 
Principles 'idea' suggests Celtic twilight and Gallic enthu- 
siasm. It is haloed with mystery and awe. 

Book iii, Part 2, of the Search opens with a clear state- 
ment of representationism, viz. 'I suppose that every one 
will grant, that we perceive not the objects that are without 
us immediately and of themselves. . . . The immediate 
object of the mind when it beholds the sun ... is not the 
sun, but something intimately united to the soul, and that 
same thing is what I call our Idea/ For Malebranche, then, 
ideas are not 'nothings', they are not merely functional parts 
of the process of thinking. Ideas differ from one another. 
This idea is not that idea. An idea can be compared with 
another idea. Again the idea differs from the thing it repre- 
sents, and, as we see from dreams, it may exist in vacuo. 
Ideas have a 'most real existence'. They are 'beings and 

i. 2, ii. 3, who probably taught Berkeley in college, vigorously 
opposes Locke's ideas of reflection. 



76 SEEING ALL THINGS IN GOD 

beings spiritual'. He almost calls them substances. 1 We 
sometimes think them 'small and despicable'. 'Merely an 
idea', we say. We think of ideas as brief and flimsy, called 
into being and annihilated by transient turns of attention. 
But it is self-flattery, says Malebranche in effect, that makes 
us depreciate the status of ideas. We mistake occasions for 
causes, and like to fancy ourselves as sharing the Creator's 
powers, as producing and destroying ideas at will. 

This stress on the actual distinct being of the idea is 
required by Malebranche 's quasi-physical view of presence 
to the mind. Representationism requires the absence of 
the thing ; therefore it demands the presence of an idea of 
the thing. The mediate object being physically absent, the 
immediate object must be actually compresent with the 
mind. For Malebranche, as for many moderns, the seeing 
eye and the thinking mind apprehend X y if they try to 
apprehend Y. From the crude form of representative per- 
ception Berkeley turned away. But he may have learned 
from this passage to take the idea as a real being, and to 
distinguish sharply between' within the mind' and 'without'. 
The quasi-physical 'without the mind' toned down later to 
'without my will' (C.P.B. 379-80). But the early entry (906) 
'all ideas come from without or from within' may well be 
an echo of Malebranche 's statement (iii, 2. i), 'Whatever 
things the soul perceives are only of two sorts ; and are either 
within or without the soul. 1 

If from Book i of the Search Berkeley learned the intern- 
ality of 'what we see', from Book iii he would learn the 
internality of 'what we know'. For as Malebranche de- 
velopes his view of the contents of the mind, the internality 
of the object receives increasing emphasis. If the bright 
sun out there and up there, the most impressive instance 
of externality, has to surrender its light and heat, parts 
1 v. Locke, Exam. Mai. 17. 



SEEING ALL THINGS IN GOD 77 

apparently of its very being, and to make them over to the 
observer's mind, as really parts of mind's very being, and 
if, further, the geometrical properties of the apparent sun, 
the circle of the orb and the line of light, prove to be 'pure 
idea' in the mind, what then is left of the real external 
object ? Berkeley in Hylas says that his purpose is to turn 
ideas into things. It is a daring attempt. It compels him 
to say among the learned that 'we eat and drink ideas, and 
are clothed with ideas' (Prin. 38). It takes courage to say 
'idea', where other folk say 'thing'. He did not rush blindly 
upon that usage. 1 He knew he was taking liberties with 
language, and the decision to do so was deliberately taken 
after debate. It is probable that he had some philosophical 
guidance and leading in his revolutionary usage. Locke 
supplied none here. The New Principle rent the old bottles 
of Lockian ideism. But Malebranche's tough substantial 
'ideas', were the proper model for Berkeley's idea-things. 
The ideas under discussion in the Search, as in the Prin- 
ciples, are the ideas of, or for Berkeley the ideas which con- 
stitute, the material world. The theory 'that we see all 
things in God', when expressed precisely becomes, as its 
author admits, 'that we see all material things in God'. 2 The 
other objects of knowledge, God, the self, other selves, and 
'Intelligences' will be under discussion in the next chapter. 
Confining here our attention to unthinking things, we shall 
outline Malebranche's classic conclusions. So far we have 
seen that the idea is there, and there for study. It is no mere 
trace, nor motion. It is to be taken seriously. It is a real 
being, distinct from the thinker, distinct from the thing 

1 The C.P.B. is full on the point. It was to have been a theme 
of the Introduction; v. 492, 510, 527, 653, 696, 770, 819, 884; 
cf. Prin. 39. 

2 Cf. Prin. 91 : 'The Eternal Mind wherein they suppose only 
Ideas of the corporeal substances . . .' 



78 SEEING ALL THINGS IN GOD 

thought about. Each idea is a spiritual entity. It serves a 
definite purpose in knowledge, enabling me to know what 
you know without losing my identity in yours . Malebranche 
carries the study into deeper waters . He examines the whence ', 
the why, the where. He tries to state the cause and proven- 
ance of ideas, and to rise thence to an understanding of 
the mystery of being and knowing. 

Malebranche follows the method of exclusion. There 
are five possibilities he says, and only five (cc. 2-6). 

(i) Ideas arise from Species Impressae y emitted by material 
objects and resembling them. This is the Aristotelian and 
Scholastic view. Malebranche rejects it for various reasons, 
amongst others the impenetrability of bodies. 

(ii) Ideas are produced by the soul on the impulse of 
impressions made by objects upon the body. This is the 
common view. It is, for Malebranche, incompatible with 
what we know of the nature of man and the status of idea. 
It overrates human powers and flatters human vanity. It 
implies that man can create and annihilate powers re- 
served to the Creator. It depreciates the idea, taking it to 
be a being 'slender and contemptible', brought to nothing 
as soon as out of mind. Further, if man made ideas, he 
must first have had an idea of what he was going to make. 
Finally, the view mistakes occasion for true cause. 

(iii) All ideas are created with us the 'magazine theory* 
(perhaps in view in C.P.B. 658: 'There are innate ideas, 
i.e. ideas created with us'). The theory errs in over-estimat- 
ing the capacity of the mind of man and his powers of 
selection. Man's finite mind cannot contain an infinite 
'magazine* ; nor if it could, would it be equal to the task of 
taking from the 'magazine' the idea appropriate to each 
occasion. 

(iv) Ideas are modifications of the soul, the soul being 
regarded as a sort of Intelligible World, containing in herself 



SEEING ALL THINGS IN GOD 79 

the ideas of external things and seeing them by considering 
her own perfect ions. This theory is a counterpart, or parody, 
of Malebranche 's 'Divine in- Vision 1 on a lower and human 
level. He sees various objections to it. It implies that each 
soul is a light to herself. It requires the soul to see in herself 
what she does not contain. It degrades pure luminous 
ideas to the level of the obscure modifications de Vdme. 

(v) There is, then, only one possibility left, viz. ideas 
representative of material things are in the infinite mind of 
God, and by our union with the Word we see them there. 1 
Malebranche's starting-point, adopted by Norris, 2 is the 
necessity that God should have in Himself the ideas of all 
the beings He has created ; for otherwise He could not have 
produced them. God sees created things 'by considering 
the perfections He includes whereunto they are related'. 
By strict union with God the mind of man can see 'what 
there is in God which represents created beings'. God's 
ideas are therefore the essences of things, and the incom- 
prehensible simplicity of the Divine Essence includes the 
deas. Malebranche warns us not to conclude that our 
ninds see the Essence of God. 'What they see is most im- 
Derfect, whereas God is most perfect.' The union is in the 
Presence of God, not in His Essence. This important dis- 
inction is sometimes overlooked. In the locus classtcus* 
Berkeley virtually admits his close resemblance to, and 
*ven partial agreement with, Malebranche, denying what 
Vlalebranche did not affirm, that we see things in the Es- 
>ence of God. Finite spirits therefore do not merge in the 
nfinite, any more than bodies do in space. 

Reflecting then on his theory, Malebranche adds five 

1 For the infinite comprehension v. C.P.B. 181, 361, and 
especially 686, 'God may comprehend all ideas', and Rand, p. 84. 

2 Theory of the Ideal World, pt. i, c. 2, especially 19. 

3 Hylas, p. 427, ist ed., and less clearly in Prin. 148. 



8o SEEING ALL THINGS IN GOD 

considerations in support of it. (a) The only plausible 
alternative is the * magazine' theory, and this is contrary 
to the principle of Divine Economy or Parsimony, (b) His 
theory has the merit of 'instating created minds in an 
absolute dependence upon God', (c) It corresponds to 
psychological fact. In order to think of 'this' we must first 
think of All. 'We first cast about our view upon all beings 
in general and afterwards apply ourselves to the considera- 
tion of the object we desire to think on.' 1 (d) The theory, 
and it alone, accounts for our 'seeing' universals. (e) God's 
purposes are self-contained. He has no other end of action 
but Himself. 'If God had made a mind and given it the sun 
for its idea . . . God, we should think, had made that mind 
. . . for the sun and not for Himself.' 

In this theory it is essential to distinguish idea from 
sensation. Malebranche, of course, denies that we have 
sensation in God. The sensation is a modification of the 
soul which God causes without having. The idea is 'joined' 
to the sensation, and is in God, and we see it there because 
He is pleased to show it us. The chapter (vi) concludes 
with an impressive passage: 'God is the intelligible world, 
or the place of spirits . . . From His power they receive 
all their modifications; in His wisdom they discover all 
their ideas; by His love they are influenced with all their 
regulated motions; and because His power and His love 
are nothing but Himself, let us believe with St. Paul, that 
He is not far from every one of us, and that in Him we live 
and move and have our being.' 

The six chapters in which Malebranche exhibits his 
theory of ideas and the vision of all things in God form an 
imposing fagade of argumentation whose inconsistencies 
Locke easily exposes without doing much harm to the in- 
tuition behind the argument. Berkeley knew the Search 

1 Cf. C.P.B. 605, and note the same phrase 'cast about*. 



SEEING ALL THINGS IN GOD 81 

when he was writing the De Ludo Algebraico and the Com- 
monplace Book. He was familiar with it when he was writing 
the Theory of Vision. Its main argument was in full view 
when he was writing the Principles. To know the argument 
was to be moved by it. 1 I would urge that the Search left a 
profound and indelible impression upon Berkeley in his 
plastic period, an impression that conferred character, an 
impression that is the key to the interpretation of Ber- 
keley's system. 

Berkeley nowhere formally denies the debt. He had good 
prudential reasons 2 for not parading it. Several passages 
virtually admit it. All his main works, Principles, Hylas, 
Alciphron lead up to the Vision of all things in God. The 
Hylas dialogues contain a dramatization of his own philo- 
sophical experience, and perhaps what Hylas in the finale 
says of himself was true of Berkeley, namely, that he did 
not thoroughly comprehend the course that brought him to 
his position ; but certainly what Hylas there says of Philo- 
nous was autobiography, * You set out upon the same prin- 
ciples that Academics, Cartesians, and the like sects usually 
do/ Berkeley advanced to ' opposite conclusions' as regards 
the senses and matter, but what Malebranche taught him 
about God remained with him, and passed into his pub- 
lished philosophy. Berkeley's disclaimers have been mis- 
understood. They are partial, not total. They concern 
final position, not origins. The extreme occasionalism, the 
doctrine of 'intelligible matter', the union of the 'pure 
intellect' with God, certain Spinozistic developments of 
Malebranche 's theory, as that we literally see and are united 
to the Essence of God these things Berkeley disclaimed 

1 Even Locke was affected by the 'many very fine thoughts, 
judicious reasonings, and uncommon reflections contained* in the 
treatise he criticized. Exam. Mai. i. 

2 See above p. 40. 

4046 M 



82 SEEING ALL THINGS IN GOD 

and disowned. Even his protest that he does not understand 
the theory should not be taken too seriously. He protests 
too much. Besides, the unintelligibility of the theory was 
the fashionable criticism in England. 1 It was a safe criticism 
because Malebranche admits somewhere that he does not 
understand parts of his own theory, accepting them on 
faith of reason. Anyhow, Berkeley's contemporaries con- 
nected his teaching with that of Malebranche, and he seems 
to have gone as near the doctrine of seeing all things in God, 
as one who was not an occasionalist could go. 

Chajracteristic of Berkeleianism is the theocentric solu- 
tion of the problems of being and knowing. It is no reflec- 
tion on Locke's deep piety to say that this mark is absent 
from his system. God is in the background of Locke's 
Essay, in the foreground of Berkeley's Principles. The two 
thinkers conceived Deity differently. Locke's deity is trans- 
cendent, Berkeley's deity is immense. 2 With the text of 
immanence (more correctly, immensity), 'In Him we live 
and move and have our being', Malebranche clinches the 
argument, as we saw above, for the Divine in- Vision. There 
are several striking occurrences of that text in Berkeley's 
writings, most of them being apropos Malebranche (Prin. 
149 ; Hylas, Works, vol. i, pp. 427, 453 ; Alciphron, iv, 14 ; 
cf. C.P.B. 839, Essays in Guardian, Works, vol. iv, p. 181, 
and title-page Theory of Vision Vindicated). The former of 
the Hy las passages is particularly striking in the first edition, 
though an inartistic 3 addition in the third edition (1734) 
has obscured the point. 

1 v. Keill, Examination of Dr. Burnet's Theory, pp. 7-9, and 
Locke, Fam. Letters, p. 45. 

2 This term, once technical, remains in the Articles of Reli- 
gion (1562-3), 'immensae potentiae, sapientiae, ac bonitatis' 
(Art i). It occurs frequently in the Search. 

3 'Inartistic* because Philonous shows full knowledge of the 
system though he has just asked for information about it. 



SEEING ALL THINGS IN GOD 83 

Hyl. But what say you? Are not you too of opinion that we 
see all things in God? If I mistake not, what you advance 
comes near it. 

Phil. I entirely agree with what the holy Scripture saith, 
'That in God we live and move and have our being.' 

The reply goes as near as possible to saying c I agree with 
that part of Malebranche's philosophy'. 

An immense, immanent Deity, substance of real ideas, 
thinker of the unthinking, will of the inactive, makes 
possible the esse percipi. For Locke, the way of ideas was 
a human path. In the Search ideas are always thought of 
in connexion with Deity. 1 Malebranche did more than 
weaken the evidence for matter. He showed Berkeley, if I 
mistake not, how to construct a system dispensing with 
matter. He substituted the God of ideas for the world of 
ideas. He showed Berkeley how our ideas can be also God's 
ideas, thereby laying a foundation for inter-subjective 
thinking. We may with confidence take 6 of the Prin- 
ciples as expressing Berkeley's primary and controlling con- 
ception of unthinking reality. For that intuition possessed 
for him 'all the light and evidence of an axiom* (ib., ist. ed.), 
and the elevated diction suggests a mystico-religious ex- 
perience as its source. The seer sees 'all the choir of heaven 
and furniture of the earth' laid out in the mind and will of 
some Eternal Spirit, like a great cathedral church in the 
mind and will of a creative architect, being there because 
thought there. 'All those bodies which compose the mighty 
frame of the world* may be called things or ideas (that point 
matters comparatively little), provided that the primary 
truth is kept, namely that they, one and all, lie there in the 
mind of God, being because thought on by Him, always 

1 In the 'Conclusion of the three first books' Malebranche 
shows in detail that all modifications and ideas are received from 
God. 



84 SEEING ALL THINGS IN GOD 

thought on by Him, sometimes thought on by me, by you, 
by us, because always thought on by Him. Instead of 
starting inquiry, as we do, with an oblong florin or some 
other little bit of a sense datum, Berkeley began from a 
vision of the total datum, of heaven and earth lying there 
passive in the vaster mind of God. Berkeley is not seeing 
this or that in the mind of man. He is actually 'seeing all 
things in God'. 

Real things, then, for Berkeley, are in God and are ideas. 
I have real ideas; therefore some of my ideas are also 
God's ideas. This twofold status of the idea comes out 
repeatedly in the Principles, and as we have seen, it is one 
of the major results of Malebranche's study of the nature 
of ideas. 1 Ideas of sense are 'our ideas' (Prin. 25 &c.), 
because we perceive them ; yet clearly they are not creatures 
of my will (Prin. 29). They are produced by the will of 
God and excited in my mind. So that also they are God's 
ideas, just as our world is also God's world. When Berkeley 
is pressed on the score of sameness (e.g. Hylas, Works, 
vol. i, p. 466) he bids the reader choose between the ab- 
stract idea of identity, which no one can grasp, and the 
familiar use of words. I have an idea, I write it in a book 
and publish it. The idea that was private becomes public 
without ceasing to be, for all practical purposes, the same 
idea. My idea does not cease to be mine because it becomes 
yours too. Sharing it makes it more richly mine, if it is true. 
Malebranche spoke of God 'discovering' his ideas to us. 
That 'discovery' is a simple corollary of the fact that 'in 
Him we live'. 2 Berkeley had the same view-point here as 
Malebranche, I think. Both thinkers agree in the prin- 

1 So Locke, Exam. Mai. 24: 'God hath the same ideas we 
have.* 

2 'Whose Will it is they should be exhibited to me.' Hylas, 
p. 428. 



SEEING ALL THINGS IN GOD 85 

ciple 'in tuo lumine lumen videbimus'. If God is left out of 
account or brought in ex machina to the system, then, no 
doubt, Berkeleianism is unsatisfactory on the score of same- 
ness. But seeing all things in God and holding that God 
maintains intercourse between spirits, Berkeley could no 
more dispute as to whether two spirits see the same thing, 
than he could dispute as to whether they see it by the same 
light. 

The same Substance is, for Berkeley, the solution of the 
problem of permanence. There is nothing so imper- 
manent as man's ideas qua fancies; but if his ideas are also 
God's ideas, they last, though 'out of mind'. If Deity be 
immense, unbounded by time and space, limitless in 
thought and act, it seems reasonable to conceive Him as 
minding the aurum irrepertum of the Commonplace Book, 
cum terra celat (say, from 1753 to 1870), and as not ceasing 
to mind it when a man's intelligent activity, drawn from 
the same abyss, 'finds' the book and makes it current coin 
in the republic of letters. 

As regards archetypes, in Dublin in 1710, as in America 
twenty years later, Berkeley was very guarded. He would 
affirm them or deny them according to the sense in which 
the word was used. The word is used in the Commonplace 
Book (701, 835) and in the Principles ( 45, 87) much as 
Locke uses it, for material originals. Two other passages 
in the Principles, 71 and 99, call for mention. The latter 
passage speaks vaguely of archetypes existing 'in some 
other mind'; the former so rejects 'intelligible matter' in 
the mind of God, as to leave open the possibility of Ideas 
in the mind of God. Berkeley believed, of course, in ideas 
in the mind of God. 1 But, dvyra </>/>ovo)v, he took little 
interest in God's ideas when they are not also our ideas, 
and did not pretend to any knowledge of the topography 

1 v. his letter to Johnson, March 24, 1730. Works, vol. ii, p. 19. 



86 SEEING ALL THINGS IN GOD 

of the 'intelligible world'. In the third of the Hylas dia- 
logues he goes rather further, and pressed for a philosophy 
of Creation he admits archetypes 'external to your own 
mind' and a 'two-fold state of things the one ectypal or 
natural, the other archetypal and eternal'. But he is always 
afraid of matter creeping back in disguise, as a Third 
Nature, or Divine attribute, or Spatium Reale, or Intel- 
ligible Matter, and his guarded admission of the archetypal 
state is always to be qualified by the governing principle, 
'No idea or archetype of an idea can exist otherwise than 
in a mind' (Hylas, Works, vol. i, p. 425). 

Ideas of the Imagination Occasionalism 

Leaving real ideas, we come to the important group of 
objects of knowledge called 'ideas of the imagination'. 
This group gives a definite and distinctive character to 
Berkeley's system. Here he may be indebted to Male- 
branche, though the difference in doctrine is very marked. 
Berkeley could hardly have framed this part of his theory 
without reference to the Search, for Malebranche devotes 
Book ii in toto to the imagination and considers the imagina- 
tion again in Book iii in connexion with the Divine in- 
Vision. Malebranche isolated the imagination and thereby 
perhaps drew Berkeley's attention to the metaphysical 
importance of the faculty. 1 Berkeley here would find no 
guidance in Locke. The peculiar character of imaginative 
perception is scarcely mentioned in the Essay. Locke speaks 
of ' Invention or voluntary putting together of several simple 
ideas' (ii. 22. 9), and he opposes fantastical ideas to real ideas, 
but he gives no systematic treatment of the imaginative 
faculty. It is simply lost among the various modes of think- 
ing. Descartes touches on the imagination in the second 

1 This, however, was a commonplace. As regards the imagina- 
tion, I merely state, without pressing, the evidence for dependence. 



SEEING ALL THINGS IN GOD 87 

Meditation, and discusses it at greater length in the sixth. But 
Descartes could hardly give more than hints. 1 The Search, 
however, devotes a whole book of three parts to the imagi- 
native faculty, and treats it as one of the primary parts 
of the human understanding co-ordinate with sense and 
intellect. 

The opening of the second Hylas dialogue affords some 
evidence of dependence. The account there given of the 
brain and animal spirits may be a summary of Male- 
branche's account of the physiological concomitants of the 
imagination (bk. ii, i. 4, 5, perhaps referred to in the petu- 
lant C. P. B. 432). Perhaps we may trace an even earlier 
connexion. For in his essay on the Cave of Dunmore (1705- 
6) Berkeley has a paragraph on the imagination reminiscent 
of the Search. 2 In the Commonplace Book the imagination 
receives considerable attention and is constantly isolated, in 
Malebranche's manner, as one of the three primary modes 
of apprehending ideas, e.g. 'the existence of our ideas con- 
sists in being perceiv'd, imagin'd, thought on' (471). Imag- 
inable and sensible ideas are contrasted in 581, 588, 667. 
From 804, 830, and 835, we see that sense and imagination 
are the only two of the three faculties left for the time being ; 
for in the meantime Berkeley has developed his doctrine of 
abstraction, and 'pure intellect \qua faculty of abstract ideas, 
is eclipsed (822). In 788 he defines idea, and it is interesting 
to see that he first wrote 'by idea I mean any sensible or 

1 Berkeley knew Molyneux's translation (1680) of the Medi- 
tations with Hobbes's objections, v. C.P.B., 807-10. 

2 'Men of a strong imagination . . . fancy the petrified water 
stamped with the impressions of their own brain* and 'the clouds 
so far comply with the fancy of a child, as to represent to him 
trees, horses, men, or whatever else he's pleased to think on.' 
Cf. 'The brains of men of strong imaginations receiving deep 
impressions of the subject* (Search, ii. 3. i) and 'chariots, men, 
lions, and other animals in the clouds' (Search, ii. 2. 2). 



88 SEEING ALL THINGS IN GOD 

imaginable or intelligible thing', subsequently erasing the 
last three words (cf. 790). So we can see that his doctrine 
of ideas of sense and imagination was well thought out, and 
its main lines may have been reached by the impact of his 
doctrine of abstraction upon the tripartite division of the 
understanding, viz. sense, imagination, and pure intellect, 
which occupy respectively Books i, ii, and iii of the Search. 
Malebranche found the fact of the imagination a grave 
difficulty. His problem was this : how can ideas, obviously 
the result of human fancy, obviously inconsistent with truth 
and reality, be in God and be seen in Him ? Accordingly 
he goes very carefully into the nature of the image, in order 
to solve this problem and save his theory. He divides the 
imagination into two faculties, active and passive. The 
passive imagination is what we should call the part played 
by the nervous system. A good deal of the physiology of 
Malebranche 's account is on modern lines. He lays down 
that memory consists in the traces left by the animal spirits 
in the brain (cf. 'you make certain traces in the brain to 
be the causes or occasions of our ideas . ' Works, vol . i , p . 42 1 , 
where Hylas expounds 'the modern way'). The active ima- 
gination receives no separate treatment (v. bk. ii, i. i). It 
is the element due to the imagination's dependence upon 
the soul. It consists in 'the action and the command of the 
will', to which the animal spirits that delineate the images, 
and the brain fibres that take the images, pay obedience. 
In bk. iii, 2. 3, he considers the imagination in connexion 
with the intellect. One great source of error, Malebranche 
thinks, is the confusion between two representative entities 
the image engraved on the brain and the idea which 
accompanies it. The idea 'regulates the image'. The idea 
tests the agreement of the image with reality. In respect 
of both entities appearances deceive us. We find the idea 
present to our mind when we will to have it. We ought to 



SEEING ALL THINGS IN GOD 89 

conclude that our will is a sine qua non, but not that our will 
is the vera causa ; our will does not show the idea to the 
mind, much less create it de nihilo. Similarly, we appear to 
produce the brain image, but we do not do so, for we do 
not know how to do so, nor how it is done. We are not then 
the cause of the image on the brain, and we may not call 
ourselves cause of ideas of the imagination. Our active part 
is to make the mind more attentive to the ideas of 'pure 
intellect'. When, for instance, we imagine a square, an 
auditory image of the sound of the word provokes a visual 
image of a shape. These images come to us. Our part is to 
go to meet them, and direct the mind towards the pure 
ideas of extension that lie behind the image of the square. 

Malebranche 's argument is a product of his occasion- 
alism. God is 'the one true cause'. Man is always trying 
to evade this truth. Man thinks, in defiance of reason, that 
a bowl in motion (cf. C. P. B. 36) communicates motion to 
the bowl it meets ; but it does not ; for non dot quod non 
habet. The bowl is in motion, not the motion in the bowl 
(Search, iii, 2. 3). Malebranche's more famous instance is 
that a man thinks he moves his arm, but does not. Moving 
the arm is a very complicated business, and a man cannot 
do it, because he does not know how. The arm is moved 
on the occasion of his willing to move it. Just so men take 
the act of applicative will which accompanies and occasions 
the production and preservation of ideas to be a vera causa, 
and overlook the true cause, the Will of God. 

Now here we have a main point at issue between Berke- 
ley and Malebranche. One wonders if this was a topic dis- 
cussed at their alleged meeting 1 in the Paris monastery, 
when Malebranche raised his voice and lost his temper, 
and according to the bon mot, Berkeley became unwittingly 
'the occasional cause of his death'. Here at any rate was 

1 See Appendix ii, p. 208. 

4046 N 



90 SEEING ALL THINGS IN GOD 

inflammable matter, a point of acute difference. Perhaps 
Berkeley was familiar with Malebranche 's doctrine of the 
image, and perhaps he was indebted to it. If so, he de- 
parted widely from it. Fortunately we are in no doubt as 
to the reason of the divergence. We have Berkeley's own 
statement of the case. Poking sly fun at the 'arm' argu- 
ment, he writes in his note-book (C. P. B. 553) : 'We move 
our legs ourselves. 'Tis we that will their movement. Here- 
in I differ from Malbranch.' It is a summary description. 
There are other differences, three of which Berkeley speci- 
fies in the second Hylas dialogue. 1 But leaving matter out 
of account, which Malebranche the monk retained but 
Malebranche the thinker discounted, there is little else to 
separate the two systems, as philosophies of ideas. This 
difference is no slight one. It is grave and far-reaching. 
Berkeley, loyal as ever to empirical fact, recognizes the 
reality of finite activity. He comes to terms with it, makes 
it indeed one of the poles on which his system turns. For 
if we move our legs ourselves, we make ideas ourselves. We 
make and unmake ideas (Prin. 28), as really as we stretch 
and relax muscles. 

Berkeley steers as it were a middle course between Male- 
branche and Locke. He granted to Malebranche that some 
(the majority) of our ideas are also God's ideas. He granted 
to Locke that some ideas are made by man. The connexion 
between idea and will comes in for repeated notice in the 
later portion of the Commonplace Book (801-45). 'Every 
idea has a cause, i.e. is produced by a Will' (843). This is not, 
as in ordinary Christian apologetic, a deduction from Divine 
attributes, but is a matter of observed fact from which the 
being of God is correctly inferred. 'When in broad daylight 
I open my eyes, it is not in my power to choose whether I 
shall see or no' (Prin. 29). Ideas of sense 'are not creatures 
1 1734 ed.; v. Works, i., p. 427. 



SEEING ALL THINGS IN GOD 91 

of my will. There is therefore some other will or spirit that 
produces them'. So the real ideas that are the real world are 
God's ideas discovered to us and becoming pro tanto our 
ideas. But there is also, for Berkeley, the plain matter of fact 
which any one can verify in his experience, that we com- 
mand, control, make, and unmake some of our ideas. 'This 
making and unmaking of ideas doth very properly denomi- 
nate the mind active. Thus much is certain and grounded 
on experience.' Thus the imagination with its ideas takes 
a defined place in the Berkeleian system. Sensation proper 
is, for Berkeley, a sphere of God's manifest operation. 
Imagination proper he views, I think, as a sphere of man's 
free activity, into which God wills not to enter. The two 
spheres are in theory distinct, but it is not likely that 
Berkeley regarded them as actually separate. The dis- 
tinctions of theory are blurred in living fact. Berkeley 
contrasts ideas of sense and ideas of imagination on the 
score of regularity, vividness, and constancy ; he contrasts 
them with regard to their source, and their metaphysical 
implications. He could scarcely have made a clean cut 
between them. In actual fact image blends with percept, 
and percept with image. So, on Berkeley's principles, 
there is no need to regard our fancies as entirely godless, 
nor our imaginings as lying outside God's general Will. 1 

1 With regard to this and to similar questions which arise out 
of Berkeley's doctrine of ideas, it would be rash to infer that 
Berkeley did not see the difficulty and had no answer ready. The 
C.P.B. contains several indications that psychological questions 
were reserved for the second Part of the Principles, which Ber- 
keley lost in Italy after making 'considerable progress in it*. On 
divine and human agency v. Hylas, p. 454. 



VI 
THE PRINCIPLES OF KNOWLEDGE 

BERKELEY'S denial of matter with its complement, the 
assertion of ideas, is his characteristic performance. 
Upon it he concentrated his attention in Part I of his 
Principles. But it is not the whole of his philosophy. There 
are for him other realities besides ideas, other objects of 
knowledge, and therefore other principles of human know- 
ledge. In connexion with these other objects he outlines his 
views on the Knowing of Knowing. 

Berkeleianism is popularly supposed to reduce everything 
to mere idea. Nothing could be further from fact. It is 
usually classified as idealism, and regarded indeed as the 
head and front of idealism. That classification is open to 
serious criticism . Fraser often speaks of Berkeley as a realist. 
It is arguable that his principles have more in common with 
our realism than with our idealism. But an adequate dis- 
cussion of the use of those labels would be a long matter, 
and perhaps not profitable or conclusive. Suffice it here to 
say that when Berkeley denied matter and refuted material- 
ism, he did not assert mentalism, nor put the human mind 
in matter's place. 

. It is not even correct to say that Berkeley lays all the stress 
upon ideas. He preserves his sense of balance. 1 Ideas form, 
for him, simply one hemisphere of things, and, even so, not 
the more important. His ideas are in their nature passive and 
inert. They do nothing, cause nothing, cannot even resemble 
an active being (Prin. 25). There is something else in the 
universe (Prin. 2). There is something entirely distinct 
from idea. There is what perceives ideas, wills, imagines, 

1 Cf. De Motu, 21 : 'Duo sunt summa rerum genera, corpus 
et anima*. 



THE PRINCIPLES OF KNOWLEDGE 93 

remembers them. There is what I call 'mind, spirit, soul, 
or myself. This rather ego-centric account passes soon 
(Prm. 27) into the account of spirit as 'one simple un- 
divided active being' whose two main operations are under- 
standing and willing. Spirit, by denotation, divides into the 
infinite Spirit and finite spirits, and in the later sections, 
more precisely, into God, myself, and other spirits. 

Thus the component parts of Berkeley's experience are 
familiar elements. His world is the world of common 
sense. So the interest for us in what he says about spirits 
is not so much his assertion of their existence as his meaning 
in calling them 'known*. Are spirits 'known' in the same 
sense as idea-things are 'known' ? The Principles is a trea- 
tise designed to express the principles of human knowledge. 
The first half of the book expounds immaterialism and 
defends it against objections. The second half fulfils the 
promise of the title by offering a co-ordinated statement of 
the wider principles of knowledge, dealing seriatim with 
knowledge of ideas r of the self, of other spirits, and of God. 
It may be noted that Berkeley in his Preface makes a point 
of inviting his readers to suspend judgement till they have 
read the whole through. The last section, that on the know- 
ledge of God, is the keystone of the fabric, and his doctrine 
of idea-things, if sundered from that section, is certainly 
one of the 'passages . . . liable to gross misinterpretation, and 
to be charged with most absurd consequences'. Before 
dealing in order with the four sections, which treat of the 
four objects of knowledge, we shall glance at the genesis of 
Berkeley's epistemology. 

