A CRITICAL EVALUATION OF G. E. MOORE’S
TEEOBY 0F EMPIRICAL KNOWLEDGE

A Blsser’I’oIlon
Ior the Degree of DII. D.

MICHIGAN STATE UNIVERSITY
Shukla Sinha
I975

 
  
 
   

  

LIBRARY

Michigan State
University

    

.i

This is to certify that the
thesis entitled
A Critical Evaluation of

G. E. Moore's Theory of Empirical Knowledge

presented by-

Shukla Sinha

has been accepted towards fulfillment
of the requirements for

Ph. D . degree in Philosophy

Major pro or

 

0-7 639

vv— 3‘ H

 

movunmvflfl’fl'mr -
_ flaw v ‘-
wtflmyi'Hzo :\ .‘ ' ‘

,. .3.- mm "
t

._ 1 u‘f‘.
«on-rfl""" ~ fl. ' .

ABSTRACT

A CRITICAL EVALUATION OF G. E. MOORE'S
THEORY OF EMPIRICAL KNOWLEDGE

By
Shukla Sinha

Since it has been not uncommon for sense data phiIOSOphers to
discredit the claims of common sense regarding knowledge of material
objects, and since it has been customary among the upholders of common
sense views either to exhibit unfamiliarity with the very notion of
sense data or to discredit the language of sense data, Moore's views
regarding empirical knowledge form somewhat of an exception, since Moore
attempts to defend the claims of common sense without either avoiding or
discrediting the language of sense data. Naturally, questions have been
raised about the compatibility of his views and the consistency of his
approach. I have considered and discussed these questions and the
grounds on which their apparent plausibility is sustained. I have
tried to show that these questions have been successfully answered
by Moore.

The main objective of Moore's theory of empirical knowledge
has been to defend the knowledge claims of common sense as expressed
in everyday perceptual statements, against the views which contradict
them--views which Moore calls skeptical views. I have argued that

Moore has been successful and consistent in defending at least one

Shukla Sinha

of the most widely accepted claims of common sense, viz., the claim
regarding the knowledge of the existence of material objects, against
the skeptics' claim that such knowledge is impossible. I have further-
more, tried to show that although Moore does not deny the existence of
sense data and accepts it as a necessary accompaniment of our percep-
tual knowledge of material objects, this does not adversely affect his
criticism of the skeptics, and does not lead him into inconsistency in

his own positive views.

A CRITICAL EVALUATION OF G. E. MOORE'S
THEORY OF EMPIRICAL KNOWLEDGE

By

Shukla Sinha

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Philosophy

I975

 

ACKNOWLEDGMENTS

I wish to thank Professors W. J. Callaghan, R. H. Kotzin,

C. McCracken, R. Suter, D. F. Koch and R. J. Hall for being instrumental
in my developing a more specific outlook in certain contemporary areas
of philosophy at various stages of my academic tenure here.

Special thanks are due Professor Rhoda H. Kotzin for her
constructive criticisms and helpful comments related to this work.

The financial support of the Fulbright Hays Binational
Foundation and the Philosophy Department of the Michigan State
University is also gratefully acknowledged.

Most of all, I wish to express my sincerest appreciation to
my advisor, Professor Richard J. Hall, for his patience, counsel, and
valuable guidance throughout this work. Without his critical but

stimulating observations, this study would not have been possible.

******

ii

PREFACE

Chapter
I.
II.

III.

IV.

TABLE OF CONTENTS

INTRODUCTION .......................

MOORE'S COMMON SENSE REALISM ...............

Section I. Moore's Realism ..............
Section II. The Criteria of Common Sense .......
Section III. Nature of Material Objects ........

MOORE'S PHILOSOPHY OF SENSE DATA .............

Section I. Introduction of Sense Data .........
Analysis of Perceptual Statements .........
Analysis of Sense Perception ............

Section II. Moore's Characterization of Sense Data . .
Sense Data Are Directly Apprehended ........
Sense Data Are Different From Sensations ......
Sense Data Are the Kind of Objects Common to

All Actual and Possible Sensory Experiences

Section III. Open Questions About the Status of

Sense Data ........ ' ..............
Are Sense Data Totally Private to the Perceiver? . .
Can Sense Data Exist Unperceived? .........

THE MAJOR OBJECTIVE OF MOORE'S THEORY OF EMPIRICAL
KNOWLEDGE AND THE PHILOSOPHIES OF COMMON SENSE
AND SENSE DATA ......................

Section I. What Is Moore's Major Objective ......
Section II. The Two Kinds of Objects of Knowledge

for Moore ......................
Section III. Questions About the Coherence of

Moore's Two Standpoints ...............
Section IV. Two Underlying Reasons for These

Questions ......................

12
18

26
26
26
3O
35
35
37
38
4O

46

54 P
54
56
59
6l

Chapter

V. MOORE'S ANALYSIS OF KNOWLEDGE AND ANSWER TO
SKEPTICISM ........................

Section I. Why We Have to Consider Moore's

Metaepistemological Views and Distinctions ......
Section II. Moore's Metaepistemological Views
and Distinctions ...................
Relation of Propositions to Knowledge and Belief . .
Ways of Knowing ..................
Certainty .....................
Types of Knowledge Proper .............
Section III. Moore's Epistemology and His Answer
to Skepticism ....................
The Common Sense View Reviewed ...........

The Metaepistemological Distinctions Appealed to . .
Moore's Defence of Common Sense Against

Skepticism ....................
Proof of the Existence of Material Objects and
of Knowledge ...................
VI. COMMON SENSE, ANALYSIS, AND SENSE DATA IN MOORE'S
EPISTEMOLOGY .......................
Section I. Why Does Moore Analyze Common Sense
Views ........................
Section II. Moore's View About the Role of Analysis
in Phil050phical Discussion .............
Section III. The Relation Between Sense Data and
Material Objects ...................
Epistemological Relation Between Sense Data
and Material Objects ...............
Ontological Relation Between Sense Data and
Material Objects .................
Section IV. Moore's Final Views About the Relation
Between the Two ...................
FOOTNOTES ............................
BIBLIOGRAPHY ...........................

iv

Page

67

67
71
71
79
89
97
99
100
102
112

141

152

154
161
164
165

PREFACE

The significance and the influence of G. E. Moore's
contributions in the fields of contemporary ethics and epistemology
is obvious from the number of questions and the amount of discussion
his writings have originated in these areas. In particular, his views
regarding perceptual knowledge and the meaning of ethical statements and
concepts have not only provided contemporary philosophy with many new
problems, but also with new trends of thought for approaching these
problems.

Despite the difference in the subject matter, his views in
these two areas share in common his analytic approach towards the
particular problems in question, and his steady pursuit of methodical
questioning so characteristic of his writings. Furthermore, his major
views in both of these areas show that his main interest has always been
to investigate the epistemological aspect of the problem.

A consideration of certain aspects of Moore's views about
perceptual knowledge in relation to certain issues in his ethical views
led to the initial planning of the present work which was originally
designed to analyze the bearings of his views regarding the direct
apprehension of sense data and those regarding the intuitive knowledge

(If intrinsic goodness relative to his views regarding the epistemolog-

 

ical and ontological status of sense data and the good.

!\
pr“ ICU-
T’s/“v.

. .

AA. 1"
2;! _I 1 MT

,3 ..-:Ju

. .
Ht'.'ln.c|‘ *
n =0.“

 

“ >-

n .. ‘~‘

-’ ' "V‘f;
"~':' C '.,.
‘9. ,kV

N .r ‘ Far:

However, in the process of reading Moore's views about
perceptual knowledge, in preparation of this work, I have been
confronted with certain questions which have seemed to me not only
equally interesting and equally complicated in character, but also
of a more fundamental nature. It seemed to me that a detailed
discussion of these questions would be necessary not only for a
proper understanding of his epistemological views but also in order
to be able to appreciate any problem which would relate to both his
epistemological views and ethical views. Hence, although the problems
this work was intended to deal with originally relate to both episte-
mology and ethics, I have confined the discussions in the present work

to problems related only to Moore's views regarding empirical knowledge.

vi

 

__‘.
(ID

r3. m m

“an.
1

I ,
I H... A
.. I

“Ely: Uf C3”

Int .I
D

and. Statf

"-
:T":F:q+ 5 .
' -‘ - c}

II: :..f; A
T" one U

"um-1

,-

I“‘"du Q +'.
_. m.

I.-
:159 jam

(/7

‘\
.s!" ,

‘1 n-

. “S <

‘

:-
t a ‘ ‘
.§
.2

C.

CHAPTER I

INTRODUCTION

It is quite clear from Moore's views regarding empirical
knowledge that his major objective has been to defend and justify the
claims of common sense, particularly those expressed in every day per-
ceptual statements, against any view that contradicts them. Among the
different types of such views--views that Moore has called skeptical
views--the one that he has been mainly concerned with is the view
according to which the only objects We can kggw_the existence of are
sense data, where it has been held conjointly by Moore and the upholders
of such a view that sense data are objects which are basically different
from material objects. Naturally, it follows from any such view that,
if all that we can ever know the existence of are sense data, then we
cannot kggw_the existence of any material objects or an external world;
whereas it is one of the most widely held views of common sense that
we all do know of the existence of material objects and an external
world.

Since Moore has asserted that such views of common sense are
known to be true beyond any question, and has claimed to be defending
them as such, his acceptance of the existence of sense data at the same
time leads his readers to raise certain questions which seem quite

plausible'at first glance. At least two such obvious questions that

21:9 ”355'

"'7" 1r33
r L‘HJVLU

. .
u one-pd r-
‘I r .

.Lu.»i'\r

O
155:: ~r r
to I

' .

n
h”.
15‘-

‘I; and 1
.‘. 1w,

:“phv r

a... a. r

"‘5!
i" 5"o"

n
1' ‘FF
.. _ :39
.
v
l‘j‘iEPq
'"\ l' 3‘
3 u-
a, P ‘
u. f *
hl‘ , Inc,
5
‘in ' .
a! :‘":l
~d~4~
‘Il-PA a.
" t we.
5.
I.

0.:
4 '3! I,”
H;
.3
Z ;‘

nit'S‘
\d'l ie t
t’;\.

' u." trz‘

may be raised in this connection are: (l) Why ha§_Moore, after all,
introduced the notion of sense data, when his main purpose has been
to defend common sense? And (2) 15 his sense data philosophy, once
he has introduced it, compatible with his philosophy of common sense?
Both of these questions are apparently reasonable. For on the
one hand, it is a fact that the very notion of sense data with all its
attendant problems and the language in which these problems are
expressed, is more or less unknown to the common sense parlour, whereas,
on the other hand, at the philosophical parlour when a discussion is
carried on through the language of sense data, it has been more or less
customary (or traditional) to deny the knowledge of those facts which
are accepted by common sense as well known. Hence the question arises:
Can these two standpoints about perceptual knowledge (or, two levels of
looking at the problem), apparently so widely apart, be reconciled with
each other relative to the main objective of Moore's theory of empirical
knowledge? And this question implicitly includes the question: Has
Moore been successful, after all, in pursuing his objective consistently?
The present work is an attempt to investigate these two questions. The
main object of this work is to show (l) that the two standpoints main-
tained by Moore--common sense and sense data--occur as relevantly
related aspects of his theory of empirical knowledge within the
framework of his main objective; and (2) that Moore's approach towards
his main objective, throughout his works, has been a consistent one
despite the two apparently divergent standpoints. I have tried to

show that when these two standpoints are considered in relation to

.. H
n' 4

43‘
t.)

.iu f‘.‘
:I w

.
.o'-‘
A PF

v I“ ‘“

..,..co
.‘ ,f

and»

«soy.-
l.-. ‘3

upper I

v v--‘

 

‘A.

I':"€

0.. :

o
I
n!

\
‘ .
c

certain distinctions that Moore has made and maintained throughout
his writings regarding empirical knowledge, the apparent incompatibility
of the two disappear. Furthermore, I have tried to show that if we
consider the arguments Moore has advanced against the skeptics together
with the fact that he recognizes analysis as an important part of the
'proper business' of any philosophical enterprise it becomes clear that
he has been successful in approaching his goal with perfect consistency.
It should, however, be mentioned here that I have chosen to
limit the scope of this work mainly to the discussion of Moore's views

regarding the knowledge of the existence of material objects and his

 

views regarding visual sense data. The main reason for this has been

 

the fact that although his views regarding perceptual knowledge include,
to a certain extent, a discussion about the knowledge of the existence

of other persons, and a reference to sense data given by senses other

 

than sight, viz., tactual sense data, auditory sense data, etc., the

 

major portion of his discussions have been devoted to the problem of
knowing the existence of material objects which he seems to have taken
as most fundamental in this context, and to visual sense data which he
seems to have taken as representative of any kind of sense data given
by any of the senses.

The discussion in the present work begins with expositions of
Moore's common sense realism (Chapter II) and his philosophy of sense
data (Chapter 111). These expositions raise certain questions regarding
the relevance and compatibility of these two philosophies relative to

Moore's main goal which is to defend and justify common sense knowledge

, .
.-~-a ‘9:
a ‘\

v.
a. -
II .,

"av-"n:
b

a. II

14:..rr9~

... bvr

u
a

.o .n. ',.

1“ .1.

' '0 ~-

 

claims against any view that contradicts it (Chapter IV). In order to
determine whether these questions are justified or not, a consideration
of the implications of Moore's analysis of knowledge and the various
distinctions he makes there in relation to his arguments to justify his
claim, has been presented in a further discussion (Chapter V). Finally,
it has been shown, on the basis of the preceding considerations of
Moore's epistemological views together with his views about the role
of analysis in philosophy, that these questions cannot be justifiably
raised against his theory of empirical knowledge. In other words, it
has been shown that his two philosophies are relevantly related to his
main objective which he has been able to pursue consistently and
successfully (Chapter VI).

I have mainly based the discussions in this work on the original

writings of Moore and only in passing have referred to the views of some

of his commentators.

CHAPTER II

MOORE'S COMMON SENSE REALISM

Section I. Moore's Realism

 

Moore's theory of empirical knowledge consists mainly of his
views regarding the nature and object(s) of perception as a cognitive
process. The maig_objective of his theory of empirical knowledge has
been to show the untenability of all those theories which imply a
skepticism about our knowledge of the existence of the different things
and beings which we find around us in this world, and which we all com-
monly believe we know about with certainty. This skeptical attitude is
generally expressed by claiming that experience is the only source of
our knowledge of the empirical facts, and that by means of experience,
we cannot know with certainty anything beyond our own existence, feel-
ings, perceptions, and thoughts. If this is the case, then se do not,
and cannot, know of the existence of the many different things which we
all ordinarily think we know of; for instance, we all believe we know
that there are different kinds of material objects, and different kinds
of living beings including human beings having feelings and perceptions,
and that all these different things are in space and time, and so on.
‘The common sense view of the world consists of such beliefs and
knowledge claims. These beliefs are tacitly assumed in such everyday

statements of ordinary language as, 'This is a table,‘ 'I see that a

'ru- Ph-
. 3.: Lu

,;.,.

“C ..|~v firm .1. r” S \J .4- I- 2. _,... lu I; .C
i 1 .I a; lvh .1 1w. .y I . Tr . . RI r» a,» S» nl... 6 s P1
r» o. w 2,. .s... 5 all Fl r, . . n A: «J .l. S S. 9|. 9: Cs .3 A)
I r.» u If. rd 9» .3 .1. n: ... . . I TV I
I . . u» a 3 ,I . 1‘ w P a a. V

-\J III. ‘ u .S v r ab.» ‘4‘ E1 n . ‘l. I" c an. .. I L

a w e) .au Fl- PM {a -1 l H u I V «Hy A~s
.1.) a. i a: a.» L . L. Ale .ms 4 . t v P -u

0 v F. nu. O i ' an. Nut 0) JUL. A v s. . PL. c.\.. I.) F. r 1‘ .t'

u s an. F d :- a\.¢ .\. o . .l.u IF I V ll o u d 9, .\u

at. n .- no. . I. , «4‘ .I.,u n. v a\- \u. n0 n h I. .V. F\u
.1. in It. n: F. .3“ J... ..... .a . .... v; u :l T. u . .. .. .. .. .5. .u

child is standing by the side of a groWn-up man,‘ 'That is an old
building,‘ and so on.

These statements are perceptual statements of common sense
which are commonly held by all of us as undoubtedly true. If the
skeptics'claim about the extent of empirical knowledge is true, then
we are not justified in holding such statements as known to be true.
Moore's main objective in his theory of empirical knowledge is to
refute this claim of the skeptics, and to establish the claim that
we are justified in holding that such perceptual statements of common
sense are known to be true. It is this enterprise that Moore's common
sense realism consists of. As such, his theory of empirical knowledge
deals mainly with the problem of perception and its different aspects.
To be specific, it mainly consists of his views regarding the gatgrg_
and object of perception as a cognitive process.

There are two aspects of his theory of perceptual (or empirical)
knowledge: (l) the aspect which consists of those views that constitute
his common sense realism; and (2) the aspect which consists of those
views which constitute his sense data philosophy. In this chapter I
shall discuss his common sense realism with special emphasis on some

of the issues with which I shall be mainly concerned later in his work.

In order to discuss his common sense realism, it seems important

to understand his basic epistemological position regarding the nature of

cognitive process in general; the particular position that he holds in
this regard is realism. Since his epistemological views mainly concern

his views about perceptual knowledge, they include his views regarding

.
.‘l..."] ”1‘?
‘ '16 w .

"u. u I
u

I ~ Fl “
I n r.‘
a p
51:0"); 5, -

'653 C5“. 91'

c“)

the nature of cognition or knowing as such. Moore's realism is a
result of his attempt to explain and analyze the nature of any
cognitive process in general.

What he has mainly intended to show is that every cognitive
process is a relation between two factors, viz., the act of conscious-
ness and the object of consciousness, and that the object of conscious-
ness can exist, in some cases at least, quite independently of the act
of consciousness. These are the two main contentions of epistemological
realism as conceived by Moore. Moore's epistemological realism came
into existence as a reaction against the idealist philosophy in general
and the idealist epistemology in particular. He started out by reacting.
critically to certain specific claims made by the idealists about the
nature of knowing and knowledge in general. Out of these critical
reactions his own epistemological theory of realism has emerged and,
in the process, has gradually taken the form of a rival theory of
knowledge on positive and independent grounds.

One of the earliest and yet, perhaps, the most important
critical reactions in this direction is to be found in his article
"The Refutation of Idealism." In this article he critically analyzes
one of the most fundamental claims of the idealists, namely, 'esse is
percipi' (i.e., to exist is to be perceived), which, according to Moore,
has very important bearings upon the nature and extent of knowledge. In

this essay he has tried to show the untenability of this claim by

.
ia':'
. I i

.. ., A l
p ' l
"I dItr'QC

analyzing the nature of cognition in general, and of perceptual
knowledge in particular. According to him any case of sense perception

is a case of knowing; to be specific, it is a cognitive relation between

 

the knower and something else which is the object of the knower's
awareness. He says that any case of sense perception involves two

factors: the object which is known and the awareness of the object.

 

This distinction between the two factors of knowledge is the basis of
his realism. He has wanted to emphasize the point that the Idealists
have overlooked that this is the case, and as such their claim is based

on a confusion: the confusion consists in their identifying the Egg.

 

distinct factors of knowledge. According to Moore the Idealists have
analyzed knowing in terms of 'content' and 'existence.‘ They say that

in each case of sensation, "we can distinguish two elements and two only,
(l) the fact that there is feeling or experience, and (2) what_is felt
or experienced; the sensation or idea, it is said, forms a whglg, in
which we must distinguish two 'inseparable aspects,‘ 'content' and
'existence.'"1 Moore holds that this analysis is false on the grounds
that their use of the word 'content' in this context is inappropriate.
According to Moore what is most commonly meant by saying that one thing

is the 'content' of another, is the general relation that holds between

 

a thing and its qualities; "and that this relation is such that to say
that the thing exists implies that the qualities also exist." For
instance, in the case of a blue flower, it may be said that blue is
part of the content of the blue flower, or a blue bead. If by the

relation expressed by 'one thing is the content of another,‘ we mean

U
u

u
«lay-c
an?

"r2" h.

n“

 

I .x. P.U .Pn II a -\ u l I
o u 2.. .h u C 111 61. r ..
. F a Ir] P\o (Ill. AF.» “ .. .- If: .S-N
r O o I. ~-— a
o L a. . p: .: o . F. :0. .5.» c, . 3. .1
A .~ . a . :~ ‘u u d nip. :‘
.U\v a. I nth» -- - Nil- ..N I. u ‘4 n 5 ~ Pt— . xv

this relation, then it follows that blue is related to the sensation
of blue exactly in the same sense as blue is related to a blue bead;
and just as in the case of a blue bead, which is a whole, there is,
at least, another content besides blue, viz., glass, so similarly in
the case of the sensation of blue, which is a whole, there is another
content besides blue, viz., consciousness. If this is what we mean by
the expression, "one thing is the content of another," then when we say
that blue is the content of the sensation of blue, what we are saying
is that blue has to the consciousness exactly the same relation which
blue has to the other parts of a blue bead. And in this sense, blue
exists only as an inseparable aspect of the sensation of blue, just as
blue (being a quality of the blue bead) exists as an inseparable aspect
of the blue bead. "The content of the thing is what we assert to exist,
when we assert that_the thing exists."2

But if this is what the Idealists have meant by saying that
blue is only a content of the sensation of blue, then we should be
able to speak of a 'blue sensation' without deviating from the common
usage, just as we can and do speak of a blue flower or a blue bead;
but the fact, according to Moore, is that although we do correctly
speak of a blue bead, we do not and cannot correctly speak of a blue
sensation or a blue consciousness (just as, blue glass, in the case of
blue bead); hence, although we can correctly speak of blue as the con-
tent of a blue bead, we cannot correctly speak of blue as the content
of the sensation of blue. This is because the relation between blue

and the blue flower or its parts does not hold between blue and the

‘ I
,. ny~a
xix-wt 3'
... b L F
. ,f H: 'T
,h .\.- u

y n

"1'3"“ ir-
II

n! v'v‘l'

v v
C:f:3"‘lq in
us‘ui.:g. l

exzerienci

I'D
(D

-. ,
:F.

.
7'“- ero'r.
w .:,,_ -;

'P 'fifi-HP .
L " ' ‘4') I

I‘- U‘fir:
“cu. ~‘

be

 

. ‘rA
\C, Q
:.

.p _

in

1O

sensation of it or its other parts, which it should if blue is to be

a content of the sensation.3 Hence, the Idealists' use of the word

'content' in this context remains inappropriate (or else unexplained).
The reason for this is that they have not been able to find

the correct analysis of sensation or knowing itself. For Moore, any

 

sensation is, in fact, a case of 'knowing' or 'being aware pf] or
'experiencing something'; and in every sensation we must distinguish
two elements: (I) the object, or that in which one sensation differs
frmn another, and (2) "consciousness, or that which all sensations have
in common and by virtue of which they are sensations." Thus, according
to Moore, in every case of knowing or experiencing something, there are

two distinct factors: the mental act which is (or consists of) the,

 

awareness of something (or being aware of something) and the gbjegt_
(of awareness), i.e., 'the something' which the awareness is 9f, And
if we look 'attentively enough,‘ we should be able to recognize further
that these two elements are perfectly distinct and utterly different
from each other, and the two are related by a perfectly distinct and
unique relation which is what is meant by 'knowing.‘ The Idealists,
in holding the content theory, have failed to recognize precisely this
fact, viz., that there is a unique relation between two distinct ele-
ments. Moore says that though they have "recognised that something
distinct is meant by consciousness, they have never yet had a clear
conception of what_that something is. "They have not been able to hold
jt_and blgngefore their minds and to compare them. . . ." The reason

for this is that "when we try to introspect the sensation of blue, all

- II. V.
no.4. :pd l
. .. ..L
——

Dp-o ‘I ~ A
~‘ mfg?

5
'9, A4
(,a ‘5
“b
'-'r o .
L1,. N'H‘
why.
1;”. “‘1"?
-,. l ,
‘ d“ I

11

we can see is the blue: the other element is as if it were diaphanous.
Yet it gag.be distinguished if we look attentively enough. . . ."“
Since the object of awareness is perfectly distinct from the awareness
gj;jt, and knowing consists in a perfectly distinct and unique relation
that awareness has to the object, there is absolutely no reason to
suppose that the object or knowledge can exist only as a content of
knowledge or awareness, and as such, only as an inseparable aspect

of experience.

What Moore has attempted to show is that the 9bjggt_of
knowledge is not necessarily related to the awareness by a relation
such that the object's nature is necessarily determined by and depen-
dent upon the awareness of it. When we analyze a sensation, what we
find is that a sensation is always a sensation or an awareness 9f.
something; i.e., the awareness cannot exist without an object. But
from this it does not follow by any means that the object of awareness
cannot exist without the awareness of it. According to Moore the object
of consciousness is neither the same as the awareness of it, nor is it
the same as an image in the mind. Hence, at least in some cases, it
can exist independently of and unaffected by the awareness of it.

This is, in brief, Moore's epistemological realism.

 

 

‘r.

c. ‘
nlr‘ .r
u. .

U "
n .
‘.-’.‘ A.'
... ii A

.n

r.-

255. .
141-i \ ‘
s

I"
If)
4"
‘I,
ll.)

12

Section II. The Criteria of Common Sense

I have already stated in the beginning of this chapter that
the main objective of Moore's theory of empirical knowledge has been
to establish the common sense claims about the knowledge of an external
world based upon sense perception. His epistemological realism as
opposed to the Idealists' theory of knowledge may be said to be the
first step in his pursuit of this objective. His analysis of knowing
into two distinctive factors provides the basis for his common sense
realism. In fact his reaction against the idealistic claims in general
stems from the fact that such claims come into conflict with the claims
of common sense. Although Moore has always made a distinction between
the common sense view of the world--i.e., the views that all Of us
ordinarily accept to be undoubtedly true, and those philosophical
views which are incompatible with the common sense views, he has not
meant to say that the common sense views are not or cannot be philo-
sophical, nor has he meant to say that philosophical views as such are
not or cannot be part of the common sense views. All that he has meant
to say is that sometimes some philosophical views are in conflict with
some views which have always been held by all of us (even by those who
contradict the common sense views in their theories). One of the
objectives pursued by Moore in most of his works has been to examine
critically and carefully such philosophical views as go against common
sense views, and to determine whether these views are based on valid

grounds or not.

.4
g 0
In

,,

.|..‘o IA~-;i
.' v av: "‘
n.” b t
as \l' ’2

u-k I"'

 

:15“; an

'2' r: @é
.‘ J= L

’.~*R

.J." ’I‘ (:9

 

C-
.
., ,. in
"F

13

It should not, however, give us the impression that Moore is
being dogmatic in defending what is commonly held by everybody, in the
sense that whatever is commonly held by everybody amounts to an authentic
philosophical view, or that any phil050phical view, in order to be sound,
must be in total agreement with what is commonly believed. On the con-
trary, he has held that sometimes what is commonly believed is certainly
false. He admits at many places that views commonly held to be true
undergo changes with time; that many examples of such cases are to be
found in the areas of ethics, theology, astronomy and geology; that
many of them have been proven to be definitely mistaken. In advocating
common sense he is not advocating any extreme position; all that he is
advocating is that most of the time it is more reasonable than not to
accept a philosophical view which is not in conflict with certain views
of common sense, in the sense that its consistency with these views may
be an additional ground of its acceptability. He seems to be suggesting
that any phiIOSOphical theory which is in opposition to the views of
common sense deserves to be treated with some additional amount of
suspicion so far as its truth is concerned.

At this point it may be asked if Moore has described any
definite set of criteria to determine whether a given view is or is
not in accord with common sense. Moore has never given a formal
specific definition of what he calls a common sense view. He has
often referred to it as the belief held by all of us, or the beliefs
universally agreed upon--something like a universal opinion; sometimes

he refers to it as what is believed by ordinary people. However, in

14

all these uses, the common sense view has been referred to as something
which is taken as certainly true--something.about which there cannot

be any doubt. On the basis of what he has said about the common sense
views in his different works, it seems that he does have certain
criteria according to which we do decide whether or not a belief

is a belief of common sense, although he has not given any such set

of criteria at any particular place in his works. White, in his book,
G. E. Moore--A Critical Exposition, enumerates five criteria which

have been used by Moore at different places of his works, though not
clearly specified and mentioned as such by him.

I. The criterion of universal acceptance: It seems clear from
what has been discussed above, that this is perhaps the only criterion
which Moore has quite often explicitly specified in characterizing a
view as a view of common sense. That is, if a belief is universally
accepted then it is a view of the common sense; for instance, our
beliefs that there are many different kinds of material objects, that
there are acts of consciousness in different living beings, that these
acts of consciousness are different from material objects, that there
are units of time and units of space, that material objects occupy
space whereas acts of consciousness do not occupy space in the same
sense as material objects do, that there are human beings having
similar thoughts and feelings, and that all these things and beings
share a common world which we call the external world; i.e., they do
not exist inside the mind or minds of any individual being or beings,

that except the acts of consciousness all other things exist outside

 

 

 

ll-v-A
0 d. 3 ‘
1‘9‘!\ 'n'
‘1': .:
' ‘t‘ :
illffi

.1.,

.4.

l I"

 

15

our minds and this locus of all these things is what we mean by an
external world.

This criterion brings out clearly that what Moore has meant
by 'common sense' is not something like what we ordinarily call the
ability to make quick and good judgment in a given situation (as in
such remarks: X is a person of sound common sense, or, It was a matter
of common sense to do that in that situation, etc.).5 Further, although
Moore has often recommended universal acceptance as a reason for holding
a view to be true, he has not meant, in general, that it should be
regarded as a proof for the validity of the view.

2. The criterion of compulsive acceptance: This refers to

 

such statements and beliefs "which we not only all do hold, but which
we often cannot help holding, even if at the same time we hold beliefs
inconsistent with them."6 However, this fact does not prove that such
a view j§_true, or its contradictory, false.

3. The criterion of inconsistency: This refers to beliefs

 

which, if denied, result in various kinds of inconsistency. For
instance, any view which denies the reality of time, also contains
propositions that presuppose the truth of the view being denied, for

such a view says that we constantly hold a belief which is not true;

 

where by saying that 'we constantly believe in things which are not

 

there,‘ it is implying that things do happen in time. That is to say,
there is no way that it can avoid doing this; for it is a fact that we
do constantly or always believe in time; they cannot deny this even if

they want to deny the truth of such a belief. Any philosopher, no

,5? "Cr-i

,uo
.

 
 

. c . .... : . {I I
3 .v .r 4 .5 (a r 1 . .c _ . l. 2. . . l 2. p s . . a . .
r Pu A o A.- o i «AM IV. as. ‘ a AE . pm. i F P. . Au. .\~ «ab V .. ‘ u .9
. . . .. r r. . c as. A . . . at r. .. r . . A r. n be -r. . .. T C c
. I. 2. . . .o.- r .n. n..- . u .4 . u p: 2. o 6 2. .IV. .3 {I P:
a A. .. .. .. ...: E a" . .. .. .. f .
. . .s. . ... . .: l. s . ale La
"Vt u: u: a . u. ..... r .T . . . I.

16

matter however sincerely he may adopt the opinion that there is no such
thing as time, will never

be able to divest himself of particular beliefs which

contradict his Opinion. He will still continue to believe,

with regard to particular things, just as certainly and

as often as the rest of us do, that some of them do exist

before others, that some of them have ceased to exist and

others not yet come into existence, and that the intervals

of time between particular events differ in length.7
Moore has talked about different variations of this type of incon-
sistency at different places of his works. For instance, at one place
he says such a view may not presuppose the truth of the very thing that
it is denying, yet may presuppose the truth of some other view which is
of the same type as the one it denies. While discussing Hume's views
about the matter, Moore says that Hume himself

declares that we cannot, in ordinary life, avoid believing

things which are inconsistent with them [that is, with

skeptical views that we never know any external facts];

and, in so declaring, he, of course, implies incidentally

that they [skeptical views] are false; since he implies

that he himself has a great deal of knowledge as to what
we can and cannot believe in ordinary life.8

4. The criterion of special kind of inconsistency: This
refers to a particular kind of inconsistency that results from the
very acceptance of the fact that a view is a view of common sense.
This is the particular type of inconsistency that Moore has referred
to in his "A Defence of Common Sense." He says that certain common
sense beliefs

have this peculiar property--namely, that if we know that
they are features in the "Common Sense View of the World,"
it follows that they are true: it is self contradictory
to maintain that we know them to be features in the Common
Sense view, and yet they are not true; since to say that

 

Lu
.7"
u "
'1!
II

«3"
E

.
n? "f‘

.r "'3"

‘ V'lbl

n~ P

.. Qn‘

.: : J
a: o ,A II
1.x: .C'
"z“: N
.. . h-

.
AVA-v...‘
. I I

l

'1‘.

3

I

use.
51

c
UT»
Him
C.
4,.1
..
DUI.

(-
\D.
._, t

‘A.
‘-
'5

17

ye_know this, is to say that they are true. And many of

them also have the further peculiar property that, if the

are features in the Common Sense view of the world Iwhether

”we" know this or not), it follows that they are true,

since to say that there is a “Common Sense view of the

world,“ is to say that they are true."9
In other words the moment a skeptic says that "we know that it is a
belief of common sense, but it is not true," he is contradicting himself
on two counts: first, in using the word "we," he is admitting that
there are persons other than himself and that he knows it; second, even
if he does not use the word "we," but accepts that certain beliefs are
mere beliefs of common sense, i.e., admits of the very existence of a
common sense view, whether or not he also admits that they are true,
he is committing himself to the truth of at least part of this view,
in the sense that he admits that it is true that there have been many
people other than himself who have held such views whether such views
are true or not is a different matter. (This move of Moore's will be

discussed in detail later in this work.)

5. The criterion of "self evidence": This refers to the "self

 

evident" character of at least some of the beliefs of common sense.
(The expression 'self evident,‘ however, is used here in a modified
sense; it is used to refer to the fundamental or primitive nature of
certain beliefs which we never question--beliefs which we all assume

to be true, and which, in a sense, form the basis of our everyday
conversation in ordinary language. And as such, the use of this
expression in this context is not to be confused with its use elsewhere
in philosophy, where it is used to refer to propositions which are

intuitively known and the opposites of which are inconceivable.) By

~ I
A
x.“
p
\
a
4;-..
"36!

“it

5

. L'

at n-~
R. a c
.A

r ’r
V
. ‘Irrra‘
'2

.
,qlfifl

.

u-IV'
:r; I“,

a)
uni

’ u

1

’8

cl

nu-
I‘VL

..'_
-v:

A"!

.4.:)

18

the self evident character of the beliefs of common sense what is meant
here, is that the truth of such beliefs does not depend on any other
factor than that they gpe_just immediately kppwp_to be true; that they
are known as such without any knowledge of any other belief or statement.

In Moore's Principia Ethica and in Ethics, the same idea has been

 

expressed in terms of intuition or inspection. The truth of such views
is not a matter of strict demonstration; we just do know them to be
certainly true, and any other proposition that can be advanced to prove
or to disprove such a view is less certain than the view in question;
and as such they cannot be proved or disproved formally, but they are
true just the same and we all know them to be true. For Moore, state-
ments like 'I know that this pencil exists,’ are known immediately, and
are much more certain than any premise which could be used to prove or
disprove their truth.'° However, he also admits that this criterion
does not pppye_that common sense statements are true; "I can only urge
that these things are self evident: I can't think of any argument to

u 11

prove that they are true. And hence, if anybody asserts "that the
contrary is evident to him . . . I do not see how it can be proved that

he is wrong."
Section III. Nature of Material Objects

In his "A Defence of Common Sense," Moore says that we all
commonly believe we know different things and different facts. In
the preceding sections I have presented various particular examples

of such beliefs. All these particular beliefs have been broadly

19

classified by Moore into four groups: '(I) beliefs about the existence
or reality of material objects; (2) beliefs about the reality of acts
of consciousness in human beings as well as in other living beings
(where 'acts of consciousness' is used in a very general or wide sense);
(3) beliefs about the reality of space; and (4) beliefs about the
reality of time. Alternatively, it may be said that we know different
kinds of facts: physical facts, mental facts, spatial facts, temporal
facts. These are most fundamental beliefs of common sense and these
beliefs are implicitly assumed and expressed in our everyday statements
of perceptual knowledge in various forms. Moore has discussed beliefs
falling under each of these classes at different places in his writings
on empirical knowledge. However, in this work, I shall be concerned

only with his discussion of beliefs about our knowledge of the reality

 

of material objects as expressed in everyday perceptual statements.
One of the reasons for this is that a major portion of Moore's theory
of perceptual knowledge has been devoted to the discussion of this
matter; it seems that he has taken the discussion of and his views
about this particular issue to be the representative of his discussion
and views of the common sense view of the world in general. According
to Moore the belief in the existence of material objects is one of the
most primitive beliefs that still continues to be held by all persist-
ently. We believe that there are in the Universe enormous numbers of
material objects; and we believe too, nowadays, "that the earth itself,
and all that is in it or upon it, huge as it seems to us, is absurdly

small in comparison with the whole material Universe." Although some

‘ .
.‘ A or s'

'. Ty Ar;

, . y r: '

or: _,

. n ‘6, V‘

 

:1. .1- i. u. r l.
r. .\. . - . u I . .- I
a.» .2 r. . .c T. .1. 5. . Rx. La tr.“ ,9. .llf “J, S F. p c . -1. r
.l s . A». p. A: a v e a 9.1., id .4. ~11 ~l- PR P 93 if. P”. 2‘ P... ‘1 ..
3 u . {a . \a .b v r .o .n . o. F. . J h: a . l . -
.Pr. . o v o . KP. . v 2. 3 5.. u i ”I .u 3 H .. fir .n n pl.» s .a 5 o C»
2..- .s. n. . ~ . .x. .. l f o u 2. :s :
.~..... ~ . \l I at.- A .- fi‘ 3.. I! ,-
.. . z. .9. u... .w u... .. . .. a J.) "2 ... .. 3. .n...

20

of the primitive beliefs about the material Universe have undergone

change, or have been proven to be false,

yet,so far as concerns the point that there are in the

Universe a great number of material objects, it has, so

far as we know, remained the same. . . . Men have believed

this almost as long as they have believed anything: they

have always believed in the existence of great many

material objects.12
In a sense all other beliefs either presuppose the belief in the
existence of material objects (for instance, beliefs in other persons
or other living beings), or, are very intimately related with the
belief in the existence of material objects (for instance, beliefs
in space and time). Naturally when the truth of such a deep rooted
belief is questioned, its discussion deserves considerable attention
and priority. Hence, in the pursuit of his rmrhi objective, which is
to refute the skeptics claim that the views of common sense are not
known to be true, the first thing that Moore has wanted to establish,
is that we do know of the existence of material objects. The question
regarding the knowledge of the existence of material objects has been
given priority by Moore also for the reason that its discussion may be
useful for the better understanding of the different ways of knowing
and also the knowledge of things other than material objects.

And first of all, I shall consider the question: How do we

know of the existence of material objects, supposing that,

as Common Sense supposes, we do know of their existence?

. . . trying to answer the principal objections of those

philosophers, who have maintained that we certainly do not.

In the course of this discussion we shall come upon a good

many conclusions as to the sorts of ways in which we know

things; and shall be in a better position to consider what
else beside material objects we can know to exist.13

21

Before discussing the main question regarding the knowledge
of material objects, it seems necessary to discuss his views regarding
the pgtppe_of material objects. In the rest of this chapter, then, I
shall discuss what Moore has meant by a material object; so far the
question of the possibility of the knowledge of material objects is
concerned, I shall discuss it later in this work while discussing
Moore's answer to the skeptics'challenge.

Moore has defined "material object" in two different ways:
the first consists of describing the essential characteristics of a
material object. He defines a material object as something which (l)
does occupy space; (2) is pp; a sense datum of any kind whatsoever and
(3) is not a mind, nor an act of consciousness.'“ Of the three, the
first is a positive prOperty, and the other two are negative properties
of a material object. By the positive property of a material object,
viz., occupancy of space, what is meant is that nothing can be a
material object except what is situated somewhere or other in space;
i.e., except what has a position in space. By saying that material
objects have positions in space or are situated in space, he wants to
include different possibilities as to the way of occupying space. For
instance, "some material objects may, quite possibly, occupy mere
points; though others, no doubt, occupy lines; others occupy areas;
and others occupy volumes."ls

By the second property, which is a negative property, viz.,
that a material object is not a sense datum of any kind whatsoever,

what is meant is that no sense datum or part of a sense datum, or a

..-.-
g

 

 

 

22

collection of sense data, is identical with a material object or a part
of one. (However, the question as to whether a sense datum is or is not
identical with a part of the surface of a material object, is an open
question which will be discussed later.)

It should be mentioned at this point that although Moore has
mainly and originally defined a sense datum, in a somewhat narrow sense,
as the object which is actually given by the different senses in differ-
ent cases of sense perception, he is using the term 'sense datum' here
in a relatively wider sense. This point may be illustrated by saying
that originally he has meant the sort of objects which we actually see
or actually hear and so on; for instance, when we look at a coloured
object, what we actually see is a coloured patch or an expanse of
colour. It is objects of this sort, with necessary changes in the
cases of different senses, that are called sense data in the narrow
sense.

The wider sense of 'sense data' that Moore is using here, can
be explained in the following way. It might be thought, on the basis
of the narrow definition of 'sense data,‘ that an object should not be
called a sense datum, unless it is actually seen or heard, etc. as the
case may be; since, it is only when an object is actually seen, and so
On. that it is, "strictly, given to the senses." Now he says that he
is not using the term 'sense datum' in this strict sense; rather, he is
so extending the term 'sense datum' to cover all those objects, actual
or possible, which resemble any sense datum, which has been actually

given to the senses. For instance, images of different kinds, are

11
1' If. 1‘ I
v " J
. uII "

\-‘.
.'I

an-
.\v

«‘5

:‘l.

Lu

AP»
m,“
A u
I)

S

“r-

 

9.

by
G c

23

never actually given to the senses, and yet they should be called sense
data by virtue of the fact that they resemble the objects which have
been actually experienced.

And just as we may say that every coloured image actually

is a sense datum, merely because it is a coloured patch,

even though it has never been actually given to the senses;

so we may say that if a coloured patch existed, which had

never been directly apprehended* at all, it would be a

sense datum, merely because it would be a coloured patch. '5

What Moore wants to emphasize is that a material object is not
a sense datum (or a collection of sense data), as some philosophers have
held, in either of the senses of the term 'sense datum' explained above.17
By mentioning the third property of material objects, which also

is a negative property, Moore has again tried to clarify and emphasize
how he wants to use the term 'material object.I By saying that a mate-
rial object is not a mind or an act of consciousness, Moore has wanted
to distinguish his view from those according to which things like wheels
and couplings of a train, which we ordinarily take to be material
objects consist of mjpg§,'° But minds do not or cannot have a position
in space at all, at least in the sense in which we talk of a material
object's having position in space. It is true that minds or acts of
consciousness are quite definitely attached, in a particular way, to
some material objects, and not at all attached to some others; they
are attached to some bodies in the sense that they all do occur at some

place or other ip_our bodies. And as such, they take place wherever our

bodies are. "My act of consciousness takes place in my body; and yours

 

*The term 'sense data' and other related terms like 'directly
apprehended,‘ etc., will be discussed in detail in the following
chapter.

24

takes place in yours; and our minds [generally, at least] go with us,
wherever our bodies go. . . . But we believe . . . no less certainly,
that to the vast majority of material objects, no acts of consciousness
are attached."19 Since there are material bodies to which no act of
consciousness at all is attached, being a mind or being conscious cannot
be a defining property of material objects, not to speak of its being
identical with mind or act of consciousness. Secondly, even though it
is a fact that minds or acts of consciousness do take place in our
bodies and are wherever the bodies are, it is also a fact that minds

or acts of consciousness gp_ppt_occupy space in the same way in which
chairs, tables, and, for that matter, even those bodies which mental
acts are said to belong to, occupy space. One main difference is that
we all do believe that material bodies, being situated in space, must
have shape; they must have the 'shape' of the part of space which they

° whereas we do not believe that minds or acts of consciousness

occupy;2
have any shape at all. We further believe that material objects are
external to our minds, in the sense that none of these are jp_any
mind(s), though some of them may be said to be with_(a) mind.

Moore's second way of defining 'material object' consists in
giving instances of a particular type of thing which we all believe
to know the existence of, and showing by that very process that there
are material objects. It seems that he wants to point out that when we
talk about these instances of a particular type of object we tacitly

assume to know that they are all instances of material objects. This

is what he calls definition by examples.

 

25

I can give a very clear definition--a definition much

clearer than phil050phers generally give. . . . It's

definition by examples: i.e., all I can tell you is that

I use "material thing" in such a sense that if there are

any chairs, or desks, or blackboards, or planets, or human

bodies, etc., then there certainly a§e_material things.
He thinks that there certainly is ppe_proper use of the terms 'material
thing,‘ 'physical object,‘ etc. which is such that a person who says,
"there are human bodies, but there are no material things" is contra-
dicting himself, just as would a person who said, "There are things
which are red, but there are no things which are coloured," or who
said, "There are greyhounds, but there are no dogs."21 I shall discuss
Moore's arguments for the existence of material objects and for our

knowledge of their existence in a later chapter.

ll.- .
.
I u u
n
."f..
.-
volE

 

 

‘-

.f‘

I. .
.vu

CHAPTER III

MOORE'S PHILOSOPHY OF SENSE DATA

Section I. Introduction of Sense Data

 

After asserting his claim, in his common sense realism, that
the perceptual statements of common sense are known with certainty,
Moore goes on to analyze the nature of a perceptual statement and the
nature of sense perception as a cognitive process. His analysis in
either case leads him to introduce the notion of sense data in his
theory of empirical knowledge. In other words, the notion of sense

data has been introduced in two different ways: by the analysis of

 

perceptual statements and by the analysis of sense perception.

 

Analysis of Perceptual Statements

In his article "Some Judgments of Perception," Moore considers
certain questions regarding what we are doing when we are making judg-
ments of perception; for instance, judgments like "This is an inkstand,"
"This is a finger," etc. He claims that in ordinary life everybody
makes many Such judgments with great certainty, and that we all ordi~
narily accept that the truth of such judgments involves the existence
of material objects. "If I am right in judging that this is an ink-

stand, it follows that there is at least one inkstand in the Universe;

26

Q
d
6
U
-1~.I
~n'
CM

 

i

U‘
95 35
:33:““: "r
P
H

n
N
v

‘7'? 6V; r1
I".

roll b “
FA.

:':r ,

s

a
n

5!-

-\b

nu.

27

and if there is an inkstand in the Universe, it follows that there is in
it at least one material thing or physical object."1 For we all accept
that things like inkstands, chairs, tables, etc., are particular in-
stances of material objects. At this point Moore says that a question
arises as to what exactly such statements are about, i.e., what it is
that is being judged in such statements. In other words, What is it
that I am judging, when I judge that that is an inkstand? According
to Moore, "What does . . . need to be taken seriously, and what is
really dubious, is not the question, whether this is a finger, or
whether I know that it is, but the question whap, in certain respects,
I am knowing, when I know that it is."2

He says that he is certain at least about one fundamental
assumption about such judgments:

it is the assumption that, in all cases in which I make

a judgment of this sort, I have no difficulty whatever in
picking out a thing, which is (quite plainly, in a sense

in which nothing else is), the_thing about which I am making
my judgment; and that yet . . . I am, quite certainly, not

. . judging with regard to it, that jt_is a thing of that
kind for which the term, which seems to express the predicate
of my judgment, is a name. Thus when I judge . . . that That
is an inkstand, I have no difficulty whatever in picking out,
from . . . my total field of presentation at the moment, ap_
object, which is undoubtedly . . . the_object about which I
am making this judgment; and yet it seems to me quite certain
that of this object I am ppt_judging that it is a whole
inkstand.3

 

Rather it is always an object of the kind which has been generally

called by philosophers a sense datum. ”Sense data are the sort of

 

things, about which such judgments as these always seem to be made--

the sort of things which seem to be the real or ultimate subject of

“‘ '.I”~P.'
. p. .
.1

MI" _W_' V

 

.: 5,? al';‘

"2:"9‘1‘31"
.‘u hi ‘w-h .

 

'l‘ [‘19 II A
' “.5 1 I

J
.II‘I

28

such judgments.”“ The point he wants to emphasize in this connection
is that although every judgment about the inkstand is a judgment about
the corresponding sense datum, in making a judgment about this sense
datum, we are certainly not judging of the sense datum that it is a
whole inkstand. He is not denying the fact that this judgment, viz.,
This is an inkstand, is, in a sense, a judgment about the inkstand
itself; he is simply maintaining that this judgment is also, in another
sense, a judgment about this sense datum which mediates the perception
of the inkstand.

The same view has been expressed in clearer terms in his later
article "A Defence of Common Sense." There, Moore says that he is
quite certain about two points regarding the analysis of such statements;
namely, whenever one knows or judges such a proposition to be type)

(I) there is always some sense datum about which the propo-

sition in question is a proposition--some sense datum which

is g_subject (and, in a certain sense, the principal or

ultimate subject) of the proposition in question, and (2)

that, nevertheless, what I am knowing or judging to be true

about this sense datum is not (in general) that it is itself

a hand, or a dog, or the sun, etc., etc., as the case may be.5
This is so because when one knows, for instance, that "This is a human
hand," the sense datum which one is directly aware of is not itself a
human hand, since we all know that the hand has many parts (e.g., its
other' side, and the bones inside it, etc.) which are quite certainly
flgp_parts of this sense datum.

The point that Moore seems to emphasize in this connection is

that the common sense statements about material objects, e.g., I am

29

seeing a chair, or this is a penny, or these are material objects,
etc., which are not (at least apparently) about any sense datum at

all, when analyzed, would necessarily involve statements about sense

 

data. To put it specifically, any attempt to analyze the perceptual
statements of common sense, ultimately leads to certain questions
which can be answered only in terms of statements about sense data.
How an analysis of ordinary perceptual statements leads

to statements about sense data can be roughly stated as follows.
According to Moore, analysis of a given proposition consists in
stating it in terms of simpler propositions. For instance, the
proposition "Material objects exist" can be stated in terms of simpler
propositions of the form, "I am perceiving a human hand," or "I am
perceiving a pencil," etc., and "Human hands, pencils are material
objects."

But even these are not simple enough. It seems to me

quite evident that my knowledge [of the proposition]

that I am now perceiving a human hand is a deduction

from a pair of propositions simpler still--propositions

which I can only express in the form "I am perceiving

this} and "Ihj§_is a human hand."6
At this point certain questions arise regarding these latter statements,
e.g., what does the word 'This' stand for? Or, what exactly is it that
I am perceiving, when I am perceiving this? Am I perceiving the whple
human hand or just a papp_of it, or, perhaps, just a part of the spy;
fgpe of it?--and so on. And an attempt to answer such questions would,

eventually, lead to statements about sense data. For instance, it

may be said that in any case of sense perception, what is actually

3O

given to the sense, i.e., what is actually perceived, is a sense datum,
and the word 'this' stands for the object actually perceived--for a
sense datum.

According to Moore any perceptual statement of the form, "I am
seeing this," or "this is a penny“ is about two objects at once-~a sense
datum and a physical surface, since the demonstrative 'this' (or 'that')
in such a statement refers to two objects at once--'this' refers to a
sense datum which is actually seen, and at the same time 'this' is a
short for a definite description which identifies the physical surface,
and as such 'this' refers to a physical surface answering that descrip-
tion. I shall discuss this issue in detail in a later chapter while

discussing the relation between a sense datum and a physical surface.

Analysis of Sense Perception

 

In his analysis of perceptual statements, Moore has been mainly
concerned with the question: What is it that such statements are about?
Or, what is it that is being judged in a judgment of perception? In
other words, What is the subject of such a statement? In his analysis
of sense perception he is mainly concerned with the question, What is

it that we are actually perceiving when we are perceiving, for instance,

 

a physical object, by means of our senses? In other words, What is the
actual object of our sense perception? In analyzing sense perception
what he is analyzing is not the physiological processes involved in
sense perception, but rather the mental act of perceiving. It is an
analysis of sense perception as a cognitive process. He says that

although the act of consciousness is a consequence of or accompaniment

 

. -\. a 3 ,)¢ II .
.C. 1d . .o 5» r n I .7 ,. C. A w . .3 u: :1... c
.1. a u 3. .PI 3, .3 at f.. Pu all a a .3 I .h 6
NET. .‘11 . , T» r: 5—» p\e Ab :9 ' 2.» LIV T 1 2 .
Wu Cs .W oil .3— r\.. J .6 . .. n.» .P .

.\.« .fi. S A a .« nuw r, S. .. v Cu Tb
Haw F. F ,. and «I9 , pol ‘ . I... 5—» p—» r PI
u . .. u 1 m. U :.. . . .3" . r- 5 . o- + . .. ..
J - m... ...1. . a n p . .
N no x-\-

1;”‘n a
"-...S d:

31

of the physiological processes of the nervous system, the mental act
of consciousness can be known or observed directly (independently of
the knowledge of the physiological processes) by the perceiver,
whereas the complicated and minute physiological processes cannot be
so known or observed. He illustrates the discussion of this subject
(i.e., analysis of sense perception) with the sense of sight and the
act of seeing or visual perception as a mental occurrence--an act of
consciousness, which we can directly observe as happening in our minds.

The occurrence which I mean here to analyze is merely the

mental occurrence-~the act of consciousness--which we call

seejpg, I do not mean to say anything about the bodily

processes which occur in the eye and the optic nerves and
the brain.7

 

The specific question to be discussed, then, is: What exactly
is it that happens when we see a material object? For instance, what
happens to each of us when we are looking at an envelope which occupies
a definite area of space and we are all looking at it at the same time?

Moore says,

I will begin by describing part of what happened to me.
I saw a patch of whitish colour, having a certain size,
and a certain shape. . . . These things: this patch of
whitish colour, and its size and shape I did actually
see. And I propose to call these things, the colour
and size and shape, sense data, things given or pre—
sented by the senses--given in this case, by my sense
of sight.8

He points out specifically what happens in such a case: (l) when we
all see the same envelope at the same time, each one of us also sees
certain sense data; (2) although the sense data that each one has seen
may be similar to those seen by the other, no two of us, in all proba-

bility, have seen exactly the same sense data; (3) although we all see

v

VI 8"

 

J.

. ‘49
n

'11.‘

wt.

in!

p‘
cull

32

the same envelope, in all probability, we all see different sense data;
each of us, for instance, sees a slightly different shade of colour,
although all these colours may have been whitish; but each is probably
slightly different from the rest, "according to the way in which the
light feel upon the paper, relatively to the different positions you
are sitting in; and again, according to the differences in the strength
of your eye-sight, or your distance from the paper."9 (4) Consequently,
we should not say that we know that any two of us see the same sense
data; whereas we should say that we know that we all see the same
envelope; and (5) from all this it seems to follow that

if_we g1g_a11 see the same envelope, the envelope which

we saw was not identical with the sense data which we saw:

the envelope cannot be exactly the same thing as each of

the sets of sense data, which each of us saw; for these

were in all probability each of them slightly different

from the rest, and they cannot, therefore, all_be exactly
the same thing as the envelope.

 

Hence, it seems very probable that pppe_of the colours seen is really
a part of the envelope (i.e., the real colour of the envelope); and
that none of the sizes and shapes seen are the size or shape of the real
envelope; at least we cannot be sure that they are. However, from this
fact that we cannot be sure whether these sense data are identiéal with
the real shape, size and colour of the envelope, it does not follow that
we are not sure whether there are anything like sense data.

Some philosophers have I think doubted 0whether there are

any such things as . . . "sense data" "sensa.

But there is no doubt at all that there are sense data,

in the sense in which I am now using that term. I am at
present seeing a great number of them, and feeling others.

33

According to Moore even in our ordinary usage, when we talk
of seeing an object, we generally do not mean that we are, strictly
speaking, seeing the whole object; we generally only mean that we are
seeing some part of it. In fact, more appropriately speaking, we
should say that we see only the surface of a physical object, and more
specifically, only a part of the surface, not the whole of it. There
is always more in any object we see, than what we specifically see,
viz., a part of it or a part of the surface of it. Similarly, there
is, according to Moore, a still stricter sense of seeing in the sense

of actually seeing or directly seeing, And in this sense of seeing,

 

 

what we actually see or directly see, whenever we see (in the ordinary
sense) the surface of a physical object or any part of it, is a sense
datum (or, are sense data); that is to say, although the object of our
seejpg, in the ordinary sense, is a material thing, say, an envelope,

(or any part of its surface), the object of our direct seeing is

 

always a sense datum, say, a patch of certain colour or a certain
shape, etc. In other words, any perceptual statement of the form,
"I am seeing a physical surface" (or to that effect in any form)
entails a statement of the form, "I am directly seeing a sense datum."11
Moore has explained this point further in the following way:
In any case of seeing an object, the perceiver has a whole visual field
which consists of at least two objects (although usually may be more
than two); for instance, if x is seeing the envelope on the table he

is also seeing, in the sense of being aware of, other things on the

table or physical surfaces (including the table itself); all these

34

things constitute x's whole visual field, although he is really looking
at only one thing, viz., the envelope. Whenever x is presented with
such a visual field which he is seeing, he is also presented with a

direct visual field corresponding to the visual field being seen, which

 

is directly seen by him.