'Certainly we do not know it. This will be plain if we 
examine what we mean by the word knowledge' (C.PJ3. 
582). Berkeley conducted this examination. He studied 
the nature of knowing, and tried to co-ordinate the varieties 
of knowledge that arise under his system of reality, making 



94 THE PRINCIPLES OF KNOWLEDGE 

critical use of the epistemological teaching of both Locke 
and Malebranche. 

The fourth book of Locke's Essay is an elaborate treatise 
upon the nature, modes, extent, and certainty of knowledge. 
It opens with the statement, 'since the mind, in all its 
thoughts and reasonings, hath no other immediate object 
but its own ideas. . . it is evident that our knowledge is only 
conversant about them. Knowledge then seems to me to 
be nothing but the perception of the connexion and agree- 
ment or disagreement and repugnancy of any of our ideas.' 
Locke goes on to consider agreement of ideas under four 
heads: (i) Identity or diversity, (2) Relation, (3) Co-exist- 
ence, (4) Real existence. Locke's controlling principle is 
clearly expressed in the words 'we can have knowledge no 
farther than we have ideas 1 (bk. iv. 3. i). The field of our 
knowledge can be no wider than the field of our ideas. 
Where there is no idea there is no knowledge. Upon this 
principle depends his account of the various modes and 
degrees of knowledge. 

Locke's view of knowledge is reflected in the first half of 
the Commonplace Book. When the second note-book is well 
advanced a break-away begins. In the formal demonstra- 
tion of the new principle more geometrico, with which 
Berkeley experiments at the end of the first note-book (903- 
24), the second on the list of axiomatic principles is, 'all 
knowledge about our ideas'. There the Lockian principle 
is endorsed. But by the time we reach the important series 
740-52 a revolution in outlook has occurred. Berkeley now 
writes 'we may have certainty and knowledge without ideas'. 
In the entry 752 he takes up Locke's four heads of agreement 
of ideas, tears three of them to pieces, establishes the fourth, 
pointing out, however, that Locke had misjudged its extent. 

Berkeley's acceptance of 'knowledge without ideas' was 
a turning-point. This meant a positive break with repre- 



THE PRINCIPLES OF KNOWLEDGE 95 
sentationism, and at least a partial assertion of direct 
awareness. Further, it released the term 'idea*. The term 
being no longer required for the tertium quid of indirect 
knowledge, can be used now for the supposed direct object 
of knowledge, usually called material thing. Now was 
Berkeley self-taught in this matter ? In asserting that we 
have knowledge without ideas, and in making that posi- 
tion the corner-stone of the second half of the Principles, 
he was abandoning Locke's first epistemological principle. 
In this grave departure had he any support and guidance 
besides his own 'arguings'? It seems probable that here 
again he is indebted to Malebranche. Berkeley must have 
read chapter vii in the second part of the third book of the 
Search. It immediately follows the exposition of our seeing 
all things in God. It is indeed a necessary supplement to the 
main theory. For the very notion of man-knowing-in-God 
obliges Malebranche to draw the line between material 
things which we know in God by representative idea, and 
other objects known, but not known in that way . Accordingly 
Malebranche marks off 'four manners or ways of knowing 
things', corresponding to the four objects, God, bodies, the 
self, other spirits. God, and God alone, is known 'by 
himself* with 'an immediate and direct view, without the 
Deputation or interposition of any creature*. Bodies and 
their properties are seen in God and by their ideas, and 
therefore 'the knowledge we have of them is most perfect*. 
The soul we know not by idea, nor in God. 'We know her 
only by conscience, and for that reason the knowledge we 
have of her is imperfect.* Other spirits, human or super- 
human, we know only by 'conjecture*. 

Malebranche *s quadrilateral of knowledge rests on the 
principle 'We see only those things in Him whereof we 
have ideas, and there are things we see without ideas* 
(Search, iii. 2. 7). That passage may have been Berkeley's 



96 THE PRINCIPLES OF KNOWLEDGE 

authority for his 'knowledge without ideas'. Allowance 
made for his peculiar use of the term idea, the head-line 
of the second half of the Principles ( 86) 'Human knowledge 
may naturally be reduced to two heads, that of ideas and 
that of spirits' seems to be an echo of Malebranche's words 
'All things in the world whereof we have any knowledge 
are either bodies or spirits.' His immaterialism compelled 
Berkeley to modify the details of Malebranche's epistem- 
ology, but the Search, I think, suggested the general plan. 

Knowledge of Ideas (Principles, 86-134) 

The second half of the Principles was, it would seem, con- 
ceived separately. It was drafted in a separate note-book. 1 
It assumes the results of the former half, but it can stand as 
an independent study of the fourfold object of knowledge. 

The first division ( 86-134) examines knowledge of 
ideas. Sections 86-91 repeat previous arguments. They 
form a strong assertion of the reality of knowledge, and an 
exposure of the sceptical alternative . Both Malebranche and 
Locke are included among those who have given a sceptical 
turn to the theory of knowledge . In 87 he has Malebranche 
in view. Using Malebranche's phrase 2 he speaks of the 
sensible qualities as 'perfectly known', assigning in effect 
the same reason 'there being nothing in them which is 
not perceived'. He then outlines the arguments of proble- 
matic idealism as expounded in the Search, bk. i. In 88 
he passes to Locke's 'assurance that deserves the name of 
knowledge'. In the draft copy 88 Berkeley first wrote 
1 intuitive or demonstrative knowledge of the existence of 
sensible things'. This fixes the reference of the passage to 
Locke's 'sensitive knowledge '(Essay, iv, 2. 14). Against such 
problematical knowledge Berkeley asserts that we know one 

1 British Museum Add. MS. 39304. 

2 i.e. for the intelligible ideas of bodies. 



THE PRINCIPLES OF KNOWLEDGE 97 

hemisphere of reality in apprehending ( 89) 'inert, fleeting, 
perishable passions, or dependent beings ; which subsist not 
by themselves, but are supported by, or exist in, minds or 
spiritual substances*. 

After some recapitulation and a brief discussion of time 
Berkeley passes from knowledge to science, that is, from 
unsystematic knowings to the two 'great branches of specu- 
lative knowledge', and he outlines an epistemology of 
physics and mathematics. He maintains ( 101-17) that 
physics as knowledge does not differ in kind from sense-per- 
ception. Science has no claim to an exclusive knowledge of 
reality. If she makes the claim, scepticism ensues. Science 
has not even an exacter knowledge of efficient causes. The 
excellence of science resides in the scientist's largeness of 
comprehension, from which springs the * endeavour towards 
Omniscience' ( 105), with its attendant dangers. In 1 18- 
34 he considers pure mathematics, discussing the genesis of 
both arithmetic and geometry. Their subject-matter too 
is grounded in sense-perception. Intelligible number and 
pure intelligible space are empty notions arising from the 
false doctrine of abstraction. 

This whole passage ( 85-134) is tantalizing. Berkeley 
says in it startling things about ideas and yet leaves much 
unsaid. We feel that he ought to have attempted a more 
systematic answer here to the big epistemological ques- 
tions which arise out of his statements about ideas. Very 
possibly he shows a calculated restraint in this matter, and 
perhaps he had a good reason for not disclosing at this 
stage his full doctrine of knowledge. 

Knowledge of ideas is, for Berkeley, first and last, sense- 
perception. It differs from sense-perception, as ordinarily 
understood, in omitting the external or, more precisely, the 
material object. 1 The psychology of the imagination and 

1 * Externality' with regard to origin is granted in Prin. 90 

4046 o 



98 THE PRINCIPLES OF KNOWLEDGE 

memory is not discussed in the Principles, but from the 
Commonplace Book it is quite clear that Berkeley held that 
the image arose from the percept. 1 Even the two great 
provinces of speculative science, natural philosophy and 
mathematics, are 'conversant about ideas received from 
sense and their relations' ( 101). The idea, then, for Ber- 
keley is the foundation of the whole fabric of human know- 
ledge, and we look to him to tell us the nature of the idea, 
and how it is related to the percipient. 2 The idea, as the 
name implies, has no independent nature. Relation to spirit 
makes ideas what they are. That relation is percipience or 
knowing, which is an ultimate, unanalysable fact, vouched 
for by every man's experience. Ideas from their birth to 
the moment of their annihilation are known. Not occu- 
pancy of space-time, but occupancy of mind is their esse. 
Yet ideas are not spirit, nor are spirits ideas. Spirit and 
idea have it in common to be immaterial, if you like, but 
that is saying little or nothing, if matter non est. My ideas, 
qua mine, are made by me, and held in brief existence by 
me; but they are not the me, nor part of it. Ideas are 
neither modes of spirit, nor states of mind. 'In' mind, they 
are not 'of mind. They are essentially unlike spirit. Spirit 
can act, ideas cannot. The Berkeleian idea is object which 
is not subject, and which, being made and conserved by 
spirit, has no independent nature. 

of things perceived by sense, i.e. of all ideas except those of the 
imagination and memory. 

1 The following en tries are explicit: 536,588,667,804,830,835. 

2 If he has not told us systematically, he has put us in the way 
of finding out. Twenty years later in a letter to Johnson he 
admits 'defects' in his early publications. He still thinks their 
teaching true, but does not pretend that his * books can teach 
truth. All I hope for is, that they may be an occasion to inquisi- 
tive men of discovering truth, by consulting their own minds, 
and looking into their own thoughts* (Works, vol. ii, p. 18). 



THE PRINCIPLES OF KNOWLEDGE 99 
Further, the idea imprinted on the sense is real, and a 
criterion of reality. The idea of the 'great square tower' on 
the bend of the River Nore, Berkeley's reputed birth-place, 
seen from the Thomastown road, is small, round, and can 
be covered with a pencil. But it is a datum. In some sense 
it is there. The visual idea is the visual datum. It is what we 
see. As seen it is perfectly apprehended. There is nothing 
in it not seen. The idea has no inside and outside, no back 
and front. We can walk away or round, and see adjacent data. 
But once we select an object for vision, what we see is all 
there is to see. Malebranche says that we should like fuller 
knowledge of ourselves and of other spirits ; but that we 
cannot desire knowledge, fuller than we have, of the repre- 
sentative idea of body. Being in God, it is perfect in its kind. 
Now Berkeley's idea is presentative. It is not, like Male- 
branche's idea, 'different from' the thing (Search, iii. 2. 7). 
But when he calls it 'perfectly known' 1 he means by the 
phrase just what Malebranche meant. 

There is no occult universality in the idea, any more than 
there is occult materiality. The idea is just what it pretends 
to be, a particular. 2 This idea is not that idea. The visual 
idea that I just cover with my pencil is not the visual idea 
that I just cover with my book. The idea imprinted on the 
senses is a real thing that really exists. So it is a particular. 
Since, however, one particular idea can stand for another 
particular idea, the idea is a proper and adequate basis for 
general knowledge. The mind is liable to be in too great a 
hurry to form general theorems ; but if we check this 'endea- 
vour towards omniscience', we may attain 'the greater large- 
ness of comprehension whereby analogies, harmonies, and 
agreements are discovered in the works of nature and the 
particular effects explained, that is, reduced to general rules' 

1 v. Prin. 87, 10 1, and elsewhere. 

2 Cf. Intr. Prin. 15 and Hylas, Works, p. 403. 



ioo THE PRINCIPLES OF KNOWLEDGE 

(Prin. 105). This symbolic universality is due to the will 
of the author of nature ; things are so made by him that we 
read general rules in particular phenomena, as surely as we 
read meaning in letters of a book (Prin. 108). 

Ideas then are the unthinking things that compose the 
world. Excited by man (36), excited by God ( 57), they 
affe ne&er excited by 'matter'. The conception of unthinking 
things that are always thought is undoubtedly difficult, but 
it is neither arbitrary, absurd, nor alien. It is no harder than 
rival conceptions, e.g. a subject that makes itself its object, 
or the too familiar notion of an external thing 'acting on' 
our minds or brains. 

Berkeley is his own best expositor. But perhaps this 
amount of exegesis may be permitted as a background to a 
brief reference to certain 'defects' in his treatment of ideas. 
The four difficulties commonly felt about his ideology are 
that the ideas are incoherent, private, impotent, and flimsy. 
The opposite features in matter attract us. This is the source 
of the dreamlike character usually attributed to the Berke- 
leian world. As regards the incoherence of the ideas, 
Berkeley just says that they are 'combined, blended, or 
concreted together' ( 99) to form objects of sense; but he 
gives no adequate treatment of the nature of body. His 
silence is marked, because we know from the Commonplace 
Book (290, 302-3, &c.) that he had considered the question. 
The difficulty of the idea's privacy is scarcely even ap- 
proached in the Principles. Yet if my idea is only mine, how 
can two people see the same thing ? The third Hylas dia- 
logue (Works, vol. i, pp. 466 sqq.) to some extent repairs the 
omission. As regards the impotence of the idea, the fact is 
stressed often, and made the basis of an argument for the 
existence of God ; but no adequate explanation is advanced 
of the appearance of causality between things (v. 51-60). 
Locke had said, 'pure matter is only passive' (ii. 23. 28), 



THE PRINCIPLES OF KNOWLEDGE 101 

but to all intents and purposes he treats matter as able to 
act, e.g. ii. 8. 10. Berkeley, one might almost say, tacitly 
transfers this part of the occasionalist teaching, viz. the im- 
potence of matter, into his own system, mutatis mutandis. 
However, the permanence of ideas is the fundamental and 
primary issue, for permanence involvessubstance. Theother 
characters of the Berkeleian idea are dependent. Berkeley 
leaves us in no doubt that God is the substance and ground 
of thepermanence of ideas, but he develops the notion hardly 
at all. All the other characters of the Berkeleian idea turn on 
his conception of the nature of deity. Because God is, ideas 
form bodies round which socialized human knowledge and 
joint action can revolve. Because God wills, ideas are 
changed by the divine operation. Now had Berkeley at this 
stage attempted the explanation of these problems he would 
have been speaking of deity in a language incomprehensible 
to most of his readers, and one likely to prejudice the accep- 
tance of the immaterialist argument, which, for Berkeley, 
is the 'principle or main point, from which, and from what 
I had laid down about abstract ideas, much may be deduced* 
(letter to Johnson, Works, vol. ii, p. 20). In fact Berkeley 
could not fully explain the knowledge of ideas until his 
view of the other three types of knowledge was familiar 
and accepted. 

Knowledge of the Self 

Berkeley had trouble with this question. Its intrinsic 
difficulty, disclosed in the conflicting views of Locke and 
Malebranche, was heightened for him by his peculiar doc- 
trine of ideas. Berkeley states his views on self-knowledge 
in the Principles, I35~44. r From the concluding words 

1 Fraser's text mixes the editions and is confusing. As we 
are studying origins, the first edition, 1710, will be our main 
topic. 



loz THE PRINCIPLES OF KNOWLEDGE 

of that section, and from the large number of neglected *S' 
entries in the Commonplace Book, we may be sure that he 
intended to carry the discussion further in the lost Part II. 
The dominant theme of this section is the bifurcation of 
knowledge. 'Our souls are not to be known in the same 
manner as senseless inactive objects, or by way of idea' 
( 142). A man sees and touches parts of his body. Those 
objects being, for Berkeley, ideas, and the 'material' body 
being non-existent, his question shapes itself as follows, 
'Can I, who know the inactive idea, also and in the same 
,sense of the word, know the active spirit?' His answer is 
an unwavering affirmative, coupled with the proviso that 
the fact that both are knowable does not mean that they are 
in any way alike. Berkeley here is not trying to apprehend 
a higher faculty resident in a material body, but a self- 
conscious centre of real finite activity. 

The novelty to many of his English readers would be 
Berkeley's refusal to recognize the term idea in connexion 
with our knowledge of the soul. In this refusal we may 
trace the influence of Malebranche. For Locke, our know- 
ledge of the soul is intuitive and of the highest degree of 
certainty, but, like all Lockian knowledge, it is knowledge 
by idea. Our idea of the soul is, for Locke, a complex idea 
of an immaterial spirit. The substance of spirit baffles us, 
he holds, but the primary ideas of spirit, thinking, willing, 
and motivity are as clear as our ideas of body (Essay, ii. 23. 
15-26). 

Malebranche 's doctrine is more subtle. Like Locke, he 
admits that there is an idea of the soul ; otherwise God would 
not have known what He was about to do when He created 
the soul. Unlike Locke, he denies that we know the soul by 
its idea. He says that God has not given us an idea of our 
own soul, so as not to weaken the union of our soul with our 
body. We know the soul ; otherwise we could not prove its 



THE PRINCIPLES OF KNOWLEDGE 103 

immateriality, immortality, and liberty ; but this knowledge, 
in fullness and clarity, falls far short of Descartes's 'first 
knowledge*. In point of fact, Malebranche thinks, we do not 
know the soul as well as we know body. We can define and 
know all the properties of extension ; x but we do not know 
all the modifications of the soul nor what she is and is not 
capable of. From this comparative ignorance arises our 
difficulty in admitting colour and light to be modifications 
of the soul. If we knew the soul by idea, we should see it 
in God and should be as gods which may not be. He gives 
various names such as sensation, sentiment inttrieur, and 
conscience, to the rush-light self-knowledge that we have 
(Search, iii. 2. 7 : tis possible that what we know of her is 
the least part of what she is'). 

There is a good deal here that Berkeley accepted or 
agreed with and a good deal that he rejected. Agreeing with 
Malebranche against Locke that we do not know the soul 
by idea, he went a stage further and said that there is no 
idea of the soul to know. In the Commonplace Book (900) 
he refers to Malebranche 's distinction between knowing the 
mind by idea and 'sensation conscientia'. He calls it 'vain'. 
But the entry is cryptic, and the reading doubtful. Berkeley 
does not characterize self-knowledge with a technical name 
in the Principles ; but as with Malebranche, it rests upon 
immediate feeling. In the Hylas dialogue (Works, vol. i, 
p. 447) he says : 'I know what I mean by the terms / and 
myself \ ari3 ITm6w~tKIs Immediately or intuitively/ Again 
(if>.) 1t my own mind and my own ideas I have an immediate 
knowledge of, and in the same context self-knowledge isl 
described as 'reflex act*. In the De Motu (1721), 21^ 
Berkeley says explicitly 'rem sentientem, percipientemi 
intelligentem, conscientia quadam interna cognovimus'. In 
the addition, made in 1734, to the Principles, 89, the term 
1 Cf. Prin. 142 'know a spirit as we do a triangle*. 



104 THE PRINCIPLES OF KNOWLEDGE 

'inward feeling' is found. So it is clear that Berkeley re- 
garded our self-knowledge, as Malebranche did, as partial 
and imperfect but immediate and quite real. Berkeley is 
tactically in a difficult position. He is fighting on two fronts. 
His first concern is to prevent self-knowledge from being 
a support for materialism. He is answering the argument 
latent in Locke's words, 'the substance of spirit is unknown 
to us ; and so is the substance of body equally unknown to 
us' (Essay, ii. 23. 30). The obscure in self-knowledge im- 
perils his consistency. 1 He has to meet the objection that 
there may be matter, though we are only dimly aware of it ; 
for there is a soul, though we have no idea of it. In order to 
repulse the threat from materialism, Berkeley has to estab- 
lish the radical difference between spirit and idea. In so 
doing he exposes his flank to the attack of scepticism, and 
repels it by maintaining that the soul is really known, though 
it is not known after the manner of ideas. 2 

Commentators usually make a great deal of the altera- 
tions in the later editions and of the stress there laid on the 
term 'notion'. It is doubtful whether the alterations repre- 
sent any change of view. Berkeley had long held that self- 
knowledge and knowledge of ideas differ in kind, and it 
seems improbable that he could ever have come to regard 
notional knowledge as a third kind. The word 'notion' had 
been used in the schools, and Berkeley would meet it in 
Malebranche, Locke, Sargent, and many other writers, and 
he uses it in a simple innocent sense 3 in the first editions of 
the Principles and the Hylas. A notion was not, I believe, 

1 'To act consistently you must either admit matter or reject 
spirit.' Hylas, Works, vol. i, p. 449, added 1734. 

2 Reid, rather superficially (Inquiry, c. vii), urges the objec- 
tion which Berkeley had anticipated. 

3 As e.g. Locke, Essay, ii, 23. 35, 'the idea or notion we have 
of God'. 



THE PRINCIPLES OF KNOWLEDGE 105 
for Berkeley, an epistemological tertium quid. In the third 
Hylas dialogue (all editions, Works, vol. i, p. 448) it is 
explained that * having a notion' of God means knowing Him 
by 'reflexion and reasoning*. In all the editions of all his 
books 1 Berkeley's doctrine of 'notions' is capable of being 
interpreted within the scheme of knowledge laid down by 
him in 1710, and we should not give to it a turn inconsis- 
tent with 'The Principles of Human Knowledge'. 

The term 'notion' is not a new departure, but an addi- 
tional safeguard against sceptical misinterpretations of 
Berkeley's doctrine of knowledge of spiritual objects. The 
critic could, and, no doubt, did say 'the bishop denies that 
we have an idea of God and the soul'. Berkeley supplies the 
retort, 'True, but he affirms that we have a notion of God 
and the soul'. He knew from personal experience the need 
of such a term. From the Commonplace Book we see that 
if he had not plumbed in person the depths of doubt, at 
least he had been to the brink of the precipice, and had 
looked over. In the early entries (25, 44) self-knowledge 
is more or less certain and complete. He seems to accept 
(552 corrected reading) by implication Locke's doctrine of 
intuitive self-knowledge. The soul is a complex idea, known 
and defined (156). However, the series 581 to 588 contains 
an important revelation of a change of view. He asserts that 
it is mere prejudice to deny that we have an idea of the soul ; 
but, as he immediately (583-7) gives a tentative account of 
the soul, indistinguishable from Hume's, it would appear 
that for the time being he had given up the fact while 
retaining the name 'soul'. Later on, he revised that page, 
adding 582 on the verso and marking with the marginal sign 
+ (? the obelus) the entries 583, 585, 586, 587. In 751 he 
expresses the view that the Cogito ergo sum is a tautology, 

1 Even, I think, the 'such are notions 1 of Stris, 308. v. infra, 
p. 176. 

4046 p 



io6 THE PRINCIPLES OF KNOWLEDGE 

and in the next entry he records the devastation in the 
Lockian theory of knowledge caused by the New Principle. 
His earlier uncertainty is thus in marked contrast with the 
certainty eventually reached. 1 His fear of scepticism is not 
feigned. So he concedes, as Malebranche had conceded, 
that 'in a large sense we may be said to have an idea of 
spirit 1 . 2 Subsequently he formally expresses this conces- 
sion by using the term 'notion*. 

Malebranche had insisted that, however imperfect our 
knowledge of the soul, it is yet sufficient to demonstrate her 
immortality, spirituality, and liberty. The same three items 
are dealt with by Berkeley. In 141 he declares that the 
soul is indivisible, incorporeal, unextended, and therefore 
incorruptible and naturally immortal. In 144 he hints at 
a projected treatment of liberty, which he intended to give 
in Principles, Part II (v. C.P.B. 511). Berkeley's firm faith 
in God and life would not permit any permanent scepticism 
about self-knowledge. In his view the soul must be real and 
really known, because God, who encompasseth us, made 
and knows it. 

Knowledge of other Finite Spirits 

Knowledge of other finite spirits is recognized by Male- 
branche, Locke, and Berkeley as a special branch of epistem- 
ology. No one of the three, however, treats it with fullness, 
though Malebranche and Berkeley are far from neglecting 
it. Berkeley has been severely criticized on this score. Reid 
wrote (Int. Powers, Essay II, c. x): 'I can find no principle 
in Berkeley's system which affords me even probable ground 
to conclude that there are other intelligent beings like 

1 As e.g. in Hylas, Works, vol. i, p. 446, *I might as well doubt 
of my own being*. 

2 Prin. 140. Cf . Search, Illustration on bk. i, c. 3 : * I have some- 
times said we had an Idea of the Soul, and sometimes denied it.' 



THE PRINCIPLES OF KNOWLEDGE 107 

myself, and he says, unjustly I think, that Berkeley 'seems 
not to have attended to' this difficulty. Dr. Dawes Hicks 
(Berkeley, pp. 147-50) appears to find Berkeley's treatment 
of the question highly unsatisfactory. The charge he makes 
seems to be one of inconsistency rather than, as with Reid, 
of neglect. I have not been able to see the full force of either 
charge, while admitting that the form in which Berkeley 
has stated his doctrine leaves a good deal to be desired. 

Berkeley was no solipsist. Few philosophers have had 
wider and more fruitful contacts with their fellow men than 
he had. It is hardly likely that his thought could belie his 
action. Might it not be the case that an interpretation of 
Berkeleianism that tends towards solipsism is neglecting 
the Malebranchian strain in his teaching ? Just before the 
sentence quoted above, Reid says, 'What I call a father, a 
brother, or a friend, is only a parcel of ideas in my own mind'. 
That is a parody of Berkeley's teaching, owing all its point 
to the Lockian interpretation of 'idea' which Reid habitually 
places on the Berkeleian 'idea'. To reach a coherent and 
sensible account of Berkeley's teaching about the 'Thou', 
we must piece together several passages, and to appreciate 
it we must read it in connexion with his account of our 
knowledge of God. 

We shall first glance at Malebranche's teaching (Search, 
iii. 2. 7). Other spirits are known, says Malebranche, by 
'conjecture'. After a visit to an exhibition of wax- works we 
are inclined to agree with him. Certainly we do not ordi- 
narily feel the existence of other spirits, in the same manner 
and with the same directness as we feel our own existence. 
Yet the term 'conjecture' does less than justice to Male- 
branche's view. He was no sceptic. His attitude towards 
'other spirits' is like his attitude to matter. He is certain of 
the existence, but he doubts the evidence. Conjecture to us 
means little more than 'not knowing'; to Malebranche the 



io8 THE PRINCIPLES OF KNOWLEDGE 

term meant analogy. He says we know by conjecture when 
we think some things (different from ourselves, from those 
we know in themselves, and from those we know by ideas) 
'are like some others that we already know'. He grants that 
conjecture is unreliable where the physical factor is promi- 
nent ; but he thinks that we reach virtual certainty in the 
spheres of law and righteousness. The truths that twice two 
make four, that it is better to be just than rich, that God 
acts equally on all spirits, are social principles, postulating 
a society of men. What 'thou' art thinking and feeling on 
any given occasion must be conjectural, being based for me 
on the precarious ground of analogy ; but God is 'the place 
of spirits', and so the existence of other spirits is, for Male- 
branche, a postulate of piety and a dictate of the moral sense. 
(Locke's few remarks on this topic are contained in Essay, 
bk. iv, c. 3. 17, and c. n. 9-12.) 

Knowledge of other spirits is one of the very few topics 
of the Principles not discussed in the Commonplace Book. 
From the entry (406) 'God knows how far our knowledge 
of Intellectual beings may be enlarged from the Principle', 
we may gather that a speculative development of the topic 
had crossed Berkeley's mind, but there are, I think, no 
entries dealing with the problem nearer home. We cannot 
then study his view 'in the making'. As we find it in the 
Principles, his account of our knowledge of other spirits, 
that is, of the spiritual or mental character of the men we 
meet, is like that of Malebranche. Both thinkers mark it off 
carefully from knowledge of ideas and from knowledge of 
God and the self. Both regard it as a complex process, and 
take similitude as its objective basis. Berkeley intended 
145 to be his formal statement. But that curt, compressed 
section should be taken along with the relevant passages 
from the flanking sections on self-knowledge and knowledge 
of God. God, the self, and other selves have it in common 



THE PRINCIPLES OF KNOWLEDGE 109 

to be spirits. So Berkeley cannot neatly isolate and treat 
separately these connected types of knowing. The relevant 
passages in the Principles are 140, 145, 147, 148, and with 
them should be read Hylas, Works, vol. i, pp. 448-9, and 
Alciphron, iv. 5. 

Knowledge of other spirits is distinguished by Berkeley 
both from knowledge of ideas and from knowledge of the 
self. Ideas are known immediately. The self is known 
immediately. Other selves are known mediately. This 
mediate knowledge is none the less knowledge ; for it is the 
product of two immediacies. Self and my ideas are both 
known immediately, and both these elements enter into 
our mediate knowledge of the 'Thou'. 'We know other 
spirits by means of our own soul ... it having a like respect 
to other spirits that blueness or heat by me perceived has 
to those ideas perceived by another' ( 140). Again ( 148) 
certain ideas 'serve to mark out unto us the existence of 
finite and created spirits like ourselves.' In Hylas (p. 449) 
the process is referred to as 'mediate apprehension by 
similitude of the one or the other' (i.e. idea or self). In the 
second edition of Prin. 89, Berkeley introduces the term 
'reason'. The new term is not a new departure. It simply 
means that we must take steps to know the existence of 
other minds. The objective basis of the transition con- 
stituting the act of 'reason' is, to use Malebranche's term, 
'likeness'. The blueness I see is to the blueness thou seest 
as I to Thou. Hearing thy voice is like, but not the same* 
as, hearing my own voice. Resemblance and difference 
combine to make me refer the idea 'thy voice' to an agent 
like myself; in fact to thee. 

The manuscript draft of the Principles breaks off in the 
middle of 145, as if perhaps Berkeley stopped to devote 
special care to the section. Fraser notes that 'this is one of 
the notable sections in the Principles, as it suggests the 



no THE PRINCIPLES OF KNOWLEDGE 

rationale of Berkeley's rejection of Panegoism or Solipsism*. 
Berkeley's world is peopled with plural spiritual existences. 
He never shows the slightest doubt of the existence of other 
minds. His assured certainty on this point is not only con- 
sistent with his immaterialism but explanatory of it, and 
perhaps even expository of it. Repeatedly (e.g. 34) he 
insists that the New Principle sacrifices no item of reality, 
sensible or conceptual. To make this evident, Berkeley has 
to pass on from knowledge of ideas to knowledge of finite 
spirits, and thence to knowledge of God. There must be 
choristers in the choir of heaven ; real men are part of the 
furniture of the earth. But the idea-things composing 
heaven and earth are not private sensa, any more than they 
are only atoms or merely electrons. The idea- things are 
mental, only if ' mental' means inter-subjective and inter- 
spiritual. My thinking * makes it so' to a minute extent and 
for a moment of time ; but our thinking (thine and mine) 
finds, not makes, a vast, stable system of idea-things, that 
could not be unwilled, unwanted, unknown by God. 

Hence it is that the knowledge of other spirits ( 145) 
leads straight on to the knowledge of God ( 146). Berkeley 
here restates, almost in terms, the well-known argument of 
29 for the existence of God, with the significant changes 
of I to the first person plural, and of 'my will' to 'the wills 
of men'. Then he passes at once to the language and the 
thought and, in a measure, to the system of Malebranche. 
'Malebranche had said 'everything proves God'. Berkeley 
here says ( 147) in effect, * everything that proves man 
proves God'. Malebranche calls God 'the place of spirits'. 
Berkeley says ( 147), 'He alone it is who "upholding all 
things by the word of his power" maintains that intercourse 
between spirits whereby they are able to perceive the exist- 
ence of each other.' There is the key of Berkeley's solu- 
tion of the problem of the Thou. The world of ideas or 



THE PRINCIPLES OF KNOWLEDGE in 

unthinking things would drop to pieces if its substance 
were removed. Just so, in Berkeley's system the social and 
spiritual order would be shattered, if God were not. 

Knowledge of God 

How God is known and what He is known to be are, no 
doubt, distinct questions; but Berkeley does not keep them 
altogether separate, though the epistemology of the subject 
is, in the main, reserved for the concluding sections of the 
Principles. 

The Principles contains over twenty divine titles. Of the 
156 sections about 70 mention God or consider the point 
at issue in relation to God. The cause of this stress on 
deity is not only the piety of the man, but the character 
of the system. Berkeley's philosophy apart from Berkeley's 
God falls. God in Berkeley's system is no deus ex machina, 
introduced as an afterthought or conventionally connected 
with the new way of ideas. He is its source and mainspring 
and secret strength. Each of the other three types of know- 
ing, as we have just seen, involves a reference to the highest 
object of human knowledge. To study Berkeley's con- 
ception of deity is to go to the heart of the system. 