 

I mean that the propositional function "x is seeing at
least two objects" entails the propositional function

"x has a direct visual field which contains at least two
objects" or "x is seeing directly at least two objects."
One can say that it is part of the very meaning of the
assertion that a person is seeing his right hand as well
as something else, that he has a direct visual field
containing at least two objects.12

To use the illustration of x's seeing an envelope, just as it can be

said that x can pick out the envelope from his visual field as the

 

object he is seeing (although he is also, in a sense, seeing many

other objects contained in his visual field), it can also be said

 

that x can pick out an object from his direct visual field, viz.,

 

a whitish coloured patch of a certain shape, which he is directly

seeing (although he is also, in a sense, directly seeing many other

 

objects contained in his direct visual field). This directly seen

 

object which has been picked out from other directly seen objects in
the rest of his direct visual field, is a visual sense datum corre-
sponding to the physical surface of the envelope which he is seeing
in his visual field. Such directly seen objects are sense data.
Thus, according to Moore, there are typ_main senses of the
word "see": the ordinary sense in which we see_physical surfaces or

parts of surfaces, and the special sense in which we directly see

 

visual sense data. The important point in this connection is that

5
eps-

'ir
..

(“1*

cu

._.

~>

Il'

35

in any case of sense perception where we perceive a physical object,
we also necessarily directly perceive an object which is different
from the physical object perceived, and this object is a sense datum.

Section II
Moore's Characterization of Sense Data

 

Sense Data Are Directly Apprehended
The most characteristic feature of sense data consists in the
particular way they are known. According to Moore sense data are the
sort of things that are directly apprehended.
. I have always both used, and intended to use, "sense
datum" in such a sense that the mere fact that an object is
directly apprehended is sufficient condition for saying that
it is a sense datum; . . directly apprehended smells and
tastes and souands are just as much sense data as directly
seen objects.
Direct apprehension, according to Moore, is a cognitive relation that
holds between knower or perceiver and the object of knowledge or per-

ception when it is actually being_experienced. It is that which

 

happens when we actually see any colour, or actually hear any sound,

 

actually smell a smell, or actually feel the so-called sensation of

 

 

hardness in pressing the palm against the table. In all these cases
the act of consciousness is exactly the same in quality;

that is to say, the actual seeing of a colour, considered

as an act of consciousness, differs in no respect at all
from the actual hearing of a sound, or the actual smelling
of a smell. They differ only in respect of the fact, that
whereas the one is the direct apprehension of ppe_kind of
sense datum, the other is the direct apprehension of another
kind: the one, for instance, of a colour, the other of a
sound.“

36

There are two points that need to be clarified in this
connection: First, when Moore defines a sense datum in terms of
direct apprehension, what he has meant is that a sense datum is an
object of the sort which must be initially or originally directly
apprehended; i.e., a sense datum, in order to be an object of cognition,
must first be directly apprehended. The reason for this clarification

is that sense data can also be indirectly apprehended. Indirect appre-

 

hension is defined by Moore as the kind of cognitive relation which one
has to a thing, when one does not directly apprehend the thing itself,
but gpe§_directly apprehend a proposition about it.15 It is the kind

of relation which is usually expressed by saying 'thinking of or about

a thing' or 'remembering a thing' or believing something about a thing,‘
etc., when the thing in question is not actually experienced or directly
apprehended. For instance, when one looks at the wall and actually sees
its blue colour, one directly apprehends the bluish colour; but, then,
if one turns one's head, or closes one's eyes, one no longer directly
apprehends it; yet one may still be thinking of the colour--thinking

of the bluish colour that one saw a moment ago; or one may still be
remembering that it was, or that one saw it a moment ago; in this case
one is, in a sense, still conscious of the colour, though one is not
directly apprehending it any longer.16 In this case the sense datum
which was formerly apprehended directly, is now being apprehended
indirectly. But even though a sense datum can also be apprehended
indirectly, the fact remains that it must still be initially directly

apprehended in order to be thought of or remembered later on. I shall

 

 

 

37

discuss the nature of indirect apprehension and direct apprehension
in detail in a later chapter while discussing Moore's views about
different forms of cognitive relation.

Second, although direct apprehension is something which,
according to Moore, characterizes what a sense datum is, in the sense
that a sense datum is initially or originally known pply_by way of
direct apprehension, there are other things also which can be known
by direct apprehension. For instance, pr0positions, direct apprehen-
sions themselves, images including after images, are all known also by
direct apprehension. Even though images and after images may be placed
under the category of sense data, in a sense, propositions and acts of
direct apprehension themselves cannot be so regarded. A question may
also be raised as to whether the term 'direct apprehension' can be used
in the same sense (or as the same relation) in the case of direct appre-
hension of propositions and in the case of direct apprehension of sense
data. Moore, himself, acknowledges both of these points but has

preferred to leave the issue undecided.17

Sense Data Are Different From Sensations

 

Sense data are ppt_sensations. In accordance with his basic
distinction between the object of x awareness and the act of awareness,
Moore has distinguished sense data from sensation. The reason for this
is that the term 'sensation' is ambiguous, since it may stand, e.g.,
for the bluish colour of the wall, and it may also stand for the
experience or the act of consciousness which consists in seejpg_the

colour of the wall. He says that although many philosophers have used

38

'sensation' without making this distinction, he prefers to use this
term only in the sense of experiencing, i.e., in the sense of an act
of consciousness; and not in the sense of that whiph_is experienced,
i.e., the object of experience, e.g., the colour of the wall. For
the objects of the latter type he uses the term 'sense data.‘

Moore offers two reasons for this distinction: (l) It is
"conceivable that the patch of colour which I saw may have continued
to exist after I saw it [i.e., even when I stop seeing it, e.g., when
I turn my head away]; whereas, of course, when I ceased to see it my_
§eejpg_of it ceased to exist." And (2) it is also conceivable

that some sense data--this whitish colour for instance--are
in the same place in which the material object--the envelope,
is. It seems to me conceivable that this whitish colour is
really on the surface of the material envelope. Whereas it

does pp; seem to me that my seeing of it is in that place.

My seeing of it is in another place--somewhere within my
body.18

§egse Data Are the Kind of Objects Common To
flfll_Actual and Possible Sensory Experiences
According to Moore there are five different classes of the

mental events called sensory experience: (l) images, (2) dreams,
(3) hallucinations, (4) after sensations, and (5) sensations pr0per.
The common feature in all these events is that an entity of some kind
or another is experienced. This entity is not identical with the event
of experiencing.

But we can speak not only of the entities which are_

experienced in the experiences of this kind, but also of

the sort of entities which are experienced in experiences

of this kind; and these two classes may again be different.

For a patch of colour, even if it were not actually expe-
rienced, would be an entity of the same sort as some which

39

are experienced in experiences of this kind. . . . In
speaking . . . of the sort of entities which are experienced
in experiences of the five kinds . . . we do not necessarily
confine ourselves to those which actually are experienced in
some such experiences: we leave it an open question whether
the two classes are identical or not. . . . I intend to call
this class of entites the class of sensibles.19

 

The class of entities that Moore calls sensibles, consists precisely
of all those entities, "whether experienced or not, which are of the
same sort as those which are experienced in the experiences of these

five kinds."

In this context, Moore has further said that all these sensibles
have some common intrinsic property, which is recognizable, but is

unanalyzable.

When we call an experience sensory, what we mean is not
only that in it some thing is experienced in a particular
E11, but also that this something has this unanalysable
property. If this be so, the ultimate definition of
”sensibles" would be merely all entities which have

this unanalysable property.2°

 

 

Secondly, he says that he prefers to use the term 'sensible'
instead of the term 'sense data' for two reasons: (l) sometimes the
application of 'sense datum' is limited to only that which is actually
M in a sensory experience; and (2) sometimes it is used to stand
fer only these entities which are experienced in sensation proper, in
which case, those experienced in the rest of the five classes will be
excluded.

However, it should be mentioned regarding this last point
(Viz., the use of the term 'sensible' instead of 'sense datum') that

it has not been made much use of nor has it been given much importance

40
in his other works; rather the term 'sense datum' has been used in
the sense in which he has intended to use the term 'sensible.‘

Section III. Open Questions About
the Status of Sense Data

 

 

On the basis of his consideration of the various aspects of
the nature of sense data, Moore thinks that the following things are
true of them (as illustrated in the case of visual perception, e.g.,

of an envelope): that no one can be sure that the directly apprehended

 

colour, shape, or size of the envelope, is the real colour, shape, or
size of it. So far as the directly apprehended shapes and sizes are
concerned, it seems quite certain that they are not the real shape and
size of the envelope, since while different people see the same envelope,
different people directly perceive different shapes and sizes at the
same time. And "these different sizes and shapes cannot possibly all_
be phe_size and shape of the envelope." The same thing is true of the
colours directly perceived by different persons. Since each of the
different persons actually sees slightly different colour at the same
time, "it is difficult to believe, though not absolutely impossible,
that all_these colours were really in the same place at the same time."
According to Moore the different areas of spape_(which seem to
be occupied by material objects) are also objects which are directly
apprehended. And in this case also, no one can be sure whether or not
any part of the sense given space which any one person directly per-
ceives, is numerically the same with parts of those sense given spaces

which other persons directly perceive at the same time. It is also

41

doubtful whether or not any part of the sense given space is
numerically the same as any part of the space occupied by the real
envelope. Since, if it is so, we will have to accept either the
hypothesis that all the different colourswhich different persons
actually see as occupying the area are really in the same place as

the real envelope, or else, that the colours only seem to be in this
sense given area but are not really there. According to Moore, so far
as the first alternative is concerned, it is difficult to believe and
explain how all the different colours be at the same place at the same
time; so far as the second alternative is concerned, he says that it

may be said that they do really occupy the sense given area;

 

. . this area, it may be said, undoubtedly is occupied,

by the colours: it is_nothing but the space over which the
colour is spread. So that, if the area, which I see, really
is numerically the same as those which you see, then it will
follow that all the different colours we see really are in
the same place. This argument . . . does not seem to me
absolutely conclusive. It does seem to me possible that

the colour I see only seems to be in the sense given area,
which I see. But it is . . . sufficient to suggest a

doubt whether any part of this sense given area seen by

me really is numerically the same as any part of any of
those seen by you.21

These are some of the reasons why many philosophers have held
the following views about sense data: (l) that absolutely every sense
datum that any person ever directly apprehends exists only so long as
he apprehends it, (2) that no sense datum which any one person directly
aPprehends ever is directly apprehended by any other person, and (3)
that no sense datum that is directly apprehended by one person can be

in the same spacejwith any sense datum apprehended by another person.

42

These three things are . . . the chief things that are
meant, when it is said that all sense data exist only ip_
the mind of the person who apprehends them; and it is
certainly the common view in philosophy that all sense
data do only exist in our minds.22

 

Any view holding these three things about sense data is called by
Moore the accepted view. 1

Consequently, two questions that have been considered in detail

by Moore, in this context are: (l) Are sense data totally private to

the perceiver and (2) can sense data exist unperceived?

Are Sense Data Totally Private to the
Perceiver?

According to the accepted view, no two persons can ever appre—
hend exactly the same sense datum even when they are looking at the same
thing, at the same time under the same conditions; although they may
perhaps apprehend exactlyysimilar sense data; "but they would say that
even though exactly alike--the same in quality--they cannot ever be

numerically the same.” They would further say that any sense datum

 

apprehended by any one person cannot possibly be in the same place as
any sense datum apprehended by another person even when both are
apprehending the sense datum in question at the same time. That is

to say, any sense datum apprehended by any one person simply doesn't
have any spatial relations whatsoever with any sense datum apprehended
by another person. They would allow, however, that the different sense
data seen by any one person have, in a sense, spatial relations 13°_°_"£
another. For instance, this corner of a particular colour patch appre-

hended by one person is really at a certain distance from the other

43

corner of the same patch apprehended by the same person at the same
time, and the same applies to other sense data presented in the visual
field of the person concerned at any one time.

But they would say that all the different sense data within
my field of vision at any one time have distance and direc-
tion from one another only within a private space of my own.
That is to say, no point in this private space of mine is
either identical with, nor at any distance from, any point
within the field of vision of any other person. The sense
given field of vision of each of us, at any moment, con-
stitutes a private space of that person's own;--no two
points in any two of these spaces, can be related to one
another in any of the ways in which two points in any ppe_
of them are related.23

 

So far as Moore's view about this issue is concerned, he is of
the opinion that though some of the arguments in favor of this accepted
view have weight, none of them seems to be conclusive. On the other
hand the objections that may be brought forward against them do not
seem to be conclusive either. For instance, concerning the argument
that if we don't accept the sense given space and the sense data
apprehended as private, then we will have to accept that all different
colours are really in the same places. He says, ". . . it seems to me
possible that any one of them might_be the truth."

‘ So far as the positive aspect of his view about this matter is
concerned, he appeals to the fact that we all have a strong tendency
to believe certain things. For instance, in a situation like this,
everyone finds it very difficult not to believe that when he looks
at this envelope, for instance, and turns away his head the next
nkmnent, the colour which he has been seeing a moment ago (but not

now) is not still existing; or that the space in which he has seen

44

it is not still existing also; or that the colour is not still in that
place. Similarly, it is also very hard not to believe that the very
same portion of the space that one sees is not also seen by other
persons; since, one can point to the space which he sees--i.e., part
of the sense given space that one sees and by doing that which one
thinks one can show to the other person that portion of the space to
which one is pointing. He says,

We all constantly assume that pointing at a thing is of

some use; that if I point at a thing, that serves to show
you which thing I am talking about; that you will see the
same thing, which I see, and will thus know what it is that
I see. And it certainly seems as if the thing at which I am
pointing now is part of the sense given space which I see;
and that, therefore, if ypp_see what I am pointing at, some
portion of the sense given space which each of us sees must
be the same. 2“

 

On the other hand, it is conceivable that one may be wholly mistaken in
believing this, i.e., it is conceivable that the space that one points
at is not a part of the sense given area that any other person sees,
and what the other person sees is not a part of the sense given area
which one sees; ". . . that the supposition that some portion of our
sense given spaces must be identical, arises from our confusion of
sense given space with the real space which we see--but see in another

sense."25

It seems that Moore has not been able to decide for certain

 

whether or not the sense given spaces and the colours, sizes, and other
data are private, so far as they are directly apprehended or actually
perceived. It seems that he has, in general, accepted that the

directly apprehended sense data are (in a sense) private--that at

45

least some of them certainly are private, while some of them perhaps
are. For instance, the data directly apprehended in dreams halucina-
tions, or in the apprehension of images or after images, certainly are
private; whereas those directly apprehended in ordinary regular and
standard cases of sense perception, which he calls cases of sensation
proper are perhaps not so private, i.e., not at least in the sense in
which the ones mentioned above are. He says that even if the three
things held by the accepted view are true of all the sense data
whatsoever,

it does not seem . . . to follow that they exist only in
my mind, or indeed are ip_my mind in any sense at all except
that they are directly apprehended by me. They are . . .
not in my mind in the sense in which my apprehension of them
is in my mind: for instance, this whitish colour, even if it
does only exist while I see it, and cannot be seen by any one
else, does not seem to me to be jp_my mind in the sense in
which my seeing of it is in my mind. My seeing of it is

. related to my mind in a way in which this which I see
is not related to it: and I should prefer to confine the
phrase "in my mind" to those things which are related to
my mind, in the way in which my seeing of this colour, and
my other acts of consciousness are related to it.26

 

Since Moore has not been able to be sure about this particular
aspect of the status of the directly apprehended sense data, one might
think that he would have to be uncertain as to whether two people could
ever have knowledge of the same material object. This would have been
so if he had accepted direct apprehension to be the pply_way of knowing.
But he has explicitly said that though direct apprehension is one of
the ways of knowing, it is ppt_the pply_way of knowing things. Accord-
ing to him many philosophers have assumed more or less unconsiously

that direct apprehension is the only way in which we ever have anything

46

before our minds, since it is much easier to observe and understand
the direct apprehension of sense data and images than it is to observe
ways other than direct apprehension, of having things before one's mind.

If you try to observe what is going on in your mind at any

moment, it is easier to see that you are directly appre-

hending certain sense data, or certain images, or both;

but it is not by any means easy to see that anything else

is happening in your mind at all. . . . It is, therefore,

very natural to suppose that all knowledge consists merely

in the direct apprehension of sense data and images; and

many philosophers have . . . constantly assumed this.27
Since for Moore direct apprehension is not the only way of knowing
things, then, even though directly perceived spaces and data are
private to the perceiver, it is quite probable that different persons
may perceive the same space according to other senses of perceiving

or ways of knowing.

Can Sense Data Exist Unperceived?

 

So far as the question of whether sense data can exist unper-
ceived is concerned, the accepted view holds that they cannot. They
have held that absolutely no sense data which one ever apprehends,
exists at all except at the moment when it is being apprehended by
one. According to them the sense given spaces and other data like
colours, sounds, aches, etc., exist only as long as they are being
apprehended by someone. That is to say, that when the apprehension
of them ceases to exist these sense data simply no longer are in the
Universe; they just cease to exist. One may apprehend the very next
moment a sense datum exactly like the one apprehended a moment ago,
but the two will not be numerically the same although the two may be

exactly the same in quality.

47

So far as Moore's views about this matter are concerned, first,
he has attempted to show that there are reasons to say that such views
as mentioned above are untenable, and second, he has further attempted
to show that there are some good reasons to say that at least some
sense data do exist unperceived.

Moore considers two a priori arguments and one empirical
argument in favor of the view that absolutely no sense data can exist
unless they are being perceived.

First a priori argument.-—According to this argument, to suppose
a sensible or sense datum to exist and yet not to be experienced is
self contradictory, since the existence of any sensible consists in
being perceived. Philosophers following this line of thinking hold
that the contents of our sensation or experience cannot exist without
being perceived, whereas our perceptions or sensations themselves may
exist unperceived; that is to say, these philosophers have held that
when we speak of such a content as a red patch with gold letters on it

'existing,‘ it (i.e., 'existing') means that the content is perceived;
but when we speak of our perception of the red patch with gold letters
on it existing, it (i.e., 'existing') does not mean that the content in
this case, viz., our perceptions, are perceived. That is, these philos-
ophers are using the word 'exist' in two different senses: in the case
of one cflass of contents, viz., colours, sounds, etc., 'exists' means
'is perceived'; but in the case of the other class of contents, viz.,
perceptions, 'exists' does not mean 'is perceived' but means something

different. According to this usage of 'exist,‘ if one asserts that a

48

sensible, e.g., a colour, a sound, etc., exists and yet is not
perceived, one is contradicting oneself, since one is really asserting,
in this case, that the sensible is perceived and yet is not perceived.
Similarly, if one says that a sensible is perceived, yet it does not
exist, what one is really asserting is that it is perceived and yet
it is not perceived.28
Moore's argument against this view is that, in the first place,

its use of the word 'exist' in two different senses is unwarranted.
This is so not simply because the word 'exist' is ambiguous but because
of the way_it is used in two different cases. He says that there is
nothing unusual about the fact that the same word should be used in
different senses, and that many words are so used.

But it would . . . be something very strange indeed, if

in the case of a word which we constantly apply to all

sorts of different objects, we should uniformly apply it

to one large class of objects in the one sense and the one

sense only and the other large class in the other sense

and the other sense only. Usually, in the case of such

ambiguous words, it happens that, in different contexts,

we apply it to one and the same object in both senses.

We sometimes wish to say of a given object that it has

the one property, and sometimes wish to say of the same

object that it has the other property.29

In the second place, there is a common sense usage of the word

'exist'--the sense in which we all ordinarly use it--where we do not
make this distinction, at least, not in the sense in which the above
mentioned view does. In fact, we don't ordinarily make this distinction

at all, but we all do use the word 'exist' significantly in ordinary

language or conversation.

49

Indeed . . . it is quite plain that we constantly do ask,
with regard to what is not a perception, whether it_exists,
in precisely the same sense, in which we ask, with regard to
a perception, whether jt_exists. We ask in precisely the
same sense: Was the Roc a real bird, or merely an imaginary
one? and, did Sindbad's perception of the Roc really exist,
or is it a fiction that he perceived a Roc? . . . the sense
in which I am proposing to enquire whether a red patch
exists, is precisely the sense in which they admit that my
perception of a red patch does exist. And in this sense,
it is plain that to suppose that a thing may exist, which
is not perceived, or that it may ngt_exist, although it
is perceived, is at least not self contradictory.3°
What Moore wants to emphasize is that such a claim cannot be self
contradictory although in a given case it may be false; and whether
it is so or not is not a matter of definition, it is a matter of
empirical investigation.

The secondiipriori argument.--According to this argument,
although it is not self contradictory to suppose that a thing is a
sensible and yet is not experienced, yet we cannot justifiably make
any such claim, viz., that a sensible can or does exist unperceived.
since, "we can clearly see that nothing can have one property without
having the other.“ In reply to this Moore simply expresses his opinion
by saying that it is difficult to deny that one may_be able to know
apriori of such a connection between two properties, but since, in
his own case at least, he 'cannot see' that there is any such con-
nection, he is not convinced about thisa.priori argument; ". . . I
cannot see that it does hold, and therefore, so far asaipriori reasons
go. I conclude that there is no reason why sensibles should not exist

at times when they are not experienced."31 (It seems that in this case

Moore is simply expressing his critical opinion, rather than giving a

50

clear-cut argument, about the matter; and his main objection is
against the claim that we can know of such a connection between two
properties, even if there is any, a priori; hence, the emphasis on the

contrast between "we can clearly see. . . ." and "I cannot see. . . .")

 

 

The empirical argument.--According to this argument there is
ample empirical evidence that the existence of the sense data which we
experience at any time in any case of a sensation proper, always depends
on the condition of our nervous system; as such, the sense data, "which
we would have experienced, if only our nervous system had been in a
different condition, certainly do not exist, when it is not in that

condition,‘ even if the external physical conditions remain unchanged.
Moore says that this argument is fallacious because it assim-
ilates two different things, viz., the existence of the sense data

which we experience and the fact that we experience them. What the

 

empirical evidence shows is that our experiencing of the sense data

 

is always dependent upon the condition of the nervous system, even
where the external physical conditions remain unchanged; but the

evidence does ppt_show that the sense data experienced also always

 

so depends. "The fact that I am now experiencing this black mark is
certainly different from the fact that this black mark now exists."
Hence, this argument does not constitute a reason for holding that
sensibles experienced in sensation proper cannot exist unperceived.
It may be argued, however, that if we accept that sense data
can exist unperceived, then it follows that the sensibles, which a

colour-blind person, if he occupied the same position as occupied by

51

a person who is not colour-blind, would experience, and "which would
certainly be very different from those which I see, are nevertheless at
this moment in exactly the same place as those which I see."32
In response to this extension of the empirical argument, Moore

says that there is no reason for the assumption that if the sensibles
exist at all, they must be in the same place. He further says that
the difficulties, if any, related to this assumption do not apply at
least to his hypothesis

which is only that they exist now, not that they exist in

the samepplace in which mine do. On this question, there-

fore, as to whether sensibles ever exist at times when they

are not experienced, I have only to say (I) that . . . there

is certainly no good reason whatever for asserting that no

sensibles do; and (2) that. . . perhaps a certain amount _of

weight ought to be attached to our instinctive belief that
certain kinds of sensibles do.

 

Moore further considers, in this connection, the question as
to whether there is any positive pgp§9p_to suppose that sensibles ever
do exist unperceived; the reason which he considers to be most important
in this context,

is simply that, in Hume's phrase, I have "a strong propensity
to believe" that, e.g., the visual sensible which I directly
apprehend in looking at this paper, still exists unchanged
when I merely alter the position of my body by turning away
my head or closing my eyes, provided that the physical con-
ditions outside my body remain unchanged. In such a case it
is certainly true in some sense that I should see sensibles
like what I saw the moment before, if only my head were still
in the position it was at that moment or my eyes unclosed. 3“

 

However, Moore has clearly said that this reason applies to only one
class of sensibles, viz., those apprehended in the class of sensory

experience which he calls the sensation proper. This class of sensory

 

experience excludes these other classes of sensory experience, viz.,

52

images, dreams, hallucinations and illusions, and, after sensations
or after images. The sensibles experienced in sensation proper are
such as would be experienced (under certain conditions which actually
exist), "if only a living body, having certain constitution, existed
under those conditions in a position in which no such body does
actually exist.“35

It should be mentioned, however, that on this issue Moore
has held different views at different times, and this fact makes it
difficult to state clearly what exactly his views on the matter are.
It seems that, so far as his earlier views on this subject are con-
cerned, he is not sure whether the accepted view on this issue is true
or not. Although he expresses his doubts about the conclusiveness of
its arguments, he thinks that they could have been true. He says,
"as regards the question whether this accepted view is true or not,
I confess I cannot make up my mind. I think it may very likely be

true "35

But while examining the arguments of the accepted view

on this matter, he seems to be 'strongly inclined' to hold the view
that at least some sense data (viz., those experienced in sensations
proper) may exist unperceived. It is this phase of his views that I
have discussed mainly in this chapter. But, this is not his final
view about this matter; in his later writings, particularly in his
"Reply," he seems to have held the opinion that sense data cannot
exist unperceived, just as headaches cannot exist unfelt. And again

there also, at other passages he says that he is not sure about either

position: that he is willing to hold both the views though they are

53

incompatible. I will discuss and substantiate this last claim in
a later chapter when I discuss his final views about the relation

between a physical surface and its corresponding sense datum.

CHAPTER IV

THE MAJOR OBJECTIVE OF MOORE'S THEORY OF EMPIRICAL
KNOWLEDGE AND THE PHILOSOPHIES OF

COMMON SENSE AND SENSE DATA

Section I. What Is Moore's Major Objective?

The main concern of Moore's epistemological writings is to
defend and justify some of the most widely accepted views of common
sense about perceptual knowledge of the external world. One such view
that he has been mainly interested in defending and justifying is the
view that we all do know that material objects exist. Although this is
a view which has been most commonly and least disputedly held by all of
us and has persisted for ages, the question of defending it still arises
because there have been philosophical views according to which it is
false to say that material objects exist or to say that we kppp_that
material objects exist.

According to Moore those views which deny such well established
views of common sense, for instance views that deny our knowledge of the
existence of material objects as such, or our knowledge of the existence
of other persons having perceptions and thoughts, are skeptical views;
such views are "those which deny something which Common Sense professes

to know, without professing to know anything which Common Sense does pp;

54

55

profess to know. I will call these, fOr the sake of a name, sceptical
yigpg," All such views hold that we do not know certain things which
common sense asserts that we do know. Some of these views, however,

do not positively deny that there are in the Universe those things
which common sense asserts to be known with certainty as existing in
the Universe; these skeptical views "only say that we simply gp_ppt_
kppp_at all whether these things are in it or not; whereas Common Sense
asserts quite positively that we gp_know that they are."1 This, however,
does not mean that these skeptical views deny the possibility of any
knowledge at all; nor that they deny the possibility of knowledge of
certain other things which common sense also asserts to be known; what
they deny is the knowledge of some of the things that common sense
asserts to know of.

Some of the things that common sense asserts to know of include
two broad classes of things, viz., material things existing in space and
time, and acts of consciousness as found in different living bodies
which also exist in space and time. Some of these skeptical views deny
our knowledge pply_of material things or bodies but not of other pipd;
and their acts of consciousness; while the rest deny the knowledge of
both of these two classes of things. In any case all such skeptical
views deny the possibility of our knowledge of the existence of material
things, which according to common sense we certainly know of.

The type of skepticism that Moore is mainly concerned with in
his theory of empirical knowledge is the type of skepticism just

explained; on his interpretation, when the skeptics deny the possibility

56

of the knowledge of an external world, what they have meant is that we
do not and cannot know of the existence of either one of the two above
mentioned classes of things. It is the refutation of this type of
skepticism that Moore has taken to be the main objective of his theory
of empirical knowledge. In his various writings, although Moore has
discussed considerably both skepticism about the knowledge of material
things and that about the knowledge of other persons having thoughts,
perceptions, etc., it seems that the major part of his views deals with
the issue about the knowledge of the existence of material objects.

Section II
The Two Kinds of Objects of Knowledge for Moore

 

In the two preceding chapters I have discussed the two different
aspects of Moore's Theory of Empirical Knowledge, viz., his philosophy
of Common Sense Realism and his philosophy of Sense Data. These two
aspects seem to represent two very different standpoints so far as
Moore's views about empirical knowledge are concerned: viz., the

common sense standpoint which is relatively less technical and less

 

analytical; and the philosophically analytical standpoint which is
technical and highly analytical. These two aspects of his theory of
empirical knowledge and the two respective standpoints which Moore
subsequently seems to be holding are so wide apart and different from
each other that, sometimes, it seems difficult to see any relation
between the two, and their relevance to his attempt to refute skepticism

about the knowledge of an external world and to re-establish the common

57

sense view of the same. (The relevance of his common sense standpoint,
however, is obvious. What is not very obvious is how this standpoint
together with his sense data stapdpoint is relevant to his purpose.)

On the one hand, his Common Sense Realism asserts that we
all do know with certainty that there are material objects and other
persons; such knowledge is constantly expressed in our everyday per-
ceptual statements like 'I see a penny,‘ 'I hear the clock ticking,‘
'There are children playing in the playground,‘ 'That is a chair,‘
and so on. On the other hand, his philosophy of sense data asserts
that when we are, for instance, seeing something, what we are actually
seeing is not a penny or a chair or a book but a patch of colour of a
certain shape and size; and further, these colours and sizes differ
from person to person even when they are all looking at the same
physical object at the same time, that these colours and shapes
actually seen by different people at the same time may be very similar
to one another but, in all probability, are not exactly the same. In
other words, in a case of ordinary visual perception where we all
ordinarily think we are seeing material objects, we are not actually
seeing material objects, but are seeing sense data of different kinds.
And the sense data are not material objects since, first, the seeing
of sense data is different from the seeing of material objects; second,
as the common sense view asserts, we all know with certainty that
material objects occupy space, whereas it is at least doubtful whether
sense data occupy space; third, as the common sense view claims, when

two pe0ple are looking at a particular object at the same time under

58

the same conditions, although they are seeing the same object, in
all probability, they are seeing slightly different sense data; and
finally, we commonly believe we know that material objects exist even
when they are not perceived, whereas it is at least doubtful, whether
sense data exist unperceived at all.

The two viewpoints seem to represent two discrepant, and, in
a sense, conflicting theses about the same issue, viz., our knowledge
of material objects, and of an external world in general, by sense
perception. It seems that Moore has given two completely different,
and apparently, incompatible answers to the same Question: "What do
we know by sense perception?“ (I) We know of the existence of various
physical objects like inkstands, tables, chairs, our own bodies, etc.,
and other persons having thoughts and perceptions, etc., and units of
space, etc., and (2) we know of the existence of sense data--objects
like patches of colour, shapes of some kind, sizes of some kind, etc.,
which are very different from the physical objects which we ordinarily
take to be the objects of our sense perception. In other words,
according to the standpoint represented by his common sense realism,
his answer is that the objects of our perceptual knowledge are material
things, not sense data; according to the standpoint represented by his
philosophy of sense data, his answer is that the objects of our per-
ceptual knowledge are sense data, not material things. These two
answers, if they are really answers to one and the same qpestion,
give the impression that two conflicting positions are being

simultaneously held by Moore so far as views about the knowledge

 

59

of the external world by sense perception are concerned: viz., his
position as upholder of a common sense philosophy and his position
as an upholder of a philosophy of sense data which is almost

inaccessible to the common sense viewpoint.

Section III. Questions About the Coherence
of Moore's Two Standpoints

Many readers of Moore have found it difficult to see how these
two extreme standpoints be combined into one coherent philosophy; for
it seems that if one maintains the sense data standpoint consistently,
one cannot maintain, at the same time, the common sense standpoint
consistently; on the other hand, if one maintains the common sense
standpoint consistently, one cannot maintain, at the same time, the
sense data standpoint consistently. Since the types of things accepted
as objects of knowledge in the two standpoints, respectively, are not
only totally different, but also of such natures that, at least appar-
ently, the acceptance of the one leads to the rejection of the other.
Hence, in particular, some of Moore's critics have been led to suspect
that the two standpoints represented by his two philosophies are in-
compatible with each other,2 and others to doubt how far Moore has
really been successful in his attempt to answer the skeptics' challenge
about cun‘lack of knowledge of material things, etc.,3 and still others
to doubt the extent of Moore's success in avoiding a similar skepticism
in his own views concerning the same.“ In other words, it has appeared
to some of his readers that although Moore has started out with the

clain1 to defend the common sense view of the knowledge of an external

60

world against the skeptics' view of the lack (or impossibility) of
such knowledge, he has not remained consistent in his attempt to
establish his claim. This fact has been expressed in various ways
by his critics: that throughout his works on empirical knowledge,
Moore has been trying to justify two conflicting viewpoints about
the knowledge of material objects; that he has not consistently
followed up his project to establish a defence of common sense;
that he has been doing two different and unrelated things when he
has originally wanted to do only one of them and ppt_the other; or
that he has wavered between two different (and perhaps incompatible)
standpoints not being able to decide which one to accept finally. I
will illustrate only one such criticism by quoting a passage from
Murphy's paper "Moore's 'Defence of Common Sense,” where the criticism
has been stated in very clear terms. In this paper Murphy defends
Moore's assertion of the validity of common sense knowledge against
the charges brought about by some of his critics on the ground that
they have misunderstood what Moore has been trying to say about common
sense knowledge. But so far as Moore's introduction of sense data is
concerned, Murphy disagrees with him.

I find it [the introduction of sense data] difficult to

accept or to reconcile with what went before. So far as

I can see the proposition Moore knew and could verify did

not have a sense datum for its principal and ultimate sub-

ject at all. No sense datum need be referred to and no

proposition of the sort discussed about its relation to

one's hand need be known in order to know with certainty
that what is observed is in fact a hand. And this is

6T

fortunate since, on Moore's showing we do not know what
this relation is. If we had to be knowing what only a
correct epistemological analysis, not yet satisfactorily
performed, would disclose when we know that "this is a
hand," there would thus be considerable ground for scepti-
cism about common sense knowledge after all. In fact, the
assumption that something of the sort must ultimately be
what we are knowing, appearances to the contrary notwith-
standing, is one of the most familiar sources of such
scepticism. Moore rejects the sceptical conclusion, but
he seems, at least at times, to have retained the assump-
tion from which it was naturally derived. Yet his own
philosophy provides the grounds for rejecting this assump-
tion as gratuitous and misleading. Once common sense
statements are interpreted in the context of their familiar
use and testable validity, the claim that they are "ulti-
mately" about the way in which sense data belong to or
represent material objects becomes extremely unplausible.

. Moore has here reverted to a theory incompatible
with the philosophical commitments of his defense of
common sense.5

 

Section IV
Two Underlying Reasons for These Questions

 

 

It seems that there are two underlying reasons for the questions
Moore's critics have raised about the coherence of his views on empiri-
cal knowledge. One of the reasons seems to be the obvious similarity
of the basic epistemological positions accepted by Moore and the
skeptics regarding the pppppg_of empirical knowledge and regarding
the pptppg_of knowledge.

The skeptics' basic epistemological position is empiricism
according to which all knowledge about the external world is founded

upon experience, i.e., experience is the only source of our knowledge

 

of any truths other than analytic truths. In other words, according
to them, the truth of synthetic propositions can be known only by

experience. They further accept in their definition of 'knowledge'

62

that absolute certainty is a necessary condition of knowledge; i.e.,

 

if we are said to kppp_something, we must know it with absolute
certainty. Hence, for the skeptics, 'knowledge' means absolutely
certain knowledge. On the basis of these two principles, the skeptics
have declared the knowledge of anything other than one's own ideas and
sensations is not possible; i.e., one cannot kppy anything beyond one's
own acts of consciousness and the data yielded by them. As such, on
their view, one cannot know of the existence of an external world--
i.e., of material things and other persons.

So far as Moore's position about the knowledge of an external
world is concerned, it seems clear that he also is an empiricist, that
he holds that all knowledge about the external world is ultimately
founded upon experience. This empiricist approach has been clearly
brought out by his repeated emphasis on the roles of observation and
sense perception in all forms of empirical knowledge and at all stages
of empirical knowledge. By the knowledge of an external world, Moore
means, roughly speaking, the knowledge of the existence of various
kinds of material objects which occupy space and are in time, and of
acts of consciousness as found in human beings and in other living
bodies which also exist in space and time; these are the two broad
classes of things that the external world consists of.

According to him the most primitive objects of knowledge, so
far as our knowledge of the external world is concerned, are the
material objects; and the most primitive way of knowing material

objects is sense perception. He is of the opinion that it is true

63

that we all commonly believe we know material objects by ways other
than sense perception, for instance, by memory, testimony and inference,
but all these other ways are based upon sense perception. He writes:

It_is, in a sense, the most primitive way of knowing material
objects: it seems, in fact, to be true, that if I had not
known of some material objects by means of sense perception,
I could never possibly have known of any others in any of
these other ways; and this seems to be true universally:

no man could ever know of the existence of any material
objects at all, unless he first knew of some by means of

his senses. The evidence of the senses is, therefore,

the evidence upon which all our other ways of knowing
material objects seem to be based.6

 

It is a part of the common sense view of the world that there are
material objects and we constantly know the existence of such instances
of material objects as chairs and tables, etc. by means of our senses.

Moore also accepts that knowledge is always absolutely certain.
To know something is to know with absolute certainty or to know for
certain.7 And yet, Moore's objective has been to come up with a thesis
which is very different from, in fact the negation of, what the skeptics
have thought to be the logical outcome of these two premises. From
this, three possibilities follow: (I) either the skeptics are wrong
or Moore is wrong about what the premises imply, i.e., either the
skeptics' argument or Moore's argument is fallacious; or (2) the
premises of the arguments in both cases are inconsistent so that any
conclusion follows; or (3) they have been taking the basic terms in
their respective premises in different senses.

In any case, the fact that these possibilities are there is

sufficient to arouse some doubt about the validity of the conclusion

64

of Moore's argument. Even if one takes an impartial and objective
stand about the whole matter, one may reasonably doubt both the
conclusions until it is clarified in what sense(s) the two parties
are using the basic terms in the epistemological assumptions.

The other reason for thinking that the two standpoints held
by Moore (or the two aspects of his theory of empirical knowledge)
might be in conflict with each other, or that Moore might not have
succeeded in answering the skeptics, seems to lie in the fact that
a considerable portion of Moore's epistemological views have been
devoted to the discussion of the problem of sense data and various
other questions generated by this general problem, the most discussed
and, perhaps, the most vulnerable one being the problem of the relation
between sense data and the physical objects. Such questions, especially
ones regarding the relation between sense data and physical object, do
not present any difficulty for the skeptics since they have claimed
(to have established) that all that we can ever know the existence of
are sense data (and our own acts of consciousness); and, since sense
data are not material objects, it follows that on their views we cannot
know of the existence of material objects.

There is, of course, little worry regarding the consistency or
validity of Moore's epistemological stand insofar as his views about
empirical knowledge consist only of a discussion of the various aspects
of the common sense views which he has wanted to defend and justify;
the problem arises as soon as he introduces the concept of sense data

in his discussion of perceptual knowledge of the external world. It

65

begins to appear as if his common sense standpoint is somehow getting
mixed up with a philOSOphy of sense data which is far from being a
feature of common sense view. Indeed, Moore's further and deeper
involvement with the problem of sense data in his various episte-
mological writings have even given the impression to some of his
critics that his common sense standpoint has been only initial and
not final.

These are the two basic reasons that seem to have led some
of his critics to suspect an incoherence between his two standpoints
about empirical knowledge. The first reason leads to the question:

How can Moore accept the same basic principles about empirical knowledge
as has been accepted by the skeptics and yet claim to reach a conclusion
so different from the skeptics'? The second reason leads to the ques-
tion: If Moore's major objective has been to defend the common sense
view then, why has he devoted so much time to the discussion of sense
data; why has he introduced the problem of sense data after all? And
further, how can his philosophy of sense data be relevant to his

defence and justification of the common sense?

In the succeeding two chapters, I shall try to show that Moore
and the skeptics have used the key terms of their basic epistemological
assumptions differently. I shall further try to show that although the
specific context in which Moore has held the sense data standpoint
should be distinguished from the specific context in which he has held
the common sense standpoint, the two standpoints are related in the

general framework of his epistemological views.

66

In the immediately following chapter I shall discuss Moore's
'views (and also those of the skeptics) about perceptual knowledge ip_

relation to Moore's analysis of knowledge and the various ways of

 

knowing. I shall further consider Moore's examination of and arguments
against the skeptics' claims about empirical knowledge. And on the
basis of these considerations I shall try to show that the two aspects
of his views regarding empirical knowledge are ppt_in conflict with
each other, and that he has been consistent in his attempt to refute
the skeptical views against common sense.

In the final chapter I shall consider and evaluate some further
questions that may still be raised about his holding to both the stand-
points. To be specific, I shall consider the question as to why Moore
has introduced the problem of sense data by analyzing perceptual state-
ments and the cognitive act of sense perception, when it appears that
he could have attempted to refute the skeptics' claims without intro-
ducing and accepting the sense data standpoint at all. I shall try to
answer this question by considering the role of analysis in philosophy
according to Moore, and by considering whether or not his introduction
of the sense datum philosophy bears against his claim to defend and

Justify common sense views against the skeptics' attack.

CHAPTER V

MOORE'S ANALYSIS OF KNOWLEDGE
AND ANSWERS TO SKEPTICISM

Section I. Why We Have to Consider Moore's
Metaepistemolpgical Views and Distinctions

In Chapters II and III I have discussed Moore's views about
common sense knowledge claims and his views about sense data, respec-
tively, which I have also referred to as two philosophies of Moore,
because of their very different natures. In Chapter IV I have pre-
sented some criticisms made by some of his critics against his theory
of empirical knowledge. These criticisms are mainly directed towards
the relatedness of the two philosophies and their joint relevance to
the pursuit of Moore's main objective.

It seems extremely important to know whether these criticisms
of his views are justified or not, because, if they are justified, then
they would lead to the conclusion that Moore's attempt to refute skepti-
cism has been a futile one, and consequently, that he has not been able
to pursue his objective successfully; and perhaps further, that instead
of providing an answer to the skeptics, he himself has been led to
skepticism--a claim, which if true, will lead one to suspect the
consistency of Moore's views about empirical knowledge (since, in

that case, his views would include both of these propositions:

67

68

skepticism about the knowledge of an external world can be avoided,
and also, skepticism about the knowledge of an external world cannot
possibly be avoided).

If these two standpoints (as represented by the two philos-
ophies) were held in two completely different contexts, then one could
perhaps say that there is no connection between the two, and their
extreme nature should not bother us; but since this is not the case,
it seems that the two philosophies must not only be related, but also
be relevant to his main objectives, and there must have been reasons
for their introduction in that context. In order to determine whether
or not it is so, and if it is so, then how it is so, it seems necessary
now to consider Moore's views regarding empirical knowledge in relation
to his other epistemological views.

Moore's epistemological views in their entirety may be said to
have two parts: the first part consists of his views about empirical
knowledge in general. This part includes his views both about common
sense and sense data. The other part consists of his views regarding
the different sense of 'knowing' under different interpretations corre-
sponding to the different ways of knowing, which may be called his

metaepistemological views; in other words, his metaepistemological

 

views constitute the other part of his entire epistemological views

in general. Perhaps it would not be incorrect to say, in this context,
that his metaepistemological views provide the form of his theory of
empirical knowledge; that is, his metaepistemological views supply,

so to say, the rules according to which the contents of his theory of

69

perceptual knowledge take shape and order, representing his two
different viewpoints about empirical or perceptual knowledge. His
epistemological views as a whole seem to be dealing with three very
fundamental questions about knowledge: (l) "What do we know?" or
"What are the objects of knowledge?" (2) "How do we know?“ or "What
are the ways of knowing?" and (3) "To what extent do we know what we
know?" or "What are the limits of knowledge?"

If we consider the contents of the two aspects of theory of
empirical knowledge only relative to questions (1) and (3), then the
two viewpoints do seem to present an almost unbridgeable gap; and it
seems very difficult to find any connections between the two so far
as their relevance to his main objective of empirical knowledge is
concerned. But this gap and the disjointedness of the two positions
seem to diminish in a considerable manner if we consider them in
relation to question (2). In that case these two positions seem
to be two related and equally important links in his theory of
perceptual knowledge, both quite relevant to the pursuit of his goal.
It seems that the different moves and shifts in his philosophical views
can be properly explained and appreciated by pointing out in a general
but comparative way how the two aspects of his theory of empirical
knowledge are related to his metaepistemological views. The apparent
incompatibility of the two philosophies, and the questions it subse-
quently presents to his readers as to whether he has really been
consistent in his attempt to refute skepticism and to avoid it so

far as Iris own philosophical views are concerned, can be better

7O

explained, and perhaps more plausibly, by referring and relating

his subsequent views about perceptual knowledge to the basic
metaepistemological views held by him. In other words, I think

that the different angles from which Moore approaches the problem

of perceptual knowledge are very closely tied down to his views
regarding the concept of 'knowing' or 'knowledge'-—i.e., different
sense of 'knowledge' corresponding to the different modes of knowing;
and I am further trying to suggest that the two aspects of his theory
of empirical knowledge do not represent two independent or two incom-
patible standpoints if we take into account the proper contexts under
which Moore has considered them (or has held them). Hence it seems to
me that it is necessary for an appropriate evaluation and/or character-
ization of Moore's views regarding empirical knowledge, that they
should be considered in relation to his metaepistemological views.

In what follows, I shall try to show how Moore's meta-
epistemological views provide the bases of the two different aspects
of his theory of empirical knowledge; and, furthermore, how the two
are related to each other, constituting two related and relevant lines
of approaching the main problem he is concerned with. I shall divide
this discussion into two sections, and the section immediately
follcnving deals with some of the main contentions of what I have

so far referred to as his metaepistemological views.

7)

Section II
Moore's Epistemolpgical Views and Distinctions

 

Relation of Propositions to Knowledge
and Belief

 

After laying the foundations of a realistic theory of knowledge
by making the distinction between the two factors of knowledge--an
awareness and the object of awareness, Moore goes on to consider the
various ways of knowing and the different senses of knowledge corre-
sponding to these ways. According to him, knowing is a relation
between a person and an object. Any case of knowing is a case of
being aware of something (i.e., some object or other). There are,
however, different ways of being aware of an object, and corresponding
to these different ways, there are various forms of knowledge. It
should be mentioned here that in all these different forms of knowledge,
the term 'knowledge' has been used in different senses. These points
will be explained in detail in the following discussions under this
section.

Moore's discussion of knowledge is based not only on the
distinction between the two factors of knowledge mentioned above,
but also on the distinction between the contents of the Universe as
a whole into two exhaustive classes, viz., things that are propositions

and things that are not propositions. Corresponding to the latter

 

distinction there are two types of facts in the Universe, viz., facts
about propositions and facts about things that are not propositions.‘
According to Moore, a proposition is what is expressed by a

collection of words contained in a sentence; it is the sort of thing

72

which is expressed by a whole sentence. He explains our knowledge of
a proposition in the following way: when we hear or read or utter a
sentence, something happens in our minds--"some act of consciousness--
over and above the hearing [or reading or uttering] of words, some

act of consciousness which may be called the understanding of their
meaning." This way of being conscious of something which, in the case
of reading or hearing a collection of words in a sentence, consists in
understanding the meaning of certain words, may be called the apprehen-
sion of the meanings of these words along with the idea expressed by
the whole sentence (in which those words occur). Thus the knowledge
of a proposition consists in apprehending what is expressed by a cer-
tain combination of words in a sentence. In other words, it consists
in understanding (the meaning of) a sentence. When we apprehend the
meaning of a certain set of words which form a sentence, we apprehend
the meaning of that sentence.

Often in the cases of two different sentences we apprehend
different meanings expressed by the two different sets of words. It
should be mentioned that in such cases although the meanings of the
two sentences are different, they are apprehended in the same way.
Moore says that a proposition is "the sort of thing which j§_
apprehended in these two cases [that is, when we understand the
meanings of two different sentences]. The two acts of consciousness
differ in respect of the fact that whpt_is apprehended in the one,

is different from what is apprehended in the other."2

73

He further points out that when we apprehend the proposition
expressed by a sentence, we often do something else also besides
hearing or reading the sentence; we ordinarily either believe or
disbelieve the proposition. Of course, we may simply consider or
understand the meaning of a sentence without either believing or
disbelieving it. In each of these cases we apprehend a proposition

in exactly the same way insofar as we do understand the meaning of the

 

sentence expressing the proposition. It is not necessary, however,
that in order to think, believe or disbelieve a proposition, we have
to read or hear the sentences expressing them. We often think of,
believe or disbelieve a proposition at times when we are neither
hearing nor reading the sentences which express them. But we apprehend
the propositions in the same sense in which we apprehend them when we
understand the meanings of sentences which we are reading or hearing.3
According to Moore, as it has been discussed before, we also
apprehend things other than propositions, e.g., a sense datum--a patch
of colour--which is not a pr0position. Now, in the context of the
apprehension of propositions, he points out that whenever we apprehend
a proposition we always p1§p_apprehend things which are ppt_propositions,
viz., "things which would be expressed by ppmg of the words," which
compose the sentence expressing the pr0position. He says that since
every proposition is about some object or other, in apprehending the
proposition we also apprehend the object(s) which the proposition is
about. For instance, the proposition, 'The wall is white,‘ may be

said to be about both the wall and also its white colour; or, the

74

proposition, 'Twice two are four,‘ may be said to be ppppt_numbers

two and four. In these cases we apprehend not only the propositions

themselves, but also the tpipg§_which these propositions are ppppt,“
There is, however, an obvious difference between the way a

proposition is apprehended and the way tppt which the proposition

is about is apprehended. In order to distinguish between these two

ways of apprehending, Moore calls the former, direct apprehension

 

and the latter, indirect apprehension. The way we are conscious of
a proposition whenever we understand the meaning of a sentence, is
direct apprehension; and the way we are conscious of tppt_(whatever
it may be) which the proposition is about, is indirect apprehension.
In discussing the way we are conscious of a pr0position in what
preceded, I have already discussed the direct apprehension of a
proposition. In what follows, I shall only briefly discuss what

is meant by indirect apprehension, and I shall discuss this relation
in detail later in this section while discussing the different ways
of knowing.

According to Moore, indirect apprehension is "the kind of
relation which [one has] to a thing, when [one does] directly apprehend
some proposition ppppt_it, but [does] ppt_directly apprehend the thing
itself."5 He says that it is, however, possible that sometimes some
or all of these objects which the (directly apprehended) proposition
is about are also directly apprehended. For instance, when I am

directly apprehending a proposition that this patch of colour exists,

 

or that it is bluish, I may also directly apprehend the patch of colour

75

itself, when I am actually looking at the colour patch. But sometimes
we may directly apprehend a proposition and yet may not directly
apprehend the thing or things that the proposition is about. For
instance, if I am no longer looking at the patch that I was looking

at a moment ago, I may still be apprehending it in the sense that I

am thinking about it, remembering that I saw a patch of whitish colour.

 

In this case although the proposition thpt_1 saw a patch of whitish
colour is directly apprehended, the patch of colour itself is not
directly apprehended. But I am still aware of the patch of colour
in a sense. This way of being conscious of something is what Moore
calls indirect apprehansion of that something.

Thus according to Moore, direct apprehension and indirect
apprehension are two different cognitive relations that hold between
a person and an object. Furthermore, whenever we are indirectly
apprehending something (whatever it may be), we mp§t_also directly
apprehend something else--"ejtpgp some proposition about it, or

"5 But when

perhaps, sometimes something other than a proposition.
we are directly apprehending something, we may or may not indirectly
apprehend something else.

At this point a question may be raised as to why Moore has
discussed the pptppg_of propositions and the ways they are apprehended
in the context of the nature of knowledge and ways of knowing. The
reasons for this seem to be the following:

a. The concept of knowledge is very closely connected with

the concepts of truth and falsity, on the one hand, and with that of

76

belief, on the other. And these concepts, viz., truth, falsity, and
belief, in their turn, are essentially connected with the concept of

proposition. A belief is always expressed in and through a proposition

 

and

a proposition is simply that in respect of which an act of
belief, which is a true act, differs from another, which

is a false one; or that in respect of which two qualitatively
different acts of belief, which are both false or both true,
differ from one another. . . . Every true act of belief
partly consists in the apprehension of a proposition; and
every false act of belief also partly consists in the
apprehension of a proposition. . . . Propositions are,

then, a sort of thing which may be properly said to be

true or false. . . . A proposition is true, if and only

if any act of belief which was a belief in it, would be

a true act of belief; and a proposition is false, if and
only if any act or belief, which was a belief in it, would
be false.7

b. ng_of the necessary conditions of knowledge is tppg_
pgligf, In order to know something we must have, at least, some jggp_
of or about something, and we must have a tppg_idea of that something.
In other words, we must have an idea thpt_something is or is not the
case. For instance, if I am looking at my typewriter I know that it
is the case that it is of blue colour. I have some idea of or about
something being the case. But can anyone have this idea without
apprehending any proposition at all? One may say that one can.

One may say that when one is looking at the typewriter, one is having

or apprehending a visual impgg, and if the image that one is appre-
hending or having before one's mind i§_like the something which it is

an image of, then one is having a true idea of that something. Suppose,

for instance, that I am having the image of a blue square thing and if

515(th Whl
square, then
:awseof tt
recanhave
htomtion.
Moor
thwto acce
fsttsed on
sending coy
tan the sat
z'mmenon
'39 phenome
'Plght the
then. we hm
hesame 55
Pitehend a
MPCSltlor
true 0r fa.
unnot hav<
0f He wr'
I must
the son
the lme
SOME DY

Obiecf?
0f the

77

the thing which the image I am having is an image of is blue and
square, then I may say that I do have a true idea of that thing simply
because of the likeness between the image and the thing. And hence,
one can have a true idea of something without apprehending any
proposition at all.

Moore says that although this may seem to be a very natural
view to accept, the fact is that it is never the case and such a view
is based on a misunderstanding of the nature of things. The misunder-
standing consists in the fact that 'having an image' has been taken to
mean the same thing as having an idea of something; in other words, the
phenomenon of having an image (of something) has been assimilated with
the phenomenon of having an idea of something. And it has been further
thought that if the image is like the thing which it is an image of,
then we have a true idea of the thing, where 'true idea' is used in
the same sense as 'true belief.I But the fact is that even if we
apprehend an image, no matter how vivid, but gp_npt_apprehend any
propositions whatsoever, we do not and cannot have any jggp_at all,
true or false, of anything other than simply having the image. We
cannot have any idea at all of the thing which the image is an image
of. He writes,

I must not only apprehend an image, which is in fact like
the something else: I must also either know or think that
the image j§_like the something else. . . . I must apprehend
some proposition about the relation of the image to the

object: only so I can be properly said to have an idea
of the object at all.8

:jrclv W as:
1:: the:l 8"
Anotl
{or SGDSE d3
leerbe abl
cf truth or
:tiscontext
reenactivel
Error 6‘
false.
which W6
bel levi l
else at

make 63.
or fals

hence, even
aMut SODEI
nuld be ir
he this is
Mainly is
reheht that
c.
Wdirect--t
inthe firs
htween us

Bendes exp

78

Simply by apprehending discrete images we can never know whether or
not they are like something else; that something is the case or not.
Another important consequence of apprehending only images

(or sense data) and no propositions at all would be that we would
never be able to make any mistake or error; consequently, the concepts
of truth or falsity as applied in the context of knowing would not have
any meaning; since, truth or falsity are properties of judgments in
this context which belong to not-erroneous and erroneous judgments,
resepectively.

Error always consists in believing a proposition which is

false. So that if a man merely apprehended something,

which was in fact unlike something else, but without

believing either that it was like or unlike, or anything

else at all about it, he could not possibly be said to

make any mistake at all: he would never hold any mistaken
or false opinions, because he would never hold any

opinions at all.9

Hence, even in order to merely believe something or to hold an opinion
about something, we must apprehend some proposition; otherwise, we
would be incapable of ever holding any beliefs or opinions whatsoever.
And this is a situation which is perhaps not logically impossible, but
plainly is not the case (even if we agree with the skeptics for the
moment that we do not and cannot kppy_certain things).

c. The discussion of the two ways of apprehending--direct and
indirect--has been necessary at this point for two important reasons:
in the first place, it explains and specifies the relation that exists
between us and propositions, viz., that it is a cognitive relation.

Besides explaining the nature of this cognitive relation, it also

79

clarifies in what specific ways we are cognitively related to

propositions, viz., that we either believe or disbelieve or simply

 

understand a proposition (understanding of the proposition being

 

common to all three cases). In the second place, these two ways

of being aware of something are, according to Moore, the most basic
forms of all cognitive relations. All the various ways of being
aware of something, i.e., all the various ways of being in cognitive
relation with something, which I shall presently discuss, can be
explained only be reference to one or the other of these two ways

of apprehending. There is also a third reason why Moore thinks it

is important to make this distinction, namely, that these two ways

of apprehending, he thinks, mgpk_the distinction between propositions
and things that are not propositions. (I do not, however, agree with
this point for the obvious reason that both propositions and non-
propositions can be apprehended--even according to his views--directly
and/or indirectly. But the distinction between these two ways of
apprehending does highlight the distinction between propositions and
not-propositions in more than one way, as I have, to some extent,
already discussed.) It will be made clearer in the following

discussion.