Early training and education would dispose Berkeley to 
take the Lockian view. Locke's doctrine of deity, for all its 
deistic tendencies, is in itself orthodox, endorsing as it 
does the traditional and rational attributes. Yet the Com- 
monplace Book shows us Berkeley profoundly dissatisfied 
with Locke's doctrine. 'God Space', followed by two refer- 
ences to Locke's Essay y is one of the very early entries. The 
danger of 'making God extended' is referred to in 299, 321 , 
and 936 (v. also 634). In 308 Locke, More, and Raphson 
are named as seeming to make God extended. In 707 he 
mentions amongst Locke's dangerous opinions 'the infinity 
and eternity of space and the possibility of matter 's thinking* . 



ii2 THE PRINCIPLES OF KNOWLEDGE 

We know that Berkeley studied King's De Origine Mali 
(v. C. P. B. 161-8) and King's characteristic title for Deity 
'Active Principle' occurs in the Principles ( 66). Berkeley 
must have known Clarke's writings, and the statement 
(Prin. 63) that God prefers to convince our reason 
of his attributes by the works of nature rather than by 
miracle is reminiscent of Clarke's Boyle Lectures on the 
Attributes. 

Traditional theism and transcendence are reflected in 
such titles as the Governing Spirit, the Supreme Agent, the 
Author of Nature titles that abound in Berkeley's writ- 
ings. But those titles are not sufficient for the needs of the 
New Principle. Immaterialism placed new stress upon 
another aspect. A Governing Spirit might conjure up 
ideas, and impose them upon men as realities. But Ber- 
keley's ideas are no fancies. The permanent ideas whose 
esse is percipi are grounded in Deity, but could scarcely be 
related in any satisfactory way to the Deity as conceived 
by Locke and by contemporary British theology. It seems 
that French philosophy supplied the decisive conception. 

We referred above, pp. 68-9, to the Commonplace Book 
entry, 'I wonder not at my sagacity in discovering the ob- 
vious tho' amazing truth . . . ' (287). One feels the half- 
suppressed exultation of the entry. It was a moment of 
enlightenment. The scales fell from his eyes. But what was 
his discovery precisely ? What was 'the obvious tho' amaz- 
ing truth' ? It could hardly have been the denial of matter. 
That principle with its corollary, the ideality of sensibles, 
he had entertained for some years as an hypothesis. Surely 
his discovery was a sudden perception of the main condition 
that made immaterialism possible, that verified the hypo- 
thesis ? If he saw in a flash the Esse Percipi against the 
background of the Immensity of God, his 'stupid inadver- 
tency* would be his previous blindness to that aspect of 



THE PRINCIPLES OF KNOWLEDGE 113 

Deity which is the sole possible basis of the New Principle. 
Now it is worth noting that the pages of the note-book pre- 
ceding this declaration show marked traces of the influence 
of Malebranche. The entry in question is 287. The entries 
239, 264, 266, 274, 278, and 296 name Malebranche and, 
probably, other entries in these pages refer to his writings. 

Locke (Essay, iv, 10. i) had called the proof of God's 
existence 'most obvious', but his laboured demonstration 
makes it seem anything but obvious. We meet the phrase 
'obvious truth* used of Deity both in the prologue and at 
the climax of the Principles ( 149). An examination of 
these two passages, the alpha and the omega, makes it pro- 
bable that the Immensity of God was Berkeley 's master-light , 
and that he thought of it in connexion with Malebranche. 
In 6, Berkeley declares, in lofty language, that God is the 
substance of all the choir of heaven and of the furniture of 
the earth, and that this truth is 'near and obvious to the 
mind*. That passage states with restraint and in untechnical 
terms Berkeley's primary vision of all things in God. In 
the second Hylas dialogue (Works, pp. 422-4) the same 
vision is painted on a larger canvas. Berkeley gives the rein 
to his fancy there, and describes the high arch of heaven, 
the abyss of space with all its 'glittering furniture', and 'the 
vast bodies that compose this mighty frame', and strikes 
the formal inference 'there must be some other mind where- 
in they exist'. The difference between this doctrine and 
the usual method of proof is explained, and at once (p. 426) 
Hylas asks the question: 'But do you not think it looks 
very like a notion entertained by some eminent moderns 
of "seeing all things in God" ?' 

At the close of the Principles ( 149) Berkeley rises to his 
climax, and there we find the same sense of discovery, 
similar language, and the same 'great truth which lies so 
near and obvious to the mind* and yet is 'attained to by 

4046 Q 



ii4 THE PRINCIPLES OF KNOWLEDGE 

the reason of so very few'. 1 That truth is, 'God ... is 
intimately present to our minds, producing in them all that 
variety of ideas or sensations which continually affect us, on 
whom we have an absolute and entire dependence, in short 
"in whom we live, and move, and have our being" '. Here 
Locke is left far behind, and the setting of the passage and 
several of its phrases are taken, apparently, from the Search. 
God as the home of 'our' ideas, God as intimately present 
to our minds, God as the Being upon whom we have an 
absolute and entire dependence, God in whom we live and 
move and have our being in this divine quadrilateral the 
theory of seeing all things in God is framed. 

For Malebranche, the knowledge of God differs intrinsic- 
ally from the other types of knowing. It is knowledge of 
* things by themselves'. It is immediate knowledge, know- 
ledge without ideas. As far as we know, God is the only 
object so known. An object must be of a most intelligible 
nature if it is to penetrate the mind and discover itself to 
the mind. None but God penetrates the mind, and dis- 
covers Himself to it. "Tis God alone that we see with an 
immediate and direct view.' What makes this possible is 
the union that our intellects have with Him. He is the 
'Master who presides over our minds without the deputa- 
tion or interposition of any creature'. A representative idea 
of God then, for Malebranche, would be impossible and 
unnecessary, impossible because an idea is a created being 
and cannot represent the universal Being, unnecessary 
because 'God is intelligible by Himself. This is Male- 
branche 's true opinion, strictly expressed. 2 In Book iv, 
c. n, however, he seems to express an inconsistent view, 
asserting that the idea of God is no fiction. He is here using, 

1 With the *sad instance of the stupidity and inattention of 
men', cf. 'my stupid inadvertency* C.P.B. 287. 

2 Search, iii. 2. 7. 



THE PRINCIPLES OF KNOWLEDGE 115 

as he says, a 'personal not a universal argument*. In one 
sense we must have an idea of God ; otherwise we could not 
answer the question 'Is God round or square ?' The argu- 
ment is ad hominem. Malebranche is defending the ontolo- 
gical argument in its Cartesian form, rather than expressing 
his own proof of the existence of God. 

Locke's account of Deity is of a different strain. Male- 
branche starts from God and works down to human know- 
ing. Malebranche J s God discovers himself to us. Locke 
discovers God (Essay , iv, c. 10. i, 12) by working upwards 
from man's ideas. First (bk. ii, 23, 33-5) there is the idea 
of God that we make. That idea is complex, though God 
Himself be simple and uncompounded. Then (bk. iv, c. 10) 
we come to the real existence of that idea. This is known 
by demonstration. It is the 'most obvious truth that reason 
discovers'. We have no innate idea of God; yet we have 
a clear proof of Him, as long as we 'carry ourselves about 
with us'. The evidence amounts to mathematical certainty. 
Man knows that he himself is. Since nothing cannot pro- 
duce a being, there must have been Something from etern- 
ity. I have power and I know. Therefore that Something 
must be powerful and knowing. We have a more certain 
knowledge of God than we have of anything not immediately 
discovered by the senses. Of the ontological argument 
Locke says that it is not the only argument, and that it is a 
mistake to concentrate attention upon it to the exclusion of 
those proofs 'which our own existence and the sensible parts 
of the universe offer so clearly and cogently to our thoughts'. 

Now Berkeley's theology goes its own way. While he 
learned from both authorities, he agreed fully with neither. 
His affinities of spirit on the whole were with Malebranche. 
Locke's cold recognition of the province of faith would 
scarcely attract one who was warmly attached to Christian 
mysteries. In passing we may note that Malebranche 



n6 THE PRINCIPLES OF KNOWLEDGE 

habitually speaks of revealed religion and the mysteries of 
faith, as things apart, requiring the submission of the intel- 
lect, just as Berkeley does. 1 In spite of differences of creed 
the conservative Protestant and the liberal Catholic shared 
many beliefs of heart and head. 

The unto mystica between man's 'pure intellect' and the 
Word of God is a hinge of Malebranche's system. Berkeley 
does not recognize the unio as such. He had not Male- 
branche's conception of the soul and body relation, and he 
was critical of the 'pure intellect'. 'Pure intellect I under- 
stand not', he wrote in his note-book (822). He never 
admits 'pure intellect' as a faculty of abstraction, but, I 
think, he never denied 'pure intellect' as a faculty of the 
supersensible. In the first Hylas dialogue (Works, p. 404) 
he recognizes 'pure intellect and its spiritual objects, as 
virtue, reason, God'. Berkeley is careful not to merge 
the finite spirit in the infinite, but there is nothing in his 
outlook formally inconsistent with the unio. Indeed the 
unio may be said to appear in Berkeleianism in the form of 
God's intimate presence. The 'Intimate Presence' is pre- 
dicated in the Principles not alone of Nature (151), but of 
our minds ( 149), and, of these, not one-sidedly, for ( 155) 
'He is present and conscious to our innermost thoughts'. 

The formation of Berkeley's natural theology can be, to 
sorpe extent, traced in the Commonplace Book. A consider- 
able group of entries has the marginal sign 'G' (= God). 
Quite early (41) he intends a 'brief demonstration of an 
active powerful being'. The characteristic proof of the 
Principles makes its appearance in 1 1 1 , 724 and 850. In 1 80 
he is pondering the difficulties in the Lockian 'complex and 

1 C.P.B. 590 and 732. v. 'Two sermons by Bishop Berkeley', 
Hermathena, vol. xxii. 1932, p. 25. Berkeley's attitude to revela- 
tion is just that of the Tory Churchman. I do not suggest that 
he owed it in any way to Malebranche. 



THE PRINCIPLES OF KNOWLEDGE 117 

compounded idea of God*. In 795 he gives up the idea 
of God completely, thereby breaking with Descartes and 
Locke and rejecting the ontological argument. In 825 he 
states that he is certain there is a God, though he does not 
perceive Him. This statement is supplemented in 850 and 
there branches out into the typical argument of the Prin- 
ciples, incorporating a notion that appears in Clarke (Dem. 
3rd ed., p. 55) and more forcibly in the Search (iv, c. 2): 
4 Every of His works is a proof of it. All the actions of men 
and beasts prove it. Whatever we think, whatever we see, 
whatever we feel, demonstrates it. In a word, there is 
nothing in the world but proves that there is a God, or at 
least may prove it.' 

In the Theory of Vision there is scarcely a reference to 
Deity, except for the hint of the Divine Visual Language 
dropped in 147. The Essay at its appearance seems to 
have been criticized on that score. 1 In the Principles the exis- 
tence of God is in the foreground throughout. In 26-9 
we have the brief proof, based on the contrast between our 
voluntary and involuntary ideas, of God as the active sub- 
stance and cause of the regular succession in our ideas. 
There is a marked resemblance between 29 and Locke's 
Essay iv. n, 5, in respect of language and the example. 
One would think that Berkeley designed the echo. Locke 
says: 'If I turn my eyes at noon towards the sun, I cannot 
avoid the ideas which the light or sun then produces in me/ 
Hence Locke infers 'the brisk acting of some objects with- 
out me'. From the same instance Berkeley brings out his 
rapier thrust, 'There is therefore some other Will or Spirit 
that produces them' ( 29). In 30-2 Berkeley considers 
the nature of the connexion between ideas and their rela- 
tion to human good, and thence he establishes the goodness 
and wisdom of the Governing Spirit. In 53, 61-6 he 
1 Rand, op. cit., pp. 72-3. 



n8 THE PRINCIPLES OF KNOWLEDGE 

turns to the power and operation of God. Apparently he 
has Malebranche in view when he speaks of 'modern philo- 
sophers who though they allow matter to exist, yet will have 
God alone to be the immediate efficient cause of all things'. 
Without here defining his attitude to occasionalism, he puts 
his finger on a weak point in Malebranche 's system. If the 
material bodies are impotent, and if God works all without 
their help, is not their very existence, if possible, at least 'a 
very unaccountable and extravagant supposition'. This is 
the argument against matter from Parsimony, to which 
Berkeley returns in 61, demanding 'To what end God 
should take those roundabout methods of effecting things'. 
It is worth noting that Berkeley here may be turning against 
Malebranche the latter 's own argument against the 'maga- 
zine' theory, viz. 'God never effects by most roundabout 
and difficult ways what can be done in ways most simple 
and easy' (Search^ iii. 2, c. 4, 6). 

Berkeley did not accept the full occasionalist doctrine of 
the divine operation; for he always recognized the finite 
spirit as a centre of real activity, but he agreed with part 
of the occasionalist creed. He accepted the passivity and 
impotence of the idea, and the absence of the second or 
material cause. For corporeal causality Berkeley substi- 
tutes the categories of sign and things signified, regarding 
ideas as the language of the Author of Nature ( 66). In 
150 he expands this notion of the divine operation, and 
there his debt to Malebranche becomes marked. Nature, 
he says, if regarded as something distinct from God, 'is a 
vain chimera, introduced by those heathens who had not just 
notions of the omnipresence and infinite perfection of God. 
. . . Fain would we suppose Him at a great distance off, 
and substitute some blind unthinking deputy in his stead'. 
Malebranche writes to the same effect and in similar lan- 
guage, 'Of the most dangerous errour in the philosophy of 



THE PRINCIPLES OF KNOWLEDGE 119 

the ancients' (vi. 2. 3) a passage which culminates in the 
daring statement, 'There is then but one true Cause, as 
there is one true God. Neither must we imagine that what 
precedes an effect does really produce it. God Himself 
cannot communicate his Power to creatures, according to 
the light of reason. He cannot make them true causes and 
change them into Gods. . . . Bodies, spirits, pure intelli- 
gences, all can do nothing . . . tis the author of our being 
that performs our desires. Semel jussit: semper paret. 
He moves even our arms when we use them against his 
orders: . . . All those little divinities of the heathens, all 
those particular causes of philosophers are chimeras.' 

No wonder that Berkeley when he reached Paris wanted 
to go and see the man who could write like this. For Berke- 
ley's theism is inseparable from his ideism, as concave from 
convex. He expounds it simply and un systematically 
throughout the Principles part passu with his doctrine of 
ideas. Apart from the Active Principle 'in whom we live 
and move and have our being', his world of idea-things is 
scarcely intelligible. But it is a mistake to approach Berke- 
leianism with preconceptions of deity. The system is diffi- 
cult, if we come to it, thinking we know already what God 
is. Berkeley in effect teaches what God is rather than 
what ideas are. Deity, as seen by Berkeley, has to be 
always everywhere, always supporting the idea- things, coi\- 
trolling their changes, discovering them to finite spirits, and 
that not in the manner of a scene-shifter or conjurer, but 
like an Immense Power, silent and pervasive as attraction, 
conscious as the human will, intelligent as reason, wise as 
law, strong as love, good as Christ. Was there a Deity in 
the philosophies known to Berkeley, outside the pages of 
Malebranche, conceived on that grand scale ? 

There follows the inevitable question How can such 
a Being be known by brief man ? Berkeley does not shirk 



120 THE PRINCIPLES OF KNOWLEDGE 

that question. He describes in outline in 146 how God is 
known. In 147-9 he classifies the knowing of God and 
compares it with other knowledges. From 150 onwards 
he explains the difficulties men have in accepting it. The 
ease, the subtlety and the fire of this exposition are re- 
markable. This is the climax to the epistemological section 
of the Principles, begun in 86 . Of the four types of knowing, 
correspondent to the four objects of knowledge, he takes the 
knowing of God last, not as most obscure but as most com- 
prehensive. We saw above that knowledge of other spirits is 
a synthesis of knowledge of ideas and self-knowledge. The 
same seems to be true on a wider scale of man's knowledge 
of God. Ideas and self and finite spirits enter into it. We 
advance from knowing idea- things to self-knowledge, thence 
to knowledge of other selves, thence 'after the same manner* 
( 148) we reach the knowledge of the Boundless Spirit. 

The concluding paragraphs of the Principles do not 
always receive from readers adequate attention. Berkeley 
attached great weight to them and devoted care to their 
composition. Perhaps their piety and homiletic undertone 
make them seem alien to metaphysics. Yet they enshrine 
the purpose of the book. Berkeleianism is a philosophy of 
God, not of mind alone. Berkeleianism is a religious philo- 
sophy, as Berkeley's was a philosopher's religion. His pub- 
lished writings, his correspondence, his private note-books, 
his public and private life all prove his absolute genuine- 
ness and unswerving conviction. From his early days in 
Trinity College to his closing days at Oxford his opinion 
never varied as to 'what deserves the first place in our 
studies . . which to promote . . was the main drift and 
design* of his labours ( 156). 

Any one trying to describe how God is known and to 
relate that knowing to other knowings is faced with a great 
difficulty. He is trying to express that which, in our 



THE PRINCIPLES OF KNOWLEDGE 121 

thoughts, is most luminous and most obscure. He is trying 
to combine the deep of inherited feelings with the surface 
of individual experience. Consequently the reasons he 
gives are not adequate to the reasons he feels. Our know- 
ledge of God is a clear dictate of reason shadowed by a com- 
plex of obscure feelings. We may detect in it the feeling 
of the part for the whole, of the branch for the stem, of the 
marching soldier for his battalion, of the man for his 
country, of the son for his father. Those intermittent, blind, 
inarticulate feelings blend with the rational Attributes, and 
the resulting apprehension may become at times as clear as 
daylight, as cogent as a Barbara syllogism. 

Berkeley faces the inherent difficulty of his task. For 
him, our knowledge of God is like sense-perception in its 
cogency, like self-knowledge in its intimacy, like know- 
ledge of other minds in the conscious reasoning involved. 
As portrayed in 146, the knowledge of God partakes of 
three elements. First, there is the intelligent study of idea- 
things, 'which are called the Works of Nature', with the 
unavoidable inference to some Spirit other than man. 
Second, there is the attentive consideration of certain qual- 
ities of value in the things, viz. order, greatness, beauty, 
purpose, laws of pain and pleasure, and animal instinct. 
Third, there is attention to the meaning and import of the 
Attributes, as they appear in rational theology. 

Having thus recognized in it reason and feeling, intelli- 
gence and instinct, Berkeley passes on to characterize this 
type of knowledge. At the outset we find a difficulty which 
is perhaps also an inconsistency. The opening sentence of 
147 appears to state that God is known certainly and im- 
mediately, but the statement is qualified almost to the point 
of withdrawal by the addition of the words, 'as any other 
mind or spirit distinct from ourselves'. Berkeley has just 
said ( 145) that our knowledge of other spirits is not 

4046 R 



122 THE PRINCIPLES OF KNOWLEDGE 

immediate. Having compared our knowledge of God to med- 
iate knowledge, he virtually states that it is immediate. Ber- 
keley has not blundered or changed his mind ; for the same 
difficulty meets us in the Hylas and the Alciphron dialogues. 
'We have at least as clear, full, and immediate certainty [of 
God] ... as of any one human soul whatsoever besides 
our own' (Ale. iv. 5). In the second Hylas dialogue (Works, 
p. 424) Philonous 'immediately and necessarily* concludes 
the beingof a God ;inihe third Hylas dialogue (Works, p. 448) 
Philonous 'has a notion of Him, or knows Him by reflexion 
and reasoning' a process which lower down is explained 
as 'collect it by reasoning from that which you know im- 
mediately'. Perhaps to some extent Berkeley is being 
swayed by the opposing influences of his primary au- 
thorities. For Locke, knowledge of God is a much medi- 
ated demonstrative knowledge. For Malebranche it is the 
clearest type of immediate knowledge. More probably 
Berkeley is simply putting common sense before technical 
accuracy. The technical distinction between immediate 
and mediate is a good servant but a bad master. When 
dealing with a process like knowing, especially the knowing 
of an Immense Object, where no clean cut between know- 
ledge of His existence and knowledge of His nature is 
practicable, a man of Berkeley's stamp, hating jargon, lov- 
ing fact, will not oppose too strictly mediacy and immediacy. 
Our knowledge of God, in fact, is immediate, in that we 
are in God and that immediate facts not representative 
ideas are the ground of our assurance. Our knowledge of 
God, in fact, is mediate, because we are not God but men ; 
reflexion on our part is required and we must 'open the eye 
of the mind' ( 154). All said and done, it is possible for 
us to neglect or to refuse to know God. So our awareness 
of Him cannot be altogether direct. 1 

1 Professor Jessop suggests to me that the vacillation between 



THE PRINCIPLES OF KNOWLEDGE 123 

God is immense and omnipresent and all encompassing, 1 
for Berkeley, and therefore in one sense we must have im- 
mediate contact with Him. The resemblance of this part 
of Berkeley's theology to that of Malebranche is too close to 
be accidental. In each of Berkeley's major works, as soon 
as he has outlined his own conception of Deity his thoughts 
pass to that of Malebranche (Prtn. 147-8; Hyl., Works, 
pp. 424-6 ; Ale. iv. 14). Nor is the debt other than empha- 
sized by the pains Berkeley takes to underline the points 
of disagreement. Berkeley is, perhaps, not unwilling that 
the disagreement should seem greater than it really is. 
From 148 we can see that there are three things in Male- 
branche 's theology to which Berkeley takes exception: 
(i) It leaves the door open to a materialist interpretation 
of the term 'seeing*. Even philosophers were saying 
'Jupiter est quodcumque vides\ (2) The representative 
ideas supposed to be seen by us. (3) A literal union of 
man's mind with the substance or essence of God. When 
he has shown his disagreement on these points, he proceeds 
to show his agreement on others. Having denied what 
Malebranche affirmed, that 'we see God by a direct and 
immediate view', he goes on at once to admit, with reserva- 
tions, that we see a man, and to conclude that 'after the 
same manner we see God'. In the next section (149) 
Berkeley, for all his disclaimers, 2 throws himself almost info 
the arms of Malebranche, using two of the other's charac- 
teristic phrases, 'intimately present', 'absolute and entire 
dependence', and quoting the text with which Malebranche 

mediacy and immediacy represents the passage from unreflec- 
tive knowledge to demonstrative justification of that knowledge. 

1 Also omniscient ; but never omnisentient. The term 'omni- 
percipient* might be allowed, subject to the proviso, 'God . . . 
perceives nothing by sense*. Hylas y p. 459. 

2 v. Rand, op. cit., p. 89. 



124 THE PRINCIPLES OF KNOWLEDGE 

closes his famous chapter (iii. 2. 6) and which Berkeley 
quotes elsewhere in connexion with Malebranche's doc- 
trine (Hylas, Works, p. 427; Alctphron, iv. 14) 'In Him 
we live and move and have our being'. 

The essepercipi is unintelligible apart from 'the intimate 
presence of an All-wise Spirit who fashions, regulates and 
sustains the whole system of Being' ( 151). The esse perci- 
pere is an idle boast, if divorced from the truth 'that He is 
present and conscious to our innermost thoughts ; and that 
we have a most absolute and immediate dependence on 
Him' ( 155). For Berkeley, then, in knowing God, we know 
Something not ourselves, a conscious encompassing Spirit 
all good and wise. From the nature of the case, every sort 
of representative entity, except signs, is excluded; there- 
fore that knowing is, to all intents and purposes, immediate. 
Thus Berkeley's world is seen in God. His idea-things are 
stable by God's law, regulated by God's ordering Will, 
luminous in His light. As a twentieth-century thinker can 
only with difficulty prevent himself from seeing all things 
in evolution, so Berkeley when he had read Malebranche 
could scarcely help seeing all things in God. 

The Vision of all things in God is the alpha and omega.qj^ 
the Principles. Berkeley never attempts to interpret material 
tKmgs by our powers of perception . Not our thinking makes 
them so. Neither does he rise by inference or analogy from 
our thoughts to God's thoughts. Rather he puts himself 
into the scheme of things and finds God there. Nor does 
he lose himself in so doing. He never forgets that he is 
a free though finite centre of real activity. He takes the 
facts of life as they come, reading in the facts what many 
people in thoughtful moments read there but cannot 
express, namely the Mind and Will of the Creator who 
'can indifferently produce everything by a mere fiat or 
act of His Will', not less because He 'produces in our 



THE PRINCIPLES OF KNOWLEDGE 125 

mind all that variety of ideas or sensations which conti- 
nually affect us'. 

Berkeley's ideism is theological realism. Its basis is not 
strictly mind but deity, and though he often speaks of God 
as Mind, in the last resort he considers God and the human 
mind as distinct as are the One Infinite Spirit and the many 
finite spirits. The New Principle of seeing all things in God 
without matter or representative idea derives in the main 
from Malebranche's Search. That was the derivation as- 
signed to it by the leaders of English thought in his day, 
Clarke and Whiston, who read the Principles in 1710 a few 
months after it appeared. 1 Fifteen years later (July 24, 
1725) Bolingbroke writing to Swift speaks of that deriva- 
tion casually, as if it were admitted matter of fact, lamenting 
that he will lose the opportunity of meeting Berkeley, *a 
man who can espouse in good earnest the system of Father 
Malebranche'. 2 

1 Rand, op. cit., pp. 87, 89. 

2 A considered judgement. Bolingbroke knew both systems 
well, and brackets them in his Philosophical Works, ed. 1754, 
vol. i. p. 16 and vol. ii. p. 141. 



VII 
ABSTRACT IDEAS 

BERKELEY'S doctrine of abstraction, through Hume's ap- 
preciation of it, made his philosophy famous at a time 
when his immaterialism was ignored. To-day the estima- 
tion is reversed. His immaterialism is admired, if at a dis- 
tance ; on all hands his doctrine of abstract ideas is cried 
down. Accordingly a somewhat full treatment of its origin 
and objective seems called for. The doctrine is primarily 
a weapon of destruction, an answer to objections ; so its use 
is of more importance than its origin, and I shall not stress 
the influence of Malebranche. But for the sake of com- 
pleteness I have stated the evidence for the connexion 
(below, pp. 143-7). Berkeley was a pioneer on this point. 
Hume at any rate took him to be such. It seems unlikely 
that he would strike out on a new line without suggestion 
and support. Berkeley was intrepid and original. Percival 
records of him that he said, 'I know not what it is to fear'. 1 
But even so he might think twice before pitting himself 
against Locke on a subject of supreme consequence to both. 
Berkeley of course had to attack abstract ideas. It is no 
good telling folk that matter does not exist, unless at the 
same time you make it clear to them why they are inveter- 
ately disposed to think matter does exist. Collier taught 
immaterialism about the same time, but he did not attack 
abstraction. So he died and Berkeley lived. 

The entire Introduction to the Principles is devoted to 
the theme. The Introduction as originally planned was 
more ambitious 2 but could hardly have been more effective. 

1 In the Egmont Papers y vol. ci, I found the note, indexed under 
Dean Berkeley, 'I know not what it is to fear, said Mr. Berkeley, 
but I have a delicate sense of danger* : see Hermathena, vol. xxiii, 
p. 28. 2 v. C.P.B. 2201 cf. 516, 592, 696, &c. 



ABSTRACT IDEAS 127 

c To say no more, it is an abstract idea* ( 13). There is his 
all-inclusive answer to objections. Matter is the abstract 
idea in chief; but there are others. Extension, time, motion, 
unity, happiness, power, will, identity, all these qua ab- 
stractions fall before Berkeley's razor. Berkeley may not 
at first have realized the full positive consequences of his 
denial of abstract ideas, but he knew that he was attacking 
roundabout methods of thinking, which issue in scepticism, 
and was defending direct awareness and true contact with 
reality. 

What doctrine of abstraction was Berkeley attacking? 
The question has been raised. Berkeley leaves us in no 
doubt. He is attacking every doctrine of abstraction, ex- 
cept that which contents itself with stating that we can and 
do consider separately separable parts. He summarizes his 
views in 10 of the Introduction to the Principles, and 
there specifies as the 'Two proper acceptations of abstrac- 
tion' (i.e. of the false abstraction which he denies) : (i) con- 
ceiving separately inseparable parts or qualities, (2) fram- 
ing a general notion by abstracting from particulars in the 
manner aforesaid. It is interesting to note that this sum- 
mary from 'To be plain* was added in the table of Errata, 
ist edition. 1 This fact does not mean that the addition 
was substantially an afterthought, which was clearly not 
the case, but it may indicate a certain degree of haste in the 
composition. I do not think that this expression of his 
views is as well considered as the succinct description of 
the 'twofold abstraction' in Principles, 99. In this latter 

1 It fully covers, I think, the ground of the addition to 16, 
1734 ed., which Professor Laird calls *a dubiously consistent 
addition', and regards as throwing up the sponge' (Hume's 
Philosophy, pp. 56-8). If Berkeley was inconsistent on the 
point, he was inconsistent in 1710 and time brought no change 
of front, v. infra, p. 156. 



128 ABSTRACT IDEAS 

passage we find, in place of two propositions separately 
denied, one denial of a twofold abstraction. This distinc- 
tion, though a fine one, considerably eases a difficulty in 
Berkeley's doctrine. As regards the false abstraction in the 
second acceptation (No. 2 above), there is neither ambi- 
guity nor inconsistency. Berkeley always explicitly denied 
a general notion framed by abstraction. The difficulty con- 
cerns the other acceptation (No. i above). It is extremely 
difficult in certain cases to draw the line between the 'con- 
ceiving separately inseparable parts' which Berkeley denies, 
and the 'considering separately separable parts' which he 
affirms. I can concentrate my attention on an inseparable 
part nearly as easily as I can on a separable part. To take 
Berkeley's instances, it is easy to consider head and body 
separately; but what are we to say of his stock instance 
of inseparables, motion and the body moving ? Since the 
body may come to rest, it seems a hard saying that we 
cannot conceive separately it and its motion. True, I can- 
not conceive that while it is moving it is not moving. But 
that is another matter. Surely a particular motion may be 
singled 1 from a particular extension, and vice versa. Con- 
sequently, with diffidence I suggest that the addition to the 
table of errata in 10 of the Introduction should be inter- 
preted in the light of other passages, as e.g. 99 of the 
Principles. When Berkeley denies that we can conceive 
separately and denies that we can frame, these are not per- 
haps two distinct denials. He may be denying that we can 
conceive separately inseparable parts in the supposed process 
of framing an abstract idea. That at any rate is the real 
substance of his double denial. For the admission that it 
is psychologically possible to conceive separately a moving 
body and its motion, or an extension and its colour, would 
not, I think, carry with it any admission of the framed 
1 i.e. be the focus of attention. Berkeley is fond of the term. 



ABSTRACT IDEAS 129 

abstract idea of motion in general, or of the 'entity of ex- 
tension' (Prin. 99). 

What historical system of abstraction was Berkeley at- 
tacking? He was, in the main, attacking the Lockian 
system. But this was no personal attack; for in the Intro- 
duction ( 17, 21) he throws a glance over the history of 
the subject, in so far as he knew it, and briefly considers the 
'ablest patrons' of abstract ideas, from Aristotle to Locke. 
In the draft introduction he calls Aristotle 'a great admirer 
and promoter of the doctrine of abstraction', and to prove it, 
he quotes in Greek from the Metaphysics (Works, vol. iii, 
p. 368). He excuses himself from the task of tracing the 
doctrine through 'the Schoolmen, those great masters of 
abstraction', concluding that of all false principles 'none 
hath a more wide influence over the thoughts of speculative 
men' (Int., 17). 

So the attack on Locke is an attack on Locke in a repre- 
sentative capacity. 'A faulty and unskilful abstraction', 
Bacon taught, 1 is one of the idols of the market-place, and 
Locke's doctrine therefore, Berkeley came to think, was 
more than a blemish in his system; it was a relapse into 
'idolatry', an original sin of human thought. Berkeley con- 
centrated his fire upon Locke as the latest exponent of an 
old error. His formal statement of the doctrine (Int., 7), 
prefaced by 'we are told', is based upon the two passages 
in the Essay where Locke states the doctrine 'exprofesso'. 2 
These passages are bk. ii. 11.9 and bk. iii. 3. 6. The ab- 
stract idea of colour is Locke's illustration in the former 
passage, and that of man and animal in the latter. Berkeley 
uses both illustrations, repeating Locke's words 'body, life, 
sense, and spontaneous motion'. Berkeley quotes several 

1 Berkeley knew the passage, v. C.P.B. 569, reading Bacon 
for Barrow, and Nov. Org., bk. i, Aph. 59-60. 

2 Locke's own term, v. Note A, bk. ii. 23. 

4046 g 



i 3 o ABSTRACT IDEAS 

long passages on the subject from the Essay in his Intro- 
duction, and makes several references thereto in the 
Commonplace Book. 