Ways of Knowing

 

On the basis of the distinction between propositions and not-
propositions and between direct and indirect apprehension, Moore

divides all cognitive relations into four classes which he calls

80

the ways of knowing. Knowing, as has been explained, is a relation

between two factors: an awareness or consciousness and an object

which the consciousness is of; in other words, it is always a cognitive

relation between a person and some object whatever it may be. Ways of

knowing pp§_the different ways a person may be cognitively related

with something which is the object of his awareness:

1. One way of knowing consists in the relation that holds, at a

given moment, between a given person and an object when the
person is not apprehending the object either directly or

indirectly.

 

2. A second way of knowing consists in the relation that holds
between a person and an object, at a given moment, when the
person is directlyAapprehending the object.

3. A third way of knowing consists in the relation that holds
between a person and an object, at a given time, when the
person is indirectly apprehending the object.

4. A fourth way of knowing consists in the relation that holds
between a person and an object, at a given time, when the

person is directly apprehending,the object ppt_is also related

 

to the same object at the same moment by some other relation

as well.

Before I explain these various cognitive relations, I would
like to mention two points about them which seem to me very important.

First, Orf these four ways of knowing, Moore regards only the fourth

milledge wit
:rree ways of
the}; are kPOk
erson and er
of something
T'tited or 9'
With
ifscussion o
l.
said to know
direct appre
thing which
:erson knows
ai;reiending
'5 1mP'Oi‘tant
the multipii
S if the mOmEn.
@these 0t
‘to know' her
although he d

'9 Writes,

81

to be knowledge proper, i.e., that which deserves to be called

 

knowledge without any qualifications whatsoever. Second, the other

three ways of knowing are regarded by Moore as knowing in a sense;

 

they are knowing insofar as they are cognitive relations between a
person and an object, i.e., insofar as they are modes of being aware
of something. But they should be called knowledge only in a very

limited or qualified sense.

 

With these points in mind, I shall now present a brief
discussion of how Moore has explained all these four ways of knowing.

1. According to the first way of knowing, a person may be
said to know something even when he is not related, either by way of
direct apprehension or by way of indirect apprehension, to this some-
thing which is the object of his knowledge. For instance, when a

person knows a poem by heart, or a multiplication table, but is not

 

at the moment conscious of them in the sense of either directly

 

 

apprehending or indirectly apprehending one or the other. But what
is important is the fact that although the person is not related with
the multiplication table or the poem by direct or indirect apprehension

at the moment, he must have had either one of the two relationships

 

with those objects in the past. For otherwise, he cannot be said to

 

know these objects at all. It seems that Moore is using the expression
'to know' here in the sense of 'having dispositional knowledge'
although he does not use the latter expression in this context.

He writes.

He conste
sense, wt
that an (
question.
with cer
my hand.
2.

tension, pla
trailedge be
news that h
e; of knee."
:crpds. I h
if this in C
and in this
between pro;
“‘9” Points
said about -
The

the sense 0.
we” the De
'i Sienna
according t1
attualiy pe
{Wally $81
{9915 the De

cases, the s

82

We constantly use this word "know'I . . . in a dispositional
sense, which is such that I can truly be said to have known
that an object . . . was my hand, provided only that, if the
question had been raised, I should have been able to affirm
with certainty with regard to the object . . . that it was
my hand. . . .1°

 

2. The second way of knowing, which he calls direct appre-
hension, plays a very crucial role in Moore's views about perceptual
knowledge both so far as his own views are concerned and so far as the
views that he criticizes are concerned. As such the discussion of this
way of knowing occupies a major portion of his entire epistemological
corpus. I have already discussed certain aspects of his explanation
of this in Chapter III, while discussing his views about sense data,
and in this chapter under Section 11, while discussing his distinction
between propositions and non-propositions. Here I will simply mention
a few points about direct apprehension on the basis of what he has
said about it at different places in his works.

The term 'direct apprehension' has been used in four

 

different senses.--(a) 'Direct apprehension' is primarily used in

 

the sense of a relation which holds between a person and an object
when the person is actually perceiving the object. In this sense
it signifies the relation between a person and sense datum, since,
according to Moore, what is actually given to the senses or what is
actually perceived is always a sense datum. For instance, when one
actually sees any colour, or actually hears any sound, or actually

feels the pain of a toothache, one directly apprehends, in all these

 

cases, the sense datum in question, i.e., the particular colour, or

83

sound, etc. In all these cases the act of consciousness is exactly
the same in quality so far as it is an act of consciousness. They
are different only in the sense that the sense datum apprehended in
one case is different from the sense datum apprehended in another
case, e.g., one is a sense datum of colour, the other is of a sound.11
(b) 'Direct apprehension' is used in the sense of an act of
consciousness which consists in being aware of images. For instance,
when one apprehends a visual image of a particular colour, which one
has actually seen before but is not actually seeing at the moment, one
is conscious of the visual image in exactly the same way as one is
conscious of a particular colour when he actually sees it. When one
thinks of or remembers a sense datum that one has actually seen (but
is not seeing at the moment), one often apprehends some faint image
of a sense datum (e.g., a patch of colour) that one has actually seen

before. In this case the image--a faint copy of the sense datum of

 

which it is an image--is directly apprehended although it is pg;
actually given to the senses. One directly apprehends the image in
exactly the same sense in which one directly apprehends the patch of
colour which is actually seen.12 According to Moore, images themselves
are sense data. Although they are not strictly given by the senses,
they are

in a most important respect, of the same kind as actual
sense data. When, for instance, I directly apprehend the
image of a coloured patch, it is itself a coloured patch
[though it is usually different both qualitatively and
numerically] from any coloured patch gf_which it is an
image . . . every coloured image actually i§_a sense
datum merely because it is a coloured patch even though
it has never been actually given to the senses.13

 

 

 

 

Since Wales a
directly at?”
eta are dire<
It sh
also in the c
af'ter inages
also are din
of sense dat
the senses. "
(C)
which we haw
are aware of
observes one
one's 539;”
the observa
itt of cons
directly a;

liege; it 1

St“‘"dins th

relation in

hare any in

84

Shme images are of the same kind as aCtual sense data, they are
cfirectly apprehended in exactly the same way as the actual sense
(Eta are directly apprehended.

It should be mentioned here that the same relation holds
also in the cases of other kinds of images, e.g., dream images,
after images, or hallucinatory images. That is, these other images
also are directly apprehended, since they themselves "are instances
of sense data, which have nevertheless pgygp_been actually given to
the senses.“

(c) 'Direct apprehension' is used in the sense of a relation
which we have to direct apprehensions themselves, i.e., the way we
are aware of our own direct apprehensions. For instance, when one
observes one's §ggjpg_of a coloured patch, one directly apprehends
one's gaging_of the coloured patch. In other words, in this case
the observation consists in the direct apprehension of the particular
act of consciousness, viz., seeing of a coloured patch. What is
directly apprehended in this case is neither a sense datum, nor an
image; it is the direct apprehension (of a sense datum) itself which
is directly apprehended.1“

(d) 'Direct apprehension' is used also in the sense in which
it is a relation between a person and a proposition which he either
believes or disbelieves or merely considers in the sense of under-
standing the meaning of a sentence. Since I have discussed this

relation in the first part of this section, I will not discuss it

here any further.

85

A question may be raised at this point as to whether in all
these cases, where we are cognitively related with an object by direct
apprehension, the expression 'direct apprehension' denotes exactly the
”same relation or whether it denotes a different relation in each case.
The answer is not very clear. Moore thinks that in the first three
cases it is exactly the same relation but he is not sure whether or
not it is so in the last case. He says,

There are . . . reasons for supposing that what I call the
direct apprehension of a proposition is something different
from the direct apprehension of a sense datum: . . . in the
sense, that the relation which you have to a proposition,
when you directly apprehend it, is different from that which
you have to a sense datum, when you directly apprehend it,

. . . But I cannot tell what the difference is, if there is
one; and the reasons for supposing that there is one do not
seem to me to be perfectly conclusive. I must, therefore,
leave undecided the question whether I am using the name
direct apprehension in two different senses.15

3. The third way of knowing consists in being conscious of
something not by way of direct apprehension of it, but by way of
indirect apprehension. There are two major contexts where Moore
talks about indirect apprehension--in the context of sense data and
in the context of propositions. By indirect apprehension Moore means
thinkingpof or about something. In the context of sense data, he
writes:

For instance, you look at this envelope, and you actually
see a particular colour: you directly apprehend that
particular colour. But, then, if you turn away your eyes,
you no longer directly apprehend it: . . . but you may
still be thinking of it--thinking of just that colour which
you saw a moment ago: you may, therefore, in a sense still
be conscious of it, though you are no longer directly appre-
hending it. Here, therefore, is one way of having before
the mind, which is not direct apprehension: the way which
we call "thinking of” or remembering.16

 

86

In such a case we are apprehending the sense datum-~the colour,
indirectly. In the context of propositions, he says that indirect
apprehension is that relation "which you have to a thing when you do
directly apprehend a proposition about it, [but] do not directly
apprehend it. . . . That is just what we mean by thinking of a thing."
It might be suggested, however, that instead of using the name
'indirect apprehension,‘ the name 'thinking of' should be used for
this kind of relation. But Moore points out that 'thinking of' can
not be an unambiguous name for this relation since it is sometimes
algg used for that of direct apprehension. Since the name 'thinking
of' cannot serve the purpose of distinguishing the kind of relation
which he wants to call 'indirect apprehension,‘ he prefers to use
the name 'indirect apprehension' instead of the name 'thinking of.‘
While discussing the ways of knowing propositions, I have

explained that whenever we indirectly apprehend something, we must

 

also directly apprehend something else--a pr0position or something
other than a proposition. In that context Moore has illustrated

this claim mainly by cases (of indirect apprehension) such that in
each case what is directly apprehended is a proposition. This gives
the impression that in the case of indirect apprehension we mpg;
QlEEXE directly apprehend a proposition. In other words, it seems
that direct apprehension of a proposition is a necessary condition
for indirectly apprehending anything whatsoever-—a sense datum, a
material object, or another proposition. Hence, a question naturally
arises as to whether one can ever indirectly apprehend something

without directly apprehending a proposition about it.

87

In his discussion of this question, Moore says that he is

not sure "that [one] may not have this very same relation to a thing
even‘when [one is] ppt_directly apprehending any proposition about it."
Indirect apprehension of an object consists in thinking of or about
an object which is not being directly apprehended. Moore says that
in order to decide whether or not we alpay§_directly apprehend a
proposition about an object whenever we indirectly apprehend it
(an object), we should be able to answer a different question, viz.,
"Is the mere thinking of [something] the same thing as directly
apprehending some proposition about [it]?" In reply to this question,
he says that he is inclined to think that it is, but he is not quite
sure. He further says that the

important point about indirect apprehension is that it is

not direct apprehension, and yet j§_a way of being conscious

of or thinking of a thing; and that it is a way of being

conscious of a thing, which certainly does constantly occur

when you are directly apprehending a proposition about a
thing, though perhaps it may occur in other cases also.17

 

4. The fourth way of knowing, which Moore calls 'knowledge
proper,‘ consists of the cognitive relation between a person and an
object, only when the person directly apprehends the object and more-

over, he is also related to it, at the same moment, in another way as

well.

There is one very important relation of this type . . .

and which can only hold between us and a proposition--not
between us and any other kind of object. It is, in fact,
tpg_relation which we commonly mean to express when we say
without qualification that we know a proposition to be true--
that we know that so and so is the case.18

 

 

88

This relation, which he calls knowledge proper so as to differentiate
it from the rest of the ways of knowing which lead to knowledge only

in a qualified sense, must satisfy four conditions: (1) one must

 

directly apprehend some proposition; (2) one must believe the propo-

 

 

sition to be true; (3) the proposition must be true; and (4) some

 

fourth condition which Moore does not specify clearly. It seems,
however, from his discussions about knowledge at different places

in his works that this fourth condition consists in having reasons

or justifications in the sense of having evidence which is definite
and certain for one's belief in a proposition. According to Moore
this is the most important sense of the word 'know.‘ In this con-
nection it is necessary to discuss Moore's views about certainty which
is regarded as a necessary condition of knowledge by most philosophers
including Moore and the skeptics. Since, according to Moore, the con-
cept of certainty is essentially related with the concept of knowledge
proper, any discussion of Moore's views about knowledge proper would
be incomplete without a discussion of his views about certainty and
particularly in this context, without comparing his views with the
skeptics' views about certainty. However, Moore's views about knowledge
Proper also includes his distinction between two forms of knowledge
proper, which has not been discussed in this work so far. Hence in

what follows I shall discuss these two subjects.

89

Certainty

I have already mentioned in a previous chapter that both
the skeptics and Moore accept that absolute certainty is a necessary
requirement for knowledge; i.e., 'to know' is 'to know with absolute
certainty.‘ But, as I have already suggested, Moore and the skeptics
have come to very different conclusions as to the extent of knowledge,
although, apparently they start with very similar views about knowledge.
Since certainty has been regarded by both as one of the basic concepts
involved in knowledge, it becomes necessary to examine their respective
views about this matter, to examine whether their views on this issue
are [gally_similar or not.

Moore's views about certainty are very closely related with
his views about perceptual knowledge in general and also with his
answer to the skeptics' challenge about our knowledge of an external
world. It is interesting to note that Moore has mainly discussed the
concept of certainty and has mainly used the word 'certain' in the
context of knowledge proper, and not in the context of direct appre-
hension where the skeptics, in general, have looked for certainty in
the sense of incorrigibility. Moor has not used the word 'certain'
in the sense of 'incorrigible' or 'indubitable' where 'indubitable'

'19 It seems that for him 'certain'

means 'not able to be doubted.
means that which is beyond reasonable doubt and ppt_that which is
beyond logical doubt.20 For Moore, the word 'certain' is always used

in context of a proposition relative to a particular circumstance.

 

Further, it is always used in the context of a knowledge-claim made

90

by somebody. Now the question arises 'Under what circumstance a
proposition is beyond reasonable doubt?‘ Moore's answer is based
upon an appeal to common usage, i.e., how the word 'certain' is used
in ordinary language. By 'beyond reasonable doubt' Moore means 'it

is absurd to doubt under given circumstances.‘ A proposition is beyond

 

reasonable doubt when doubting the truth of the proposition involves an
absurdity. By 'absurdity,‘ however, Moore does not mean a logical
absurdity in the sense of self contradiction or logical impossibility,
since the kind of absurdity that Moore is talking about is always
dependent upon and relative to circumstances--to empirical facts.
That is to say that the same proposition is beyond reasonable doubt
under one set of circumstances may not be beyond reasonable doubt under
a different set of circumstances. Moore illustrates this point by
considering certain assertions made under a particular circumstance,
viz., a standard classroom phenomenon where the following assertions
are made by the teacher: "I am at present . . . in a room and not in
the open air; I am standing up . . .; I have clothes on . . .; I am
speaking in a fairly loud voice . . .; there are a good many other
people in the same room in which I am; and there are windows in that
wall and a door in this one." A few lines later he writes:

. . and I have made these assertions quite positively,

as if there were no doubt whatever that they were true.

. . Not only true but also certain . . . I implied,

though I did not say, that they were in fact certain.

. . On the contrary, I should have been guilty of

absurdity if, under the circumstances, I had ppt_spoken

positively about these things, if I spoke of them at all.

Suppose that now, instead of saying ”I am inside a building,"
I were to say "I think I'm inside a building, but perhaps I'm

 

 

9]

not: it's not certain that I am," or instead of saying "I
have got some clothes on,” I were to say "I think I've got
some clothes on, but it's just possible that I haven't."
Would it not sound rather ridiculous for me now, under
these circumstances, to say "I think I've got some clothes
on" or even to say "I not only think I have, I know that it
is very likely indeed that I have, but I can't be quite sure"?
For some persons, under some circumstances, it might not be
at all absurd to express themselves thus doubtfully. . . .
But for me, now, in full possession of my senses, it would
be quite ridiculous to express myself in this way, because
the circumstances are such as to make it quite obvious that
I don't merely_think that I have, but I know that I have.2r

 

 

 

 

Having illustrated what he means by 'certainty,‘ Moore
clarifies certain ambiguities about the use of the word 'certain'
in common usage. There are four main usages of the word 'certain'

in ordinary language: (l) I am_certain* that something is the case (p);

 

(2) I feel certain that p; (3) I know for certain that p; (4) It is
certain that p. A semantical difference between (T) and (2) on the
one hand, and (3) and (4) on the other, is that whereas (3) and (4)

entail the truth ofpp and also that p is known to be true to the

 

speaker, (l) and (2) do not entail either. This is so because the

conjunction of either (1) or (2) with the negation of p, is not self

 

contradictory; but the conjunction of either (3) or (4) with the
negation of p, is self contradictory, for at least one common use
of those phrases. For, in the latter two cases, the fact that not-p

proves that I did not know that p, and hence, it was not certain that

 

p; whereas it does ppp_prove that I did not feel certain that p. "In

 

*The expression "I am_certain that . . . ," however, is
ambiguous since it is sometimes used to mean the same as "I feel
certain that . . .” and sometimes to mean the same as "I know for
certain that. . . . Hence the implications would change accordingly
depending upon the sense in which it is used.

92

other words, "I feel certain that p" does not entail that p is true
(although by saying that I feel certain that p, I do imply_that p is
true), but "I know that p" and I'It is certain that p” do entail that

p 1; true; they can't be true, unless it is." The point of this
clarification is that when Moore is using the word 'certain' in the
context of knowledge (as in the case of the above discussed assertions),
he is using it in the senses (3) or (4). Thus, it is true that Moore
agrees with other philosophers in holding that certainty is a mark of
knowledge, and that 'to know' is 'to know for certain' (or 'knowledge =
'certain knowledge'); but by 'certainty' he does not mean 'incorrigibil-
ity' in the sense, for instance, the direct apprehension of sense data
is incorrigible, or 'indubitability' in the sense 'not able to be
doubted' as opposed to 'indubitable' = 'known.' For him 'to know

that p' = 'to know for certain that p' or 'it is certain that p'

where p is beyond reasonable doubt, or where doubting p involves an
empirical absurdity. It is true, Moore says, that in the case of a
contingent proposition like "I am standing up now'--to call it p--it

is possible that not p; but it should be noticed at the same time that
p is not incompatible with the fact that p may be known with absolute
certainty, or the fact that it is absolutely certain that p. "Because
if any person whatever does at a given time know that a given proposi-
tion p is true, then it follows that that person could say with truth
at that time "It is absolutely certain that p." Thus if I do know now
that I am standing up, it follows that I can say with truth "It is

1122

absolutely certain that I am standing up. In this case, on Moore's

93

view, it would be absurd (though not logically), under the
circumstances, to say that 'I kppg_that p but it is not certain
that p.'

In his discussion of certainty, what Moore seems to emphasize
is the fact that under certain correct usages of 'certain,' although it
does not mean logically certain, it is essentially related with the
concept of knowledge (i.e., knowledge proper); and similarly, that
there are certain correct usages of the word 'know' where such an
usage does not satisfy the condition of certainty in the sense of
logical certainty, yet it satisfies the condition of certainty or
absolute certainty in a different sense. One such usage of 'certain'
and 'know' pointed out by Moore needs mentioning in this connection.

For Moore, the proposition 'It is certain that p' entails that 'p is

 

kpgyp to be true'; 'that p is known to be true' is a necessary condition
for the truth of 'it is certain that p.' And this is a fact which also
clearly shows the difference between 'It is certain that p' and 'I am
(or feel) certain that p.' For instance, it is not self contradictory
to say that, e.g., I'I fglp_(or was) certain that it would rain, but in
fact it didn't"; but it would be self contradictory to say that, "It

was certain that it would rain, but it didn't," for at least one common
use of that phrase. Since, the fact that it did not rain ppgyg§_that

I did not kppy_that it would rain, and it was not certain that it would,
whereas it does not prove that I didn't fggl_(or I wa§_not) certain that
it would. “It is, indeed, obvious . . . that a thing cannot be certain,

unless it is known: this is one obvious point that distinguishes the

94

use of the word "certain" from that of the word "true"; a thing that
nobody knows may quite well be true, but cannot possibly be certain."23
It seems to me that Moore is making a distinction between
something like a logical certainty and empirical certainty, although
he has not used these two terms; but when he talks about the skeptics'
stand on certainty and contrasts that with his own, he does talk, in
this context, about logical doubt, logical possibility and impossibility,
on the one hand, and reasonable doubt and empirical possibility, on the
other.2“ It seems that what he is trying to say is that while these
philosophers have accepted logical certainty as a mark of knowledge,
they have somehow either completely ignored the fact that there is an
essential relation between knowledge and empirical certainty or they
have not recognized this kind of certainty as certainty at all. When
they say that to know that p is to know for certain thatpp, what they
mean is that if one knows that p then it is logically impossible that
not-p, i.e., p is logically indubitable. For instance, they would say
that one can know p (i.e., with absolute certainty) where p has been
formally or deductively proven, i.e., p has been shown to follow
formally from a set of premises. Whereas Moore wants to say that
we do not need this kind of certainty for all knowledge whatsoever
(or for knowledge as such); he says that we may know that p (with
certainty) where it is based on a set of premises but does not follow
deductively or formally from those premises; from the fact that p does
not §p_follow from the premises, it does not follow that p has not been

proven by the premises on which it is based, nor does it follow that p

95

is not known with certainty, though it is logically possible that not-p.
He writes elsewhere to the same effect:

from the fact that two things have been conjoined, no matter
how often, it does not strictly follow that they always are
conjoined. But it by no means follows from this that we may
not know that, as a matter of fact, when two things are
conjoined sufficiently often, they are always conjoined.25

 

In all these cases when Moore talks about knowing, by 'knowing' he
also means 'knowing with certainty.‘ What he wants to suggest is that
these philosophers have used 'certainty' in a very strict or narrow
sense so far as they take it as a necessary mark of knowledge.

On the other hand, sometimes these philosophers, he suggests,
have used the word 'certainty' in a sense, where it seems that they
think that something can be certain without its being known at all.
For instance, when they characterize the common sense views about
external world as mere matters of Faith--meaning not of Knowledge--

what they mean to say is: "It is certain that we all do believe many

 

such propositions, but none of us know any of them to be true." And

what is meant by "it is certain that . . ." is “I and many other men
0

know that we all do believe many such propositions, but none of us know

 

any of them to be true." That is to say, that when they are saying

'it is certain that we all believe . . ." they are, in fact, simply

 

saying "we all know that we all believe. . . ." But they are not

realizing that in saying that somethingyjs certain, they are also

 

acknowledging that that something is also known (for otherwise it
cannot be certain). That is, in saying so they are themselves

admitting that at least one common sense view is known to be true,

96

viz., the view that there are many human beings and we know that

there are many human beings. This is so because in saying that 'it

is certain that we believe . . ." they are admitting that they know
that there are many human beings who have held such beliefs (which

they call mere matters of faith not of knowledge). Moore says that
some philosophers have overlooked this point "in the most amazing way."
Since they have held that "it's quite certain that mpn_believe these
things, that they are beliefs of Common Sense; and yet that it's not

11 25

certain that there are any men. What Moore wants to suggest here

is that to say that something is certain, is to say that something
is known, i.e., there is a sense in which something cannot be certain
without its being known; whereas, in the case mentioned above, what
these philosophers are saying, according to Moore, that something
can be certain without its having anything to do with knowledge.
They seem to me to constantly betray the fact that they
regard the proposition that those beliefs are beliefs of
Common Sense, . . . as not merely true, but certainly true;
and certainly true it cannot be, unless one member, at least,
of the human race, namely themselves, has known the very

things which that member is describing that no human being
has ever known.27

 

 

 

In other words, what Moore is suggesting is that, these
philosophers, on the one hand, have accepted certainty to be an
essential mark of knowledge in such a stringent sense that one can
hardly be said to know anything with certainty, and have thus limited
the scope of knowledge very considerably in the sense that we can be
said to really 5ppp_very few things; while on the other hand, they

have sometimes used the word 'certainty' without recognizing that

97

under certain usages, at least, the concept of certainty essentially
involves (the concept of) knowledge in the sense of knowledge proper;
they have failed to recognize the fact that if we say 'that something
is certain,’ it implies, as has been suggested by Moore, 'that we knpw_
something to be the case.‘ In other words, according to Moore, the
proposition 'It is certain that p' entails or implies the proposition
'Somebody knows that p.' (However, it is to be remembered that Moore
uses here the word 'certainty' not in the sense of logical certainty,
but in the sense of empirical certainty.) Moore has criticized both
of these attitudes of the skeptics, on the grounds that, in the former
case, either we do not mean by 'certainty' what they have meant, or at
least, it is not necessary to mean by 'certainty' what they have meant;
and, in the latter case, their use of the word 'certainty' in some
cases is not proper, since it violates, at least, one correct usage
of the term.

Later, in the next section, I shall discuss the application of
his views about certainty to his views about perceptual knowledge,
which forms a part of his justification of common sense knowledge

claims.

Types of Knowledge Proper

 

Moore talks about two different forms of knowledge proper:
immediate and mediate. They are forms of knowledge proper because
they are propositional knowledge, i.e., they always consist in knowing
a proposition, and secondly they meet all the conditions of knowledge

proper.

98

Immediate knowledge consists in directly apprehending a

 

 

proposition to be true, without knowing any other proposition whatever

 

from which it follows. This form of knowledge, though immediate, is
quite different from direct apprehension in more than one way. In the
first place, direct apprehension can be of propositions as well as of
objects that are not propositions, e.g., sense data, images, acts of

consciousness. Immediate knowledge is always of some proposition or

 

other. In the second place, it is also different from direct appre-
hension even when a direct apprehension is an apprehension of propo—
sitions; one may directly apprehend a proposition without believing

or disbelieving it to be true (i.e., in the sense of simply under-
standing the meaning of a sentence), and further one may directly
apprehend a true proposition or a fal§g_proposition. Whereas immediate
knowledge always consists in apprehending a proposition which is
believed to be true and is in fact tppg, However, immediate knowledge
of a proposition is quite compatible with direct apprehension, in the
sense that, while knowing a proposition immediately, one may also
directly apprehend something else--a proposition or a sense datum,
etc., and the converse is also true, i.e., while directly apprehanding
a sense datum or a proposition one may also, at the same time, imme-
diately know a proposition about the directly apprehended objects.28

Mediate knowledge, on the other hand, consists of knowing a

 

proposition on the basis of knowing some other proposition from which
it follows. "We give the name mediate knowledge to all cases in which

.you know a proposition, because you know some other from which it

99

follows. . . ." One of the consequences of thus distinguishing
between mediate and immediate knowledge is that one may know the same
proposition at the same time both immediately and mediately:

. . it is important to insist that even when you do know

a proposition immediately, you may_also at the same time

know some proposition from which it follows: you may know

it both immediately and also because you know some other

proposition from which it follows.29
This consequence also marks the difference between immediate knowledge
and mediate knowledge on the one hand, and, direct and indirect appre-

hension on the other. We can never apprehend the same thing at the

 

same time both directly and indirectly. (Mediate knowledge is differ-
ent from indirect apprehension for the same reasons, mutatis mutandis,
for which immediate knowledge is different from direct apprehension.)

Section III. Moore's Epistemologyyand
His Answer to Skepticism

 

With this background in mind I shall attempt in the rest of
this chapter to explain how the various epistemological works of Moore
in their entirety form one consistent and complete attempt to answer
the skeptics' challenge about the knowledge of an external world. The
entire plan of the epistemological work of Moore seems to me to consist
of three parts: The first part consists in stating and explaining the
common sense views about knowledge. The second part consists in
explaining the meaning or meanings of 'knowledge' relative to dif-

, ferent ways of knowing. It further includes an explanation of the

common sense view about the knowledge of the existence of material

100

objects and an explanation of the skeptics' views about perceptual
knowledge, relative to these different meanings of 'knowledge.‘ The
third part consists in justifying the common sense knowledge claims
in various ways.

I will try to explain each part separately but, as I have
already explained the first part in detail in an earlier chapter,
and the first half of the second in an earlier section of this chapter,
I will just mention them without going into the details. I shall

mainly discuss in detail the second half of the second part and the

third part.

The Common Sense View Reviewed

By common sense views Moore means views or beliefs held by
everybody; i.e., certain definite views about the various aspects of
the Universe are held by everybody nowadays and that have been held
for a long time and have not undergone any substantial changes, though
they may have been modified by way of improvement in certain respects.
By common sense views about knowledge or common sense knowledge--claims
Moore means general views about the various aspects of the Universe

which are assumed to be true; which are further assumed to be known

 

to be true with certainty. These beliefs which are accepted by every-
body tx: be known to be true are not to be assimilated with mere preju-
dices; they differ from mere prejudices in the respect of the degree
of their'acceptance and also in the respect of the consistency and
the persistence with which they are accepted. The beliefs that are

accepted as known with certainty have usually been accepted universally

lOl

irrespective of temporal, geographical and cultural distinctions.
Further, they have been maintained in a consistent and persistent
manner through ages without ever losing their identity or basic
character even if they have gone through modifications in certain
respects. They are in a sense the fundamental and primitive beliefs

of mankind. In various works Moore has given many illustrations of
such beliefs; for instance, our beliefs about the existence of material
objects which are situated in space and are different from what we call
mental acts; or, our beliefs about the existence of acts of consciouss
ness which are not situated in space in the sense in which we believe
that material objects are. These acts of consciousness are believed to
exist in our minds which are generally believed to belong to some par-
ticular kinds of material bodies called human beings or living beings.
We further believe that the minds are somehow related to the material
bodies they belong to though we may not know the exact nature of their
relationship. We also believe that we know many of these objects by
means of our senses--by perceiving; that sense perception does really
give us knowledge about certain things--knowledge which is certain and
objective or verifiable and we believe that there are also various
other ways of knowing--the ways by which we know that we have dreams,
or imaginations, or feelings and emotions, or the way we know that two
is different from five and so on--i.e., ways other than sense percep-
tjgpl. These are some examples of the various things that we all
ordinarily'claim to know of. For Moore, these are the kinds of beliefs

that constitute the common sense view of the world.

102

The Metaepistemological Distinctions
Appealed to

 

As it has been already discussed, this part of Moore's
epistemological views consists of an analysis of various cognitive
relations between a person and an object, explaining the different
sense of knowledge corresponding to the different relations. Accord-
ing to Moore, any cognitive relation necessarily involves two factors,
viz., consciousness and an object which the consciousness is of.
Second, there are four different types of cognitive relations which

he calls the ways of knowing and which he considers to be fundamental

 

to all other ways of knowing and types of knowledge yielded by them.
That is, all different ways of knowing that we ordinarily take into
consideration are ultimately founded and based upon these four ways

of knowing and can be explained in terms of them. For instance, we
all recognize that testimony, inference, observation, memory, etc.,
are various means of knowledge; they are all, in the final analysis,
based upon one or more of these four types of cognitive relations.
Third, of these four ways of knowing only one is recognized by Moore
to be knowledge proper and the other three are knowing or knowledge in
a restricted and qualified sense of the term 'knowledge.‘ These four
ways of knowing are: (1) direct apprehension, (2) indirect apprehen-
sion, (3) dispositional knowledge, and (4) knowledge proper. (It
should be mentioned here that in his entire discussion about perceptual
knowledge Moore has seldom talked about dispositional knowledge. As

such, the different issues will be discussed here in relation to only

103

three ways of knowing: (1), (2), and (4). In this part of his
epistemology Moore has further done the following things: First,
he has explained in what ggp§g_of knowledge we ordinarily claim to
know of the existence of material objects and how far this sense of

knowledge corresponds to any of the three ways of knowing mentioned

Wl

above. Second, he has related the skeptics' views on knowledge with
the various ways of knowing in order to determine whether their views
denote any one or more of these, and whether or not any (or more) of

these ways may be said to be leading to the knowledge of an external

 

 

world in the sense in which it has been explained. l,_
(l) Common sense claim about knowledge of material objects

and the ways of knowing,--The main issue of the common sense knowledge

 

which Moore's entire discussion of empirical knowledge centers around
is the issue regarding the knowledge of material objects by means of
sense perception. So far as the question regarding the knowledge of
the existence of other human beings capable of acts of consciousness
is concerned, it arises, and can be settled, only when the question
regarding the knowledge of the existence of material bodies can be or
is settled.

Moore thinks that the existence of material objects is one
of the most primitive and most persistently held beliefs of common
sense; and as such when this very belief is in question, the discussion
of it needs maximum priority. It is also one of the most primitive and
persistently held beliefs of common sense that we know of the existence

of material objects primarily and originally by sense perception; and

 

104

as such the question as to whether we do or can really have knowledge
of material objects by sense perception is also of primary importance.
He writes:

This way of knowing material objects, by means of the

senses, is, of course, by no means the only way in which

we commonly suppose we know of their existence. For instance,
each of us knows . . . material objects by means of memory;

. by the testimony of other persons. . . . And we know
also, . . . by means of inference. . . . But all these other
ways do seem, in a sense, to be based upon sense perception,
so that jt_is, in a sense the most primitive way of knowing
material objects: it seems, in fact, to be true, that . . .
no man could ever know of the existence of any material
objects at all, unless he first knew of some by means of
his senses. The evidence of the senses is, therefore, the
evidence upon which all our other ways of knowing material L.~
objects seems to be based.30 '

 

Hence, the main question that Moore addresses himself to
throughout his discussion of empirical knowledge is whether or not
we do or can know of material objects by means of the sense. In this
context, however, he discusses whether or not any of these three ways
of knowing may be said to be the knowing of material objects by sense
perception.

One of the things that Moore points out here is the distinction
between 'perceiving' (and all other words to denote different forms of
sense perception) as commonly used in ordinary language, and 'perceiv-
ing' as used in the context of sense data. "When applied to sense data
they simply express direct apprehension; but when applied to material
objects or their surfaces or any part of the space occupied by material
objects they express something quite different."31 In order to decide
what sort of relation is denoted by 'perceiving' as applied to material

objects when we say that we know of their existence by sense perception,

 

105

we have to discuss first in what sense we are talking about knowledge

of material object.

 

Moore points out that when we talk about the knowledge of the

existence of material objects, we talk about knowing that so and so is

 

the case, i.e., that there apg_material objects, or, that material
objects exist; and when we use the term 'knowledge' in this sense,
i.e., in the sense of knowing that so and so is the case, use it in

the sense of knowledge proper. Knowledge proper, as has been already

 

discussed, involves four conditions: (l) that we must directly appre-
hend some proposition, (2) we must also believe the proposition thus
apprehended, (3) that the proposition must be true, and also some
fourth condition (which Moore does not specify but he seems to have
meant conclusive evidence or good reason). Coming back to the question
regarding the knowledge of the existence of material objects by sense
perception, we have seen so far that it partly consists of direct
apprehension of sense data; e.g., when we look at an envelope, we
directly apprehend certain sense data. But the knowledge of the
existence of a material object cannot consist merely in direct appre-
hension of sense data, for in that case, one is simply experiencing or
having certain sense data, e.g., certain colour, shape or size, etc.,

and merely direct apprehension of sense data is not knowing that so and

 

so is the case. In order to know that something exists, we must also

 

 

directly apprehend a proposition, and besides these two, viz., direct

 

apprehension of sense data and of some proposition, we must also know

that something other than sense data exists which is not directly

lO6

apprehended. For instance, when we look at an envelope and directly

 

apprehend certain sense data, we must also know, if we are to know that
the envelope--a material thing--exists, that these sense data which we
are directly apprehending are in fact connected with the envelope; in
order to know that, we have to know something other than mere sense data,
i.e., we have to know that something with which the sense data are in *
fact connected. Hence, in order to know of the existence of a material

thing, we have to know at least this much, although we may have to know

something more.

 

According to Moore what really happens when we know that there Q.

is an envelope which we are perceiving, is something like this: we

 

directly apprehend certain sense data and this is accompanied by a
pgligf, may be an obscure one, about the existence of something else
with which these sense data are somehow connected. In having this
belief we directly apprehend a proposition. A belief or a proposition
is always about something; and this something, whatever it is, which
the belief or proposition is about is, as earlier discussed, indirectly
apprehended by us. In the case of our perceiving the envelope, our
direct apprehension of colour, shape, etc., is accompanied by a belief
or proposition about the envelope--a material thing--which is other
than the sense data; and the envelope is, in this case, indirectly
apprehended by us.

So that sometimes, at all events, seeing an envelope cannot

merely mean seeing certain sense data with which it is con-

nected; it must mean also indirectly apprehending this some—

thing else--the material envelope itself, with which these
sense data are connected. . . . That you do, when you

 

lO7

directly apprehend certain sense data, pftep thus believe

in the existence of something else is . . . certain. And

if this something i§_a material object, then you really are,
whenever you do it, indirectly apprehending a material object.32

The question that needs to be considered now is this: Can
we know of the existence of material objects by indirect apprehension?
Moore has discussed this question while explaining the skeptics' views
about perceptual knowledge relative to the different ways of knowing.

(2) Skepticism and ways of knowing.--The skeptics Moore is
considering are empiricists and accept that experience is the only
source of factual knowledge. 'Sense perception' and 'observation' are
two very commonly used terms to denote the forms of experience. Sense
perception and the various specifications or forms of it, e.g., seeing,
hearing, etc., are regarded to be the only means by which we can, if at
all, know of the various facts of the external world.

A question arises at this point as to what the skeptics could
have meant by the words 'perceive,' 'observe,' 'feel,' and all the other
words to express sense perception if these words are to be applied to
material objects at all, for instance, when one says, 'I am eeeipg_an
envelope.‘ In other words, what sort of way of knowing could have been
denoted by 'seeing an envelope'? Clearly enough, it cannot denote
direct apprehension, since in this case, by direct apprehension we
would be apprehending certain sense data only. One may say that in
seeing the envelope--a material object--one is directly apprehending
a proposition about the envel0pe in the sense of having a vague belief

that there is something else existing besides the sense data, which we

apprehend when we see the envelope, and this something is the envelope

 

 

108

or a material object, and hence, we are indirectly apprehending the
envelope. Hence, seeing an envelope Would denote indirect apprehension.
Moore says that it is quite possible that the words 'seeing,‘
'observing,' etc., as or when applied to material objects can denote
indirect apprehension and the skeptics could have meant exactly this
relation when they apply these words to material objects. But he
points out that, even if, whenever we see an envelope which is a
material object, we indirectly apprehend something, say, the envelope,

it does not follow that we know that there is an envelope; nor does it

 

follow that even if there is an envelope, we kppp_that it is a material
object. The reason for this is that, when we talk, as it has been
mentioned already, of knowing the existence of material objects by
sense perception, we are talking of knowledge proper; since, when we
are talking of knowing that something exists, or, that something is a

material object, etc., we are talking of knowing certain facts--that

 

such and such is the case; and knowing facts denotes knowledge proper.33

And according to Moore, in order to know whether or not something is the
gpgga, we have to apprehend a proposition directly, we must believe that
proposition, the proposition must be true, and we also have to meet some
fourth condition.

So far as indirect apprehension is concerned, although it
forms apjyypp of knowledge proper, it cannot be said to be a way of
knowiru; the existence of something--knowing that_something exjptg, or
that something _i_§ a material object. It forms a M; of knowledge

proper in the sense that one can only know of the existence of a thing,

109

Le., know that it exists, in the proper sense of the term, when one
is apprehending the thing in question at least indirectly. Hence, if

byseeing or perceiving a material object, we mean a process by which

we do get to know of their existence, we must indirectly apprehend the

 

nmterial objects with which the sense data, which we directly apprehend,
are, in fact, connected.

But indirect apprehension of something cannot amount to
knowledge proper, or, to knowing that so and so is the case. Moore

says that even if we indirectly apprehend something else when we

 

directly apprehend a sense datum, it does ppp_follow that we know
of the existence of a material object by means of the senses, for
the following reasons: first, when we indirectly apprehend something,
we simply believe that there is something besides the sense data that
we are directly apprehending; and from the mere fact that we believe or
think that there is something, it by no means follows that there jg, in
fact, any such thing as we think or believe that there is. Second,
even if we believe that there is something else besides the sense data,
and there, in fact, is something else besides the sense data, it will
only amount to true belief; but true belief constitutes only part of
knowledge proper, but not the whole of it. Third, even if we not only
believe that there is something but kppy_that there is something, it
does not follow that we know that this something is-a material object.y’
Hence, although the direct apprehension of certain sense data
along with the indirect apprehension, in the sense of a belief about

the existence of a material object, take us only half way through to

110

the knowledge of the existence of material objects, it still does not
enable us to say that we know of the existence of material objects.
Neither direct apprehension nor indirect apprehension can lead to the
knowledge of the existence of something or knowledge of a fact——tpat_
something is the case. In order to be able to say that we know of the
existence of the material objects, we have to know that this belief
that something else than sense data exists, or the belief that there
exists a material thing which is different from sense data and yet

connected with them, is true and is known to be true, and also some-

 

thing else. In any case, we have to know at least that the proposition

 

expressing such a belief is true. The question as to whether something
exists or not is a factual question, and in order to settle it, it is
necessary to show whether the proposition expressing any such claim or
belief is true or not; since, the proposition expressing such a belief
is not such that simply by understanding the meanings of the words in
the statement one can decide whether it is true or false as one can do
in the case of an analytic or inconsistent statements. The skeptics
also have recognized that the issue of the knowledge of the existence
of material objects is an issue regarding the validity of a proposition;

they have declared that we cannot ever know for certain any statement

 

eXpressing a belief in the existence of material objects to be true;

whereas Moore asserts that we can and do know for certain that at least

 

some propositions expressing belief in the existence of material objects

are true.

1]]

According to Moore when the skeptics talk as to how much of
our knowledge is based upon observation and try to substantiate this
claim with the illustrations they give, they seem to have used the term
'observation' in the sense in which it is commonly used. Observation,

in this sense, does not mean the relation of direct apprehension.

 

 

Whore says that ordinarily by observation we mean "the relation which ?_3‘
we have to material objects themselves, ppep_we directly apprehend '
certain sense data: we do not mean by it the relation of QIEEEE.

apprehension. . . ."35 For instance, we observe movements of material ;

 

objects, or positions of material objects; and by 'observe' as applied
in such cases, we do not mean that we directly apprehend movements or
positions of material objects. But when we consider the conditions
that have been accepted by them as necessary for knowledge, we find
that by 'observation' or 'perception' they have meant direct apprehen-
sion only. The main reason for this has been the fact that they thought
that absolute certainty is a necessary mark of knowledge; for them
'knowledge' means 'absolutely certain' knowledge. And by 'absolute
certainty' they have meant 'indubitability' in a sense in which it is
not possible to doubt. They would say: "You don't know for certain,
aWthing which you pap_doubt. And you can doubt that this is a black-
board." It seems that they are using 'indubitable' to mean 'not able
to be doubted' as differentiated from 'indubitable' which means known
(i.e., beyond reasonable doubt). And for them as long as it is logi—
cally possible to doubt a proposition, one cannot be said to know that

DPOposition for certain to be true.

 

112

As a result of such considerations as these, they seem to have
supposed that all knowledge consists mepely in the direct apprehension
of sense data and images. And all those who maintain this have also
commonly maintained (l) that absolutely every sense datum that any
person ever directly apprehends exists only so long as he apprehends
it, (2) that no sense datum which any one person directly apprehends Fr
ever is directly apprehended by any other person, and (3) that no sense
datum that is directly apprehended by one person can be in the same

space with any sense datum apprehended by any other person. It follows

 

 

that, according to this view, all that we can ever know with certainty i
are sense data and images. On these hypotheses, nobody can possibly
know of the existence of anything at all except his own private sense
data and images.36
Moore's Defence of Common Sense
Against Skepticism

This part of Moore's theory of empirical knowledge consists
in justifying the common sense view which he wants to defend against
the skeptics' attack on it, by giving various reasons to show the
plausibility of the common sense view on the one hand and the implau-
sibility of the skeptics' view on the other. His discussion of this
specifically centers around the issue of knowledge of the existence
of material objects through sense perception, which, according to
Moore, is the most basic and fundamental view of common sense. Moore
claims that we all do know for certain that there are material objects

in the Universe. In other words, he claims that this belief of common

ll3

sense is known to be true belief and there are good reasons for this
claim. The skeptics, in contrast, claim that no such knowledge is
possible. Hence what needs to be established is that the statement
expressing the belief in the existence of any material object is a
type_statement. Hence the question to be settled is this: Can we
know for certain that any such statement is a true statement? According [fig
to the skeptics we cannot know the truth of any such statement with

certainty; and as long as we do not, the statement remains at the level

of mere belief, and belief is not knowledge.

 

In this situation, in order to appreciate either of the claims,
Moore's or the skeptics', it is necessary to discuss first the condi-
tions that the skeptics accept as necessary for knowledge. According
to Moore the conditions that skeptics accept as necessary for knowledge
are originally based upon the conditions laid down by Hume as necessary
for knowledge. Although different skeptics differ from one another in
their individual views so far as it concerns the forms in which they
express these views, but they all agree on the conditions of knowledge
and these are basically Hume's.

Moore's statement of Hume's rules of knowledge is as follows.
Hume divides all propositions into two classes--(i) those which assert
the existence of something and (ii) those which do not assert the
existence of anything. Propositions belonging to class (ii) may be
known to be true intuitively where their truth is self evident; or
may be known to be true by demonstration by a strictly deductive

method where their truth follows from the truth of some other

ll4

proposition. Prppositions belonging to class (i) cannot be known

by either of these ways. Propositions falling in this class can

 

further be divided into two sub-classes: (a) propositions asserting
the past, present, or future existence of something where this some-
thing pap_pe_directly apprehended; and (b) propositions asserting the
past, present, or future existence of something where this something
cannot be directly apprehended. Hume lays down two sets of rules for
knowing the truth of either of the two sub-classes of propositions
belonging to class (i).

There are two conditions for knowing the truth of any
pr0positions belonging to (i)(a): (l) A man may really know that
a thing does exist, if, at the very moment when he believes that it
exists, he actually is directly apprehending the thing in question.
And (2) A man may really know that a thing did exist in the past, if
he did directly apprehend it in the past and now remembers it.

There are two conditions for knowing the truth of any
propositions belong to class (ii)(b): (l) Nobody can ever know of
the existence of anything which he has not directly apprehended, unless
he knows that something which he has directly apprehended is a sign of

its existence. And (2) Nobody can ever know that the existence of any

 

one thing A is a Sign of the existence of another thing B, unless he
himself (or, under certain conditions, somebody else) has experienced
a general conjpnction between things like A and things like B; and,
nobody can be said to have experienced a conjunction between any two

things, unless he has directly apprehended both the things.37

115

These last two conditions of knowledge, which are called the
first and the second rules by Moore, constitute the main topic of
discussion, since, the issue in question is the knowledge of the
existence of material objects and material objects are never directly
apprehended. According to Moore, the skeptics have denied the possi-
bility of the knowledge of the existence of material objects on the
grounds of these two principles laid down by Hume. Their argument
consists of two parts (T) If Hume's principles are true, then none
of us ever knows of the existence of any material objects; and (2)
Since Hume's principles are true, none of us ever knows of the exis-

38 Moore agrees with the first part

tence of any material objects.
of this argument, i.e., if_Hume's rules are true, then we cannot know
of the existence of any material objects. For, on the basis of these
two rules, one may kppg, at the most, pf_the present existence of
something very much like, in many respects, the material object, in
which one commonly believes. For instance, if I am looking at a pencil
at this moment, Hume's rules "would ppt_allow me to know of the exis-
tence of exactly that, in which I believe--the material object, the

pencil." Since what it would allow one to know of are only patches

of colour of a certain shape, patches of hardness, and smoothness or

 

roughness of a certain shape. But these patches of colour, of hardness,
etc., certainly do not constitute the whole material object which one

believes one knows. Moore writes:

The pencil, in which I believe, certainly does not consist
solely of colours and of tactual qualities: what I believe
when I believe that the pencil exists is that there exists
something which really is cylindrical in shape, but which

ll6

does not consist merely of any number of patches of colour
or of smoothness or hardness, or any other sorts of sense
data which I have ever directly apprehended. Even if sense
data of all these kinds really are now in the same place
where the pencil is . . . I certainly believe that there
is in that place something else besides. This something
else, even if it be not the whole material object, is
certainly a part of it. And . . . if Hume's second rule
were true, I could not possibly know of the existence

of this something else. For I have never directly appre-
hended in the past anything that was like it: I have only
directly apprehended sense data which had a similar shape
to that which it has.39

 

According to Moore these skeptical views, which are essentially
based on Hume's principles regarding the conditions of knowledge, in
denying our ability to know of the existence of material objects, flatly
contradict common sense according to which we all do know of the exist-
ence of material objects. Hence, in order to be able to defend and
justify the view of common sense, it becomes absolutely necessary to
refute the skeptics' argument first. Accordingly, Moore's first step
towards a defence of common sense consists of a close examination and
evaluation of the skeptical views in their different aspects. In what
follows I will discuss Moore's main m0ves against the skeptics' views.

The statement of the skeptics' main argument.--The skeptics'

 

argument for the denial of our knowledge of the existence of material
objects is this:

If Hume's principle (regarding the conditions of
knowledge) are true, then we cannot know of the

 

existence of material objects; lst premiss
Hume's principles are true; 2nd premiss
Therefore, we cannot know of the existence of Conclusion

material objects.

117

l. Moore's first move consists in showing that even if we
accept the first premiss of their argument, it does not contribute to
the strength of their argument at all. This is so because this premiss,
if true, would also validly lead to a conclusion which is not only very
different from their conclusion but also has a very significant and
yet very destructive bearing on their own argument. This alternative
argument which can also be validly formulated with this very premiss

(and which Moore has suggested so far) is this:

 

If we do know of the existence of material lst premiss
objects, then Hume's principles are false.

We do know of the existence of material 2nd premiss
objects.

Therefore, Hume's principles are false. Conclusion

This argument being equally consistent destroys the strength of
their argument considerably since it contradicts their basic assumption
(expressed by their second premiss) which has to be true to establish
the conclusion of their argument; since, the first premiss of their
argument, by itself, even if true, cannot establish their claim without
also establishing an alternative claim equally consistently, which
falsifies their basic assumption.

I gp_know that this pencil exists; but I could not know

this, if Hume's principles were true; therefore, Hume's

principles . . . are false. . . . In other words, . . .

the fact that, if Hume's principles are true, I could

not know of the existence of this pencil, is a reductio

ad absurdaum of those principles.“

However, since both the arguments are consistent, and yet are

in conflict with each other, neither of them can be said to establish

118

its claim conclusively, unless it can be determined which one of the
two satisfies the conditions of a conclusive argument.

According to Moore, in order to decide whether or not an
argument is a good or conclusive argument, one has to consider not
only the fact that the conclusion follows validly from the premisses,

but also whether each premiss of the argument is known to be true or

 

not.

The fact that one proposition coheres with or follows from
another does not by itself give us the slightest presumption

in favour of its truth. My conclusion coheres with my premiss,
exactly as strongly as my opponent's coheres with pig. . . .
Both equally satisfy the condition that the conclusion must
follow from the premiss. . . . And yet obviously this mere
fact does not give the slightest presumption in favour of
either."1

In order to know that the conclusion of a given argument is
true, one must know the premiss or the premisses to be true. In the
case of the two arguments being considered, the respective premisses
other than the one already considered are, 'Hume's principle are true'
(skeptics') and 'I do know of the existence of this material object'
(Moore's). According to Moore the point that must be emphasized in
this context is that the premiss must be absolutely known to have the
equal degree of certainty or probability as the conclusion is known to
have. In other words, the premiss must be at least as certain as the
conclusion; the mere fact

that the premiss l§_true will not, by itself, enable any one
whatever to know that the conclusion is so. If anybody what-
ever is to be enabled by the argument absolutely to know the
conclusion, that person must himself first absolutely know
that the premiss is true. And the same holds good not only

for absolute certainty but also for every degree of proba-
bility short of it. . . . No argument is a good one, even

ll9

in the sense that it enables us to know its conclusion to

have any probability whatever, unless its premiss is at

least as certain as its conclusion: meaning by "certain,"

not merely true or probably true, but known to be so.“2
Hence, it seems that the only way it can be decided between the two
arguments, as to which one is better or is really conclusive, is by
deciding which of the two respective premisses in question is abso-
lutely known to be true. Moore writes,

My opponent's premiss is that Hume's principles are true;

and unless this premiss not merely l§_true, but is abso-

lutely known to be so, his argument to prove that I do not

know of the existence of this pencil cannot be conclusive.

Mine is that I do know of the existence of this pencil;

unless this premiss not only l§_true, my argument to prove

that Hume's principles are false cannot be conclusive."3
So the question arises: what are the conditions under which a
pr0position can be known to be true?

Before giving his own view, Moore considers a view about the
conditions under which one can know a proposition to be true. He says
that some philosophers seem to have held that the only way in which a
premiss or proposition can ever be known to be true is by having a
conclusive argument in its favour; what they have meant by 'having
a conclusive argument in favour of a premiss or proposition' is that
the proposition in question validly follows from some other proposition
(or a set of other propositions) which is already known to be true.

According to Moore this view cannot be true unless a certain
qualification is added to this method; but those who have held this
view have failed to notice that without this qualification this view

cannot be true. In other words, he does not deny that we do sometimes

know a certain proposition to be true on the basis of certain other

lZO

propositions which also we know to be true and from which the
proposition in question really follows; what he denies is that this
is the pply_way we can know the truth of a proposition. And his reason
for saying this is that this method can work, i.e., we can know the
truth of a proposition by this method, only if there is some other way
also by which we can know alproposition to be true. So he seems to be
suggesting two things here: (l) knowing a proposition to be true by
having a good argument in its favor is not the only way of knowing
the truth of a proposition; there is some way other than this way,
of knowing a proposition; and (2) the way which is being considered
cannot be a way of knowing the truth of a proposition unless there
is also some other way of knowing a proposition to be true.““

Hence, according to Moore the qualification which must be
taken into account if we are to know of the truth of a proposition
by this method is the following: In the case of any proposition P,
if we are to know the truth of P by knowing that P validly follows
from some other proposition or a set of other propositions, i.e., by
knowing that we have a conclusive argument in favor of P, then we must
know some proposition S to be true by some method other than the method
in question; where S is either the only other proposition from which
P follows, or S is one of the propositions in the set of other
propositfionS‘from which P follows; i.e., S must be a premiss of
the argument that we have in favor of P.

According to Moore, if we do not add this qualification

to the method of knowing the truth of a proposition by having a

 

l2]

conclusive argument in its favor (i.e., by knowing that the proposition
in question follows from some other proposition or a set of pr0posi-
tions), then we would never be able to kppy_any proposition to be true--
not even to be probable. The reason for this is that

if I cannot know any proposition whatever to be either true
or probably true, unless I have first known some other
pr0position, from it follows, to be so; then, of course,

I cannot have known this other proposition, unless I have
first known some third proposition, before it; nor this
third proposition, unless I have first known a fourth
before it; and so on ad infinitum. . . . And it is quite
certain that no man ever has thus known a really infinite
series of propositions. . . . So that if . . . the view
that we can never know any proposition whatever, unless

we have a good argument for it were true, then it would
follow that we cannot ever know any proposition whatever
to be true, since we never can have a good argument for it."5

 

Hence, there must be some condition or method other than this
under which a proposition can be known to be true. So far as the
question of knowing the truth of the respective premisses of the two
arguments--Moore's and the skeptics'--is concerned, it still remains
unanswered. It remains yet to be decided whether or not either one
of the premisses in question can be known to be true. So far as the
premiss of Moore's argument is concerned, he says that it can be known
and is known to be true by epme_way of knowing; although it cannot be
known to be true by the deductive method discussed above, i.e., by
having a good argument for it.” According to Moore, this way of
knowing consists in knowing the truth of a proposition without knowing
the truth of any other proposition. This is what Moore calls the

immediate knowledge of the truth of a proposition.