But did Berkeley mistake his target ? Did Locke actually 
hold the view that Berkeley attacked as his ? Berkeley in 
his lifetime was charged with misrepresenting Locke's 
doctrine of abstract ideas, and the charge has been recently 
repeated. 1 

Locke's doctrine of abstraction is undoubtedly complex, 
developing as he wrote. But that consideration, though it 
lends colour to the charge against Berkeley, is really beside 
the point. For certain parts of Locke's doctrine of ab- 
straction, as e.g. the singling of separable parts, Berkeley 
accepted and made his own. The issue really is: Does the 
doctrine of the fourth book, in particular the passage about 
the abstract general triangle (c. 7), the exposure of which 
Berkeley styled his 'killing blow' (C.P.B. 699), represent 

1 In Geometry no friend to Infidelity, 1734, by Philalethes 
(Jurin), pp. 71 sqq. and by Professor Aaron, Locke's Theory of 
Universahj Arist. Soc., 1933. In addition to using the argument 
answered above, Dr. Aaron, like Philalethes, argues that Berkeley 
puts a misconstruction upon Locke's words 'some parts of several 
different and inconsistent ideas are put together*. I doubt if 
Dr. Aaron's two arguments are on all fours. For he wants to 
defend Locke's statement, at the same time pleading that it is 
not typical of Locke's teaching. His point here is that Locke 
means 'consistent parts of inconsistent ideas', whereas Berkeley 
makes him mean 'inconsistent parts of inconsistent ideas', and 
thereby depicts the triangle as 'made up of manifest staring 
contradictions' (T.V. 125). I do not think that Dr. Aaron's case, 
though ably pleaded, is made out. Locke seems to have meant 
by the words what Berkeley said he meant. For if the abstract 
triangle were made up of judiciously selected consistent parts, 
what is the point of the passage ? Locke's whole point here is 
that the abstract idea is a mass of inconsistencies, and is therefore 
difficult to frame, imperfect and non-existent. If in saying so, he 
gives his case away, it is Berkeley's right and duty to point it out. 



ABSTRACT IDEAS 131 

Locke's considered views ? The reader will be able to judge 
for himself, after reading the sketch of the doctrines of 
Locke and Berkeley which I give below. Here I will only 
say that there is no indication that any part of Locke's com- 
plex doctrine escaped Berkeley's notice. Berkeley was a 
fighting Irishman, and when he saw a head to hit he hit it. 
The triangle argument was his 'killing blow', but he had 
given plenty of other shrewd blows before administering 
that coup de grace. It is sheer paradox to say that Locke did 
not hold the doctrine. Berkeley takes it from the Essay in 
Locke's own words. Locke did say that the 'absurd triangle' 
must be 'all and none of these at once', and he meant it. So 
far from the 'triangle' passage being an aside, it appears to 
represent the culmination of Locke's thought on the subject. 
For in the fourth book he has reached the conception of 
mathematics as the true type of knowledge, and the contem- 
plation of abstract ideas as the main method of advancing 
knowledge. He would naturally therefore look for an illus- 
tration of his doctrine in the field of pure quantity. Surely 
then , a mode of extension such as the abstract general triangle , 
would be for Locke, the abstract general idea par excellence. 1 
The Analyst controversy called forth from Berkeley in 
1735 an admirably lucid commentary upon his own attack 
on abstraction and his relation to Locke. 2 At times he 
speaks with some heat. It was natural that he should do so, 
for Philalethes was, in effect, attempting to 'save Locke's 
face' at the expense of Berkeley's judgement and candour. 
Philalethes had said in effect, 'this account of a general 
triangle was a trap which Mr. Locke set to catch fools'. 
Berkeley in reply gives a patient exposition of Locke's 
teaching, and then sums up in the words, 'This doctrine 
of abstract general ideas seemed to me a capital error, 

1 v. e.g. Essay , iv. 12. 12-15. 

2 Defence of Free-thinking in Mathematics, 45-8. 



i 3 2 ABSTRACT IDEAS 

productive of numberless difficulties and disputes, that runs 
not only throughout Mr. Locke's book, but through most 
parts of learning. Consequently, my animadversions there- 
upon were not an effect of being inclined to carp or cavil 
at a single passage, as you would wrongfully insinuate, but 
proceeded from a love of truth. . . .'* 

Locke on Abstract Ideas 

The chief passages are Book ii. 1 1 . 9 and Book iii. 3. 6-9 
and Book iv. 6 and 7, and 12. 7-15. The above passages in 
Books ii and iii are stated by Locke to be his official or 
ex professo expositions. They are mainly psychological 
accounts of the process of abstracting, while the fourth 
book carries the doctrine out to its epistemological con- 
sequences. There are many incidental references to and 
sidelights upon the doctrine in these books, and, remem- 
bering that Book i (cc. 2. 14 and 4. 24) on 'Innate Ideas' 
finds the solution of its main problem in the faculty of 
abstraction, we see that Berkeley's statement that the doc- 
trine runs through the whole Essay is correct. 

Locke's abstract idea vacillates between subjective and 
objective, between function and product. At its first occur- 
rence (i. a. 15) the term is 'abstracting', the process, one 
of the steps by which the mind attains truth. In the note 
on ii. 2, when Stillingfleet advances 'substance' as a rational 
concept not derived from sense or reflection, Locke coun- 
ters with 'the abstraction of the mind', which from ideas 
of sense and reflection derives the obscure vague idea of 
substance. In the first of the two official accounts, abstract- 
ing is simply one of the given functions or operations of 
the mind, like comparing, compounding, and naming. It 
arises in fact out of naming. There are not enough names 

1 Browne (Procedure p. 187) and Bolingbroke (op. cit. vol. i. 
p. 114) treat Locke's triangle as a typical passage. 



ABSTRACT IDEAS 133 

to go round, and so the mind makes particular ideas to 
become general. The process of generalizing is called 
'abstraction'. There is no substantial difference between 
the two official accounts. The first is the fuller as to the 
psychological process ; the second shows in greater detail 
the function of the abstraction in making sorts and species. 
The term 'appearance' for idea occurs in the first and is 
dropped in the second. The second introduces the doubt- 
ful term 'framing'. In both accounts the essence of ab- 
straction is the notion of separating. The abstract idea is 
a separated mental appearance; it is separated from the 
here and the now, and from other ideas that in actuality are 
connected with it. These 'precise, naked appearances' are 
thus fitted for generality. They are named, and laid up in 
the mind 'as the standards to rank real existences into 
sorts'. 'Thus universals, whether ideas or terms are made.' 
Thus far the abstract idea represents a return to the pure 
datum, and the agency of the mind is conceived as entirely 
selective. In the course of the Essay this simple and fairly 
innocent notion of abstraction takes on far-reaching develop- 
ments. As the work proceeds, the abstract idea becomes 
more and more concrete. Even in Book ii (c. 32, 7-8) the 
abstract idea is 'something in the mind, between the thing 
that exists, and the name that is given to it'. It must have 
a 'double conformity', to thing and to name; otherwise 
there will be neither valid thought nor intelligible speech. 
In Book iii the subsistence of the abstract idea becomes 
more marked. In the second ex professo passage the term 
'framing' is used, but Locke is careful to point out that 
the framing results in 'nothing new'. If abstract ideas are 
framed, they are framed as a postage stamp by perforation, 
or as a field by a sunken fence. This is framing by diminu- 
tion, and Locke repeatedly uses the term 'partial' as equi- 
valent to abstract, and against Stillingfleet (Note A on bk. ii, 



i 3 4 ABSTRACT IDEAS 

c. 23) he insists that Abstracting and enlarging' differs from 
'complication of ideas'. However, a framing is a making. 
Locke has to relate the abstract idea to the word in Book 
iii, and to assign its place in Knowledge in Book iv. So 
naturally the idea takes shape and substance, and the mind's 
contribution becomes more positive as the Essay advances. 
In one passage, indeed, Locke speaks as if he were conscious 
of the development. He writes (bk. iii. 5. 16) of 'an argu- 
ment that appears to me new, and a little out of the way 
(I am sure it is one I thought not of when I began to write)'. 
The reference is not quite certain. But since the term 
'nominal Essence' comes into use in Book iii for abstract 
idea, and he speaks here of 'the pudder made about es- 
sences', it seems probable that he had grown convinced as 
he worked at the problem of abstraction, that the abstract 
idea was an entity, an essence actually made by the mind, 
mistaken by scholasticism for real essence, and restored to 
its true status by his slow-motion picture of the mental 
process. 

Locke, no doubt, began to realize that abstraction raises 
in an acute form the problems of universality and of the 
one and the many. Abstracting then, for him, is no longer 
one among several activities of the mind, but takes a place 
apart. It has a distinctive value; for brutes abstract not, 
and the faculty puts a perfect difference between them and 
us. The abstract idea becomes the goal of clear reasoning 
and the sine qua non of intelligible discourse. The original 
notion of separating what is already there becomes that of 
framing or constructing what was not there. Abstract ideas 
are 'fictions and contrivances' ; they are 'put together', and 
with great difficulty. If the abstract idea were simply the 
sense datum isolated, abstraction would be easy for children 
and possible for animals ; for they attend to the sheer datum 
and owe little or nothing to memory, imagination, and sug- 



ABSTRACT IDEAS 135 

gestion. The stress laid upon the difficulty in Book iv, c. 7 
is a symptom of the development of the * precise naked 
appearances' into 'fictions and contrivances of the mind*. 

The study of words as signs in Book iii gives Locke 
occasion for his closest examination of the abstract idea, 
and the term there is usually followed by 'to which the 
name is attached' or similar phrase. The name would be 
of little use, Locke thinks, without the abstract idea. It 
would be a sign signifying nothing. When Adam said 
'niouph', Eve might think 'kinneah', and both be wide of 
the mark as regards Lamech's trouble (bk. iii. 6. 44). The 
sign is arbitrary in that there is no resemblance between it 
and its meaning, but it may not be capricious ; for caprice 
in the use of significant words is the death of knowledge. 
By receiving a name the abstract idea attains its full status 
and becomes the nominal essence (bk. iii, c. 6) which in 
substances is distinct from the real essence, and yet is of 
tremendous significance for human knowledge. Thus in 
Locke's system an appearance has flowered into an essence. 
Ultimately the paradox springs from the attempt to com- 
bine the fact of generality in a representationist system 
with the traditional principle that all existents are par- 
ticulars, a principle which is set as a headline in both of 
Locke's ex professo accounts of abstraction. 

The theory of universals now (bk. iii) passes into a theory 
of species. It had generally been held that species and 
genera were made by nature, and that 'general natures', 
otherwise called 'substantial forms', constituted the kind 
after which things were created ; and that in apprehending 
the 'general nature' the mind finds its satisfaction. Locke 
is not concerned to deny that there is a general nature in 
the sense of real essence ; only he suspects that it is not the 
mysterious entity written up by the scholastics. We can- 
not know it in the case of substances, but it probably consists 



i 3 6 ABSTRACT IDEAS 

in the arrangement of minute material parts. For us the 
nominal essence is all -important. This is what makes the 
sorts. This is what we know when we have general know- 
ledge. By its aid we 'shorten our way to knowledge* (bk. ii. 
32. 6), and learn to think correctly and talk intelligibly. 
The denomination per se will not make the unlike like. 
Locke expressly recognizes objective similitude (passim, 
especially bk. iii. 3. 13). Natural things are made alike. 
Organisms are the most obvious instances. It is, however, 
also evident in the cases of artificial things like watches, and 
of mixed modes like covetousness that the making and 
naming of the sort is the workmanship of the mind. Nature, 
Locke holds, did not make the abstract idea of man, and 
in the cases of monsters and changelings, &c., real essence 
fails us. The real essence of rain, snow, and water may be 
the same, but the nominal essence or abstract idea of each is 
different, and Locke draws the corollary that there is a dis- 
tinct sort for every named abstract idea. The abstract idea 
is what the sortal name stands for, the hypostatization of 
the standard notion, the boundary of the man-made species. 
The controversy with Stillingfleet brought out some 
other points (Note on bk. iii, c. 3). Stillingfleet saidthatthe 
abstract idea was the work of the mind, but not the creature 
of the mind, and that though there be only one individual 
in. the sort, e.g. the sun, yet there is a real essence, which 
if there were more suns would still constitute them. Locke 
replies that abstract general essences have no being outside 
the understanding, that we know not the real essence of 
the sun, and that the only essence of the sun he talks about 
is the nominal. Stillingfleet sees that the nature of knowing 
is involved, as well as its objects, and maintains that 'the 
general idea is not made from the simple ideas by the mere 
act of the mind, but from reason and consideration of the 
nature of things'. Locke's reply is that reason and con- 



ABSTRACT IDEAS 137 

sideration are acts of the mind, mere acts of the mind. Con- 
ceding that there is an 'internal constitution of things on 
which their properties depend', he argues that if we wait 
till we know it, we shall have to wait a long time without 
species and sorts. Our thinking cannot alter the nature of 
things, but it can and does alter the meaning of their names, 
i.e. make or remake the boundaries of the species. God 
makes the real essence, man the nominal. To Stillingfleet's 
contention that real essences are immutable, Locke rejoins 
that that holds solely of the essence of God Himself. His 
creatures are in essence mutable. What is grass to-day 
may be sheep to-morrow and man the day after, the nominal 
essence of grass remaining unchanged. 

An important distinction is then made between sub- 
stances on the one hand and simple ideas and modes on 
the other (bk. iii. 3. 18. and 5. 14). In the case of substances 
real and nominal essences always differ. In the case of 
simple ideas and modes they are identical. The distinction 
has consequences for his theory of knowledge. I know or 
can come to know what men call a gold ring; but I can 
never know the real nature of this gold ring. On the other 
hand, the definition of a triangle is the essential being from 
which all its properties flow. It is both nominal and real 
essence. It is the abstract idea of the triangle. So the 
general knowledge of substances attainable was for Locke 
very limited ; whereas mathematics seemed to offer the true 
type of method and the sure prospect of certain knowledge: 
This teaching, in physics, put conventional rules in place 
of scientific law. Indeed what Locke says about antimony, 
sulphur, and vitriol shows the deplorable state of chemistry 
at that time. That two sorts of sulphur may show qualities 
very different is accepted by him as simple fact, not as an 
urgent problem. The true method of advancing knowledge 
is, Locke holds, to consider our abstract ideas, and 'by 

4046 



i 3 8 ABSTRACT IDEAS 

what steps we are to proceed is to be learned in the schools 
of the mathematicians' (bk. iv. 12. 7). He goes so far as to 
state that all general knowledge 'consists barely in the con- 
templation of our own abstract ideas'. For instance, our 
working knowledge of man is based upon an idea of a 
shaped body with powers of sense, voluntary movement, 
and reason. This is the abstract idea of man and the 
essence of our species. All universal knowledge of man is 
the perception of the agreement or disagreement of another 
abstract idea with this. But we do not know the real essence 
of man, and so cannot affirm, for instance, that all men sleep 
by intervals, but simply that these men do so. Opinions and 
judgements we may have through 'hints well laid together'. 
But 'this is but guessing still'. We must glean by particular 
experiments and may thus gather a little knowledge of 
quinine, the compass, &c. ; but it is idle to attempt to know 
'general natures' and to 'grasp at knowledge by sheaves 
and bundles'. We never can know the minute changes in 
material substances, nor how they affect our senses and 
produce secondary sensations. Natural philosophy then 
cannot know for certain ; and so Locke comes back to mathe- 
matics with its clear abstract ideas of triangles, angles, &c., 
and its art of finding out intermediate ideas as the true type 
of epistemological method. Locke's sceptical tendencies 
and his doctrine of abstraction were closely knit, and in 
attacking his doctrine of abstract ideas Berkeley was rais- 
ing a fundamental issue. 

The Commonplace Book and Abstract Ideas 

Berkeley was trained 1 in the Lockian system. In his 
early college days he took the faculty of abstracting as a 
matter of course, was rather proud of it, and regarded the 

1 Not uncritically. Provost (afterwards Bishop) Browne took 
an independent line as regards the Essay y see above, p. 74 n. 



ABSTRACT IDEAS 139 

abstract idea as the fine flower of intellect. This was one 
of the few principles firmly held at the start of the Common- 
place Book and discarded before the end. In fact the rejec- 
tion of the abstract idea is one of the chief results of Berke- 
ley's 'wonder year' (1707-8). 

In his earlier writings he had spoken contemptuously of 
'specierum supellectilem' (Arithm., pt. 2, c. 5), and of 
'lapidific virtue' (Dunmore Cave Essay), but all adherents 
of the New Learning had, since Bacon wrote, broken with 
that type of jargon. Half of the Commonplace Book is 
written before Berkeley comes fairly in sight of the problem 
created or reshaped by Locke's Essay. The first entry on 
the subject is *Qu. Is it not impossible there shou'd be 
General ideas?' (329, corrected reading). The implied 
answer is 'Yes', and the next entry gives the reason, namely, 
that all ideas are particular. The line between General 
ideas and Abstract ideas is fine, but here, I think, we must 
draw it. For in the first note-book abstraction is scarcely 
mentioned and was not originally discussed, and the for- 
mulation of the Principle at its close (904-24) gives no 
hint of the doctrine. The editions might give the con- 
trary impression, but the references to abstraction in 53, 
113, and 385 occur on the verso, and are therefore late; 
while 'abstractive' ideas in 55 (we owe the reading to 
Professor Aaron) is a later insertion (N.B. 'or both'). .In 
141 'abstracting' merely equals 'singling'. 'No idea of 
circle, &c., in abstract' (247) may be an anticipation of his* 
later doctrine; but I doubt if there is much conscious 
thought about abstraction behind it. The entry concerns 
the problem of the infinite divisibility. The fact is that 
during the first part of the work Berkeley's interest in ideas 
was confined to simple and complex ideas. Not till that 
problem sank (it is scarcely mentioned in the publications, 
and the C.P.B. entries are almost all marked with the 



140 ABSTRACT IDEAS 

obelus, v. Appendix, pp. 186-8), did the problem of abs- 
traction rise. 

Berkeley wrote in the draft Introduction that there was 
a time when he did not doubt in the least that he had the 
faculty of abstracting. 1 To that period we may assign the 
first note-book. His doubts seem to have come gradually 
and to have begun with the problem of General ideas, 
which with a sweep are denied in the second entry of the 
second note-book (397). In his publications he rejects abs- 
tract ideas and accepts general ideas. It was the other way 
about during the transition period. Setting aside the re- 
considered or correcting statements on the verso, the first 
clear statement about abstraction is 437. Here his horizon 
is limited to extension, but as far as that instance goes, his 
doctrine is complete ; for he not only denies the abstract idea, 
but in the words 'as any one may try' he expresses his ulti- 
mate court of appeal in the finished argument. He then in 
the next twenty pages concerns himself successively with 
abstract ideas of colours, tastes, sounds, substances, and 
existence. Here he uses the term 'abstract* quite neutrally, 
as we may suppose Locke did at first. Abstract means 
'separated'. He speaks of abstract simple ideas (498). He 
asserts that all abstract ideas are particular, and distin- 
guishes abstraction of an idea from another of a different 
kifld, and abstraction of an idea from all particulars of the 
same kind (499). At the same time he follows up the denial 
'of general ideas. He seems to have reached this conclusion 
first in the case of figure. In 247 he had denied the circle 
in abstract, and now in 485 he denies general figure and 
in 5 1 5 general body. About this time the breach with Locke 
widens. In 496 he raises an important new question, pre- 
facing it with Qu., 'how can all words be said to stand for 
ideas ?' A short time before (371 = 418) it had been to him 
1 Works, vol. iii, p. 361. 



ABSTRACT IDEAS 141 

an axiom that no word is to be used without an idea. This 
principle is used by Berkeley in his essay Of Infinites, 1 and 
he owes it no doubt to Locke's repeated advice to lay aside 
the word, or at least to keep the word as a fixed sign of a 
definite idea. This principle had led Berkeley into elabor- 
ate speculations and indeed into an experiment about the 
solitary man, 2 a case designed to do for words and ideas 
what the Molyneux problem had attempted for sight and 
touch. In the sketch demonstration of the Principle the 
same axiom (904) 'All significant words stand for ideas' 
stands first on the list of premises from which the New 
Principle is deduced. Yet before long we shall find it 
classed by Berkeley as a chief cause of the false doctrine 
of abstraction. 

About the same time (541, 566) Berkeley turned his 
attention to essences and sorts. He writes that the dis- 
tinction between real and nominal essence is fruitless. If a 
general idea be the unknown real essence, it is an unknown 
idea, a nothing. If it be the nominal essence, a man-made 
thing, clear as crystal, open to all to discuss and compare 
with personal experience, then call it 'denomination', but 
drop the obscurantist, scholastic term 'essence', Berkeley 
would say, for its label 'nominal' does not remove the tang 
of its cask. 

The climax of the debate is reached in the notable series 
(566-72), entries which fixed, I believe, for life Berkeley's 
thought and to some extent his terminology. Here he 
selects his foe and decides his objective. He will oppose 
not general ideas, but abstract ideas, and the abstraction 
specified is that of Locke's fourth book, not as differing 
from the earlier books, but as carrying their teaching to the 

1 Works, vol. iii, p. 410. 

2 C.P.B. 571, 594, 598, 613, 657, 739 and Dft. Int. in Works, 
vol. iii, pp. 379-80; cf. Hylas, Works, vol. i, p. 467. 



i 4 2 ABSTRACT IDEAS 

natural and logical outcome. Sharply distinguishing genera 
and species (the sorts) from abstract general ideas, he re- 
jects the latter as including a contradiction in their nature, 
and he refers to the Essay (bk. iv. 7. 9 which is the famous 
passage about the abstract or general triangle). Berkeley 
tries to form the idea of the triangle according to Locke's 
recipe. He fails. He notes that Locke himself describes it 
as 'something imperfect that cannot exist*. If it cannot exist, 
it cannot, for Berkeley, be thought of. It is indeed 'made 
up of manifest staring contradictions' (T.V. 125). The 
general idea had been to Berkeley the cause of confusion 
in mathematics. He sees now that it is a psychological ab- 
surdity, and the cause of Locke's sceptical tendencies. That 
which for Locke is the magnet of our thoughts, the currency 
of communication, the impulse of progresses proved hollow, 
unreal, self-contradictory. From this point Berkeley never 
looked back. The impression made on him by this passage 
is well evidenced. This is 'the killing blow' (699), and the 
index letter in the margin 'I etc* is unique. Berkeley acted 
upon the 'etc'. The Lockian triangle figures largely in the 
Theory of Vision ( 125) and in the Introduction ( 13) to 
the Principles ; a distant allusion to it occurs in the Hylas 
(Works, vol. i, pp. 403-4). Express references to it are in 
the Alciphron, Dial, vii, 5, ist ed., and in the Defence of 
Free-Thinking in Math. 45, and a clear allusion in Siris, 

323- 

The remainder of the Commonplace Book shows several 

of the consequences of this decisive change of view. Berke- 
ley now takes a non-Lockian view of demonstration (592 ; 
cf. 220), which soon passes into a depreciation of demon- 
stration. 1 He devotes attention to the doctrine of words 
(650-1, 668, 671, &c.). He has to seek other differentiae 
between man and beast (600) in lieu of Locke's 'perfect 
1 870; cf. Prin., 107 *I do not say demonstrate*. 



ABSTRACT IDEAS 143 

distinction'. The analysis of substance is made (712-13, 
736-7), in which Locke is bracketed with the Schoolmen. 
Berkeley's razor is applied not only to extension, figures, 
motion, existence, substance, but to psychological entities 
like will and understanding; also to the scientific object, 
substantial forms, plastic virtue, Hylarchic Principle, and 
even to 'eternal and immutable' morality. He razesjudicium 
intellectus, indifferentia, uneasiness ; the Aeternae Veritates 
vanish (748), and the very faculty 'pure intellect', long sup- 
posed to deal with abstract ideas, is suspect, its occupation 
apparently gone (822). 

Malebranche on Abstract Ideas 

The evidence connecting Malebranche and Berkeley on 
this subject falls short of proof. We know that Berkeley 
examined Malebranche's teaching on the kindred subject 
of the 'sorts' (C.P.B. 296). And he repeatedly speaks as 
one familiar with the whole of Malebranche's system. He 
could hardly fail to see and ponder Malebranche's forceful 
attack on abstraction, at the time he was confronting his 
Goliath 1 and was in need of an ally. 

The decisive passages in the Search are iii. 2. 8 and 9, 
vi. 2. 2 and 3, and the Illustration (i.e. excursus) on iii. 2. 8. 
In these passages Malebranche repeatedly warns his readers 
against 'general and abstract ideas'. The first of these (iii. 
2. 8) is of special interest, because it follows close upon the 
exposition of the theory of seeing all things in God, and 
because matter is the point at issue. Malebranche dis- 
tinguishes nine attributes of matter. He takes five of them 
away because they are separable. He isolates three as 
secondary though inseparable, and thus by a reductio ad 
unum he is left with extension as the essence of matter. 
Why then, he asks, do people persistently hanker after 
1 Locke, v. C.P.B. 689. 



144 ABSTRACT IDEAS 

something else in matter ? There may be in it, he admits, 
some unknown super-extensional quality, and if the 
Church so decides he will withdraw his contention. But 
at this point he offers a reason why men seek an occult 
quality in matter against the weight of evidence. The 
reason is a queer one, and not very lucidly expressed. 
From two passages in the Principles ( 17, 74) we can see 
that Berkeley's attention was drawn to it. Berkeley studies 
in 'the most accurate philosophers' the meaning of material 
substance. He finds it to be 'the general idea of Being'; 
and, again, he shows that the last resource of the beaten 
materialist is 'I know not what abstracted and indefinite 
notions of being or occasion'. Now Malebranche in the 
above passage abstracted from matter hardness, softness, 
fluidness, motion, rest, figure, divisibility, and impenetra- 
bility, and found that extension is the sole remainder. In 
the minds corrupted by false abstraction, he holds, the 
clear idea of extension blends with the general idea of being 
that accompanies all our thoughts and produces a refusal 
to admit that there is nothing in matter but extension. It 
is a curious doctrine, because the general idea of being 
which, for Malebranche, is the source and almost the fabric 
of the disorderly abstractions, flows directly, he says, from 
the mind's union with God. In fact he comes near re- 
presenting the general idea of being, as blending our 'idea' 
of God with our idea of limitless extension. We make a 
bad use, he says, of the best things. His position is that the 
inescapable general idea of being makes mere abstractions 
impose upon us, by giving them an appearance of reality. 
He instances act, power, cause, effect, substantial forms, 
faculties, occult qualities, &c. These by themselves excite 
no determinate ideas in the mind. We should see their 
hollowness at a glance, but for the background of being 
in general. 



ABSTRACT IDEAS 145 

Malebranche does not reject all abstract ideas, any more 
than he rejects matter by sapping the evidence for matter. 
But it is quite possible that in both cases he put into Berke- 
ley's hands the weapon he himself forged but used half- 
heartedly. From 'idols are useless' to 'idols are nothing' is 
no far cry. The young iconoclast after reading Malebranche, 
assisted by Leclerc's systematic manuals of the New Logic, 
could hardly avoid the question , ' Can I truly form an abstract 
general idea ?' If Malebranche does not reject the abstract 
general, at least he tirelessly recommends the concrete par- 
ticular. He writes, in this chapter, 'If the vulgar philoso- 
phers . . . would give those men leave to be quiet who affix 
to these terms distinct and particular ideas, we should have 
nothing to reprehend in their conduct. But they set up 
themselves for the explaining nature by general and abstract 
ideas, as if Nature were herself abstract.' After instancing 
in the case of matter he concludes the chapter with the 
words, 'Real ideas will produce real science, but from 
general and logical (de logique, i.e. abstract) ideas can pro- 
ceed nothing but a random superficial and a barren science. 
Wherefore we ought with serious reflection to attend to the 
distinct and particular ideas of things.' 

Passing on to Book vi of the Search we find a striking 
development of the movement initiated by Bacon against 
pseudo-scientific objects masquerading as abstract ideas. 
In bk. vi. 2.3 Malebranche argues that such abstract ideas 
are mere relics of animism ; they are pagan deities usurping 
the rights of the one true God. Causality imputed to 
matter is what he has chiefly in view, but at times he goes 
further afield, and represents abstractions of the mind as 
a main obstacle in the pursuit of truth. He says: 'There 
are two things which I cannot too much mistrust (i) the 
impressions of my senses (2) my readiness to take abstracted 
natures and general ideas of logic for real particulars.' 

4046 U 



i 4 6 ABSTRACT IDEAS 

Quotations to the same effect might be multiplied. Book 
vi in fact reaches this conclusion both from the study of 
method and from the study of causality. True method 
requires us to seek clear and distinct ideas, and to eschew 
abstractions as loose and obscure. His opinion of causality 
finds expression in the words, 'philosophers explain the 
effects of nature by some beings of which they have no 
particular idea . . . the mind is a pagan, whilst the heart 
is a Christian'. 

If Berkeley's polemic against abstract ideas be uncon- 
nected with Malebranche's tremendous sayings on the sub- 
ject, the coincidence is startling. It is just possible that the 
urge of his own immaterialism, combined with the obvious 
weakness in Locke's doctrine, may be sufficient explanation. 
Yet Malebranche's Illustration upon the eighth chapter of 
the second part of the third book greatly strengthens the 
evidence for connexion. Its aim is to enable readers 'to com- 
prehend what I have said in some places, how that they give 
not the reasons of things who explain them by logical and 
general terms'. The gist of the argument is that such terms 
must signify being or mode of being, or they mean nothing. 
Consequently speakers and hearers must examine carefully 
their meanings, sticking closely to clear ideas. Then Male- 
branche goes on to say that we may and sometimes must 
(as when speaking of the soul) use significant words with- 
out distinct ideas. Still, whenever they are available we 
must give the preference to clear terms. For instance, we 
ought to say that God created the world by His Will, rather 
than by His Power. For Power is a term de logtque y signi- 
fying no distinct and particular idea. Natural philosophy is 
the chief offender, and he mentions as offending terms 
gravity, levity, nature, faculty, qualities. 

Several of the elements in this argument are found in 
Berkeley's works. At the outset of the Introduction ( 7-9) 



ABSTRACT IDEAS 147 

to the Principles Berkeley separates the cases of modes and 
beings, and towards the end ( 18-24) in the use and 
misuse of words he locates the origin of false abstraction. 
Again, the proviso that sometimes words have to be used 
without distinct ideas was a departure that became of deci- 
sive importance for Berkeley. As we saw above (p. 141) 
during the writing of the Commonplace Book he reversed 
his judgement on the point. With Malebranche's statement 
in this connexion that we cannot always put the definition 
in place of the defined we may compare Berkeley's change 
of view about the definition of the soul. 1 Again, when one 
thinks of the stress laid by Locke upon 'power* and the great 
use that he makes of it, and notices that Berkeley made 
similar use of it in the early part of the Commonplace Book 
but dropped it towards the end (N.B. 495, 814) and marked 
several of the * power' entries with the obelus, it may be 
legitimate to connect the change with Malebranche's con- 
demnation of that term as a misleading abstraction. 2 

1 Cf. C.P.B. 155-6 with 182. 

2 As stated above, p. 126, 1 have not laid]stress on the possible 
connection between Berkeley and Malebranche with regard to 
abstract ideas. I have worked out the evidence for connection, 
and must leave the reader to judge its value. The source of 
Berkeley's doctrine of abstraction is a matter of literary interest. 
The source of his doctrine of ideas is, if I mistake not, a matter 
of practical moment. 