 

l22

Immediate knowledge, as I have already discussed in the
previous section of this chapter, is a form of knowledge proper,
and it is different from direct apprehension. According to Moore,
the premiss of his argument can be known with certainty to be true
by immediate knowledge. In this case we directly apprehend a propo-
sition, and at the same time we know that it is true without knowing
any other proposition from which it may follow. And, further, the
evidence for it is that we know it with more certainty than that with

which we can know any proposition which may be advanced for or against

 

it; or which may be used either to prove or disprove it. Thus accord-
ing to him, when, for instance, we see a pencil, or touch one we know
immediately that it exists; the truth of the proposition that it exists,
is known immediately; and it is known with certainty because it is
beyond reasonable doubt under the given circumstances, although it is
not beyond logical doubt. He says that the proposition which forms the
premiss of his argument, namely, '1 do know that this pencil exists' is
known to be true immediately. And it is more certain than any other
proposition which can be used to prove this proposition to be true
or to be false. He writes:

That is why I say that the strongest argument to prove that

Hume's principles are false is the argument from a particular

case, like this in which we do know of the existence of some

Inaterjal object. And similarly, if the object is to prove

in general that we do know of the existence of material

(ijects, no argument which is really stronger can, I think,

be brought forward to prove this than particular instances

in which we do in fact know of the existence of such an

objectu . . . However much more convincing it may be, it

is . . . sure to depend upon some premiss which is, in fact,
less certain than the premiss that I do know of the

 

123

existence of this pencil; and so, too, in the case of any

arguments which can be brought forward to prove that we

do not know of the existence of any material object."6

Moore has considered some objections that might be brought

against his claim and have given reasons why they do not affect his
claim adversely: (i) it may be said that the question here is regarding
whether or not the main ppemiss of his argument is known to be true; the .f““
main premiss of his argument is: I do kppp_that this pencil exists;
hence what he is claiming to know immediately is not the proposition,

that this pencil exists; but the proposition, '1 know that this pencil

exists,‘ and this cannot be known to be true unless the proposition,

 

'the pencil exists' is known to be true. Hence, it can be said that
even if one can know immediately that this pencil exists, one still
cannot be said to know that one knows that this pencil exists imme;
diately, since the knowledge of this proposition depends on the
knowledge of another proposition, viz., that the pencil exists.
Hence, the proposition in question must have been known only mediately.
In reply to this, Moore says that from "the mere fact that I should not
know the first, unless I knew the second, it does not follow that I
know the first me;ely_because I know the second. And, in fact, . . .
I do know pptp_of them immediately.”7

(ii) It may also be said that the proposition 'the pencil
exists' cannot be known immediately either, since in order to know
this, one has to directly apprehend certain sense data and has also
to know that these sense data are signs of the pencil's existence.
In reply to this objection also Moore gives more or less the same

answer:

l24

Of course I admit, that I should not know it, unless I were

directly apprehending certain sense data. But this is again

a different thing from admitting that I do not know it imme-

diately. For the mere fact that I should not know it, unless

certain other things were happening, is quite a different

thing from knowing pply_because I know some other

pr0position."8

It should be mentioned here that Moore's views about the

immediate knowledge of the truth of a proposition have a significant
bearing on Hume's classification of propositions and the corresponding
ways of knowing them. As has been discussed already, Hume has divided
all propositions into two classes: (i) propositions asserting the
existence (past, present, or future) of some particular thing(s);
and (ii) propositions which do not assert the existence of anything.
He has further divided the propositions belonging to (i) into (a)
those which assert the existence of things which are being directly
observed at the moment, or things which had been directly observed in
the past and now are being remembered; and (b) those which assert the
existence of things which are not being directly observed at the moment
nor had been directly observed in the past, or if observed in the past,
are quite forgotten. So far as propositions belonging to (ii) are
concerned they can be known to be true either duductively, as in the
case of mathematical theorems, or intuitively, as in the case of
propositions like, 2+2=4, white is different from black, if two
things are equal to a third thing, they are equal to each other;
irnnritively known propositions are just 'eeepf to be true immediately

without the help of any other propositions. In any case, according to

those who accept this classification of propositions, class (ii) of

125

propositions gp_ngp include any proposition asserting existence
of something.
It seems that Moore is suggesting that class (ii) of the
above division should not be limited only to the type of propositions
that the skeptics accept to be in it; it seems that he would say that
it should also include at least some propositions of a particular type,
viz., those which do assert the existence of something which is not
being directly observed at the moment, nor have ever been so observed
in the past, and yet are known to be true immediately without the help
of any other proposition. He says:
Just as we know matters of fact, which we have observed,
without the need of any further evidence, and just as we
know, for instance, that 2+-2==4, without the need of any
proof, it is possible, that we may know, directly and
immediately, some facts which we never have observed."9
It should be mentioned, however, that Moore has not clearly
discussed this matter in his epistemological writings. In what follows
I will try to mention certain points about the immediately known
propositions by way of qualifying Moore's suggestion mentioned above.
The immediately known propositions are, indeed, very different
in nature from those which are taken to be intuitively known, since
intuitively known propositions do not assert the existence of anything
whatsoever; and furthermore, they are self evident in the sense that
their opposites are inconceivable. In any case, such propositions are
pep contingent since their denial would be self contradictory. They

are like necessary or analytic or tautologous propositions. Whereas

the propositions, which Moore claims to be immediately known to be

126

true, are such that they do assert the existence of something. Moore
has never pajg_that they are self evident, and, even if they are, they
have to be so in a quite different sense. In any case, they are not
self evident in the sense that their opposites are inconceivable; Moore
has explicitly said that such propositions are all contingent since it
is always logically possible to deny them without any self contradiction,
and yet they are known with absolute certainty to be true.

Such propositions as Moore claims to be immediately known to

be true are different from the intuitively known propositions also in

 

the respect of their mode of knowing. It seems to me that Moore would P"
not say that intuitive knowledge is a form of knowledge proper; since
in knowledge proper besides knowing a proposition to be true we also
have to fulfill some other condition (which Moore does not specify but
seems to have meant sufficient evidence or good reasons); but in intui-
tive knowledge the propositions known are self evident, and as such
there is no need of having any evidence or reasons for knowing them--
they are self justifying. Whereas so far as immediate knowledge is
concerned, Moore says that it is a form of knowledge proper, and as
such in knowing them we must satisfy the other condition of knowledge
PrOper as well.

If these propositions, which are immediately known, are to be
included in the skeptics' class (ii) of propositions, then their clas-
sification would have to be modified, and consequently it would con-
siderably bear upon one of their major claims, viz., we cannot and do

not know of the existence of material objects. It would modify their

127

classification of propositions considerably, since, if such propositions
are to be included in their class (ii) of propositions, then they

would have to revise their classification of propositions of class (i);
since, in this case, at least some of the propositions of their class
(i)(b) would now belong to their class (ii) of propositions.

Secondly, if these propositions, which are immediately known,
are included in their class (ii) of propositions, then they would have
to admit of the knowledge of the existence of material objects, since
their conditions of knowing the existence of material objects would
not hold good at least in those cases as are expressed by immediately
known propositions. At any rate, whether or not immediately known
propositions are to be put together with the deductively and intui-
tively known propositions in the same class, Moore's arguments,
showing that the propositions asserting the existence of material
objects can be known immediately, do considerably diminish the force
of the skeptics' claim.

2. Moore's second move against the skeptics' claim consists
in showing that they could ppt_have known the truth of the premiss of
their argument; and in subsequently showing that they themselves could
not have known the truth of the conclusion of their argument. (This
move is related with the first move just considered; in the first move

Moore has tried to show that his argument, viz.,

128

If I know that this pencil--a material object-- lst premiss

 

exists, then Hume's principles are false;
I do know that this pencil exists; 2nd premiss
Therefore, Hume's principles are false. Conclusion

is a conclusive argument, since, the basic premiss of his argument, I
do know that this pencil exists, can be known to be true immediately.
In the second move, he considers whether or not the skeptics can make
a similar claim about their argument, since if they can then it would
defeat Moore's first move. On the other hand, if they cannot, then
it would not only establish Moore's claim, it would also defeat the
purpose of their own argument.) The basic premiss of the skeptics'
argument is: Hume's principles are true. The question that Moore
raises here is this: Can they, according to the conditions of knowledge
accepted by them, know that this proposition is true? To put it in a
different way, he is asking: "By which of the ways of knowing that
ypp_have accepted, do you know that these principles are true?" It
is to be noticed that in order to evaluate the proposition, viz.,

"Hume's principles are true,I epistemologically, i.e., whether this
particular proposition is or is not known to be true, one has to con-

sider Hume's principles (i.e., propositions stating these principles)

 

themselves. It seems that Moore's move here is rather complex, or,

at least, not as straightforward as his first move, for the reason

that Moore is not directly attacking their premiss in question, viz.,
"Hume's principles are true"; he is attacking the principles themselves

(or the propositions stating these principles). He is not directly

129

asking, "How do you know that the proposition, 'Hume's principles are
true' is true?" What he is asking, instead, is, "How do you know that
the propositions stating (or expressing) these principles are true?"
(These principles have been stated earlier in this section.)

The reason for this explanation of Moore's move here is that
even if Moore's argument that I am going to state, does show that the “i“

skeptics do not know that their conclusion is true on the basis of the

 

conditions of knowledge they accept, yet it may be said that Moore has
not shown, after all, whether or not the basic premiss of their argu-
ment is known to them, and as such he has not shown which of the two
arguments, his and the skeptics', is a better argument; so far it has
been proposed to be decided on the basis of the knowledge of the
respective premisses. Before I discuss his argument, I will briefly
mention some of the reasons why Moore may have preferred to take this
indirect and rather complex route to show that the skeptics' argument
is not a good one.

a. Moore could have attacked the skeptics' argument directly
by asking: "How do you know that the proposition 'Hume's principles
are true' is true?" He could have further argued that since this
proposition is not self evident (as its denial is conceivable without
self contradiction), it could not have been known by intuition. And,
since this proposition does not assert the existence of anything, it
could not possibly have been known by the principles which Hume has
laid down to know the propositions asserting the existence of some-

thing. But the skeptics could still have said that they know this

130

pr0position to be true by deduction from the truth of the principles
themselves. In that case, one can always reasonably question the

truth of the principles themselves, unless they are already proven

to be true. And if they are not proven to be true then their deduction
from these principles would not be justified.

b. In order that a person may justifiably say that he kppg§_
that the proposition 'Hume's principles are true,‘ is true, he must
know that Hume's principles are true; since the proposition 'Hume's
principles are true' can be true only if Hume's principles ape_true.
And if one does not know that Hume's principles are true, then
obviously one does not know that the proposition, 'Hume's principles
are true' l§_true. Hence, in any case, it is the knowing or not
knowing of the principles themselves which is crucial. It is crucial
for the skeptics to show that they do know the principles to be true;
and it is crucial for Moore to show that the skeptics could not have
known them to be true according to the conditions of knowledge that
they accept.

The principles being considered in this context are those laid
down by Hume as conditions under which certain propositions asserting
the existence of external facts (in this context specifically,
material objects), can be known to be true. The general form of
these principles, as stated by Moore, is this: One can only know
of the existence of external things (in this case, material objects)
by inference; and one cannot be justified in inferring the existence

of any particular thing, except by Experience--"meaning by 'experience'

131

what Hume meant; namely, that you have directly apprehended in the

past conjunctions between things similar to something which you now

n 50

 

directly apprehend and things similar to what you now infer.
The question here is whether this proposition stating the
principle can be kppwn to the skeptics by the methods of knowing
accepted by them. Moore has given two arguments to show that this
proposition cannot be known to be true by the conditions of knowledge
which they accept.
Moore's first criticism is based on Hume's own argument;

it is, in a sense, as Moore acknowledges, a restatement of Hume's own
criticism of this principle. (It must be mentioned here, by way of
clarification that according to Moore's interpretation, Hume, at dif-
ferent places of his two works dealing with epistemology, have held
three different views regarding the knowledge of the existence of
'matters of fact' [and external objects or facts which represent one
kind of matters of fact], with varying degrees of skepticism. A very
brief summary of these three views is as follows: According to his
first view although we are incapable of knowing a good many religious
and philosophical propositions asserted in books of "divinity and school
metaphysics" which have been claimed to be known, we can and do know
"the vast majority of facts peypng our own observation, which we all
commonly suppose that we know." His second view, on the other hand,

is that it cuts off at once all the possibility of our

knowing the vast majority of these facts; since he implies

that we cannot have any basis in experience for asserting

any external fact whatever--any fact, that is, except facts
relating to our own actual past and future observations.

132

And his third view is more skeptical still, since it
suggests that we cannot really know any fact whatever,
beyond the reach of our present observation or memory,
even where we have a basis in experience for such a fact:
it suggests that experience can never let us know that
any two things are causally connected, and therefore that
it cannot give us knowledge of any fact based on this
relation.5')

 

In his first argument here, Moore is referring to Hume's
argument for this tplpg_view. Moore says that after laying down his
principle that we cannot know any fact, A, to be causally connected
with another, B, unless we have experienced in the past a constant
conjunction between facts like A and facts like B, Hume goes on to
ask, "what foundation we have for the conclusion that A and B ape_
causally connected, even when we paye_in the past experienced a constant

conjunction between them." For instance, it is quite conceivable, that

 

even though in the past, whenever I was close to a fire, I experienced
a sensation of heat, "yet in the future this should never happen again.
What is there to prove that it will happen or is likely to happen in
any single instance?" It can be proven only if one already knows some
such principle as mentioned above. But the question is, how can one
possibly know any such principle as this?

It is a principle which cannot be known in any of the ways

which Hume allows to be possible. It is not self evident;

it cannot be deduced from anything which is self evident;

and it cannot be known by Experience, since on the contrary,

experience can never be known to prove anything whatever,

unless this principle is first known to be true.52
Hence, the only foundation, according to Hume, we have for this
principle is custom; “It is nothing but custom which induces us to
believe that, because two facts have been constantly conjoined on

many occasions; therefore they will be so on all occasions."53

133

Regarding this argument of Hume's, Moore says: “I do not see

how it can be answered." He thinks that this argument leads the
skeptics to a dilemma which it is hard for them to escape; for if they
accept that they do kppp_that experience is a sound basis for inference,
since they do know, in fact, that when two things have been conjoined
in the past, they are likely to be conjoined again, then they should
also admit that they do know of the existence of material objects. For
if it is a question of factual certainty based on experience, then it is
no less certain that we do know of the existence of material objects.
If the inferential knowledge of probable conjunction is admissible then
there is no reason why inferential knowledge of material objects should
not be admissible, since criterion of absolute certainty in the sense of
indubitability is not satisfied in either case; the proposition express-
ing knowledge of experience as sound basis for inference is no more
certain than the propositions expressing the knowledge of material
objects.

If, on the other hand, the skeptic admits that he does not know
if experience is a sound basis of any inference, then he has to admit a
totally solipsistic situation in which case he does not and cannot know
the existence of any thing beyond his own existence and sensations (and
that also in a limited sense, for he would not know of his future
existenceeand sensations even if that be for the next moment). And,
in that case, the skeptics cannot be entitled to lay down any such rule
as Moore is attacking, namely, that IE man knows of the existence of

material objects. For, if they do not even know whether or not there

134

are any other human beings, they cannot have claimed as to how much
other human beings do or do not know. For all that they can have
claimed is that lj_there are any other human beings they may_(or
perhaps, may not) know of the existence of material objects.

In other words, if you once accept this sceptical argument

of Hume's as showing that Experience [in the sense in which
he defines it] is ppt_a sufficient basis for inference, while
at the same time you hold that it is the only basis you have,
all attempts to lay down general propositions about the limits
of human knowledge become absurd. You can only at most be
entitled to lay down rules as to what you yourself do know
and do not know; even with regard to yourself you cannot
possibly be entitled to assert that you may not, the very
next moment, know of the existence of some material object:
you can have no basis for the assertion that there is the
very smallest probability that you will not know this.5“

Moore's second argument to the same effect consists in con-
sidering the conditions that must be fulfilled in order to be able to
lay down any general principle, and in showing that the general prin-
ciple that has been accepted by the skeptics, and which is being
considered, is such that it does not fulfill those conditions. All
such propositions about the limits of human knowledge, as is expressed
by the principle being considered, are, in fact, arrived at and sup-
ported by the consideration of particular instances, in which we do in
fact seem to know or not to know something. Moore says,

They are in fact based--and this is the best basis they can
have--upon an attempt to collect all the various instances,
in which we obviously do know something, and all those in
which we obviously do not know something, and to discover,
by comparison, what conditions are common to the cases in
which we gp_know, and absent from all those in which we do
not know. In fact any general principle to the effect that

we can never know a particular kind of proposition, except
under certain conditions, is and must be based upon an

135

empirical induction: upon observation of the cases in which

we obviously do know propositions of the kind in question,

and of those in which we obviously do not, and of the

circumstances which distinguish one class from the other.55
From this it follows that the particular instances always have greater
(or at least equal) certainty than what any general principle may have.
Once the general principles are thus established they are often thought
to have greater authority, but initially by itself no general principle
has any greater certainty, or even equal certainty or perhpas any cer-
tainty in comparison to what these particular instances have. It is
completely on the basis of the certainty of the particular instances,
and constant continuation of such a support, that any general principle
is established and achieves authority. Now, it seems clear that the
skeptics' principle, viz., No man can know of the existence of anything,
unless certain conditions are fulfilled, does not meet this condition,
i.e., it does not seem to be based upon, or to be established by
particular instantiations. Unless it is obviously instantiated by
particular cases where we do not know of existence of a material object,
no principle which asserts that one cannot know of the existence of
anything except under conditions which are not fulfilled in the case
of material objects, can be regarded as established.

The mere fact that in any particular instance I glg_know

of the existence of a material object, in spite of the

fact that the conditions named were not fulfilled, would

be sufficient to upset the principle, and to prove that

it was not true. . . . It must first be known, with a

greater certainty, that I do not, in fact, in any par-

ticular instance, know of [the existence of any material

object]. And is this, in fact known with any certainty?
It seems to me it certainly is not.56

136

It seems that the skeptics committed the mistake of reckoning as
instances of the cases where (according to them) we obviously do not
know something, eppp_instances in which it is by no means obvious that
we do not know the thing in question, whereas the claim that we do know
of the existence of material objects is one that can be shown to be
based upon many such particular instances in which we obviously do
know of the existence of some material object or other.

3. Moore's third move against the skeptics consists in
considering some of the consequences that follow from their denial
of the common sense views; and in showing that these consequences
exhibit, in one form or another, inconsistency in their views. In
other words, his move consists in showing different types of incon-
sistencies that the skeptics get involved in as a result of their
denial of the common sense views.

According to the view of common sense, propositions asserting
the reality of material things, time, space, and other human beings
can be known to be true; for instance, we all believe we know that
there are things like chairs and tables, etc., which we take to be
instances of material things; similarly, we all believe we know that
there are many human beings each of whom have had some experiences
similar to those that each of the rest has had; we also believe we
know that these material things and living bodies occupy space and
exist in time. According to Moore some philosophers have held views
incompatible with those of common sense, in two ways: (a) either

they have held that some or all the proposition accepted by common

137

sense to be true, are falge; (b) or, they have held that pppe_of us
kppp§_for certain that any of the propositions accepted by common sense
as kpgyfl_to be true, although they have not said these propositions ape_
false.

Moore's main argument about (a) is that all those philosophers
who have denied the truth of the propositions of common sense, have also
held other views inconsistent with their denial of the truth of common
sense propositions. One way of showing this is the following: accord-
ing to common sense there have existed (or do exist) many human beings
each of whom has known, for instance, that there are material objects
and there are human beings other than the knower and who have similar
experiences; if anybody denies that such a view is true, then, he is
denying that there were or are any human beings who have known the
existence of material objects and other human beings. If he is right,
then it follows that he is denying that there were or are any philos-
Ophers who have held or does hold any such view, viz., the statement
"there were or are many human beings who have known such and such
things" is false. But the fact is that there are some philosophers
who have held such views, since, at least he (i.e., one who denies)
himself holds it; for in holding the view that he does, he is also
admitting the existence of other human beings who have bodies and
thoughts, etc., as he does, from whom he anticipates the views that
he is denying; and it is also a fact that his views are known to
somebody, or else, his views would not be discussed by anybody. And,

if all ‘this is true, then he cannot be right in saying that there were

138

or are no human beings who have known . . . etc.

In other words, . . . if any phil050pher has denied it, it
follows from the fact that he has denied it, that he must
have been wrong in denying it. . . . Since, if I know that
they [i.e., such philosophers as one mentioned abovel have
held such views, l_am, 'pso facto, knowing that they were
mistaken; . . . since I am more certain that they have
existed and have held some views . . . than that they have
held views incompatible with it [common sense].57

 

Another way in which, according to Moore, they have betrayed
this inconsistency, is by alluding to the existence of other philos-
ophers in particular, and of the human race in general by using the
word "we" in the sense in which we all ordinarily use it. He writes,

Any phil050pher who asserts that "we" do so and so, e.g.,
that "we_sometimes believe propositions that are not true,"
is asserting not only that he himself has done the thing in
question, but that very many other human beings, who have
had bodies and lived upon the earth, have done the same

. they have, therefore, been holding views inconsistent
with propositions which they themselves knew to be true. . . .58

Moore's argument for (b) is that, whenever philosophers have

denied the knowability of propositions asserting the existence of

 

material things or other persons, their denial entails two mutually
incompatible propositions, viz., 'I do not know that p' and 'Ilgypy
that p.' (This position is different from the preceding position

in the respect that from the preceding position this consequence does
not follow. Those who hold the preceding position simply hold that
the statements of the common sense are not true; there is no contra-
dicticn1 involved in thinking that what they say is true, e.g., it
might have been true that there were no material objects, or there
were»rw) human beings, etc.; their mistake consists in simply holding

also other "views inconsistent with propositions which they themselves

139

kney_to be true" and in their failure to notice this fact. But this
type of view, according to Moore, is not, in itself, a self contradic-
tory view in the sense that it entails both of two incompatible
propositions (which the following view does)).

~The position against which Moore's second argument is
directed, can be briefly stated as follows: These philosophers admit
that we all do in fact believe such propositions of common sense as
asserting the existence of material objects, etc., and that such
propositions may even be true; but they deny that we can ever kppp_
them, for certain, to be true; that they are just matters of Faith
or mere belief, but not of Knowledge.

The point that Moore wants to emphasize in this connection
is that these philosophers are making assertions about human knowledge
in general. Whenever a philosopher makes any such assertion, he is
making an assertion about all_human beings including himself (or about
himself as well as about many persons other than himself). In other
words, he is ngt_making the assertion pply_about himself-~it is always
about others too. When such a philosopher says, for instance, "No human
being has ever kpppp_of the existence of other human beings or material
objects,' what he is in fact saying is that 'There have been many other
human beings, who have had material bodies as I have and who have been
capable~of having similar experiences as I am, and none of them, jpplpgy
irniinyself, has ever known of the existence of other human beings and of
inaterial bodies.‘ In any such case, the philosopher is actually making

a statement about 'g'uhe says, 'None of ll; knows,‘ or 'W_e_ do not know,‘

140

or 'Nobody including myself knows'; he does not simply mean to say,
'l_do not know.‘ And any such uses of words like 'us' or 'we' involves
allusion to the existence of persons other than the speaker.

However, if by the expression 'Nobody knows,’ one really means
only 'I don't know, and lf_other men have existed, they don't know
either,’ then there is no contradiction involved. Moore says, "There
is only self contradiction in holding l_knpw_there have been other men,
and they have made such judgments, but they haven't known them to be
true." Moore thinks that this is exactly what these philosophers are
doing, when they are denying the knowability of the common sense state-
ments about the existence of material objects, etc.; since, philosophers
who deny that such propositions of common sense are ever known to be
true, almost all of them "not only admit but insist that ye_do make
such judgments as 'This is a desk' and so on." What they deny is that

we do know that such judgments are true. These statements, according

 

to them, are just part of the system of common beliefs; they are, in
all cases, only believed, not ever kppwp_for certain; these are just
matters of faith. "But what does 'we' mean? It means 'men with human
bodies living on this earth'; and this means men related with material
things related to ppj§_material thing (i.e., the earth)."59
Furthermore, the admittance, on their part, of the very
existence of common sense beliefs is self contradictory; since if

such a philosopher says: "These beliefs are beliefs of Common Sense,

 

but they are not matters of knowledge," he is saying that 'There have

been many other human beings, besides myself, who have shared these

141

beliefs, but neither I nor any of the rest has ever known these to

be true.“ But the fact is that if these beliefs are beliefs of common

 

sense, then they must be true; since the proposition that they are (or

there are) beliefs of common sense is such a proposition which implies

 

(because of the very meaning of 'beliefs of common sense') the pr0po-
sition that many human beings, besides the philosopher who denies the
knowledge of common sense statements, have had bodies, which have lived
upon the earth, and have had experiences, including beliefs of this kind.
According to Moore,

when people say: "We believe such things" they commonly
mean, "It's certain that we believe them; and by this, ngp_
only, l_know that we believe them. . . . This means "it's
certain that other men have existed, but none of them have
known.” Nobody, in fact, means only: "l_don't know, and,
if, other men have existed similar to me, they don't know
either."6°

Proof of the Existence of Material
Objects and of Knowledge of Them

 

 

Moore's final justification of the claim that the common sense
propositions asserting the existence of material objects are known to
be true, consists in appealing to the facts of common experience (and
to the validity of the authority of the ordinary language. His appeal
consists in pointing to particular facts of life which cannot be

reasonably denied or even doubted, and particular usages of ordinary

 

language which we all accept to be significant. However, in order to

 

justify the claim that certain propositions are known to be true, it
is necessary to show that the pr0positions in question are true; since

when we are talking of the knowledge of facts, i.e., that something j§_

142

the case, we are talking of knowledge proper; and one of the conditions
of knowledge proper is that the proposition believed to be known must
be type, Hence, in order to show that the propositions asserting the
existence of material objects are kpgwn, it is necessary to show that
such propositions are true, showing that material objects exist.)
Accordingly, (T) Moore first argues that there ape_material objects,
and subsequently, that it can be kppyp_that there are material objects;
and (2) then argues to the same effect that if the ways we use certain
words in common usage or ordinary language are ways of using those words
meaningfully, then we will have to admit also that we know of the
existence of material objects; for otherwise, our ways of using those
words in the ways we do use them would be reduced to absurdity.

Everyday facts appealed to.--(l) By material objects we

 

generally and ordinarily mean objects like chairs, tables, different
kinds of bodies-~some animate, some inanimate, trees, mountains, stones,
flowers and the like. In everyday life we do not all the time make
statements like: I am sitting on a material object, I am now writing
on a material object, I am seeing a material object running, or I am
seeing a material object reading a material object; the statements that
we do make are like these: I am sitting on a chair, I am writing on a
piece of paper which is on this table, I am seeing a cat running, or I
am seeing a person reading a book. But we all commonly believe that
all these objects are material objects; that all these different types
0f objects are different examples of material objects. This is one way

0f defining the term 'material object,‘ what Moore calls definition by

 

143

examples. If we further try to determine what is the common feature
that we all accept about those things which we ordinarly accept to
be instances of material objects, then we find that we all accept that
these are objects which exist outside of us, i.e., not inside our bodies
or minds, and occupy space. In other words, we all accept that they are
external objects. To say the same thing more specifically and techni-
cally, by a material object we mean something which does occupy space,
is not a sense datum and is not a mind, nor an act of consciousness.
According to Moore, one way of showing the ppptp_of propositions

asserting the existence of material objects, is to show that such propo-
sitions as 'This is a chair,‘ or 'This is a human hand,‘ and so on, are
true; i.e., if we can show that there are things like chairs, human
hands, etc., where these things ape_accepted as particular instances
of material objects, then it would be shown, by that very fact, that
there are material objects. In other words, if it can be shown that
one chair or one human hand exists, then the proposition, 'There exists
a chair,‘ or the proposition, 'There exists a human hand,‘ is true.
Further, Moore says that

it is also true that from the proposition that there is at

least one "thing" of that kind there follows the proposition

that there is at least one thing external to our minds . .

[for instance, from "There is at least one chair," it follows

"There is at least one external thing," i. e. , "There is at

least one material object]. "51

Moore claims that he can prove quite rigorously that there are

many such things--viz., chairs, tables, human hands, and the like. He

writes:

144

I can prove now, for instance, that two human hands exist.
How? By holding up my two hands, and saying, as I make a
certain gesture with the right hand, "Here is one hand,"
and adding, as I make a certain gesture with the left,
"and here is another." . . . But did I prove just now
that two human hands were then in existence? I do want
to insist that I did; that the proof which I gave was a
perfectly rigorous one; and that it is perhaps impossible
to give better or more rigorous proof of anything
whatever.62

Moore says in order to decide whether or not it is a proof, we simply
have to find out whether it fulfills all the conditions that a proof
must satisfy.