VIII 
THE ALLEGED WITHDRAWAL 

THIS chapter calls for some apology ; for it lies outside 
the direct scope of the title and sub-title of the book. 
Yet perhaps there is justification for including it. The 
chapter is an attempt to resist the disintegration of the 
Berkeleian philosophy. In a short time it will not matter 
to any one what Berkeley's sources were, or what the spirit 
of his philosophy, if our young students are taught that there 
were two, if not three Berkeleys, and that the challenging 
extracts from 'juvenile' writings they are required to read 
were answered in mature life by their author. To make a 
philosopher refute himself is the easiest and cheapest refu- 
tation. In Berkeley's case there may be legitimate lines of 
refutation. There are certainly legitimate lines of criticism. 
But the short way with him, which says that Bishop Berkeley 
ceased to be a Berkeleian, is not, I believe, legitimate. 1 One 

1 The supposed opposition between the teaching of Berkeley 
and that of Bishop Berkeley is graphically expressed by Dr. 
Kemp Smith (Mind, July 1933, p. 363), who asks, 'Do not nearly 
all Berkeley's second thoughts run counter to his first thoughts ?' 
I should give the unexpected answer to that question. To me 
it is simply incredible that Berkeley 'ceased to be a Berkeleian'. 
Do bishops change their minds ? Berkeley taught his early philo- 
sophy and presented his early books to Johnson and his other 
.American friends, and corresponded with them and others on 
philosophy, inter alia, throughout the Cloyne years. I cannot 
recall a hint in his correspondence of any change of view. Surely 
the De Motu and the Alciphron contain Berkeley's early philo- 
sophy? He republished those works six months before his death. 
In case Bishop Berkeley suffered an unperceived conversion, the 
question deserves examination point by point. But mere differ- 
ences of tone and emphasis in the Siris do not prove a conversion. 
For Berkeley in 1710, esse was concipi (by God) as well as percipi 
(by man). I believe the same to have been true in 1744. 



THE ALLEGED WITHDRAWAL 149 

way of showing the solidarity of Berkeley's thought is to 
trace through all his philosophical works his pivotal doc- 
trine of abstraction. Accordingly this section will outline 
Berkeley's discussion of abstract ideas in the Theory of 
Vision, will proceed to a detailed analysis of the doctrine as 
it appears in the Principles, and will then briefly trace the 
subsequent history of the doctrine in Berkeley's other philo- 
sophical works. 

A limited criticism of abstraction had been in the fashion 
for some time before Berkeley wrote. Newton's preface to 
his Principia mentions the passing of substantial forms and 
occult qualities. Cheyne opens his Philosophical Principles 
of Natural Religion (1705) with a rejection of the Universal 
Soul of Plato, of the Substantial Forms of Aristotle, of the 
Omniscient Radical Heat of Hippocrates, of the Plastic 
Virtue of Scaliger, of the Hylarchic Principle of Henry 
More. Berkeley recorded in the Commonplace Book (626) 
this charming gallery of ghosts, and reproduced it in the 
Hylas dialogue (Works, vol. i, p. 479). We have now to 
watch the steps by which this limited criticism of the 
pseudo-scientific abstraction was extended by Berkeley to 
the whole field of knowledge. Berkeley's studies of ideas 
and words had led him to see in the abstract idea the chief 
bar to the acceptance of the New Principle, and the chief 
buttress of .atheism, scepticism, and materialism. He 
desired direct awareness, and rejected the abstract medium, 
which, plausible as a mediating function or operation of the- 
mind, became a self-confessed figment and phantom, when 
he tried to give it shape and substance as Locke had done 
with his general triangle. If the geometrician is to be allowed 
to say that he can frame such an idea and that his science is 
built upon it, nothing can invalidate the titles of the abstract 
ideas of extension, of external existence, of matter. Berkeley 
is not simply attempting to discredit Locke. He had the 



i 5 o THE ALLEGED WITHDRAWAL 

greatest respect for Locke. But in fair fight for truth he 
takes Locke's triangle as the typical development of neo- 
scholasticism. The abstract triangle is typical. If it has to 
be 'all and none of these at once', so too has the abstract idea 
of colour. Abstract colour must be red, blue, and green, all 
and none of these at once. Probably Berkeley was correct 
in seeing the abstract idea in Locke's fourth book as the 
inevitable goal and grave of Lockian epistemology. 

The New Theory of Vision Abstract Ideas 

Sections 122-5 lk like an addition to the original 
design of the Theory of Vision. 1 They deal with the abstract 
idea of extension, as the supposed object of geometry, and 
the supposed idea common to both senses, sight and touch. 
They aptly round off the argument, and are in no way 
an inapposite addition. Only, when Berkeley began the 
Commonplace Book and soon afterwards planned the Theory 
of Vision , he still held the Lockian view of abstraction. 
Berkeley in 122-5 deals explicitly with but one abstract 
idea, that of extension, but it is clear that when he wrote the 
passage he had already reached his final position, and was 
prepared to condemn all abstract ideas. At the end of 125 
he speaks of the source of the error (as to extension) as being 
also the source of errors in all parts of philosophy and 
science, and he hints at a fuller treatment of the vast and 
comprehensive subject. The remainder of the book is con- 

1 v. above p. 37, and note that the last two sentences of T.F., 
124 must have been written after Nov. 29, 1708, on which day 
he wrote (Works, vol. iii, p. 368) of 'all knowledge is about uni- 
versals', 'I could never bring myself to comprehend this doc- 
trine*, subsequently correcting to *I can by no means see the 
necessity of this doctrine*. The guarded admission of the doc- 
trine in T.V., 124 should be compared, but compared carefully, 
v. infra, p. 156, with the downright 'to which I fully agree* of 
Intro, to Prin. t 15. 



THE ALLEGED WITHDRAWAL 151 

cerned with other matters, but it contains more than one 
reference to abstract extension as disproved and one strik- 
ing mention of abstract motion, previously not named, as 
if all abstract ideas were already dead and buried ( 137). 
The formal disproof ( 122-3) f extension in abstract 
is simply an abbreviation of the disproof of abstract ideas. 
There is first the statement of what the term means. Ab- 
stract extension is extension stripped of all sensible qualities. 
The space that is to be both seen and touched must be a 
mass of paradoxes like Locke's triangle. It must be stripped 
of colour, of roughness or smoothness ; otherwise it would 
be this concrete space. Yet it must have colour to be seen 
and tangible qualities to be touched. It must be thought 
to have colour in order to be thought to be seen, and so on. 
It must be stripped of figure and size ; yet it must contain 
all figures and sizes, being the supposed object of geometry. 
After stating what the term means, Berkeley simply says 
he cannot answer for other folk, but for himself such an 
abstract idea is perfectly incomprehensible. He has tried 
and failed. He has tried 'all the ways'. 1 An empiricist must 
go by experience. This appeal to experience is decisive for 
Berkeley and recurrent in his works. We find it in the 
Commonplace Book (437) about the abstract idea of exten- 
sion, just as in this passage from the Theory of Vision. We 
find it about all abstract ideas in the Principles (Intro., iq), 
and twenty years later (1730) writing to Johnson he says 
of the abstract idea of existence, 'I cannot find I have any 
such idea, and this is my reason against it'. He points out 
in passing that the universality required for general know- 
ledge is not endangered by his rejection of abstract exten- 
sion. He does not pursue the topic, but passes at once to 

1 He can neither perceive, imagine, nor any wise frame in his 
mind such an abstract idea. T.F., 123. The point is important 
because attempts have been made to limit Berkeley's idea to image. 



152 THE ALLEGED WITHDRAWAL 

the 'killing blow* Locke's abstract idea of a triangle ( 125). 
He gives more space to it in the Principles. But he is quite 
as effective here, perhaps more effective because here he 
quotes Locke against himself. 1 

Berkeley knew quite well that the problem of the sorts 
is bound up with the problem of abstraction. It is tan- 
talizing to find him coming up to the problem of the sorts, 
and sheering off ( 128). Light and colours are mentioned 
as a species or sort, and the terms, homogeneous, hetero- 
geneous, specify, denominate, patterns, and kinds are pro- 
minent. The main theme in this part of the Essay is 
that of the two sorts, visible and tangible ideas. He 
says ( 128), 'When, upon perception of an idea, I range 
it under this or that sort . . .' and there he seems to be 
on the point of giving his theory of sorts. Yet nothing 
clear issues, and the paragraph leaves the reader with the 
impression that the mind both makes the sorts and finds 
them. It would seem that when Berkeley was writing the 
last part of the Theory of Vision (probably in the autumn 
of 1708) he had already reached and stereotyped his attitude 
towards the Lockian abstract ideas, but that while interested 
in the sorts and assuming their reality, he had found no 
theory as to their relation to the particulars of sense. 

The 'Principles' and Abstract Ideas The Introduction 
and the Draft Introduction 2 

Locke (iv. 6. 13, 16) placed the sole source of general 
knowledge in the contemplation of our own abstract ideas. 

1 iii . i o . 3 3 . The reference shows that Berkeley gave careful con- 
sideration to Locke's words in bk. iv. 7. 9, 'an idea wherein some 
parts of several different and inconsistent ideas are put together'. 

2 The Draft Introduction is in autograph in the Chapman 
MS. note-book (D. 5. 17) in the T.C.D. Library. References to 
it here use the abbreviation Dft. Int., followed by the page- 
number of Eraser's vol. iii of the Works (1901) in which it is 



THE ALLEGED WITHDRAWAL 153 

Berkeley takes up the phrase 'contemplate abstract ideas' 
ironically in the Commonplace Book entry 761, and repeats 
that entry closely in the Draft Introduction (p. ^i^prima 
manu). Noting also that the previous entry 1 expresses the 
spirit of the exordium of that Introduction and to a large 
extent its words, 2 we can see clearly the dominant thought 
in Berkeley's mind on November 15, 1708, on which day 
he began to draft his famous attack upon abstract ideas. 
The 'something in every drop of water, every grain of sand* 
is the real essence of scholastic philosophy. The Abstract 
Idea is the Lockian substitute, whose other name, 'nominal 
essence', betrays its parentage. Therefore 'the opinion that 
the mind hath a power of framing abstract ideas or notions 
of things' is the chief cause of perplexity in speculation, and 
of error in almost all parts of knowledge. 

The plan of the Introduction is as follows : 

1-5. Exordium, placing the problem in perspective. 

6-10. Analysis of the alleged faculty of abstraction. 
Reason for rejection. 

11-17. Criticism of Locke's arguments in favour of 
abstract ideas. 

18-20. Language the source of the error. 

21-25. Mainly about remedies. 

In the analysis of the faculty, qualities or modes are first 
treated, then beings. 3 As regards qualities, two stages of 
the process are recognized. There is first the singling, i.e. 
the considering separately the qualities combined in a* 

printed. References to sections of the Draft are references to pas- 
sages in the Draft which correspond to the sections of the Intro- 
duction. 

1 760, like 761, marked in text and margin for Introduction. 

2 The opening paragraph of Dft. Int. contains the words 
'Something they imagine to be in every drop of water, every 
grain of sand', which now appear in Prin., 101. 

3 v. supra, p. 146. 

4046 x 



i 5 4 THE ALLEGED WITHDRAWAL 

thing, as colour or extension. Then there is the framing 
or putting together the common qualities of several things. 
This second stage goes beyond the mere observation of a 
quality in repeated instances. The mind is here not only 
selective but constructive. It takes abstracts of abstractions 
and frames them. The two stages are roughly abstraction 
without the sorts and abstraction with the sorts. In the 
case of beings, the mental separation is stated to be the same ; 
but here the singling is not so much in evidence, and the 
stress falls on the putting together of the common qualities 
of several beings, and thus forming abstract ideas as e.g. 
man, animal, body. It would not seem that Berkeley mis- 
represents or distorts Locke's account. But neither Locke 
nor Berkeley gives crystal clear accounts, and indeed we 
may say in excuse for the obscurity of two masters of plain 
English, that, supposing Berkeley right, it must be hard to 
give lucid accounts of how a nonentity is made. 

Berkeley at once 1 denies that he has this faculty of 
abstracting. He has something like it, constructive imag- 
ination, but that is not the same. He can single and imagina- 
tively combine the separable. He cannot single the insep- 
arable, nor work it up into a composite entity. 2 

Berkeley passes on to consider Locke's defence of abs- 
traction. In the first place the alleged faculty puts a 
perfect distinction between man and beast. In the Com- 

1 In the ist ed., and it would be absurd to take the omission 
of the words from the and ed. J I dare be confident I have it not* 
as a withdrawal. Berkeley was fond of making small stylistic 
omissions and insertions in later editions. Here the equivalent of 
the words omitted is emphatically stated more than once in the 
remainder of the paragraph ( 10). 

2 'The two proper acceptations of abstraction', inf., 10, should 
be bracketed with the * Twofold abstraction* of Prin., 99. This 
latter passage gives, I think, Berkeley's clearest, if briefest, account 
of the false abstraction; v. supra, p. 127-8. 



THE ALLEGED WITHDRAWAL 155 

monplace Book (766) Berkeley had taken this point seriously 
and had suggested an alternative differentia. Here he 
passes it off with a joke; he is entitled to do so because 
he is proceeding to discuss the connexion between speech 
and abstraction, which is Locke's real point. 

The strongest argument is drawn from the use of words. 
There are general names; so there must be general ideas. 
Berkeley replies that the ideas signified by these names are 
not general, in the abstract sense of the word, but general- 
ized. They are particular ideas deputizing for other par- 
ticul^r ideas. He is careful to point out that he is not 
denying general ideas absolutely, but those framed by 
abstraction. Particulars become generals, he holds, not 
by abstraction, but by signification. They are made to 
signify other particulars. To illustrate further the mean- 
ing of abstraction he brings in again Locke's abstract tri- 
angle, and again administers 'the killing blow'. He then 
scores a few points off Locke with reference to the diffi- 
culty of abstracting. If abstract ideas are so difficult, why 
should they be so necessary for communication, easy and 
natural as it is ? Again, neither youth nor maturity can be 
the age at which the difficult task is learned (14). 

Locke's contention that abstract ideas are necessary to 
the advancement of knowledge is considered next. Berkeley 
meets it by distinguishing the abstract from the universal. 
He denies that scientific knowledge is knowledge of the 
abstract idea. If that were true of any science it would be 
true of geometry, 1 and he proceeds to show that it is not 
true of geometry. When we prove of a triangle the equality 
of its angles to two right angles, we prove it not of the 
abstract triangle, but of a triangle which in fact may be 
right-angled or equilateral. In the proof such facts are not 
regarded, nor relied on. Hence the proof is universal, and 

1 Dft. Int. 368. 



i 5 6 THE ALLEGED WITHDRAWAL 

need not be repeated in other particular cases. A compari- 
son of this section (15) with the corresponding passages in 
the Draft Introduction, p. 368, is of interest. In 15 he 
speaks of 'a point much insisted on, that all knowledge and 
demonstration are about universal notions, to which I fully 
agree'. For these words the Draft prima manu reads 'apoint 
much insisted on in the Schools, that all knowledge is about 
universals, yet I could never bring myself to comprehend 
this doctrine', the last words being corrected to 'I can by 
no means see the necessity of this doctrine'. 1 

At first sight it looks like a pronounced change of mind 
at short notice. But perhaps the change is rather of termin- 
ology than of doctrine. Are we to infer that Berkeley came 
round and accepted a scholastic tenet that he had previously 
rejected ? Certainly the words 'to which I fully agree' should 
not be taken as an admission that knowledge of the parti- 
culars of sense is not knowledge. The addition of the 'and 
demonstration' is marked. Probably 'all knowledge and 
demonstration' is a hendiadys, whose real import is 'all 
demonstrative knowledge' ; and if universal notions be taken 
in Berkeley's sense and not in Locke's then truly they are 
the goal of scientific knowledge. Universality, for Berkeley, 
is not the product of abstraction, nor does it consist in the 
'absolute, positive nature or conception of anything, but in 
the. relation it bears to the particulars signified or represented 
by it ' . The universality of a geometrical demonstration does 
not argue the existence of abstract extension, but, on the 
contrary, the representative character of the particular. 
The section added in the second edition at the end of 16 
does not alter Berkeley's position at all. 2 It is due, no doubt, 
to philosophic balance, and the anxiety to avoid one-sided- 
ness. The 'So far he may abstract' is a formula that occurs 
elsewhere, e.g. Prin. y 5, Ale. vii, 5. 

1 Cf. T.F., 124. 2 See above p. 127 n. 



THE ALLEGED WITHDRAWAL 157 

The section (18-20) on language as the source of the 
error about abstraction is given at much greater length 
(pp. 371-80) in the Draft, and more convincingly. A few 
months before he penned the Introduction, it had been 
axiomatic to Berkeley that every significant word stands for 
an idea. He had accepted the charge laid by Bacon, Locke, 
and others upon the New Learning to 'lay aside the word' 
and penetrate to the one true meaning and determinate 
idea. At that stage of his thought the words triangle, man, 
etc., seemed to postulate concrete universals, and his prob- 
lem was to retain them, while giving up the abstract ideas 
supposed to correspond to the words extension, existence, 
matter. Further reflection showed him that he had been 
confusing definition and idea. Where names are definable, 
we must keep to the definition ; but neither name nor defi- 
nition guarantees the one idea. Names have other uses 
besides that of evoking ideas. They determine action and 
touch the chords of feeling. Often words are no more than 
counters, having a potential but unrealized meaning. In 
arithmetic and algebra long trains of reasoning can go on 
and conclusions be reached by the aid of the symbol, with- 
out actual reference to the ideas symbolized. 

In the Draft Berkeley puts more strongly the divorce 
between word and idea. He there denies that words are 
necessary for apprehending ideas, and states that every 
appellative name has a diversity of meanings. He also ex- 
pands the notion of speech as the algebra of thought. He 
then gives an important analysis of prepositional form and 
the status of the predicate. His design is to show that its 
two terms do not require or admit two corresponding ideas. 
'Melampus is an animal' simply means, he says, 'Melampus 
has a right to be called by the name animal'. 1 Try to make 

1 'has a right to*. Perhaps a convenient Dublin idiom of 
import baffling to logicians. 



158 THE ALLEGED WITHDRAWAL 

the proposition mean more, and it becomes either self- 
contradictory (Melampus is an abstract universal), or 
tautologous (Melampus is Melampus). Another very in- 
teresting passage which has completely disappeared in the 
revision is that on the solitary man in a world of particulars 
(pp. 379-80). This experimental conceit had been much 
in his mind, as we see from the Commonplace Book (571, 
etc., see above, p. 141). It does not appear in the Principles 
but is alluded to in the third Hylas dialogue ( Works > p. 467). 
The absence of any treatment of the sorts or species in 
the Principles is noteworthy. We know from the Common- 
place Book that he saw the relevance and importance of 
the question, and had weighed the position of the sorts in 
the system of Malebranche (296). We saw him coming up 
to the problem in the Theory of Vision ( 128). Is then the 
silence of the Principles the silence of embarrassment? 
'He that knows he has no other than particular ideas' (Int. 
Prin., 24) would seem reducible, on the principle of esse 
percipi, to 'there are no other than particular ideas'. Does 
then Berkeley hold that sorts exist? A good deal can be 
learned of his views on the topic by comparing what he 
wrote in the Draft (pp. 365-6) with the corresponding 
sections of the Introduction as we now have it. Dealing 
with communication of ideas, he finds, as Locke did, that 
since there are so many ideas and so few words, one word 
has to stand for many particular ideas 'between which 
there is some likeness and which are said to be of the same 
sort'. He then wrote and erased the following judgement 
about the status of sorts. 'But then these sorts are not 
determin'd and set out by nature, as was thought by most 
philosophers. Nor yet are they limited by any precise ab- 
stract ideas settl'd in the mind, with the general name an- 
nexed to them, as is the opinion of the author of the Essay, 
nor do they in truth seem to me to have any precise 



THE ALLEGED WITHDRAWAL 159 

bounds or limits at all.' He adds that precise bounds to 
the sorts are not necessary in ordinary speech, and that if 
such there were, the sorting of particulars would not be 
so much disputed as it is. On the opposite page we find 
a note interesting as being one of the very few references 
by Berkeley to nominalism. 'Every one's experience may 
convince him that this is all that's meant by general names, 
and that they do not stand either for universal natures distinct 
from our conceptions as was held by the Peripatetics and 
generality of the Schoolmen, nor yet for universal notions 
or id^as as is the opinion of that sort of Schoolmen called 
Nominals and of the author of the Essay. 9 I think it is 
clear that Berkeley recognized nominalism (strict) and was 
anxious to avoid it. Clearly also he held the fact of objective 
similitude, and the 'existence' of the sorts, i.e. that some 
things are alike because God made them * after their kind'. 
Opinions will differ as to why he cut the passage out. Per- 
haps the printer told him the Introduction was too long, 
and he excised what was least to his main purpose. In any 
case his belief in the sorts, as there expressed, seems quite 
consistent with his rejection of abstract ideas. His silence 
may be the silence of wisdom. He has a very sober, sen- 
sible, and, may we say it, an evolutionary view of what the 
sorts are, but he knows that when they are much talked 
about they become fetishes and worse, abstract ideas. 

Other changes made by Berkeley in the Draft itself or 
in subsequent revision may be briefly noticed. They turn 
mainly on the sharp distinction between abstract and 
general that Berkeley tried to establish, but failed to main- 
tain in his terminology. In revising the Draft he makes 
a point of substituting the word 'abstract' for 'general'. 
Section 12 with its distinction between 'general' and 
'abstract general' is mostly new. He erases from 15 
his earlier denial of universal notions, and from 7 his 



160 THE ALLEGED WITHDRAWAL 

statement that abstract ideas, genera, species, universal 
notions 'all amount to the same thing*. 

The use made in the main work of the position estab- 
lished in the Introduction is direct and indirect. Indirectly 
the rejection of abstract ideas induces the modern attitude 
of realism and fosters the habit of direct awareness. The 
direct use is to guillotine misleading concepts. Existence 
( 5), External world ( 6), Matter ( 1 1), Unity (13), Being 
( 17), Presence ( 68), Occasion ( 74), Quiddity, Entity 
( 8 1 ) , Time, Place, Motion ( 97) one after another they are 
brought to summary execution. The Cartesian paradoxes 
of matter are traced to a twofold abstraction ( 99). Nor do 
ethical concepts escape. Happiness prescinded from parti- 
cular pleasure, goodness prescinded from things that are 
good, abstract justice and virtue, these notions have marred 
the truth and usefulness of ethics ( 100). Turning then to 
natural philosophy ( 101-2), he prefixes to the section the 
passage which formed the original opening of the Introduc- 
tion (v. supra, p. 153). It is directed against the Lockian 
and neo-scholastic doctrine of the real essence which causes 
the properties and yet is unknowable. He classes together 
occult qualities, mechanical causes, and real essences. He 
pays especial attention to the vis attractrix* 'the great mech- 
anical principle nowin vogue'. It is a general name for similar 
appearances, but is not an efficient cause. There is only one 
efficient cause of phenomena the will of a spirit. Thus 
Berkeley in all but words says that the New Physics is built 
on the doctrine of abstract ideas. When he comes to mathe- 
matics ( 1 1 8) he is quite explicit. The secret error in all 
branches of the science springs from the doctrines of abstract 
general ideas and external existence . In the case of arithmetic 
the opinion of the abstract idea of number has led to mysti- 
cism and idle speculation. Number is nothing apart from 
1 Cf. C.P.B. 488, 626-8. 



THE ALLEGED WITHDRAWAL 161 

numerable things. Glancing at arithmetic 4 in its infancy', he 
finds that utility invented strokes and points, and the arabic 
notation. Numbers came into use in imitation of language. 
They are, like letters, symbols that dispose us for action. 
The existence of words with no particular meaning gave 
rise to the belief in abstract general ideas. Just so, neglect 
of the symbolic character of figures has led to belief in the 
abstract unit or collection of abstract units ( 121-2). The 
most obvious error in geometry, due to abstraction, is, he 
thinks, the doctrine of the infinite divisibility of finites. 
Men ^re ready to predicate of the mysterious abstract idea 
of extension what obviously does not belong to the concrete 
ideas of sense. He grants that the theorems and demon- 
strations of geometry are about universal ideas, but refers 
to the Introduction ( 15) for the sense in which that term 
is to be understood. Again, he makes it quite clear that he 
recognizes the abstracting that issues in the generalized 
particular, deputizing for other similar particulars, but re- 
jects the abstracting supposed to issue in the abstract 
general idea. Finally (143) he notes the evil effect of the 
doctrine of abstract ideas in metaphysics and morality, 
instancing 'powers' and 'acts of the mind' as 'dark and 
ambiguous terms, presumed to stand for abstract notions'. 

Three Dialogues between Hylas and Philonous 1 

In the first dialogue all the main features of Berkeley's 
doctrine of abstraction reappear. The Preface speaks of 
the abstractions that occur in the very entrance of the 
sciences, and echoes the opening of the Introduction to the 
Principles on real essences. The subject of the abstract idea 
is formally introduced on p. 403, where the usual distinc- 
tions are drawn and the usual refutation from experience 

1 N.B. References are to the pages in Eraser's ed. of the Works 
(1901), vol. i. 

4046 Y 



162 THE ALLEGED WITHDRAWAL 

is given. The whole of the first part of the dialogue is an 
analysis of particular abstractions (colours, sounds, heat, 
pain, pleasure) singled and separated; Berkeley shows 
that, abstracted from the mind, they want reality. Passing 
to the primary qualities, he takes up in turn the wider 
abstractions, extension, figure, motion, denying that they 
can exist apart from the mind or from other sensible 
qualities. He shows the sense in which general ideas form 
the subject-matter of mathematics (403). In that context 
we find Berkeley's first published mention of 'pure intellect', 
the supposed faculty of abstract ideas. Berkeley does not 
deny 'pure intellect', but denies that it is the faculty of 
abstract ideas, and he gives it concrete 'spiritual objects, 
as virtue, reason, God'. 

In the second dialogue (436) after the review of matter 
as object, substratum, cause, instrument and occasion, the 
final appeal is to matter as the general abstract idea of 
entity, which in its turn 'hath quite vanished out of sight'. 
In the third dialogue there are occasional references to 
the doctrine, such as 'unknown natures' and 'philosophical 
quiddities' (455), but nothing systematic on the topic, and 
nothing new, except the attempted solution of the problem 
of sameness, by reference to the abstract ideas of identity 
and diversity (467). It is interesting to see that the drama 
ends with a recognition by Hylas, that Unknown Natures 
and Absolute Existence are to be rejected as the fount of 
'scepticism, and one might almost say that Berkeley has in 
the Three Dialogues dramatized the conversion of an ab- 
stractionist as well as the conversion of a materialist. The 
rejection of abstract ideas was to him no superficial or 
passing topic of interest, but one that had cut deep into his 
thoughts, and, in his view, was the corner stone of his 
philosophy. 1 

1 The section added, p. 427, in the 3rd edition, 1734, on 



THE ALLEGED WITHDRAWAL 163 

De Motu 

The De Motu was written in 1720 and published in 1721 . 
Berkeley reprinted it in the Miscellany (1752), and in his 
letters to Johnson of 1730 he refers to it more than once, 
as an important and specific part of his philosophy. It was 
much more to him than an essay submitted for a prize 
offered by a French Academy. The De Motu does for 
motion what the Principles had done for matter. The De 
Motu was, we might say, Part III of the Principles, and 
there is evidence in the Commonplace Book that his original 
plan for that work was tripartite. 1 The De Motu discloses no 
substantial change of mind. The ideist terminology is kept 
in the background ; he here speaks of 'things' where the 
earlier work spoke of 'ideas'. But his immaterialism stands, 
though naturally under the circumstances of the competi- 
tion it is not stated aggressively. He says (21) 'Duo sunt 
summa rerum genera corpus et anima'. This statement 
is not a recognition of mind and matter, but is, in effect, 
a repetition of the 'spirit and ideas' of the Principles. For 
when he has spoken of corpus as rent extensam &c., he adds, 
'Loquor de rebus cognitis : de incognitis enim disserere nil 
juvat'. 

Berkeley's doctrine of abstraction, though not formally 
stated in the De Motu> is there assumed, and treated by 
implication as of capital importance. The essay opens with 
a warning against 'voces male intellectae', such as nisus> vis y 

Malebranche's philosophy contains the rather severe judgement, 
'He builds on the most abstract general ideas, which I entirely 
disclaim'. 

1 C.P.B. 589, 865, especially when the latter entry is taken 
along with the apparatus of marginal signs, v. Appendix, p. 184. 
Apparently the *S. Mo* entries were designed mainly for Part II, 
and the < N > entries for Part III ('our Principles of Natural 
Philosophy'). 



164 THE ALLEGED WITHDRAWAL 

orgravitas, and on * Voces generates et abstractae' definitely 
states ( 7) that they are not accommodated to the natures 
of things, which are singular and concrete. They are indeed 
useful to describe concrete phenomena. But they are not 
realities, apprehensible by sense, imagination, or intellect. 
As abstract ideas they are the false explanations that obs- 
cure our view of the causes and nature of motion, and hide 
from us the vera causa, God. He instances the impetus of 
Torricelli and the vis activa of Leibniz, and he says that 
Newton did not adduce attraction as a physical quality but 
as a mathematical hypothesis. The method of sound philo- 
sophy is to abstain as far as possible from abstract and 
general notions ; and of his three formally expressed prin- 
ciples ( 66), the second is 'Cavere ab abstractionibus', the 
other two being part and parcel of the same caution. The 
result of observing these rules will be 'motus contemplatio 
a mille minutiis, subtilitatibus, ideisque abstractis libera 
evadet'. Those who place the principle of motion in bodies 
are building on obscure and general terms, and those who 
place it in mind are fortified by experience, and have be- 
hind them the authority of great thinkers from Anaxagoras 
to Newton. Similarly with regard to the nature of motion, 
if we take it as a 'simple and abstract idea* ( 43) apart from 
all that makes it sensible, we fall into the obscurities of 
Aristotle and the Scholastics. Many sections of the De 
Motu contain references to Berkeley's views on true and 
'false abstraction, and the whole work might be aptly styled 
an appeal to Vera natura' against 'abstracta mathesis' ( 70). 

Alciphron 

The Alciphron was written in Rhode Island and was 
published in 1732 (first and second editions) on his return 
to London. Berkeley's travels on the Continent did not 
alter his views on abstraction. Did crossing the Atlantic 



THE ALLEGED WITHDRAWAL 165 

affect them? No. In the first six dialogues abstract ideas 
receive little notice, the subjects treated scarcely permit- 
ting it. In the fourth dialogue the treatment of the Divine 
Visual Language involves a notice of semeiology which 
forms part of the explanation of general ideas. The ethical 
theories of Shaftesbury and Clarke, respectively, are in view 
in the phrases 'those heroic infidel inamoratos of abstracted 
beauty' (iii. 12), and 'abstract idea of moral fitness' (vi. 17). 
In the seventh dialogue the whole doctrine is formulated 
in the clearest possible manner. Each division of the sub- 
ject receives specific and adequate treatment, and the whole 
is cunningly worked into the fabric of his defence of theism 
and Christianity. The doctrine of abstraction is prominent 
throughout the dialogue. Berkeley starts from a study of 
words as related to ideas, and, in the first two editions, he 
gives a set statement of the Lockian view, with a refutation 
which is the argument of the Introduction to the Principles 
in shortened and dramatized form. Semeiology receives 
close attention, and other signs besides words are instanced, 
which raise no determinate ideas. He states that the con- 
trary opinion seems to have produced the doctrine of ab- 
stract ideas, and he draws his usual distinction between the 
true and the false abstraction. He studies closely and 
together the abstract ideas of force and grace, impugning 
the fairness of those who 'maintain the doctrine of force 
and reject that of grace ; who shall admit the abstract idea 
of a triangle, and at the same time ridicule the Holy Trinity'. ' 
Practical faith or assent is contrasted with 'abstracted faith', 
i.e. the requirement of abstract precise distinct ideas of the 
nature of the Godhead, of the person of Christ, of original 
sin. These and other references to the 'wire-drawing' of 
abstract ideas show that his rejection of the abstract had 
entered into the fabric of his faith and into his outlook 
upon life. 



i66 THE ALLEGED WITHDRAWAL 

Having to compare the bases of faith and science, he 
gives a clearer and more emphatic statement than he had 
as yet given of the place of signs in knowledge ( 11-14). 
Using the terms 'idea' and 'thing' indifferently, he asserts 
that general rules are not to be reached by mere considera- 
tion of the original ideas or particular things. The activity 
of the mind, he holds, lies in the realm of signs. 'It is not 
therefore, by mere contemplation of particular things, and 
much less of their abstract general ideas, that the mind 
makes her progress, but by an apposite choice and skilful 
management of signs/ He is here attempting a methodo- 
logy that will avoid the errors of active abstraction and 
passive particularism. We see that his rejection of abstrac- 
tion is no longer (perhaps it never was so) merely a polemic, 
but is a motive in a constructive methodology of signs. 
Applying it to arithmetic and algebra, he advances to the 
view that all sciences, 'so far as they are universal', are con- 
versant about signs as their immediate object. The things 
we know are used as steps to the unknown. The lower 
faculty deputizes for the higher, sense for imagination, 
imagination for intellect. Turning to geometry he traces 
infinitesimals and other paradoxes to abstraction, and says 
that the supposition of abstract ideas creates difficulties 
throughout the several branches of human knowledge. He 
extends this view ( 18) to the problem of freedom, and 
says that to make judgement, will, power, act, indifference, 
freedom, necessity, and the like into distinct abstract ideas 
seems to 'ensnare the mind into the same perplexities and 
errors which in all other instances, are observed to attend 
the doctrine of abstraction'. 

Berkeley's two letters to Johnson, printed by Fraser in 
his Preface to the Alciphron (Works, 1901, vol. ii) belong 
like the Alciphron to the Rhode Island period. These 
letters should be read along with the two letters of Johnson 



THE ALLEGED WITHDRAWAL 167 

that evoked them. All four are given in the Schneiders' 
Samuel Johnson, vol. ii, pp. 263 sqq. The Schneiders print 
the Johnson letters from copies kept by Johnson. I found 
the original 1 of Johnson's letter of February 5, 1730 in the 
Berkeley Papers in the British Museum. The fact that 
Berkeley kept this letter (or had a copy made), and that 
Johnson kept a copy too shows the value both sides set on the 
correspondence, which is indeed notable as being a very 
early sustained criticism of the Berkeleian philosophy by an 
acute mind, with Berkeley's systematic defence of it. In 
both letters, it will be noticed, Berkeley throws the weight of 
his defence upon what he has written in his books about 
abstraction. He regards it, as Hume did, as an original con- 
tribution to philosophy. He writes, 'Abstract general ideas 
was a notion that Mr. Locke held in common with the 
Schoolmen, and I think, all other philosophers; it runs 
through his whole book of Human Understanding', and he 
speaks of immaterialism as 'a principle or main point, from 
which, and from what I had laid down about abstract ideas, 
much may be deduced'. 

Thus from the Akiphron and these almost contemporary 
letters, it is certain that Dean Berkeley in America still held 
toj^is early views on abstraction, and in one direction had 
pushed them further. He is more conscious than he was 
of the mischief caused by abstract ideas ; at the same time 
he has become critical of the alternative. He sees the doc- 
trine of abstract ideas as right in postulating mental activity 
in the sciences, as wrong in postulating the 'black hat that 
is not there'. He sees sense particularism as rightly standing 
out for apprehension of the actual, as wrong if requiring 
intellectual quiescence. Between these extremes Berkeley 
extends and deepens his own doctrine of signs, in which 

1 I presume it is the original, but the last page is lacking. See 
Proc. R. Irish Acad. xli, sec. C. 4. p. 142. 



i68 THE ALLEGED WITHDRAWAL 

the higher faculties of mind are called into play by the 
concrete particulars of sense. 

The changes made in dialogue vii in the third (1752) 
edition call for a special study. For on the strength of them 
the absurd legend has grown up that in mature life Bishop 
Berkeley sat loosely to his early doctrine of abstraction, 
and in 1752 made a death-bed recantation. I have already 
pointed out 1 how precarious inferences from Berkeley's 
corrections in the text may be. Those who argue from the 
1752 textual omissions usually turn the blind eye upon the 
small but significant insertions of the same date, which 
point the other way, and, Berkeley's works being more 
written about than read, completely ignore those refuta- 
tions of abstract ideas which stand unaltered in all editions 
of the Alciphron. The most striking change in the third 
edition of the Alciphron is the omission from dialogue vii 
of three complete sections, 5, 6, and 7, whose titles in the 
original table of contents were : 

5. Abstract Ideas what and how made. 

6. Abstract general Ideas impossible. 

7. In what sense there may be general Ideas. 
Eraser's first comment is very guarded (Works, vol. ii, 
p. 323). He writes 'the omission is significant if it means 
dissatisfaction with his former method of assailing "ab- 
stract ideas" '. In his vol. iii, p. 91 , he gives the inconsistent 
and, I submit, misleading footnote, 'Note also Berkeley's 
'reasonings in the first and second editions of Alciphron and 
his withdrawal in the third edition against abstract general 
ideas'. A priori such a withdrawal is highly improbable. 
A philosopher who published a tenet at the age of 25 might 
well at the age of 47 see fit to withdraw it. But if, as Ber- 
keley did, he published it at the age of 25 and renewed it 
in an intensified form at the age of 47, reasserting it at the 

1 See above p. 154, n. i. 



THE ALLEGED WITHDRAWAL 169 

age of 59 and later, a withdrawal at the age of 67 would be 
scarcely credible. In this case the tenet was no triviality, 
but one that, as we have seen, had entered into the deep of 
Berkeley's faith and attitude to life and into his outlook on 
science. If his intellect had wanted to withdraw it, would 
his heart have consented ? And have we any solid reason 
for thinking that he wished to withdraw ? The De Motu 
contains, indeed constitutes, a slashing attack upon abstract 
ideas. If Berkeley in 1752 wished to withdraw his polemic 
against abstract ideas, why did he republish the De Motu in 
that year? Lastly, had he become convinced that his early 
teaching was wrong, he was man enough to withdraw it 
openly. A clandestine withdrawal of so supreme a tenet is 
unthinkable in the case of one who had 'dedicated his age 
as well as youth at the altar of Truth'. 

But hypothetical reasoning is unnecessary here. In point 
of hard fact the 1752 edition of the Alciphron does not with- 
draw Berkeley's doctrine of abstract ideas. It omits three 
sections that expounded the doctrine ; but that is a different 
matter. Whatever the reason for the omission of the three 
sections the doctrine contained in them is not withdrawn. 
To withdraw that doctrine, Berkeley would have had to 
withdraw the bulk of the dialogue. Of its thirty-one sec- 
tions, fourteen sections (those on which the weight of the 
argument falls) assert or assume the falsity of the abstract 
idea. In 5 (1752) Alciphron asks 'Will you not allow then 
that the mind can abstract?' Euphranor replies, 'I do not 
deny it may abstract in a certain sense : inasmuch as those 
things that can really exist, or be really perceived asunder, 
may be conceived asunder, or abstracted one from the 
other; for instance, a man's head from his body, colour 
from motion, figure from weight. But it will not thence 
follow that the mind can frame abstract general ideas, 
which appear to be impossible. (See the Principles of 

4046 z 



iyo THE ALLEGED WITHDRAWAL 

Human Knowledge^ 135, and the Introduction, 20.)' 
That question and answer with his formal reference to the 
Principles seems proof positive that Berkeley's doctrine of 
abstraction was the same in 1752 as it had been in 1710. 
Further, far from withdrawing his line in old age, he con- 
solidates it. In 1752 he inserts in 12 and 14 sentences 
that bring relations expressly under his doctrine of signs, 
taking care to point out that 'these relations are not ab- 
stract general ideas'. 

Why then did Berkeley omit these three sections? He 
does not tell us, and we can only surmise. Perhaps he cut 
them out because they are redundant and can easily be 
spared. Perhaps because they argue a point that to his 
mind was long ago settled and past argument. He did not 
want to be flogging a dead horse. Most probably, I should 
think, he cut them out in order to make his book more effec- 
tive. He was not writing for metaphysicians. He was writ- 
ing to convince folk of the truth of Christianity. These 
sections occur at the climax of the whole work. The Free- 
thinker has promised a demonstration that Christianity is 
false ( i). The kernel of his 'demonstration' is that the 
word 'grace' on which Christianity depends is a mere word, 
and that no idea in the mind answers to it ( 4). As we saw 
above, Berkeley's discovery that there may be significant 
wojds without ideas was epoch-making in his Commonplace 
Book days, and gave a great impetus to his attack on abstract 
ideas. So it was natural that as he wrote under the ' Hanging 
Rocks' at Rhode Island, the old argument should remind 
him of the old enemy, and that he should pen the three 
sections, 5, 6 and 7, unfolding the refutation that had meant 
so much to him twenty years before. But it was no less 
natural that on revising the work amid the myrtles at Cloyne 
in the last year of his life, he should regard the same three 
sections as inartistic, and as breaking the continuity of the 



THE ALLEGED WITHDRAWAL 171 

main argument. The Free-thinker's argument is so telling 
that it calls for a speedy and a crushing retort. That retort 
is forthcoming in the fine adhominem argument of 7, 'That 
which we admit with regard to force, upon what pretence can 
we deny concerning grace ?' I suggest that Berkeley cut out 
the three sections in order to hurry to that 'killing blow*. 
The three sections give a technical exposition and a techni- 
cal refutation of a metaphysical doctrine. They are 'cap and 
gown' philosophy. They are of 'that dry, formal, pedantic, 
stiff and clumsy style, which smells of the lamp and the 
college' (dial, v, 20). They interrupt and delay the im- 
portant argument about grace and force, and weaken its 
appeal to the average cultured reader. The dialogue reads 
better without them. 

The Analyst (1734). A Defence of Free-Thinking 
in Mathematics (1735) 

These mathematical works are more closely connected 
with Berkeley's metaphysic than appears at first glance. 
In both of them his doctrine of abstraction is decisive, if 
not prominent. The immediate purpose of the Analyst is 
pithily expressed in the words ( 7) 'He who can digest a 
second or third fluxion, a second or third difference, need 
not be squeamish about any point in divinity'. But the 
scope of the work extends beyond that argumentum ad 
hominem. The Analyst does not mention immaterialism, 
but is in fact a continuation of his subtle defence of that* 
doctrine, begun a quarter of a century previously. Its con- 
cluding section (50) connects it with the 'hints given to the 
public about twenty-five years ago', and all but says that he 
has here brought to fruition previous studies 'after so long 
an intermission'. The 'hints given to the public' were 123 
-32 of the Principles on the infinite divisibility, infinitesi- 
mals, and 'amusing geometrical paradoxes', a division of 



172 THE ALLEGED WITHDRAWAL 

the work which ends with a promise of fuller treatment. 
The 'studies' intermitted must have been those of the 
Commonplace Book which centre round these mathematical 
problems. There are well over a hundred of these entries, 
on fluxions, infinitesimals, surds, angle of contact, quad- 
rature of the circle, &c., a medley of topics, all dealing with 
the mathematicians' argument for the existence of matter. 

There is indeed a personal link between the Analyst and 
those early studies which I have not seen noticed. The 
'infidel mathematician' to whom the Analyst is addressed is 
Halley, 1 second only to Newton in fame. He is mentioned 
by name in the Commonplace Book (445), ' Halley 's Doc- 
trine about the proportion between Infinitely great quan- 
tities vanishes. When men speak of Infinite quantities, 
either they mean finite quantities, or else talk of (that 
whereof they have) no idea; both which are absurd.' Now 
when Berkeley was penning the sixty-seven Queries with 
which the Analyst concludes, he might have had the Com- 
monplace Book open before him ; for a score or more of them 
echo its entries ; and it is interesting to see that the very first 
of these queries takes up the point of the entry which men- 
tions Halley. The query is, 'Whether the object of geometry 
be not the proportions of assignable extensions? And 
whether there be any need of considering quantities either 
infinitely great or infinitely small?' 

The Analyst aims at showing that the method of fluxions 
"rests upon abstract ideas, impossible to conceive, because 
compounded of inconsistencies ( 4). Berkeley instances 
'velocity prescinded from time and space' (30), 'nice ab- 
stractions and geometrical metaphysics', 'evanescent incre- 
ments', and similar 'ghosts of departed quantities' ( 35). 

1 I have assumed, perhaps rashly, the truth of the tradition, and 
should add that Professor Jessop (A Bibliography of George Berke- 
ley y p. 10) has shown that a great part of Stock's story is unreliable. 



THE ALLEGED WITHDRAWAL 173 

We find in 37 the usual appeal to personal experience as 
to the impossibility of forming an abstract idea (of velo- 
city). Since the method of fluxions was held to be the key 
of geometry and therefore of Nature ( 3), it is clear that 
Berkeley is dealing not only with a mathematical method, 
but with the nature of existence, and his early philosophy 
is present by implication in several of his concluding 
Queries, especially in No. 7, 'Whether it be possible to 
free geometry from insuperable difficulties and absurdities, 
so long as either the abstract general idea of extension, 
or absolute external extension be supposed its true ob- 
ject?' 1 

In A Defence of Free-Thinking in Mathematics ( 12) the 
continuity of the Analyst with the Principles is clearly 
stated: 'My late publication of what had been hinted 
twenty-five years ago'. Its argument is for the most part 
narrowly controversial, and need not detain us. In two 
sections, 20 and 49, Berkeley lists the 'arcana of the modern 
analysis'. He writes, 'Every reader . . . knows . . . what 
idea he frames or can frame of velocity without motion, or 
of motion without extension, of magnitude which is neither 
finite nor infinite, or of a quantity having no magnitude 
which is yet divisible, of a figure where there is no space, 
of proportion between nothings, or of a real product from 
nothing multiplied by something*. It is quite easy to recog- 
nize here a more pointed and practised use of the old 
weapon. Perhaps 45-48 form the most interesting part 
of the tract; for they show us that his opponent, Dr. Jurin 
of Cambridge, recognized that the doctrine of abstract ideas 
was the real point at issue in the Analyst controversy, 
attacking Berkeley on that point of metaphysics. Accord- 
ingly Berkeley with great clearness restates his old doctrine, 
distinguishing between abstract general ideas and general 
1 v. also Nos. 8, 9, 14 and cf. T.V. 124. 



174 THE ALLEGED WITHDRAWAL 

ideas, giving his reasons for rejecting the former, and 

elucidating his relation to the Lockian doctrine. 1 

Siris 

Siris was first published in 1744. It contains no formal 
treatment of abstract ideas or kindred subjects. This need 
cause no surprise. The surprising thing would be if he had 
returned to labour an argument which he regarded as 
already established. The third (1752) edition of the Alci- 
phron makes it clear that his early views on abstraction 
remained unchanged to the close of his life. 2 The Siris 
contains no suggestion of any such change. We find in it, 
indeed, explicit recognition of the doctrine of the Intro- 
duction to the Principles. Plato's ideas are distinguished 
from 'abstract ideas in the modern sense* ( 335). Having 
to mention Plutarch's term for Deity, ^cu^otarov efSos, sep- 
arated Form, Berkeley adds, 'not an abstract idea com- 
pounded of inconsistencies, and prescinded from all real 
things, as some moderns understand abstraction' ( 323). 
The allusion to Locke's 'absurd triangle' is unmistakable. 
The Siris contains express approving references ( 249 , 27 1 ) 
to the De Motu and the Analyst. Berkeley then in 1744 
takes for granted the main results of his earlier work. An 
opinion to the contrary would have to maintain that in 
the Siris he has made substantial alterations in the doctrine 
of the Principles. It would be hard, if not impossible, to 
establish that contention. Certainly it is not the professed 
intention of the Sins to recant, correct, or even modify the 
teaching of his early works. 

As to the merits of the book in general, the true estimate 

must lie somewhere between Eraser's judgement that it is 

the most profound of Berkeley's works (Works, vol. iii, 

p. 117) and its author's own statement that it is a 'rude 

1 v. supra, pp. 130-2. 2 v. supra, pp. 168-71. 



THE ALLEGED WITHDRAWAL 175 

essay 5 ( 297). It is certainly not designed as the exposition 
of a new metaphysic or of any metaphysic. It is a treatise 
on tar- water combined with a meditation, whose one un- 
faltering purpose is to show that God worketh, if not all, 
at least in all. No doubt on that account Berkeley does not 
trouble to make clear his position as regards the authors he 
quotes. Like a bee among flowers he lays antiquity under 
tribute, and wherever he can suck the honey of the universe 
its unity in God thither he flits. 

The abstractions of physics, the elastic ether, gravity, and 
force, are attacked in Sirisjust as they were in the De Motu, 
and are coupled with hamate atoms', as explaining ignotum 
per ignotius. The particular phenomenon is still the reality 
( 23 1). Knowledge is still the interpretation of signs ( 253). 
Absolute space and absolute motion are still phantoms 
(271). The idea- things are still there in all but name, part 
of the cosmic chain, but not receiving emphasis dispropor- 
tionate to their value. Opposing Cudworth's statement 
that 'The Democritic hypothesis doth much more hand- 
somely and intelligibly solve 1 the phenomena, than that of 
Aristotle and Plato', he states that all phenomena are appear- 
ances in the mind, and denies that external bodies, figures, 
or motions can produce an appearance in the mind ( 251). 
Berkeley has simply substituted the Latin 'appearance* for 
the Greek 'idea*. In 292 he says of these 'appearances' 
just what he had said of 'ideas' in the Principles. 'They are 
such as we see and perceive them. Their real and objective 2 * 
natures are, therefore, the same: passive without anything 
active; fluent and changing without anything permanent 
in them/ This passage is in the nature of a climax; for 
Berkeley passes at once to the contemplation of a new 

1 Sic. Cudworth has 'salve', i.e. <7e6eiv. 

2 'objective* here means perceived. It is almost equivalent to 
subjective in modern usage. 



176 THE ALLEGED WITHDRAWAL 

and distinct class of objects, 'the Mind her acts and 

faculties'. 

What Berkeley says in Sins about the human mind is 
not extensive or profound or very clear. He says at least as 
much in the Principles, and writes there certainly more 
systematically about it. It is more than doubtful if he 
accepts the * native inbred notions of the Platonic philo- 
sophy.' 1 In Berkeley's last public pronouncement that re- 
mains, a sermon on the Will of God, 2 we find a remarkable 
passage on innate ideas. This passage is modelled on, and, 
in part, reproduces Alciphron, dial, i, 14, and echoes 
certain thoughts and phrases from the sections of Siris to 
which we have just now referred. There Berkeley, ex- 
pressly in the draft copy of the sermon, by implication in 
the British Museum fair copy, rejects actual original notions 
in the mind, and substitutes 'natural inbred dispositions' 
for the 'native inbred notions' of Siris. 

In the final sections of Siris Berkeley soon passes from the 
psychology of the human mind into 'remote inquiries and 
speculations'. Here too there is nothing advanced as his 
own, inconsistent with his earlier writings. Plato was to 
him the prince of ancient philosophers. 3 But Berkeley was 
no Platonist. He no more accepts the Platonic Ideas, be- 
cause he expounds them appreciatively, than he accepts 
Plato's doctrine of matter, or Aristotle's, because he ex- 
pounds them with obvious relish and a degree of sympathy 
intelligible in an immaterialist. Berkeley finds an 'imper- 
fect notion' of the Holy Trinity in Plato and Plotinus. He 
no more understood the Church's tenet in a Platonist sense, 

1 Siris, 308-9. N.B. 'Some perhaps may think. . . .' 

2 Preached Whitsunday 1751, published Hermathena, vol. 
xxii, 1932; v. especially pp. 9, 10, 30-1. 

3 N.B. 'whose hoary maxims scattered in this Essay are not 
proposed as principles, but barely as hints', 350. 



THE ALLEGED WITHDRAWAL 177 

than he attributed efficacy, force, or material extension to 
tar, when, speaking with the vulgar, he claims that tar- 
water cures diseases. 

The closing problem of Siris is the nature of the Supreme 
Being. Making 'the best of those glimpses within our 
reach', Berkeley conceives that 'God may be said to be All 
in divers senses'. He tentatively discusses the relations of 
the Persons in terms of TO "Ev and Novs. But the stress 
falls on his unwavering conviction that the Substance of 
Deity is 'the cause and origin of all beings . . . the same 
which comprehends and orders and sustains the whole 
mundane system' ( 328). The argument of this section 
and, in fact, of the whole of Siris expresses and, some may 
think, justifies the intuition that inspired his youth. In 
his last book Berkeley delineates, in the light of enriched 
experience and with some colours of sunset, his early vision 
of all things in God, 

'That Spring of Life which this great World pervades, 
The Spirit that moves, the Intellect that guides, 
Th' eternal One that o'er the Whole presides.' 1 

As a young man in Trinity College Berkeley read much, 
thought much, saw clearly, and wrote as for far off future 
days. In the quiet of Cloyne he read much, thought much, 
learned much, but unlearned little or nothing. He was in 
the year 1744, as he had been in the year 1710, and as -he 
remained till the closing scene at Oxford, an unrepentant, 
immaterialist, anti-abstractionist, theist, and Trinitarian. 

1 From verses prefixed to some copies of Siris, 2nd ed., 
Dublin 1744, found by Professor Jessop, and reprinted in Proc. 
R. Irish Acad. y 1933, vol. xli, sec. C. 4, p. 159. 



4046 A a 



APPENDIX I ' 
BERKELEY'S COMMONPLACE BOOK 2 

THE date, the purpose, the structure and the marginal signs of 
the Commonplace Book are closely connected subj ects . The argu- 
ment as to date turns largely upon the structure and purpose of 
the work. The apparatus of marginal signs, which, for its own 
sake, deserves more attention than it has yet received, can also 
throw much light upon the questions of structure and purpose. 

If the Commonplace Book be a book, that is a purposeful 
composition, 3 a fairly precise date for it is a desideratum. If 
otherwise, if, that is, Add. MS. 39305 be merely a pair of old 
note-books filled with random jottings, its date would not 
matter much, and the accepted date, 1705-8, would be near 
enough for current uses. At present, writers take quotations 
from the Commonplace Book to adorn their tale, much as 
theologians of the old school used quotations from Scripture. 
A serious use of the work in a study of the sources and the 
growth of Berkeley 's system is nearly impossible while its date 
remains uncertain and its purpose obscure. 

Dr. G. A. Johnston and Professor R. I. Aaron have pub- 
lished valuable studies, to which I am indebted : but with a new 
angle of approach and some new evidence, we may perhaps 
carry the solution of these problems somewhat beyond the 
point they reached, 

The Purpose 

The Times has styled the Commonplace Book 'Berkeley's 

1 Reprinted by permission without substantial alteration from 
Hermathena, vol. xxii, 1932. 

2 Johnston's numbering of the entries is followed in this 
Appendix. 

3 Among the definitions of 'composition* the Oxford Eng. Diet. 
gives (21 b) *A train of ideas put into words'. In that sense the 
term is, I think, a just description of the C.PJE?., and one needed 
to correct the first impression left on the mind by the apparent 
discontinuity of the entries. 



BERKELEY'S COMMONPLACE BOOK 179 

Scrap-book*. That term seems to represent the prevailing view. 
In a similar vein, commentators describe the entries as 'jottings' 
or 'aphorisms'. Now there is systematic purpose in the book, 
in a higher degree than these terms would suggest. Misconcep- 
tions arise partly from overlooking the purpose, partly from 
assigning the wrong purpose. It has been supposed that the 
work could be accounted for, partly as a desultory companion 
to Berkeley's private reading, and partly as a record of subjects 
debated in his College Society. This debating Society has 
clouded the issue. It is a red herring across the trail. Berkeley 
has written a good deal in Add. MS. 39305 that has no claim 
to inclusion in the Commonplace Book proper. To draw the 
line between relevant and irrelevant material is an essential 
preliminary. We must know precisely what we are trying to 
date before we date it. I have given reasons elsewhere 1 for 
thinking it improbable that there was any real connexion 
between the work of the Society and the Commonplace Book. 
It is surely a mere accident that 'That the Junior begin the 
Conference' and 'Time a sensation, therefore onely in y e mind' 
occur on successive pages. One has only to look at page 103 
of the manuscript, to read what is there and, if I may say so, 
what is not there, to be convinced that after December 7, 1706, 
the note-book ceased to be a Society rule-book. The Common- 
place Book, in the accepted meaning of the term, begins on 
page 104, and to suppose that it or any of the pages that follow 
were filled in before page 103, while the note-book was a 
Society book, would be a gratuitous assumption. 

If one leaves the College Society out of account, the -in- 
adequacy of the scrap-book hypothesis is manifest. The entriea 
were not casually made ; they are not 'jottings'. For Berkeley 
they were sections. He calls them so on one occasion (208). 
The Commonplace Book externally is a very formal work. 2 The 
original entries are for the most part confined to the right-hand 
page, the opposite page being reserved for later comments 
and corrections. A definite margin is left throughout. The 

1 Hermathena y vol. xxi, 1931, p. 157. 2 See frontispiece. 



i8o BERKELEY'S COMMONPLACE BOOK 

beginning and end of each entry are clearly marked by capitals 
and spacings. Though the handwriting is, for the most part, 
impetuous, there are scarcely two cases of serious doubt as to 
whether entries are separate or continuous. Of course some 
diarists keep their log with great care. But if one considers this 
sustained attention to external form, along with the nature of the 
entries, and the fact that scores of them reappear in Berkeley's 
published works, in some cases verbatim, one must conclude 
that the sole and adequate purpose of the Commonplace Book 
was authorship. Even those few entries that seem trivial and 
personal have some bearing upon his writing. The entries 
seem disjointed at first sight: but familiarity and close study 
will often bring to light the connexion of thought and the latent 
argument. Berkeley modified his literary intentions as the 
work progressed, but 41, 49, 60 are indications that right from 
the start he had publication in view. The Commonplace Book, 
in my view, was a definite piece of work, undertaken by Berkeley 
as a preliminary stage in working out the argument of the 
Theory of Vision and of the Principles, just as, apparently, he 
wrote another Commonplace Book as a prelude to the Querist 
(Fraser, Works, vol. iv, p. 567, note). 

The Marginal Signs 

Of this thesis, which is not new, the apparatus of marginal 
signs furnishes new proof. As a whole, this apparatus has 
received little attention. Signs set against some hundreds of 
the entries have never yet been printed. For reference I tabu- 
late them at the end of this sketch. They might be difficult 
to print in the text, and I do not wish to exaggerate their 
importance. But is an edition of the Commonplace Book com- 
plete without them? The marginal letters are in the printed 
texts: but they look lost and lonely, because detached from the 
rest of the apparatus. Few readers know that in the manuscript 
the letters form part of a system in which practically every entry 
has a marginal sign, and some have more than one. Over 500 
entries are marked with a letter, some 300 are marked X , and 
nearly 200 are marked +. Berkeley has left an index of the 



BERKELEY'S COMMONPLACE BOOK 181 

letters. I give below my view as to the meaning of the other 
symbols and their variations. 1 

Before dealing with the details of the system, I wish to point 
out that the mere fact that the apparatus is there shows that the 
Commonplace Book is no rag-bag of philosophical odds and 
ends. The 950 entries must have been written under the con- 
trol of some design, or they could never have been brought 
within the compass of so simple a system. Further, the appa- 
ratus shows that Berkeley made a definite, careful, and calcu- 
lated use of the Commonplace Book when it was written. The 
Commonplace Book was not only his storehouse, but his work- 
shop. The initiated reader, if he understands the sign system, 
can often tell in the case of a given entry, merely by looking at 
its sign, whereabouts in the published works to find it, or at 
least where not to look for it. Most paragraphs of the Theory 
of Vision and of the Principles still show the influence of the 
Commonplace Book. The same influence is most marked in the 
case of the original (draft) Introduction to the Principles. 2 And 
it is natural to infer that the lost Part II 3 of the Principles made 
use of the large group of entries dealing with morals and 
psychology. 

Again, apart from what the marginal apparatus meant for 
Berkeley, it serves us a good turn. It provides an objective 
criterion as to what constitutes the Commonplace Book proper. 
There is a good deal of material in one of the note-books 
that has obviously no connexion with the metaphysical entries. 

1 Most of them are shown on the photograph of MS., p. r 18, 
which forms the frontispiece of this book. 2 See below, p. 202 % 

3 The Preface to Hylas, which mentions Part II, adds point 
to this suggestion. This Preface has made clear use of the six 
entries on p. 4 Add. MS. 39304, which Fraser published as part 
of the Commonplace Book (Works , vol. i, p. 92). Four of these 
appear in substance, the first almost in terms. For * Giant and 
Dwarf ' cf. 689. The third dialogue in several places shows the 
influence of the Commonplace Book: e.g., Works, vol. i, p. 479, 
'substantial forms, hylarchic principles, plastic natures . . . 
possibility of Matter's thinking': cf. 626, 707. 



182 BERKELEY'S COMMONPLACE BOOK 

There is some border-line material. And editors have been 
somewhat subjective in forming their canon. I suggest that 
no material to which Berkeley has not applied the marginal 
apparatus should be regarded as part of the Commonplace Book. 
This test would exclude two sections that Johnston includes: 
(i) The * Queries'. 1 These are simply odd notes on Locke's 
Essay. They occupy page 102 of the manuscript, recto and 
verso. This was a blank page between the two codes of Rules. 
The 'Queries' were written after December 7, 1706: for page 
103, which contains the statutes of that date, has subsequently 
been used as a blotting-pad for the verso of 102, which contains 
part of the * Queries'. Probably they were written before the 
Commonplace Book was begun: but in any case the absence of 
marginal signs shows them not to be part of the Commonplace 
Book. (2) The same applies to 946-53. 951 has been brought 
into the Commonplace Book as 151, and is there duly indexed 
S. Mo. These entries too are merely notes on Locke's Essay. 
Eraser did not include them. They have little claim to inclu- 
sion. They occur on p. 164, verso: but they could not have 
been written or read continuously with the main body of 
entries: for they are upside down. 

Thus the apparatus of marginal signs supplies a useful 
criterion. It serves to distinguish the Commonplace Book proper 
from the other contents of the famous note-books. It confirms 
the view that Berkeley wrote it as a prelude to his published 
works. Now to keep such a book and to invent such an apparatus 
might naturally occur to any author. But if there is no conse- 
quence, it is a curious coincidence that shortly before he began 
his Commonplace Book (Eraser gave it that name) Berkeley came 
across John Locke's New Method of a Commonplace-Book. 
Locke says that he had himself used the method for five and 
twenty years. It is a highly elaborate system of entries and 
index signs. The tract is the last item in the Posthumous Works 
of Mr. John Locke (1706). Berkeley names this work at the end 
of his De Ludo Algebraico, published early in 1707, and praises 
it, or parts of it, highly. 

1 Numbered i-xxiv by Johnston. 



BERKELEY'S COMMONPLACE BOOK 183 

The marginal sign system falls into two divisions, approxi- 
mately equal : letters of the alphabet and mathematical symbols. 
As regards the letters, I can be brief. The system is self- 
explanatory. Berkeley has prefixed a table, showing the mean- 
ing of each letter, to the second of the two note-books (page 3 
of the manuscript). The sign system brings out the distinction 
between the two note-books, and I shall have occasion to refer 
to them separately. So I here note that, following Johnston, 
I designate as A the note-book now appearing as the first 
(MS., pp. 3-95), and as B the other note-book (pp. 96-180). 
The index of letters is in its proper place: for the existing letter 
system is clearly original in A, and has been extended to B, 
and one might perhaps argue that the mathematical symbols 
were original in B, and have been extended to A. 

The distribution of index letters shows the distribution of 
subject-matter, so the following table may be of some interest : 

No. of Entries 

Index Letter Note-book A Note-book B 

I. Introduction ... 58 4 

(i.e., to the Principles). 

M. Matter .... 66 62 

P. Primary and Secondary Qualities 18 17 
E. Existence .... 29 

T. Time 3 n 



S. Soul Spirit . . .120 15 

G. God 15 4 

Mo. Moral Philosophy ... 44 5 

N. Natural Philosophy 26 8 

The line between T. and S. is Berkeley's. It would seem to 
mark off the subjects of Principles , Part I, from those of the 
projected Part II (and Part III ?). It will be seen that the only 
'lettered* topic at all equally represented in both note-books is 
matter (and its qualities). There is an interesting vestige of an 
earlier sign system in B. S., which now means Soul or Spirit, 
originally in B meant Space. There are over 40 cases in B of 



1 84 BERKELEY'S COMMONPLACE BOOK 

S. stroked out, 1 and for the most part M., sometimes P., has 
been substituted. In a few cases, e.g. 98, 123, the original S. 
has been left standing: for Berkeley has omitted to stroke it 
out. But any one can see that these entries have nothing to do 
with Soul or Spirit. 

In addition to the 58 shown in the table, about 25 more 
entries in A were originally designed for the Introduction. 
The I. has been stroked out, 2 and for the most part S. Mo. 
substituted. Abstract ideas now monopolize the Introduction, 
but this theme came gradually to Berkeley in the course of his 
Commonplace Book construction. He first planned to give in 
the Introduction a theory of demonstration (592), truth, know- 
ledge, and certainty (687). It is interesting to note that there 
are no cases of E. (Existence) in B. The meaning and nature 
of Existence was the New Principle, or the most positive form 
of it, which he discovered with exultation when he was three- 
quarters of the way through note-book B, and which he pro- 
ceeds to develop in the new note-book A. Very few, if any, of 
the 'lettered* entries appear in, or are represented in, the Theory 
of Vision, and as the letters belong primarily to note-book A, 
it is clear that B was a preliminary to the Theory of Vision , and 
A a preliminary to the Principles. This is only a broad 
generalization as to ground plan. It cannot be pressed in the 
case of particular entries. For instance, the group on Time with 
which B opens is represented in the Principles, in, and the 
very first entry in A is embodied in the Theory of Vision, 145. 

I come now to the two mathematical symbols. I found the 
clue to the meaning of X in the entries 687 and 865. 

687. N. Mo. X. Truth, three sorts thereof natural, 
mathematical, & moral. 

865. N. Mo. X . Three sorts of useful knowledge: that of 
coexistence to be treated of in our Principles of Natural Philo- 
sophy, that of relation in Mathematiques, that of definition, 
or inclusion, or Words ... in Morality. 

It is quite clear that X was originally the mark for one of 
his three primary divisions of subject-matter. It is the symbol 

1 See frontispiece. 2 See below, pp. 202-3, note. 



BERKELEY'S COMMONPLACE BOOK 185 

for mathematics in the widest sense, covering geometry, exten- 
sion, and all pure quantity. There are over 300 entries so 
marked, and they all come under this broad heading. Possibly 
the little note on the back of page 9, * X this belongs to 
geometry', 1 is Berkeley's own note of the meaning of the symbol. 

From these 300 Mathematical entries' Berkeley then pro- 
ceeded 2 to select those with a bearing on the themes of his 
first book. He did so, I suggest, by two stages 3 ; (i) Against the 
selected entries he set the figure i , for the most part in an angle 
of the X (he also set the i opposite some of the M. entries). 
(2) He then distributed those so selected between the three 
main sections in the original plan of the Theory of Vision , 
marking them respectively i, 2, 3 in an angle of the X . The 
meanings thus are 

1 X 1 Distance T.V. 1-51 ; 

1 X 2 Magnitude T.V. 52-87; 

J X 3 Heterogeneity of sight and touch T.V. 121-46. 

The present third section ( 88-120) is in the nature of an 
afterthought. Nominally it is on 'situation': but in fact it is 
monopolized by the problem of the inverted retinal image. 
The marginal signs bear out this view. There is a clearly 
marked group of seven entries 4 dealing with this problem, of 
which 235-6 are remarkable for their close correspondence with 
T.V. 103, 108. All the seven, and they only, have the special 
sign, * X 3 . It would seem that Berkeley became aware of the 
capital importance of this problem as the Commonplace Book 
progressed. Years after, in his Theory of Vision Vindicated, he 
writes (52): 'The solution of this knot about inverted images 
seems the principal point in the whole Optic Theory . . . the 
surest way to lead the mind into a thorough knowledge of the 
true nature of Vision'. It is worth noting that the pages of 

1 Printed as part of 438. 

2 I am presuming that he did not insert the signs as he went 
along. The letters certainly seem subsequent to the entries, a 
few of which are marked Int., Introd., Pre., &c. 

3 I speak with less confidence about the first stage. 

4 e.g. 128, shown on frontispiece. 

4046 B D 



186 BERKELEY'S COMMONPLACE BOOK 

the Commonplace Book most charged with the joy of discovery 
are those which come immediately after his intensive study of 
this problem. The problem, I suppose, supplied him with the 
clearest proof from fact that we do not properly see what we 
think we see. 

Reference to the table on page 204 will show that X 2 Magni- 
tude and X 3 Heterogeneity are definitely earmarked for their re- 
spective subjects. X * Distance is not so unmistakable. Yet there 
are enough instances to establish the meaning, e.g. 172, 504 (the 
Harrovian case, T.V. 29, &c.); 306 (see T.V. 26, 34). Of 
those marked X + several deal with distance, and I suggest that 
in these cases the original i has been stroked through by 
Berkeley, possibly because of the confusion caused by the 
figure i doing double duty (if my conjecture is sound) for the 
first book and for the first section. 

Several entries are marked for two sections. 240 is interest- 
ing. It is marked for all three sections, and its principle, i.e. 
'unlike signs', occurs in all three sections of the Theory of Vision 

(9.33,65, H3-5)- 

The asterisk * which occurs occasionally I have not treated 

as part of the apparatus. It is always, I think, merely a refer- 
ence to a qualifying entry on the opposite page. 

I turn now to the symbol +. It occurs against 123 entries 
in note-book B, and against 65 in A. In the last quarter of A 
it is rare. My suggestion is that Berkeley used it as we might 
use a minus sign, as a mark of omission. Cajori 1 shows that 
the usage of these signs was not fixed in the seventeenth cen- 
tury. In any case arbitrary symbolism would appeal to Berkeley. 
At one time I thought it might be '-' subsequently stroked 
through. I see now that that explanation is unnecessary, and 
in most cases evidently not true to fact. It is clear 2 that 
Berkeley would need some sign to set against those entries 
which represented positions tentatively occupied and subse- 

1 History of Mathem. Notations, vol. i, 208 sqq. Note that 
4- and x and six other 'operationum logisticarum symbola' 
appear in the Tabula Lusoria: Eraser, Works, vol. iv, pp. 55-7. 

2 e.g. 'I must cancel!', 870. 



BERKELEY'S COMMONPLACE BOOK 187 

quently abandoned in the course of further thought, or against 
entries which for any other reason he decided not to use. This 
sign + is as good as any other for that purpose. Over against 
486 he has written the caution, 'this I do not altogether approve 
of, and over against 473 we find the flat contradiction 474. 
Both 473 and 486 are marked +, and it is clear that in the 
majority of cases the symbol is neater and more efficient than 
such verbal cautions and contradictions. 

I do not pretend that this conclusion is established. Those 
who have patience to look through the list of entries so marked 
(pp. 204-5) may find a better explanation. I will only say I have 
had it in mind during several months' close examination of the 
entries, and have not found a clear case of breakdown. Trivial, 
personal, irrelevant, or merely unsuitable entries are marked +- 
The sign in fact seems proper to all those entries which would 
be of no service to him in writing his books. This view is not 
inconsistent with the view advanced above as to the purpose 
of the whole work. Material amassed by an author often 
proves unsuitable as the work progresses. I draw attention to 
424. It provides, on my theory, an instructive instance of the 
working of Berkeley's mind and pen. Beside the M. appears 
in the manuscript, and the reader must remember that the 
second half of the entry is a later comment written on the 
opposite page, and marked M. Berkeley when he first made 
the entry wrote boldly, but consistently with the Principle. 
On later revision he is afraid of his own courage, and marks the 
entry + . Subsequently he reflects upon the distinction between 
mediate and immediate object. Courage returns. He recants 
his recantation by ringing the + , adds the opposite page entry., 
and marks both entries M. 

The groups 143-6, 158-62 (?i63) are marked +. I suggest 
that the reason for the sign here is prudence. All these entries, 
in my view, were made in direct criticism of W. King's De 
Origine Mali. The 'A.B.' of 144 and 161 is King, 1 the powerful 

1 I am interested to hear from M. Maheu that this explanation 
of A.B. occurred independently to him, and that he observed, 



i88 BERKELEY'S COMMONPLACE BOOK 

Archbishop of Dublin, who was subsequently to pass the severe 
censures on the Theory of Vision answered by Berkeley in the 
appendix to the second edition (see Berkeley's Letter to 
Percival, March i, 1710), and to demonstrate the imprudence 
of giving 'the least handle of offence to the Church or Church- 
men' (727) by citing Berkeley before his ecclesiastical court 
on the charge of receiving illegal ordination. 1 

If my explanation of + is correct, the symbol is the most 
important part of the marginal apparatus for students of the 
development of Berkeley's thought. For the large group of 
entries so marked will constitute a * Black List' of metaphysical 
positions tentatively held and definitely rejected. Of these the 
most important are 

The Soul knowable 25, 44, 156, 182. 
Simple Ideas 53, 135-6, and 14 others. 
Doctrine of Powers 41, 52, and 7 others. 
All Ideas from without 330, 906. 
No word without ideas 371, 904, and 5 others. 
No soul apart from ideas 583, 585-7, 631, 646. 

The much quoted 'Mind is a congeries of perceptions' is on 
the Black List. So too are most of the tasteless and unworthy 
references to mathematicians facts which illustrate the un- 
fairness of indiscriminate and uninformed quotation from the 
Commonplace Book. 

Biography and the Date 

This mass of detail must seem, I fear, trivial and unimpor- 
tant, but it all goes to build up a view of the Commonplace Book, 
essential perhaps to a sound judgement on those larger ques- 
tions of date and structure to which I now turn. The Common- 
place Book is thus shown to be no day-book, no journal for 
jottings, no scrap-book, nor collection of philosophical bric-a- 

as I did, the C A.B. of Cashel' in the Chapman MS. which proves 
Berkeley's use of A.B. as abbreviation for 'Archbishop'. 

1 See Berkeley's autograph letter to King preserved in T.C.D. 
Library. 



BERKELEY'S COMMONPLACE BOOK 189 

brae. It is on the face of it a composition with a defined 
purpose, undertaken and carried through as the scaffolding of 
Berkeley's published works. 

The composition of the Commonplace Book must, then, have 
required serious attention, and a sustained labour of the 
intellect. I do not see its author, arm-chaired, opening his note- 
book just before going to bed, and jotting down in it casual 
reflections upon his reading. He surely gave the best working 
hours of his day to it. I see him sitting up to the table, with 
the immortal note-book in front of him, and Locke, Male- 
branche, Molyneux, and all the other books piled around, 
the books being companion to the note-book, rather than it 
to them. 

Now at what period of his life could Berkeley have given 
the requisite time and pains ? And what length of time would 
one expect the work to occupy ? Can we find a period in which 
Berkeley was free from the pressure of other work, with mind 
untroubled by fear of want or by anxiety about his career? 
I think we can. I think there is a period that fits all the facts; 
a period not too long and not too short; a suitable period of 
some twelve or thirteen months. The traditional two or three 
years is, to my mind, too long: for then it is not so easy to 
account for the underlying unity of the work and its continuity 
of thought and interest. If, on the other hand, we compress 
the period unduly, say, to six months, we do not allow sufficient 
time for the real and substantial development shown in course 
of the work, nor even for the mere marshalling of the host 
of authorities consulted by this mind encyclopaedic, critical, 
well-educated, mature, and scholarly. 

Now the Fellowship Examination seems to me to be an 
important and neglected factor in the question of date. I could 
not be positive in the matter, but it seems to me unlikely that 
Berkeley could begin to write the Commonplace Book with that 
ordeal hanging over his head. Stock, in his Life, says : 'He was 
admitted fellow of that College June 9, 1707, having previously 
sustained with honour the very trying examination which the 
candidates for that preferment are by the statutes required to 



IQO BERKELEY'S COMMONPLACE BOOK 

undergo.' The examination was 'trying', because it was on a 
wide course, competitive, open practically to all comers, and 
occasional. If held, it was held by statute, as at present, 
immediately previous to Trinity Monday. If no Fellowships 
were vacant, it was not held. Some years three or four Fellow- 
ships might be thus filled : then there might be no examination 
for three or four years. Death, resignation, and, especially, 
preferment by 'going out on a College living', made the vacan- 
cies, if any. I have searched in vain for early records of the actual 
examinations. Nor can I find any details as to courses read. 
The statute prescribes: 'Quattuor diebus praecedentibus elec- 
tionis diem, ab hora octava antemeridiana ad decimam, et ab 
hora secunda pomeridiana ad quartam, omnes electores dili- 
genter exquirant . . . primo die, in dialectica, et mathematicis : 
secundo, in philosophia turn naturali, turn morali; tertio, in 
linguarum peritia, in historiis, et poetis, et in toto genere 
humanioris literaturae: quarto, in scribendo de themate aliquo, 
et versibus componendis.' 

This statute must have been the controlling factor in 
Berkeley's reading after his graduation in 1704. One can read 
its conception of the scibile even in the ground-plan of the 
Commonplace Book. 

Interesting sidelights upon the working of the Fellowship 
Examination system are to be had in the Locke-Molyneux 
correspondence. 1 Molyneux had undertaken to secure a young 
Trinity scholar to translate Locke's Essay into Latin. He 
secured a Fellowship candidate. He writes, on May 7, 1695: 
'As to the translation that is going on here, tis undertaken by 
one Mr Wm Mullart. He proposes to finish it in half a year 
or nine months at farthest: for he cannot wholly disengage 
himself from some other studies.' And again of Mullart: 
'Aug. 24th, 1695. I formerly told you how he designed for a 
Fellowship, had any at that time hapen'd vacant, as there did 
none. But very lately there are 2 fellowships become void and 
a 3rd like to be so before the time which is next June 1696 & 

1 Locke, Some Familiar Letters , pp. 112, 122. 



BERKELEY'S COMMONPLACE BOOK 191 
he tells me plainly he must endeavour to get one of them & 
that there will be at least 5 competitors if not six & therefore 
he must use his utmost diligence application and study in 
the intermediate time to fit himself for the examination they 
undergo ; and this, he says, will take up so much of his time that 
he knows not whether he shall have any to spare for the 
translation/ 

This Mr. Mullart just touches Berkeley's life. So I will 
briefly give the relevant facts. Mullart 's 'diligence, application 
and study' were rewarded. On June 8, 1696, he was elected 
to the third of the four Fellowships filled on that day. In the 
College Register we may read Mr. Mullart 's appointment to 
various lectureships in almost each succeeding year till we reach 
the following important entry: 

'1706 Sep 24 th The living of Clinish in y e Diocese of 

Clogher being void by y e death of M r Robert Smyth, 

M r Mullart was nominated to it.' 

Clearly it was Wm. Mullart 's 'going out* that let George 
Berkeley in. 

Let us reconstruct the situation. Berkeley coming up for 
Michaelmas Term, which opens on October i, 1706, is met 
by the news that there will be an election to Fellowship on 
Trinity Monday, 1707 (the sole election for six years). There 
has been no election since 1704, and, as it proved, the next 
election after 1707 was in 1710. He has eight months for inten- 
sive preparation. Hfe cannot afford to take chances. He is a 
brilliant student, but he has brilliant and hard-working friends, 
like Synge 1 and Madden, 2 who no doubt were his rivals, because 
they were subsequently his colleagues. Berkeley was a com- 
paratively poor man, an Erasmus Smith Exhibitioner. True, 
there were poorer men in his College set (574). But there is 
only one vacancy, for which the best men of three academic 
years are competing. Clearly, Berkeley had every reason for 

1 Mentioned in Chapman MS. N.B. 'tristem in musaeo solitu- 
dinem, duramque eorum qui vulgo audiunt Pumps vitam': De 
Lud. Alg. Works, vol. iv, p. 59. The context explains the collo- 
quialism Pumps. 2 C.P.B. 574. 



i 9 2 BERKELEY'S COMMONPLACE BOOK 

concentrating on the Fellowship Examination of 1707. His 
career may depend on winning that one vacant Fellowship. He 
has now little time to devote to his Society, with its cast-iron 
rules. One piece of literary work must be done at once. The 
Arithmetica and Miscellanea Mathematica, which has been on 
the stocks for nearly three years, is finished off, and prepared 
for the press. Early in 1707 these 'first-fruits' are published, 
as a Fellowship thesis, conciliating influential opinion, in- 
directly and diplomatically, by the dedications, answering 
damaging rumours (as that he hated mathematics), showing off 
his powers of composition in the Latin tongue, indicating his 
absorption in sterner disciplines, and holding out the promise 
of greater things. 

Once he had won a Fellowship, his career was safe, his posi- 
tion secure. For the first year or two a Junior Fellow's College 
duties would not be exacting. Here, then, is a golden oppor- 
tunity for settling down to authorship. He can now solutiore 
animo examine, arrange, express those high thoughts which 
have long been beckoning. I submit then that the biographical 
facts point to June-July, 1707,* as a probable date for the 
beginning of the Commonplace Book, and that they make an 
earlier date next door to impossible, especially during the 
eight months immediately preceding the Fellowship Examina- 
tion of 1707. 

The Structure of the 'Commonplace Book'. 

Before discussing the internal evidence as to the date, I must 
deal with the question of structure. While doubt remains as 
to which is the first page and which the last, we cannot draw 
any conclusion about the date which would be precise enough 
for purposes of scholarship. Here I have something new to say : 
but as, in the main, it provides new support for old belief, I can 
be brief. 

1 He proceeded M.A. July 15. The Long Vacation began 
July 9. It is important to realize that the immaterial hypothesis 
(19) was clear-cut in his mind before he began to write the C.P.B. 



BERKELEY'S COMMONPLACE BOOK 193 

Lorenz was the first to show that the Commonplace Book is 
composite. It consists of two note-books bound together. The 
fact is visible and palpable. Lorenz further argues that the 
note-books have been accidentally bound together in wrong 
order. His view is generally accepted. I do not dispute the 
substantial fact. At the same time, in justice to precision and 
to the book-trade, it should be pointed out that the time-order 
of writing is not necessarily the right order for binding. It is 
not impossible that Berkeley himself had the books bound. 1 
Probably, however, the present binding of the Commonplace 
Book is after Berkeley's time. It is uniform with 39304 and 
39313 in the Berkeley Papers, and with those only. 39313 is 
inscribed, 'Hugh James Rose 1828'. On the flyleaf is written, 
apparently in Rose's hand: 'These MS. sermons and charges 
of Bishop Benson were found among the papers of Dr. George 
Berkeley, son of the celebrated Bishop Berkeley. They 
probably came into his possession through his friendship with 
Mrs. and Miss Talbot, with the former of whom Bishop 
Benson's sister resided till her marriage with Archbishop 
Seeker.' One might infer that either to Berkeley's son or to 
H. J. Rose we owe the present binding of the Commonplace 
Book. However that may be, the present front page (page 3) 
has been a front page for a long time. Father and son have 
recognized it as such. It may be that *G: B: Coll: Trin: Dub: 
alum' 2 is the father's signature, and I am sure that the signa- 
ture half-way down the page is that of George Berkeley, Jr., 
in his young days, at least before he took his M.A. (1759). He 
has inscribed himself with ornamental flourishes, displaying 'his 
Bachelor's Degree and his Oxford College. 3 So there is no. 
proof that the work came to Berkeley, Jr., or to Rose in the form 

1 Since writing this, I came across an indication that Berkeley 
may have used the C.PJ3. later in life. See above p. 172. 

2 Or did the son matriculate at T.C.D., April 25, 1752, on his 
way to Oxford (June 4) ? 

3 The A.B. is disguised by flourishes and the ex aed. Xti is 
microscopic; but I happened to notice the son's 'A.M. ex aed. 
Xti' elsewhere in the Berkeley papers. On comparison I do not 

4046 C C 



194 BERKELEY'S COMMONPLACE BOOK 
of two detached note-books. Even if Bishop Berkeley left them 
so, still even he might not consider the present manuscript 
order 'wrong'. It is at least as 'right* as that of some printed 
editions. 

To return to the question of the time-order. Lorenz and 
Johnston put note-book B before A. In the main I agree with 
them; but I wish to advance a new reason for agreeing with 
them in the main, which is also a reason for differing from them 
in certain not unimportant details. The evidence as presented 
by Johnston is not convincing. He writes : x 'A contains the date 
August a8th, 1708. B contains the dates January 10, 1705-6, 
and December 7, 1706. There is no doubt as to these dates, 
consequently A must be later than B. This is absolutely con- 
clusive.' The dates per se will not bear this stress. The dates 
in B occur in matter unconnected with the Commonplace Book 
proper. The same might hold of the date in A (but see below, 
p. 201). When Dr. Johnston says that he has read the Common- 
place Book for years, and considers the order to be B A and not 
A B, he gives an argument morally conclusive, and one to which 
all Berkeleian scholars will defer. But, none the less, there is 
need for an objective argument such as the following, which 
seems to me to be absolutely conclusive. 

The Orthography of 'Idea' . 

I noticed that during the course of writing the Commonplace 
Book Berkeley has made a marked change in the orthography 
of the pivotal word of his philosophy. I will tabulate the facts 
shortly. In outline the change is this. In the first third he 

think there can be any doubt as to this signature which has 
puzzled readers of page 3. 

1 Development of Berkeley's Philosophy, p. 22. Subsidiary 
arguments are unconvincing, e.g. Fraser on Newton's knight- 
hood. In a private note-book where Newton occurs passim, 
'Sir Isaac' and 'Mr. Newton' are equal and opposite accidents. 
Take a parallel case such as 'Eddington* in a 1932 note- 
book! 



BERKELEY'S COMMONPLACE BOOK 195 

writes 'idea' invariably. In the central third he writes 'idea* 
and 'Idea' indifferently. In the last third he writes 'Idea* 
consistently. That the change of practice was from small 
to capital, and not vice versa, needs proving. I do so along 
three lines: the evidence of the Commonplace Book itself, 
the evidence of his published works, and the evidence of 
his other writings of the period. If the change were from 
capital to small, the order of note-books could not be B A, for 
B opens with 40 pages of unbroken 'idea'. But no more could 
the order be A B. For, if so, we should have to suppose that 
Berkeley first wrote 'idea' and 'Idea' indifferently, then fixed 
on 'Idea', then changed abruptly to a steady 'idea', then 
reverted to his original indifference. Whereas if the order be 
B A, and the change be from small to capital, a simple, natural, 
and accountable change occurred. He first wrote 'idea' uni- 
formly. When he discovered the Principle, idea takes on a new 
meaning, and the word a new importance. He is out to 'change 
ideas into things' (Hylas, Works, p. 463). He studied the nature 
of the idea. It became concrete and objective. He decides to 
show to the eye of his readers this added importance. Such a 
change is not to be made all at once. So we find a transition 
period, in which both styles compete, the capital gradually 
superseding the small letter. Finally, perhaps under the influ- 
ence of preparing the Theory of Vision for the press, 1 'Idea' 
became the fixed orthography. 

The evidence of his other writings, both unpublished and 
published, points beyond a doubt to the same conclusion. The 
following table analyses all the occurrences of the word in the 
Berkeley manuscripts of the period. Not much survives prior 
to the Commonplace Book] but the two essays that do survive, 
it is worth noting, are formal, representative writings, care- 
fully revised, and preserved in the Molyneux Papers. This 
table proves, I think, that the orthographical change was 

1 I think that a good portion of the Theory of Vision was 
written, no doubt, in draft, before the completion of the Common- 
place Book, but I do not express a final opinion. Surely 754 
refers to the Dedication to Percival? 



196 BERKELEY'S COMMONPLACE BOOK 

deliberate, purposive, and in the direction idea-Idea, not vice 
versa: 

Date Manuscript idea Idea 

Jan. 1706 The Cave of Dunmore 

Essay (Dublin Copy A) . i none 

probably 1706-7 Of Infinites . . .13 none 
probably 1706-7 The Queries and MS. page 

164 (946-953), verso . 9 none 1 
Jan. ii, 1708 Sermon on Life and Im- 

mortality ... 3 none 
(Commonplace Book, Note- 
book B ... 120 6 
Commonplace Book, Note- 
book A . . .69 211 
Nov. i5-Dec. 18, f Draft Introduction to the over 
1708 \ Principles . . . none 200 
Draft of Principles, 85- 

145 .... none 89 

I pass to the evidence of the publications. 'Idea (spelt with 
a capital) in Sins is very different from idea in the earlier works.' 
So Johnston writes. 2 It would not be fair to take this casual 
remark au pied de la lettre. Still, the remark is unfortunately 
phrased. For the orthographical facts are the other way about. 
The earlier works Theory of Vision, Principles, Passive Obe- 
dience, Dialogues, also the Analyst and Alciphron print with 
capital. Siris prints 'idea* (even of Plato's), 3 and it is, I think, 
the only major publication to do so. A closer scrutiny of the 
books of the early period shows that Berkeley at that time took 
great pains in the matter. It should be remembered that he 
would find both usages in the books he read. Now, here is a 
curious fact. In the Theory of Vision (1709) he prints the word 
invariably in italics with initial capital Idea. In the Principles 
(1710), retaining the capital, Idea, he drops the italics in the 

1 953 has 'Ideas', but as it begins the entry, it does not enter 
into account. All entries begin with capitals. 

2 Development, p. 256. 

3 The only 'Idea* I have noticed is in Table of Contents. 



BERKELEY'S COMMONPLACE BOOK 197 

case of realities; he retains the italics (with capital), however, 
in the cases of abstract, unreal, or otherwise suspect ideas. In 
view of these facts we cannot regard the marked orthographical 
change that occurs in the Commonplace Book as undesigned or 
insignificant. It is, surely, an outward and visible sign of an 
abiding change of intellectual orientation. 

To show graphically the bearing of these facts upon the 
questions of structure, I must inflict yet another table upon 
the reader. It pursues the analysis straight ahead through the 
pages of the manuscript in the B A order. Its page divisions 
at 143, 28, and 57 are more or less arbitrary, and struck in the 
interests of my argument. To a reader of the manuscript the 
broad fact is objective and independent of such arrangement. 
But I arrange the details thus to assist those who have not the 
advantage of reading the manuscript, and to enable them to 
visualize the three periods the 'idea* period at the outset, the 
'Idea* period at the close, and the middle or transition period, 
during which the new practice comes in gradually, competes 
with the old, and finally ousts it. 

Entries idea Idea 

1-276 Note-book B, pp. 104-43 72 none (excluding 

i on verso, ob- 
viously a later 
addition, 1 79). 

J277-395 dio pp. 144-64 48 5 (+2 on 

L903-45 verso). 

396-524 Note-Book A, pp. 4-28 46 6 
525-681 ditto pp. 29-57 18 96 

682-902 ditto pp. 58-95 5 107 ' . 

I submit that the above analysis constitutes an independent, 
new, and objective proof that note-book B is, in general, prior 
to A. It may be said that this is a laborious proof of a proposi- 
tion that nobody doubts. I do not fear that reproach. For there 
are doubters, and will be. Moreover, this line of proof, if 
sound, corrects the prevailing opinion on certain points, and 
thereby facilitates a much needed restoration. Thus we may 



i 9 8 BERKELEY'S COMMONPLACE BOOK 

reach a clear view of the structure of the Commonplace Book 
as a whole. 

I have claimed for B priority in general. Strictly, what is 
proved is that the first part of B was written before the last 
part of A. I have not yet met any argument that compels us 
to think that Berkeley finished B before he began A. And there 
are indications that towards the middle of the period he had 
the two note-books in use simultaneously. 1 

A is not simply a continuation of B . B begins may we say 
it? with a 'try-out* of the 'immaterialist hypothesis'; A begins 
with a 'try-out ' of the New Principle. I grant that the very first 
entry of A (396) links to B and the Theory of Vision. But the 
entries 397-411 i.e. the first three pages of A do not look 
back to B. They look ahead. They reveal the author on a peak 
in Darien, descrying new territory, and mapping its main 
features, or, in prose, show us Berkeley taking stock of the 
position reached in his previous study and projecting the 
Introduction and scope of the Principles. There is further proof 
that A does not continue just where B left off. It seems to me 
that when he was on page 6 of A he was also on page 158 of B. 
For at that point in both books we have a duplicated series of 
eight or nine entries; 2 not identical, but very close indeed; cf. 
365-73 with 411-20. These entries, with the exception of the 
last two in each group, concern Infinitesimals, and are closely 
related to, probably copied in from, the essay Of Infinites. 
If A were a simple continuation of B, a double transcription of 
such length and so close in time would surely be a blunder hard 
to explain. Whereas, if at that stage he were filling both note- 
books simultaneously, we have the easy solution that he entered 
them in both note-books because he had not then decided 
whether to use them for the Theory of Vision or for the 
Principles, or for both. 

The orthography of 'idea' towards the end of B and the 

1 Locke, New Method of a Common-Place-Book , p. 321, advo- 
cates two or more note-books. 

2 An occasional entry is duplicated. This is the only case of 
a duplicated series. 



BERKELEY'S COMMONPLACE BOOK 199 

beginning of A is that of the transition or middle period. He 
makes such entries as 392 : * By thing I either mean Ideas or that 
which has ideas/ From 904 to 923 'idea* predominates: but 
923 has 'neither our Ideas nor anything like our ideas'. We 
must, therefore, keep entries 903-53 in their manuscript posi- 
tion (relative). Johnston, 1 'in agreement with Lorenz, Erd- 
mann, Rossi, and Hecht', takes them out of B and prints them 
after A. This unnatural surgery is indefensible. These entries 
record several positions 2 that Berkeley had held, but definitely 
abandoned, weeks before he finished A. The section is 
ludicrously isolated when transposed to the end of A, as in 
Johnston's text. Moreover, it is very much wanted at the end 
of B. For the demonstration of the Principle more geometrico 
(903-24) is the decisive factor in the structure 3 of this part, and 
forms the transition to the theme of A. Finally, the argument 
from orthography given above clinches the matter. 

My view as to the structure of the Commonplace Book can 
be expressed shortly in terms of the pages of the manuscript 
and in terms of Johnston's numbering of the entries. With the 
proviso that probably the closing pages of B and the opening 
pages of A overlap in point of time-order, my view is this: the 
Commonplace Book proper consists of MS. pages 104-1 64 (recto) 
and pages 3-95, and was written approximately in that order. 
I would, therefore, 'top and tail' Johnston's text, relegating the 
'Queries' and entries 946-53 to an appendix. I would re- 
transpose entries 903-45 to their original place after 395. Thus, 
900 would be the concluding entry, with 901 and 902 as 
Berkeley's flyleaf finale. Immediately before 396 I would insert 
Berkeley's own index-table of the marginal letters. I woiilcl 
restore the apparatus of marginal signs throughout, and I 
believe we should have the Commonplace Book substantially as 

1 Berkeley's Commonplace Book t p. xxi n. 

2 e.g. 904, 906, 908, 911, 916 (all marked +) 

3 The compression at end of B is explained, if, at foot of p. 
159 (378), he went ahead to p. 161 and sketched out 'my Principle 
with a demonstration', subsequently filling all adjacent blank 
pages. 



200 BERKELEY'S COMMONPLACE BOOK 

Berkeley wrote it, and with as close an approximation to the 
original time-order as we need or can attain. 

The Date 

I have indicated above my view as to the nature and precise 
contents of what we are trying to date. We are trying to date 
a purposeful homogeneous composition, of determined struc- 
ture, continuously written, and systematically used by Berkeley. 
The whole work is contained in the two note-books which now 
constitute B.M. Add. MS. 39305. Our problem, then, is two- 
fold: (a) to determine the date of the opening page of the work 
(104 of that manuscript); (b) to determine the date of the clos- 
ing page (94 of the manuscript). 

(a) The Commonplace Book could not have been begun 
before December 7, 1706. That is the one precise and certain 
fact. Page 103 is headed with that date, very formally written 
out in extenso to fix an agreement. Therefore, page 104 was 
not written before that date. This argument seems to me con- 
clusive. I cannot conceive any circumstances under which 
Berkeley would have first made the tremendous entries on page 
104 and sqq. and then gone back to page 103, supposed blank, 
and used it as a record of a trivial agreement, not completed, 
about a debating society. 1 

As to the further question, 'How long after December 7, 
1706, was page 104 written ?' there is evidence, but, I think, 
no rigorous proof. Whatever may have been the reason why 
the 'underwritten persons' did not subscribe on page 103, it is 
natural to think that Berkeley would thereupon, after a decent 
interval of time, convert the Society rule-book to his private 
purposes. I have given biographical reasons above for thinking 
it highly improbable that Berkeley could have undertaken the 
Commonplace Book till after the Fellowship Examination. 

1 It will be noted that I now regard both codes of rules as in 
Berkeley's handwriting: see Hermathena, vol. xxii, p. 7. The 
curious spelling 'buisiness' is found in both sets of rules (pp. 97, 
103), and also in the sermon of Jan. n, 1708. 



BERKELEY'S COMMONPLACE BOOK 201 

There are indications that during the early months of 1707 
Berkeley used this note-book occasionally for rough notes; 
several pages have been torn out at the far end; pp. 168-9 
certainly contain rough work for the De Ludo Algebraico, 1 
which he was preparing for the press early in 1707. On page 
165 there are odd notes for the sermon of January n, 1708. 
On page 166 laws of motion are listed, apparently from KeilPs 
Introductio. On page 164 there are rough notes on Locke's 
Essay, which, with the Queries, look like 'tip' questions for the 
Fellowship Examination. The Queries themselves (page 102) 
were almost certainly written in this period (see above, page 
182). The Cave of Dunmore Essay, an inaugural address, is 
proper to the Society rule-book period, and it occurs right at 
the far end ; but most of the casual entries mentioned above 
must belong to the interval between December 7, 1706, and 
the opening date of the Commonplace Book. Once that work 
began, the note-book would be kept, and clearly has been kept, 
for the one purpose. All these facts make it very probable that 
page 104 was not written till June or July, 1707. If we con- 
sider the minimum period needed for the composition, and 
work backwards from the autumn of 1708, by which time the 
last page was certainly written, we should infer that the work 
was begun in or soon after June, 1707. 

(b) As to the date of the conclusion of the Commonplace Book 
we can be fairly precise within a small margin. Professor Aaron 
has noticed that a book published in 1708 is quoted in the last 
quarter of the work 2 ; but that fact does not carry us far, as 
Professor Aaron shows. The decisive factors seem to me to 'be: 
(i) the date on page 95; (2) the draft Introduction to the 
Principles. Of these two, the latter gives the only evidence that 
can stand alone as virtual proof. 

(i) I incline to the opinion that Aug. 28, 1708, at the top 
of page 95, is Berkeley's dating of the completion of the 
Commonplace Book. But I admit that it and 'The Adventure 
of the [Shirt]', which seems to go with it, may have no more 

1 So Fraser, Works, iv, p. 54, note. 

2 721, Locke, Some Familiar Letters. 

Dd 



202 BERKELEY'S COMMONPLACE BOOK 

significance for us than the small sum which Berkeley has 
scribbled at the foot of the same page. 

The latter half of page 94 is crowded with entries, written 
small, yet neatly. The writer has not allowed himself to over- 
flow into either of the adjacent verso pages, as he did at the close 
of his othemote-book. Outwardly page 94 (889-900) has all the 
appearance of a deliberate conclusion it contains twelve entries , 
while page 95 simply contains 901, 902. Upon 902, 1 think, no- 
thing turns. It may have been written immediately after page 94, 
or before it, or much later. It may be a quotation or Berkeley's 
own reflection. The sentiment is of the type to be found in the 
early correspondence with Percival. It forms a fitting epilogue 
to a work by a young, ambitious author with lofty ideals. 

901 may, of course, be merely the remembered date of some 
holiday adventure. 1 Yet the entry is put prominently at the 
head of a page, and the month is written out in full. In the 
autumn of 1708 Berkeley had some reason (if only habit), no 
doubt literary, for dating his writings precisely. In the Draft 
Introduction to the Principles, which he began on November 15, 
1708, and finished on December 18, 1708, he has entered in the 
margin not only those days, but almost all the days in between. 
He has put the date at the end of each day's work in November- 
December. He might have had the same reason for doing so 
a couple of months earlier. 

(2) If August 28, 1708, be not the actual date we seek, it must 
be near it. I know no reason for putting the date of completion 
earlier. The draft Introduction prevents us from making it 
much later. This draft, begun on November 15, 1708, must have 
been based on a previous draft. For in parts it is written out 
fair, as if for the press, the long quotations from Locke being 
actually penned in italics. So the original composition of the 
Introduction to the Principles must be put back to the early 
autumn of 1708. Subsequent revisions have obscured the 
dependence of the Introduction upon the Commonplace Book. 
But originally the Introduction embodied whole entries 

1 I do not think that 'shirt* is the correct reading; but I have 
puzzled in vain to find a better. It is nearly 'ship', but not quite. 



BERKELEY'S COMMONPLACE BOOK 203 

verbatim, as well as taking over from it and developing the 
experimental conceit of the 'Solitary Man', who was to do for 
the Principles what Molyneux's 'Blind Man* had done for the 
. Theory of Vision. I select the most striking instances of depen- 
dence, adding in brackets the pages of Fraser's Works, vol. iii, 
where the entries may be found. These are: 310 (383), 606 
(380), 651 (380), 750 (380), 760 (357), 761 (371). This is evi- 
dence and, I think, proof, that the Commonplace Book up to 761 
was written before he began to draft the Introduction to the 
Principles. Now entries marked for the Introduction run on to 
the last page but two (872). In fact, it is only at 870 that he 
finally abandons his original intention of including a theory of 
Demonstration in the Introduction (592), and for which, no 
doubt, he had prepared a specimen 'Demonstration* months 
before (903-24). It is, therefore, certain that the first 761 
entries, and almost certain that the remaining entries, were 
written before the Introduction to the Principles was begun, 1 
that is, before the early autumn of 1708. 

We may, I think, regard it as established that the Common- 
place Book proper was not begun before December 7, 1706, and 
that it was finished before November, 1708 : and there are good 
grounds for holding that Berkeley began the Commonplace Book 
proper shortly after his election to a Fellowship (June 9, 1707), 
and completed it on August 28, 1708. 

Note. This paper is based on a first-hand study of the actual 
manuscript. My brother, Mr. F. M. Luce, has kindly checked 
a good deal of the detail for me. 

1 870, 'I must cancell all passages/ It would be possible {o 
take this of passages in an Introduction actually written. But 
then would he not take up pen and cancel them, instead of 
recording his intentions? It seems to me more natural to inter- 
pret it of passages in the Commonplace Book. Some twenty-six 
entries have the marginal sign 'I* ( = Introduction) cancelled. 
The great majority of these concern Demonstration and Cer- 
tainty, e.g. 537, 554, 556, 590, 592, 710, 731, 74<>-7 752~3, 7^3- 
770 with T cancelled shows another change of plan. In 870, 
after 'pretend*, B. at first wrote 'at least near the beginning*. 



204 BERKELEY'S COMMONPLACE BOOK 
CLASSIFICATION OF ENTRIES MARKED WITH 
MATHEMATICAL SYMBOLS (x +) 

N.B. (i) In a few cases the symbols are combined with 
letters; for the letters see Johnston's edition. 

(2) Some of those marked x + appear to have been originally 
x 1 = Distance. 

(3) The position of the numbers in the angles is not consistent. 
They may even be at a little distance from the symbol. 

x 21, 26, 31, 55, 61, 72, 75, 88, 103, 106-7, no (?), 119, 
121-3, 170, 226, 244-7, 251, 256-63, 267-70, 272-3, 
276, 284, 295, 300, 317, 319-20, 324-6, 328, 331-6, 
340-50, 352-9, 365-70, 375, 377, 382-3, 385, 387, 
410-17, 423, 425, 429, 435-7, 442-3, 445, 447, 456-7, 
459, 461-3, 465, 468-70, 475, 482-5, 489, 494, 503, 
512-14, 5i7~i9, 523-4, 529, 531-6, 545, 550, 556, 
559, 562-3, 579-8o, 611, 620-1 (?), 642, 657, 687-8, 
7io, 734, 742-3, 745-7, 752, 763, 7?i-8i, 783-5, 79i, 
815-16, 846-7, 849, 858, 865, 880, 885, 889, 892-3, 
895, 925-30, 933-4, 939-41- 

1 x 56, 609, 738. 

1 x 1 58,172,306,313,453,504. 
x+ 97, 99, 173-4, J 86, 223, 225, 386, 391, 444, 458. 

l x a u, 90, 118, 124, 127, 142, 152, 171, 176-7, 185, 194, 

201-5, 209-10, 212-13, 2l6, 2l8-I9, 221-2, 227-8, 
238, 241-2, 253, 265-6, 277, 28l-3, 285, 312, 314, 762. 

*x 8 27-8, 32, 35, 43, 49 (4 times), 54, 59, 69-70, 79, 102, 
108, 116, 126, 139-40, 229, 249-50, 252, 271, 305, 
307, 396, 439-40, 722. 
1 x 3 104, 128, 150, 233-6, 255, 286, 318. 
11 X 2 291-2 (may be x x 2 ), 438. 

., r X 123 240. 

1 x 23 93 , 105, 88!. 

+ I, 2, 5-8, 10, 12, 14-16, 24-5, 29, 33, 38, 41-2, 44, 
52-3 (?), 62-3, 65-6, 73, 77, 83-4, 86, 89, 95-6, 98, 
101, 109, 112-13, 125, 131-2, 134-6, 141, 143-6, 149, 
153-6, 158-64, 166-7, 169, 182, 184, 187, 191-3, i95~ 
7, 199, 206-8, 215, 217, 220, 230, 248, 254, 264, 290, 
293-4, 302-4, 321, 330, 337, 339, 362-3, 371, 373, 381, 
384, 389-90, 392-4, 398, 403, 405-6, 409, 4i8, 420, 
422, 430, 433, 448, 451, 460, 464, 473, 486-7, 490-1, 



BERKELEY'S COMMONPLACE BOOK 205 

495-9, 505-9, 530, 548, 551, 5^4, 567, 5?o, 573, 575, 
583, 585-7, 601, 604, 614, 628, 631-4, 646-8, 669, 
672, 674, 723, 733, 754, 904, 906, 908, 911, 916-17, 
931, 942-5- 

The Arithmetical and Miscellanea Mathematica 

Hone and Rossi, Bishop Berkeley, p. 14, note, speak of 771 as 
written about the same time as the Arithmetica (written from 
1704, published early in 1707). If they are right, my view as to 
the date of the C.P.B. is clearly wrong. I do not think they are 
right. The series 771-83 is preparatory for the treatment of 
Mathematics in the Principles ( 120-25). Possibly 410 and 929 
show recollection of what he had hoped to do for Arithmetic and 
Algebra by his first publication : but he had similar hopes with 
regard to the Theory of Vision and the Principles (see title-page, 
&c.). I have examined the Arithmetica carefully, and can find 
in it no trace of evidence for the above suggestion made by Hone 
and Rossi. All the evidence points the other way. In the C.P.B. 
Berkeley outlines a study of the nature of number and the philo- 
sophy of the numerical sciences. His outlook is mature. In the 
Arithmetica his aim and objects are much humbler, and his out- 
look is 'juvenile' (to use Eraser's damning and often misplaced 
epithet). Contrast his early rhapsodies about Algebra in his 
De Ludo Alg., e.g. 'Ars magna mirabilis, supremus cognitionis 
humanae apex', with 779-81. Arithmetic and Algebra 'are 
sciences purely Verbal & entirely useless but for Practise in 
Societys of Men. No speculative knowledge, no comparing of 
Ideas in them'. Surely years of thought and his doctrine of 
abstract ideas separate the two works. 

My view that the Arithmetica was an ad hoc publication 
hurried through the press in view of the Fellowship election is 
supported by the following facts. Three copies are in the T.C.D. 
Library one marked 'ex dono authoris,' another with pen and 
ink corrections, I think, by Berkeley's hand. This latter (oo. g. 
55) differs from the other two in the following points. The 
impression is much clearer. It is without the Tabula Lusoria. 
It is without the table of Errata and Addenda, which Eraser's 
text incorporates, and the ne quis note (Works, vol. iv, p. 72). 
It looks like an advance presentation copy : for it has a crimson 
ornamental end-paper. The page containing the Errata and 



206 BERKELEY'S COMMONPLACE BOOK 

Addenda has been slipped into the other copies at a later date. 
It is headed : 'Quae potissimum prae absentia Authoris irrepse- 
runt sic corrigantur.' 

THE TEXT 

The manuscript is perishing, in parts rapidly. So it is a matter 
of urgency to fix the text now. Already we are dependent, for 
a few readings, upon Eraser's work of sixty years ago. Professor 
Aaron has published in Mind (October, 1931, April, 1932) lists 
correcting Johnston's text. I venture to add the following correc- 
tions and textual notes, giving the MS. page and the entry 
number. 

MS. Entry. 
102 backxvi/or 'we names' perhaps read 'we've names'. 

112 78 for'S. 61, 65, 66'rawTS. i, S. 5, S. 6'. 

113 88 for 'general' read 'severall'. 
134 225 for 'thought' read 'thoughts'. 
141 262 for 'make' read 'made'. 

145 287 N.B. after 'witchcraft to see* occurs, lightly 
penned through, 'we know nothing but our 
thoughts or w* these (?) think'. 

150 309 for 'say I* read 'says I' a colloquialism. 
4 399 omit '[Natural]'. 

7 414 for 'lato' read 'dato'. 

8 421 N.B. on verso against 420. 
12 451 for 'sin' read 'sic'. 

15 461 after 'arguing' insert 'against'. 

18 471 for 'discussed* read 'discors'd'. 

474 the second 'rationis' is abbreviated 'roms*. 

20 482 for 'sensible' read 'sensibile'. 

23 494 for 'intended' read 'intimated*. 

29 525 for 'a fool' read 'afoot'. 

30 533 for 'difficulty truly' read 'difficulties'. 

31 538 probably one entry. 

33 547 The MS. here is much perished: but I prefer 

'Principle* to 'principles', and 'understanding* 
to 'considering*. 

34 550 I prefer 'Relations' to 'relation'. 
36 561 before 'none' insert 'few or'. 

38 569 for 'Barrow's* read 'Bacon's*. This is quite certain. 



BERKELEY'S COMMONPLACE BOOK 207 

MS. Entry. 
41 587 for 'and empty words with us' read 'empty sounds 

without a meaning*. 
48 626 for 'form 1 read 'forms': for 'HylascmV read 

'Hylarchic'. 
628 for 'attractivae' read 'attractrix' (used by Berkeley 

in De Aestu Aeris). 
54 664 for 'viz' read 'v.g.' (so also in 671, and in 683 and 

699 for 'e.g.'). 

56 676 for 'abstract general' read 'general*. 
677 for 'really* read 'on*t*. 

65 729 for 'and & 4th* read '2 ist*. 

66 730 for 'words* read 'beards* (cf. Locke, iv. 10. 9). 
66 732 for 'fact* read 'Text*. 

7 757 f r 'Both* read 'Res* (Respondeo ; cf. Misc. Math. 

Works, vol. iv, pp. 17, 44, and Resp., p. 32, ib.). 
71 761 for 'information* probably read 'informations' (so 

Draft Intr., Works, vol. iii, p. 371). 
762 for 'systems* probably read 'spheres'. 

74 782 'severest* is doubtful. 'Gospels' is quite illegible 

now. 

75 784 omit '[b. 4, c. 8]'. 

788 N.B. after 'imaginable* B. wrote and erased 

'thing or intelligible thing*. 
78 80 1 before 'volitions* insert 'the*. 
80 813 after 'bodies* insert '&*. Note capital 'Our', as 

often for emphasis. 
87 845 for 'seeing* read 'being* (the same archaism occurs 

in 672 and in B.'s letter to Percival, Jan. 19, 

1711). 
89 859 for "'we', our 'selves', our 'mind'" read 'we' or 

selves or mind*. 
92 871 for 'cavil* read 'covet*. 

872 for 'diseases' read 'discovery*. 



APPENDIX II 

THE PROBLEM OF BERKELEY'S VISITS TO 
MALEBRANCHE 

THE following are the data : 

(a) On November 24, 1713, Berkeley wrote to Percival, 
'Today he (1'abbe d'Aubigne) is to introduce me to Father 
Mallebranche, a famous philosopher in this city. 1 (Rand, op. 
cit. p. 129). 

(b) On November 25, 1713, Berkeley wrote to T. Prior, 
* Tomorrow I intend to visit Father Malebranche, and discourse 
him on certain points/ (L.L. p. 67). 

(c) Stock's Life (1776) contains the following account: 'At 
Paris, having now more leisure than when he first passed through 
that city, Mr. Berkeley took care to pay his respects to his rival 
in metaphysical sagacity, the illustrious Pere Malebranche. He 
found this ingenious father in his cell, cooking in a small pipkin 
a medicine for a disorder with which he was then troubled, an 
inflammation on the lungs. The conversation naturally turned 
on our author's system, of which the other had received some 
knowledge from a translation just published. But the issue of 
this debate proved tragical to poor Malebranche. In the heat 
of disputation he raised his voice so high, and gave way so freely 
to the natural impetuosity of a man of parts and a Frenchman, 
that he brought on himself a violent increase of his disorder, 
which carried him off a few days after.' 

(d) Berkeley arrived in Paris for the first time on November 
17, 1713 (Rand, pp. 128-31) and stayed there about a month, 
expecting 'every minute' the arrival of Lord Peterborough, 
whom he was to attend as Chaplain. On the homeward journey, 
Berkeley reached Paris about July 10, 1714. He gives no record 
of his doings on that visit. He appears to have stayed there 
about ten days, and to have reached England, via Brussels, in 
August. (Rand, p. 138). 

Inference from these data is tempting, but precarious. All 
we knowfor certain is that Berkeley twice, namely, on November 



BERKELEY'S VISITS TO MALEBRANCHE 209 

24 and 25, 1713, had the intention of visiting Malebranche, and 
was on the point of doing so. We cannot be sure that the philo- 
sophers met, nor that they did not meet. If we are to consider 
.the probabilities, there is room for difference of opinion. On the 
one hand it could be argued, firstly, that the proposal on the 
25th of a visit for the 26th is an indication that the introduction 
proposed for the 24th did not take place, and, secondly, that 
Berkeley's subsequent silence is an indication that even the 
interview proposed for the 26th did not come about. On the 
other hand, it may be reasonably asked, 'How could so polite 
and punctilious a man as Berkeley propose to call on Male- 
branche, apparently by himself, on the 26th, unless an intro- 
duction had already been effected, and how could he state 
positively that he was going to discuss 'certain points', unless 
the subject had been broached and an appointment made?' 
The objection from Berkeley's silence might be met by pointing 
out that no letters of his survive for the latter fortnight of the 
stay in Paris, 1 that the silence cannot be proved, and that in any 
case it could be explained, e.g. by a disagreement at the interview. 

The story (c) in Stock's Life, as it stands, is unreliable, and 
possibly it should be regarded as entirely worthless. Male- 
branche died in October 1715. Berkeley was then in England, 
and had been there all that anxious summer and autumn 
(Rand pp. 139-54). His second continental tour did not begin 
till the early autumn of 1716. So there is no foundation for the 
witticism that he was 'the occasional cause of Malebranche 's 
death'. Does the story raise any presumption that the philo- 
sophers met? Dr. Robert Berkeley, the Bishop's brother 'and 
Vicar- General, supplied Stock with most of the particulars fqr 
the Life. If we could be sure that this story came from that 
source, we might reasonably say that it looks like an oral account 
given by Berkeley of a meeting either in November 1713, or, 
rather, as the opening words would suggest, in July 1714, the 
story being embroidered later by the London wits. On the 
other hand, as Professor Jessop points out to me, the story is 

1 After the letter of Nov. 25 to Prior the next extant letter is 
that to Percival dated Dec. 28, 1713 from Lyons. 

4046 



zio BERKELEY'S VISITS TO MALEBRANCHE 
full of improbabilities and errors other than the date. Accor- 
ding to Lyon (Uidfalisme en Angleterre au XVIII 6 siecle, 1888, 
p. 136), Malebranche did not like oral discussions, and his last 
illness was slow and painful. Again, no translation of Berkeley's, 
works had been made at the time. Readers may like to be 
reminded that De Quincey, in his essay, Murder considered as 
me of the Fine Arts, gives a version of Stock's story. 



INDEX 



Aaron, R. I., 2, 47, 130, 139, 178, 

201. 

Acta Eruditorum, 9. 
Alciphron, changes in 3rd ed., 

i68ff. 

Algebra, 13, 15, 21, 46, 1 66, 205. 
America, 7, 28. See Rhode 

Island. 

Anaxagoras, 54, 164. 
Archetypes, 85. 
Aristotle, 24, 48, 129, 149, 164, 

175 f. 

Arithmetic, 13, 23, 161, 166. 

Arnauld, 71. 

Articles of Religion, 82 n. 

Bacon, 15, 129, 145, 157. 

Balfour, A. J., 3. 

Barrow, I., 26, 35, 38, 50, 53 n., 
54, 186, 206. 

Baxter, A., 9. 

Bayle, P., 53 ff., 64-5. 

Beattie, J., 9. 

Bergson, H., 5. 

Berkeley, G., his relation to 
Locke and to Malebranche, 7 
and passim; his originality, 7; 
an Irishman, 10; his experi- 
ments, 17; his religion, 56 f.; 
his education, n, 56 f., 189 ff.; 
his proposed visits to Male- 
branche, 8, 89, 119, 208 ff.; 
his criticism of Malebranche, 
8, 81, 117 f., 123, 16371.; his 
method, 13 ff., 166; on vision, 
3 off.; on matter, 61 ff. ; on 
ideas, 70 ff . ; on imagination, 
86 ff. ; on knowledge, 92 ff. ; on 
Deity, 115 ff. ; on abstraction, 
126 ff. 

Berkeley, Canon G., 193. 

Bibliotheque Choisie, 12. 



Bolingbroke, Lord, 9, 125, 132 n. 

Briggs, W., 26. 

Browne, P., n, 74 n., 132 n. 

Cajori, F., 186. 

Carteret, Lord, 56 n. 

Cartesians, 5, 48, 52, 56, 65, 81, 
115, 160. 

Cave of Dunmore, description of, 
2, 87, 139, 196, 201. 

Cheyne, G., 50, 54, 149. 

Christianity, 165, 170. 

Clarke, S., 9, 48, 52, 66, 112, 125, 
165. 

Clayton, R., 9. 

College Debating Society, 179, 
200. 

Collier, A., 7, 15 n., 27, 40, 47, 
48 n., 126. 

Commonplace Book, the, purpose 
of, 2, 178 ff. ; structure of, 2, 
i92ff.; date of, 3, 29, 200 ff.; 
its marginal signs, 3, 30, 36, 
62, 163 n., iSoff., 204 f.; and 
Malebranche, 41, 44, 57 f., 
103 ; on idea, 70 ff. ; on imagina- 
tion, 87 ; on knowledge, 93 f . ; 
on knowledge of self, 105; on 
knowledge of other spirits, I, 
1 08; on knowledge of Deity, 
116 f.; on abstract ideas, i38'ff.4 
on Principles Pt. II (and III), 
163, 183; and Analyst, 172; 
its 'black list', 188; textual 
corrections, 206 f. 

Compound sensation, 41. 

Creation, 44, 86. 

Cudworth, R., 175. 

Defoe, 21, 26. 

Deity, 74, 79 ff., 123, I77J 
Essence and Presence of, 79, 



EC2 



212 INDEX 

81, 116; knowledge of, 95, 
in ff.; as substance, 101 ; 'the 
place of spirits', 108 ; existence 
of, no, 117; titles of, in f. 

Demonstration, 18, 34, 142, 170, 
203; of the New Principle, 
20 ff., 94, 199, 203. 

Descartes, 9, 14 f., 36, 38, 40, 44, 
48, 54, 60, 65, 86, 103. 

Digby, Sir Kenelme, 48, 

Distance, 34 ff., 42, 45, 185. 

Draft Intro, to Principles, 2, 8, 
25, 129, 140, 150 n., i52n.fi., 

l8l, 196, 202. 

Draft of Principles S.S. 85-145, 

2, 96, 109, 196. 
Dublin Society, the, 1 1 . 
Dungannon Royal School, n n. 

Enthusiasm, 40. 

Erdmann, 199. 

Essence, real and nominal, 134 ff., 

141, 153, 160. 
Ethics, 1 06, 1 60, 163 w. 
* Excuse', the, 24. 
Externality, 97. 

Fellowship Examination, 57, 

189 ff., 200, 203. 
Figure, 42. 
Fraser, A. Campbell, 2 f., 5, 11, 

25, 74 *>> 92, 109, 168, 174. 
Freedom, 106, 166. 

Gassendi, 40. 

Geometry, 13, 46, 55, 69, 97, 

155, 161, 166. 
Gilbert, C., n. 
God, v. Deity. 
Grace, 165, 170. 
Graham, W., 9. 

Halley, E., 15, 172. 

Hamilton, Sir W., 9, 55, 61, 64 . 



Hecht, A., 199. 

Heterogeneity of sight and touch, 

7, 32 ff., 185. 
Hicks, G. Dawes, 3 n., 5 n., 6, 10, 

25, 47> 67, 72, 107. 
Hippocrates, 149. 
Hobbes, T., 40, 52, 86 n. 
Homoeomeriae, 54. 
Hone, J. M., 205. 
Hook, R., 70. 
Hume, D., 6, 105, 126, 167. 

Ideas, abstract, 21, 126 ff.; and 
words, 22, 140, 147, i55ff., 
165, 170; of sight and touch, 
32 fl., 42 ; and mode, 67 ; 
passivity of, 67, 92, 100, 118; 
orthography of, 70, 194 ; nature 
of, 70 ff ., 98 ; simple and com- 
plex, 72, 139; of reflection, 73 ; 
and things, 77, 83, 95, 163, 166, 
175; double status of, 84; in 
the mind of God, 85 ; real, 84, 
99 ; of imagination, 86 ff ., 97, 
154; presentative, 99; and 
body, 100; privacy of, 100, 
1 10 ; Platonic, 1 74 ; and appear- 
ances, 175. 

Immanence, Text of, 82, 84, 114, 
123 f. 

Immaterialism, 20, 26 ff ., 47 ff ., 
56 ff., 67, 112, 163, 167, 171. 

Immensity, 49, 82, H2f., 119, 
122. 

Inextensionism, 49 ff . 

Infinite divisibility, 20, 47, 50 f., 

54, 161, 172. 
Infinites, Of, 2, 49 f-> 54, 141, 196, 

198. 
Infinitesimals, 21, 47, 49, 55> i?2, 

198. 

Intellectual Beings, 108, 119. 
Intelligible World, 78, 80, 85. 
Intuition, 65, 68 f., 83. 
Ireland, 7, 10 f., 40. 



INDEX 



213 



Jessop, T. E., 122 ., 177 n. 
Johnson, S., 28 w., 49 f., 98 w., 

101, 14871., 151, 163, 167. 
Johnston, G. A., 2 ff., 72 ff., 178, 

183, 194, 199. 
Judgement of Sense, 41 n. 
Jurin, J., 130, 173. v. Philalethes. 

Keill, Jo., 4 w., 50, 53 n. f., 82 w., 

20 1. 

Kemp Smith, Dr., 148 n. 
Kilkenny College, n n. 
King, W., u, 26, 51, 62 w,, 112, 

i8 7 . 

Knowledge, without ideas, 74, 
94; mediate and immediate, 
103, 109, 121 ff.; notional, 
104 ff. ; of ideas, 96 ff. ; of self 
(the soul), 74, 95, 101 ff.; of 
other spirits, 74, 95, 106 ff.; of 
God, 1 1 1 ff . 

Laird, J., 127 n. 

Language of Nature (Divine 
visual), 38, 43, 117, 165. 

Leclerc, 12, 16 w., 71, 145. 

Leibniz, 12, 164. 

Locke, J., 3, 6ff., 39, 52, 54, 
6 if., 150, 167, 182, 190; Exam. 
P. Malebranche' s Opinion, 
75, 80 f. n.\ on ideas, 70 f.; on 
knowledge, 94, 105, 108; on 
Deity, in, 113, 115; on 
abstraction, 129 ff. 

Lorenz, T., 2, 193 f-> 199- 

Lucretius, 69. 

Madden, S., 191. 

Magazine Theory, 78, 118. 

Magnitude, 35 f., 42, 44, 185. 

Maheu, R., 187 n. 

Malebranche, in Commonplace 
Book, q.v.\ as Berkeley's 
source, 6 ff . et passim ; on 
method, 1 3 ff . ; on vision, 40 ff . ; 



his Illustrations, 46, 55, 58 ff., 
146; on idea, 63, 71, 75 ff.; and 
Arnauld, 71 ; on seeing in God, 
77 ff.; on Deity, 81 ff., no, 
H4ff.; Berkeley on, v. sub B; 
on imagination, 86 ff.; on self- 
knowledge, 103; on other 
spirits, 107; on unio mystica, 
q.v.; on abstract ideas, 143 ff.; 
on sorts, 158. 

Matter, 47 ff., 56 ff., 104, 127, 
143, 162. 

Matter, Intelligible, 81, 85 f. 

Modifications of the soul, 63 ff., 

78. 

Molyneux Problem, 7, 28, 33, 

141, 203- 
Molyneux, W. (Dioptrics), n f., 

26, 36 ff., 87 n., 190. 
Moon, apparent size of, 35, 40, 

45- 

More, II., 49, in, 149. 
Mullart, W., 190 f. 

Newton, Sir I., 10, 26, 38, 50, 

54, 149, 164, I94W. 
Nominalism, 159. 
Norris, J., 7, 9, 48, 79. 
Notion, 104 ff., 122, 176. 

Occasionalism, 43, 81, 86 ff., 101, 

118. 

Outness, 46. 
Oxford, 120, 177. 

Palliser, W., n, 12 n. 

Parsimony, principle of, 80, 118. 

Percival, Sir John, letters to (edi 
Rand), 2, 25 n. y 28 f., 44, 57, 
68, 126, 188, 195 n., 202. 

Philalethes, 131. v. Jurin. 

Plato, 149, 174 ff. 

Plotinus, 176. 

Plutarch, 174. 

Power, 146 f., 1 66. 



214 



INDEX 



Principle, the New, 20 f., 30, 58, 

68, 106, no, 125, 141. 
Principles, its relation to Theory 

of Vision, 25 ff. 
Principles, Sec. i, interpretation 

of, 72 ff. 

Problematical Idealism, 65, 96. 
Pure Intellect, 87, 89, 116, 162. 

Qualities, sensible, primary and 
secondary, 20, 42, 54 f., 61 ff., 
96, 162. 

Queries, 16 f. 

Rand, v . Percival. 
Raphson, J., 49 ff., 54, in. 
Reid, T., 9, 29, 70 n. y 104 n., 106 f . 
Representationism, 75, 88, 94, 

99, 123. 

Retinal image, 36, 39, 69, 185. 
Revelation, 116. 
Rhode Island (v. America), u n. t 

164, 166, 170. 
Rose, H. J., 193. 
Rossi, M. M., 199, 205. 
Royal Society, 38, 40, 49 f., 70. 
Rules of Method, 15 ff., 68. 

Sameness, 84, 162. 

Sargent, J., 104. 

Scaliger, 149. 

Schneider, H. and C., 167. 

Sermons, Berkeley's, n6n,, 176, 

196. 

Shaftesbury, 165. 
Signs, 23, 118, 124, 135, 155, 161, 

165 f., 168. 
Singling, 128, 139, 154, 162. , >a 



Situation, 36, 42, 185. 
Solitary man, 141, 158, 203. 
Sorts (species), 133, 136, 141, 

143, 152, 158 f. 

Soul, self, spirit, v. Knowledge. 
Species Impressae, 78. 
Spinoza, 9, 47, 49, 51 f., 81. 
Stephen, L., 10. 
Stillingfleet, 132 f., 136. 
Suarez, 48. 

Swift, J., u, 56 n. y 125. 
Synge, E., 191. 

Tar- water, 177. 

Taylor, T., 4, 14, 46. 

Teape, C. R., 10, 34 n. 

Temple, Sir W., 15. 

Theory of Vision, its relation to 

Principles, 25 ff. 
Torricelli, 164. 
Transubstantiation, 48, 61. 
Triangle, Locke's 'absurd', 

130 ff., 142, 150 ff., 174- 
Trinity College, Dublin, n, 38 n. t 

120, 177. 
Trinity, the, 165, 177. 

Unio Mystica, 79, 81, 114, 116. 
Universals, 80, 99, 133, 156, 166. 

Vortices, 48. 

Wallis, J., 40. 
Whiston, W., 9, 125. 
Will, Divine, 89 f., 146; human, 
89 f., 166. 



PRINTED IN 
GREAT BRITAIN 

AT THE 

UNIVERSITY PRESS 
OXFORD 

BY 

JOHN JOHNSON 
PRINTER 

TO THE 
UNIVERSITY