According to him, any rigorous proof must satisfy three
conditions: (a) the premiss which is given as a proof of a conclusion,
must be different from the conclusion; (b) the premise must be kppyp_to
be true; and (c) the conclusion must really follow from the premiss.
Moore claims that his proof, in fact, satisfies all the three condi-
tions:

(a) The premiss which I adduced in proof was quite certainly
different from the conclusion, for the conclusion was merely
"Two human hands exist at the moment”; but the premiss was
something far more specific than this--something I expressed
by showing you my hands, making certain gestures, and saying
the words "Here is one hand, and here is another." . . . In
asserting the premiss I was asserting much more than I was
asserting in asserting the conclusion. (b) I certainly did
at the moment know that which I expressed by the combination
of certain gestures with saying the words "There is one hand
and here is another." I knew that there was one hand in the
place indicated . . . and that there was another in the
different place indicated. . . . And finally (c) it is quite
certain that the conclusion did follow from the premiss.
This is as certain as it is that if there is one hand here
and another here neg, then it follows that there are two
hands in existence npg,53

 

 

The point to be emphasized here is that even though this proof is

different from a strictly formal proof in certain ways, we, in fact,

145

do constantly accept proofs of this sort as absolutely conclusive
proofs of certain conclusions--as perfectly reliable for finally
settling issues as to which we were previously in doubt. If this

can be accepted as a proof for the existence of a material object in
the present, then by a similar proof it can also be shown that material
objects existed in the past and were known to be so. But this proof
may not be regarded as a conclusive proof by some philosophers and this
fact leads Moore to present further considerations to the same effect.

Ordinary language appealed to.--Moore has further presented

 

consideration of a similar sort as arguments to show that propositions
asserting the existence of material objects are, and can be, knpgn_to
be true. He says that some philosophers, however, may refuse to accept
such arguments as constituting proofs at all for any propositions what-
ever: whether it is a proposition asserting the existence of material
objects, or, whether it is a proposition asserting the knowledge of the
existence of material objects. According to Moore these philosophers
seem to have accepted only two kinds of proofs as real proofs for any
conclusion: deductive and inductive. Of these two, according to them,
only the conclusion of deductive proof can be absolutely known or known
for certain to be true; the conclusion of an inductive proof can never
be known for certain, since, the conclusion of such a proof is at the
inost highly probable, but it can never be absolutely known or abso-
lUtely certain. Further, they have held that any argument intended

to establish an existential proposition about physical objects must

be inductive in nature, since no existential proposition about physical

146

objects is self evident and as such they cannot be intuitively known
to be true. Hence any argument about the existence of physical objects
must be inductive in nature and as such its conclusion can never be
gpppp_to be true.

Moore's reply to this sort of argument is that such pr0positions

as "this pencil exists now" can be known immediately to be true without FT

 

knowing the truth of any other proposition from which it follows. (I
have discussed this issue in detail earlier in this chapter, hence I

will not discuss it here.) Moore claims that although propositions like

 

these are not deductively arrived at, nor are they self evident in the L
sense that their opposites are inconceivable, yet they are known with
absolute certainty to be true because it will be absurd to doubt them
under certain circumstances, although it is logically possible to
doubt them.

In order to prove his point Moore draws instances from every
day facts where we not only use the word "certain" or "know" quite
significantly, but also where, if we hesitate to use those words or
doubt their appropriateness under those circumstances, it will be,
in a sense, absurd, unless we absolutely change the meanings of those
words in which they are used commonly. For instance, if I am looking
at my'rjght hand now while I am also seeing that I am writing with it,
i.e., making certain movements with it, then I kppp_that I am looking
at a human hand. In other words, in such situations, I kn_o_w that it
j_s_ a human hand, and I know this with absolute certainty. However,

Moore admits that it is not a matter of logic; it is always, perhaps,

 

147

logically possible that it is not a human hand. Those who doubt the
truth of such statements might say: “You don't know for certain any-
thing which you pap_doubt. And you pan_doubt that this is a human hand.
But gap_1 doubt it? or even if we can, does it prove that I don't know
it?"5“ Moore writes:

How absurd it would be to suggest that I did not know it,

but only believed it, and that perhaps it was not the case!

You might as well suggest that I do not know that I am now

standing up and talking--that perhaps after all I'm not,

and that it's not quite certain that I am!65

The point Moore wants to make is that if propositions of this

 

nature are not certain, i.e., known with absolute certainty then
nothing of the kind is ever certain: “if ppey_[are] not certain,

then no pr0position which implies the existence of anything external

to the mind of the person who makes it is ever certain." If it is
true, for instance, that I do not know now (when I am standing up and

I also think that I am), that I am standing up, then certainly I do not
know anything of the sort at all, and never have known anything of the
sort; and, not only so, but nobody else ever has. But this is not a
matter of logic.66 Since, it may be argued that, it is logically
possible, e.g., that I am dreaming now and yet claiming that I am
standing up. It is logically possible also that I am dreaming now

and while dreaming I am thinking that I am standing up, and am actually
standing up. Hence, if I am not certain that I am not dreaming now,
then, even if I am thinking that I am standing up and I am, in fact,
standing up, I cannot be certain that I am standing up; i.e., if I do

not know for certain that I am dreaming then, I do not know for certain

148

that I am standing up. Moore says that he does not wish to dispute
that, from the hypothesis that one is dreaming, it might follow that
one does not know that one is standing up; "though I have never seen
the matter argued, and though it is not at all clear to me how it is
to be proved that it would follow." However, even if it is granted for
the moment that if one does not know that he is not dreaming, then one
does not know that one is standing up (though one actually is standing
up and thinks that one is), it is doubtful how much force such an
argument really has. Since, as Moore points out,

. . . [the] first part of the argument is a consideration

which cuts both ways. For, if it is true, it follows that

it is also true that if I gp_know that I am standing up,

then I do know that I am not dreaming. I can therefore

just as well argue: since I do know that I'm standing up,

it follows that I do know that I'm not dreaming. . . . The

argument is just as good as the other, unless my opponent

can give better reasons for asserting that I don't know

that I am not dreaming, than I can give for asserting

that I do know that I am standing up.67

The reason that Moore invokes is the evidence of the senses.

By evidence of the senses for an assertion 'what is meant is that the
assertion is, at least in part, based upon the things that I am seeing
or hearing or feeling (or, in other words "having certain visual,

auditory, tactile or organic sensations," or a combination of these.

This evidence may not be necessarily conclusive, or the only evidence

 

available or possible; but one can certainly claim that

I certainly have ppy_the evidence of my senses in favour
of the proposition that I am_standing up, even though this
may not be all the evidence that I have, and may not be
conclusive . . . if a man at a given time is only dreaming
that he is standing up, then it follows that he has npt_

149

at that time the evidence of his senses in favour of that

proposition: to say “Jones last night was only dreaming

that he was standing up, yet all the time he had the

evidence of his senses that he was" is to say something

self contradictory.68
Hence, the view that I do not know for certain that I am not dreaming,
entails also that it is possible that I have not the evidence of my
senses that I am standing up. Even if it is not certain that I have
at this moment the evidence of my senses for the proposition that I am
standing up, it still is quite certain that I am having some sensory
experience, though it is not quite sure whether it consists of only
dream images, or, sensations that really constitute the evidence of
the senses in waking experience. The point that Moore wants to empha-
size here is that these philosophers cannot possibly discriminate
between the two (experiences), if they say, as their argument surely
contends to, that it is not certain that my sensory experience now
consists only of dream images. That is to say, none can be in a
position to say that somebody's sensory experience now consists of
only dream images, unless one knows that dreams have occurred, and
that the contents or data of sensory experiences in dream are dif-
ferent from those in waking life, i.e., the very state of dreaming
is different from the state of waking experience. Moore writes:

. . . can he consistently combine this proposition that

he knows that dreams have occurred, with his conclusion

that he does not know that he is not dreaming? Can any-

body possibly know that dreams have occurred, if at the

same time, he does not himself know that he is not dreaming?

If he l§_dreaming, it may be that he is only dreaming that

dreams have occurred; and if he does not know that he is not

dreaming, can he possibly know that he is get only dreaming
that dreams have occurred? .‘. . I don't think that he can.69

150

Hence, Moore concludes that anyone who says that dreams have occurred
and also asserts that nobody ever knows that he is not dreaming, is
guilty of an inconsistency. Since, by asserting the former proposition
he implies that he himself knows that dreams have occurred; "while, if
his conclusion is true, it follows that he himself does not know that
he is not dreaming, and therefore does not know that he is not only
dreaming that dreams have occurred."
Moore advances a similar argument against all those who argue

that nobody can ever know for certain that pr0positions like 'This is
a blackboard,‘ or 'I am seeing a blackboard,’ etc., on the ground that
we might be hallucinating, or under illusion, and hence, might be mis-
taken. Moore refers to those cases of perceptual statements of ordinary
life where it would be unreasonable to say that one does not know for
certain, or to say that one is hallucinating. But if one still says,
even about such cases, that we might be mistaken, then such a statement
will be inconsistent in the sense that it will imply one's knowledge of
those cases where one might ppt_be mistaken, whereas what they really
want to say that we cannot have any such knowledge. Moore argues,

Is it ossible I am mistaken in [saying] "This is a black-

board" Iwhen I know that I am_looking at one]? Np; if I

know it. What does [their] argument prove? There are

experiences which resemble this in a certain respect and which

are false beliefs, [therefore] this one may be a false belief.

I.e., some of its characters are compatible with its being a

false belief. I don't see how this can give any good reason

for th1nking that this is not a case of knowledge. If such a

v1ew says; We know that in some cases we are mistaken, it

is self inconsistent: for we can certainly only know that

in a particular case "This exists" was not true, by inference
from cases in which we know that it is true.7°

 

151

The point that Moore wants to emphasize is that these
philosophers think that a statement can be known for certain, only
if it can be known by formal deduction; i.e., if it is not formally
proven, then it is not proven at all. But Moore does not think it
necessary (as it has been earlier pointed out), that in order to know
a statement, we always have to have a formal proof for it; for Moore,
under certain circumstances, we can know a statement to be true, even
though that statement does not formally follow from a set of premisses.
"From the fact that it's not known formally, it doesn't follow that is
is not known. . . . Why shouldn't premisses which don't prove a conclu-
sion formally yet prove it?"71 He thinks that one can know a proposi-
tion in consequence of certain premisses (which constitute the evidence
for that proposition), in spite of the fact that it doesn't follow
formally from them.

The consideration of these arguments shows that Moore has
consistently pursued his objective to refute the skeptics' challenge
about our lack of knowledge of material objects. His arguments, as
stated and discussed in this chapter, show that he has been able to
defend and justify the common sense claims about the knowledge
material objects; and that his acceptance of the existence of sense
data does not affect either the consistency of his procedure or the

validity of his arguments.

CHAPTER VI

COMMON SENSE, ANALYSIS, AND SENSE DATA
IN MOORE'S EPISTEMOLOGY

Section I
Why Does Moore Analyze Common Sense Views

From the discussion of Moore's epistemological views in the
previous chapters, two points become quite clear: the first is that
Moore has never cherished any doubts either about the existence of
material objects or about the existence of sense data; he has expli-
citly admitted the existence of both sense data and material objects,
and yet has come up with a theory of perceptual knowledge very different
from that of the skeptics' in certain important respects. The main dif-
ference, in this context, between his views on the one hand, and those
of the skeptics on the other, is that the skeptics have mainly contra-
dicted the views of common sense on the ground of the existence of sense
data; i.e., the philosophers whose views Moore has tried to refute, have
all accepted certain views about sense data such that on the basis of
those views they either have denied the knowability of material objects,
or, have denied both the knowability and the existence of material
objects; whereas these are two of the most important views accepted
by common sense, viz., that there are material objects and that we all

know that there are material objects. In short, all those philosophers

152

'51.; 'VIT- ad (.7

 

153

(i.e., philosophers that Moore is concerned with), who have accepted

the existence of sense data have at least denied the knowability of

 

material objects. Whereas Moore, while fully accepting the existence
of sense data, has never denied either the existence of material
objects or the knowability of material objects.

The second point is that Moore's philosophy of common sense

._u:,5_.‘

and his philosophy of sense data are not two conflicting views as some
of his critics might have thought; that despite the two philosophies

Moore has been quite consistent in the pursuit of his main objective,

 

‘ 911mm

viz., to refute skeptics' claims against the claims of common sense;
and that his sense data philosophy has not invalidated his arguments
against the skeptics.

In this chapter, I shall discuss some further questions that
may still be raised about Moore's maintaining the two standpoints and
certain consequences that follow from this. In particular, I shall be
concerned with two questions: first, although the two standpoints are

not in conflict with each other, it may still be asked, Why has Moore

 

introduced the concept of sense data, after all, when what he has
mainly wanted to do is to defend and justify common sense against
skeptical views about external world? For it seems to be a fact that
the notion of sense data does not form a part of the common sense views

about perceptual knowledge. And second, since Moore accepts the exis-

 

tence of both sense data and material objects, how does he explain the

relation between the two?

154

Section II. Moore's View About the Role of
Analysis in Philospphical Discussion

It has been already indicated in Chapter III that the notion
of sense data has been introduced mainly in two ways, viz., by way of
analyzing the perceptual statements of common sense and by way of
analyzing sense perception as a cognitive act. Hence the question:
why has Moore brought in a philosophy of sense data in order to refute
skepticism about the external world, when he already has a philosophy
of common sense realism? can be replaced by the question, why, after
all, does Moore analyze the perceptual statements of common sense, and
sense perception as an act of consciousness, when such analyses lead
to the introduction of sense data--an issue which does not seem to be
required for the pursuit of his main objective?

It seems to me that this question can be answered by pointing
out the fact that although it is true that the main objective of his
theory of empirical knowledge has been to establish the validity of
the common sense knowledge claims as expressed in everyday perceptual
statements (against the skeptics' attack on them), from this it does
not follow that this has been the only objective of his phiTOSOphical
pursuit in this context; or that he has not or cannot have pursued any
other objective simultaneously.

From the different writings of Moore it seems to be a fact that

he has accepted the analysis of concepts and propositions, not only

155

as the most appropriate metppg_of doing philosophy, but also as one

of the objectives of philosophical pursuit. According to Moore the
main task of philosophy consists in providing a general description

of the Universe by classifying the different things that are or are not
in the Universe, in describing how and to what extent we know these
things, and in discussing what things are good and what are not. But
more importantly, according to Moore, any discussion of these problems
also leads to the discussion of a great number of questions or problems
regarding the meanings of the different concepts and propositions
involved in such discussions. And it is part of the object of philos-
ophy to answer all these questions regarding the meanings of concepts
and propositions. As such analyzing not only forms the method of
philosophizing but also becomes part of the subject matter of philosophy.
While discussing the subject matter of Philosophy--the sorts of ques-
tions Philosophy deals with--Moore says that the class of philosophical
problems includes not only that of classifying the different things we
take to be in the Universe, but also the problem of determining

whether these classes of things do exist or do not, or whether we

are simply ignorant as to whether they do or do not; and also of trying
to define these classes and considering how they are related to one
another--to define more clearly what is the difference between these
various sorts of things: for instance, what is the difference between
a material object and an act of consciousness, between matter and mind,
etc.; what sort of a thing one's knowledge is, what the relation

between knowledge and truth is, and so on. It is a function of

156

philosophy to answer these questions, including the questions apppt_the
meanings of concepts and propositions; and any discussion dealing with
these problems involves analysis.1 Quite in accordance with this, Moore
says, while discussing the nature and kinds of philosophical questions,

So far I've distinguished two main kinds of questions which
it seems to me be the business of philosophy to discuss.

(1) Questions about the meaning of particular words, phrases
or forms of expression in common use (only some). (2) Ques-
tions about the nature of Reality as a whole--understood in
a very wide and a very vague sense.

And he further adds two more questions to these:

(3) A number of questions abouthuman knowledge,(4) Still more
questions about what it's reasonable for us to believe and

in what degree. I don't see how it can be denied that all_
these sorts of questions do belong to philosophy. Of course,
people may say that they ought not to: but as philosophy is
now used they g9,2

According to Moore 'analysis' or, as he alternatively called
it, 'phil050phic definition,' does not consist in providing the verbal
definition of words or the meanings of words as dictionaries provide
them--reportive definitions or how a certain word is used in a par-
ticular language. Philosophic definitions are

all of them definitions of words or forms of expression

you already understand--you already know their meaning,

in the sense that you attach the common meanings to them
when you use them or hear or read them, though you may not
know their meanings in the sense of being able to make true
propositions of the form "this means so and so" what philos-
Ophy tries to analyse is concepts which you p§e_in ordinary
life. (To use a concept for which a word stands = to use
the word, and attach certain meaning to the word.)3

 

 

 

Analysis of a concept consists in describing the real nature of the
Object or notion denoted by a word by reducing the object in question

to, and enumerating, its simplest constituent concepts which cannot be

157

further analyzed.“ Similarly analysis of a proposition consists in
breaking the proposition into simpler and more fundamental propositions
involved in it.5

So far as the usefulness and justification of doing analysis
in philosophy is concerned, either as its method or as a part of its
subject matter, it may be said, in a general way, that the usefulness
of analysis consists in producing clearness of thoughts when one is
thinking about the various questions concerning the various aspects of
a particular problem and the solution of the problem. According to
Moore, "the chief use of analysis in the way of clearness, is only

clearness which it produces when you are doing_philosophy itself."

 

In this connection he distinguishes between two different ways in
which analysis produces clearness: (a) the sort of clearness one
attains in understanding a particular definition; "the sort of clear-
ness you enjoy when you arrive at a good definition of this sort . . .
is, it seems to me, something worth having for its own sake." And
(b) the sort of clearness which understanding a given definition may
bring to one in understanding other philosophical propositions--either
another definition, or a philosophical proposition of the kind which is
not a definition._ This sort of clearness does not help very much in
answering philosophical questions, or in solving philosophical problems;
but it does help in having a clear understanding of the questions or
problems.

Philosophical clearness is not necessary to answer such

a question as: Am I nearer to the blackboard than you are?
And does it really help to answer philosophical propositions?

158

I doubt if it helps you at all to say whether they are

true; but only to see more clearly what the question

means.6
But for Moore, the clear understanding of a philosophical question or
problem is no less important than providing the answer or solution
to it, since he is of the opinion that most of the difficulties and
disagreements in philosophy are because of the fact that the questions

or problems that philosophers try to provide answers or solutions to,

are not clearly understood. In the preface of Principia Ethica, he

 

writes to this effect:

I do not know how far this source of error would be done

away, if philosophers would ppy_to discover what question

they are asking, before they set out to answer it; for the
work of analysis and distinction is often very difficult:

we may often fail to make the necessary discovery, even

though we make a definite attempt to do so. But I am in-
clined to think that in many cases a resolute attempt would

be sufficient to ensure success; so that, if only this attempt
were made, many of the most glaring difficulties and disagree-
ments in philosophy would disappear. . . . They are constantly
endeavouring to prove that 'Yes' or 'No' will answer questions,
to which neither answer is correct, owing to the fact that
what they have before their minds is not one question, but
several, to some of which the true answer is 'No,' to others
'Yes.'

Since Moore has always recognized that analysis is a part of the
function of philosophical discussion of a problem, as has been shown
above, he has, accordingly, thought it necessary to analyze the differ-
ent aspects of the problem of the perceptual knowledge of the external
world, the problem which has been the main concern of his epistemological
views. In particular, he has thought it necessary to analyze concepts
like 'material object,‘ 'existence,‘ 'knowledge,‘ 'perceiving,‘

'seeing,' and so on; and statements like: 'This is a table,‘

159 -

'I am seeing a human hand,‘ 'There are material objects,' and so on.
The analysis of concepts and statements by way of clarifying their
meanings and by raising and answering various questions about them,
forms an important and major part of his entire philosophical enter-

prise--both from the viewpoint of method and from that of subject

 

matter. As has been discussed in Chapter III, it is in the context p+=
of analysis that he has thought it necessary to introduce the notion

of sense data. According to the skeptics we cannot know with certainty

 

the truth of such statements as 'That is a table,‘ where by 'table' we

 

mean a material object; whereas according to Moore we can and do know
with certainty that that is a table and that it is a material object
in the sense in which we ordinarily use the expression 'material object,‘
and if we are asked how we know that it is a table, surely in most
cases we would say that we know that it is a table because we are
perceiving it; in other words we know the table by means of sense
perception and we do know with certainty that the statement 'That is
a table' is true. Hence, so far as the truth of such common sense
statements is concerned, we do know it. But this claim should not be
confused with the claim that we know the correct analysis of such
statements,’ and furthermore, the claim that we all perceive the table,
should not be taken as implying that there is only one sense of perceiv-
ing, since 'perceiving' (and words expressing different forms of it) can
be analyzed into different senses.

When Moore claims that he is defending the common sense view,

he is not claiming that he is defending the common sense analysis of

160

such views, since there aren't any common sense analyses of such views.
The common sense view, as Moore takes it, consists of certain knowledge
claims which are expressed in everyday perceptual statements and not
the analyses of such claims, and as such the question of defending or
not defending the common sense analyses or claims of analyses of state-
ments does not arise. It is only when the common sense views or claims
become the subject matter of philosophical discussion, as they do in l.
Moore's theory of empirical knowledge, that the question as to their

analyses arises.

 

The fact that Moore's analysis of the common sense views in 'l
certain respects leads to the introduction of the notion of sense data
in his views regarding empirical knowledge, does not affect adversely
either the coherence of his views or the consistency of his procedure
in his theory of empirical knowledge for the following reasons:

(l) Although the analysis of the common sense views does not
constitute a part of the common sense view, it provides a clarification
of such views and the consequences they lead to when and if analyzed;
for instance, it clarifies that though we know such views to be true,
we do not know their correct analysis; it also clarifies how an attempt
to analyze such views leag§_to the introduction of the notion of sense
data. As such the analysis of the common sense views is not only in-
directly related to any discussion of such views but also occurs within
the general framework of Moore's views about perceptual knowledge.

(2) Although the analysis of certain aspects of the common

sense views does lead to the introduction of the notion of sense data,

161

the latter does not generate a skepticism in Moore's views as it does
in the case of the skeptics'; i.e., Moore's acceptance of the fact that
there are sense data does not lead to a skepticism regarding the knowl-
edge of material objects or an external world; he does not think, as
the skeptics have thought, that it is necessary to deny the knowledge
of the existence of material objects, if one has accepted the existence
of sense data. His arguments for this position have been given in the
preceding chapter.

Section III. The Relation Between Sense Data
and Material Objects

 

 

So far as the problem of the relation between sense data and
physical objects is concerned, it is closely related with the analysis
of the perceptual statements of common sense (as discussed above).
After asserting that we all do know with certainty that perceptual
statements of common sense, like, 'I am seeing a human hand,' or
'This is a human hand,‘ are true, Moore raises two questions as to
what could have been peapt_by such statements. In this connection
he points out that although one may kppp_or understand the meaning
of a statement (in the sense in which we all ordinarily understand
it), and yet one may not know the meaning of a statement in the sense

of knowing the correct analysis of that statement;8 his questions here

 

refer to this last point.
The first question that he raises is: What is it that I am
actually perceiving when I am perceiving that this is a human hand or

that this is a part of the surface of a human hand? According to Moore,

 

162

what is actually perceived is always a sense datum such that in

perceiving this sense datum one is ppp_perceiving the whole human hand.

In other words, “what I know, with regard to this sense datum, when I

know 'This is a human hand,‘ is not that it is jpeelf_a human hand. . . ."

That this is so is, according to Moore, quite certain because one knows

that a hand has many parts (e.g., its other side, and its bones inside 5—:
it), which are quite certainly ppp_parts of this sense datum. i
This leads to his second question which is: If this sense i
datum which one is actually perceiving in perceiving a human hand is i
p“-

 

pgt_the ppgle_of the human hand, then can it be a part of the surface
of the hand? Moore says that any proposition of the form "This is a
human hand" includes in its analysis a proposition of the form "This
is part of the surface of a human hand" (where "This” has a different
meaning from that which it has in the original proposition, viz.,
"This is a human hand”). But the important point is that

this proposition also is undoubtedly a proposition about the
sense datum, which I am seeing, which is a sense datum pf_my
hand. And hence the further question arises: What, when I
know "This is part of the surface of a human hand," am I
knowing about the sense datum in question? Am I, in this
case, really knowing about the sense datum . . . that it
itself is part of the surface of a human hand? Or, . . .
that even here I am ppp_knowing, with regard to the sense
datum that it is itself part of the surface of a human hand?
And, if so, what is it that I am knowing about the sense
datum itself?9

 

Before I discuss the various answers to this question that
Moore considers, it seems important to me to make a distinction between
two different aspects of the relation between sense data and physical

objects. From Moore's discussion of this particular problem of the

163

relation of sense data to physical object, it seems clear that this
problem admits of two different aspects: according to one aspect,

the question is, "What is the epistemological relation between sense

 

data and physical object?“ and according to another aspect, the question

is, "What is the ontological relation between sense data and physical

 

objects?"

In what follows I shall simply briefly mention some of the main

 

points of Moore's views regarding the epistemological relation between
the two, since I have discussed the contents of these points in detail
in Chapter III (although there I have not classified them as belonging
to this aspect of the problem; rather I have discussed them under the
more general topic, viz., Moore's views regarding sense data). The
main concern of the rest of this chapter, however, is the question

regarding the ontological relation between sense data and physical

 

objects. It seems to me, nevertheless, that some of the main points
regarding epistemological relation between the two should at least be
mentioned here; since Moore's views regarding the ontological relation
between the two are not only very closely related with his views regard-
ing the epistemological relation between the two, but also are, some-
times, overlapping. (However, Moore has not, in his discussions,
explicitly distinguished between these two aspects nor has he used

these two names to classify them.)

 

. s7

164

Epistemological Relation Between Sense
Data and Material Objects

 

 

So far as the epistemological relation between sense dat and
material objects (or physical surfaces) is concerned, the specific
question it presents is this: How is the knowing of material objects
related with the knowing of sense data? Or, conversely, how is the
knowing of sense data related with the knowing of material objects?

In other words, in what sense is the knowledge of the one related to
the knowledge of the other? Moore has given a fairly clear and defin-
itive answer to it. His account for this relation is that the perceiv-
ing of material objects necessarily involves the glpegt_perceiving of
certain sense data; that whenever, for instance, we gee-a material

object, we necessarily directly see certain sense data. In such cases,

 

the direct seeing of sense data is, what he calls, a function of seeing

a material object. That is, in a situation where we see a material

 

ppjept_in the ordinary sense of 'see,' we necessarily directly see or
apprehend something el§e_as well, and that something else is (are) a
sense datum (sense data).1°

Furthermore, in a case where two or more people are looking at
the same physical object under normal or ordinary conditions, although
we can say that we kppy that these people are seeing the same physical
object, we cannot or should not say that we kppy_that they are seeing
the same sense data.

Finally, although it is true without any qualifications, that
the seeing of a physical object necessarily involves the direct seeing

of sense data, it is ppt_true that the direct seeing of sense data

 

 

165

necessarily involves the seeing of a physical object; since in many

 

sensory experiences, for instance, in dreams, hallucinations, after
sensation, etc., though we directly see sense data, we do not see any
physical objects; however, as Moore would say, in the cases of sensa-

tions proper, the direct seeing of sense data do involve, as a matter

 

of fact, the seeing of physical objects. Since, I have already dis-
cussed these issues earlier in Chapter III, I have not discussed them
in detail here. What I have tried to suggest here by mentioning these
points is that since Moore has talked in fairly definitive terms about
the epistemological relation between sense data and physical objects,
it does not seem that there is any skepticism involved in his views
regarding the epistemological status of sense data, or of physical
objects, or regarding the epistemological relation between the two.
The reason why I have mentioned this is that it may be thought that
although Moore's sense data philosophy does not lead to a skepticism
regarding our knowledge of the existence of material objects, it still
leads to a skepticism about our knowledge of the relation between sense
data and the physical surfaces (that sense data are said to represent,
or to be signs of or identical with).

Ontological Relation Between Sense
Data and Material Objects

 

 

So far as the ontological relation between sense data and
material objects (or physical surfaces) is concerned, the specific
question it presents is this: Are sense data identical with the

corresponding surfaces of physical objects. And, if they are not

 

I W
1

166

identical with physical surfaces then what sort of a relation can
they have to them?

Moore considers three alternative answers to this question,
which, according to him, are the only answers so far offered; but he
finds that each of them can be objected to.

(1) According to the first view, when one knows "This is a
part of the surface of a human hand," what one knows about the sense
datum in question, is that it is itself a part of the surface of a

human hand. If this is so then one has to say that though one does

 

not directly perceive one's hand, yet one does directly perceive part
of its surface; "that the sense datum itself l§_this part of its surface
and not merely something which (in a sense yet to be determined) "rep-
resents" this part of its surface"; in other words, the part of the
surface of a hand is directly perceived exactly in the same sense in
which a sense datum is directly perceived.ll

But Moore finds this view untenable, because if this view is
true, then either we would have to give up the view which has been held
to be true by most philosophers, viz., that sense data really do have
the qualities which they appear to have, or else, we would have to
admit that incompatible qualities (in a particular respect) can inhere
the same thing at the same time. What is meant by this is that if it
is a fact, as has been so far agreed upon, that different people expe—
rience slightly different sense data, in a particular respect, e.g.,
in the respect of colour, while seeing the gape_physical surface, then,

if the above mentioned view is true, all these slightly different

 

167

colours directly perceived by different people would have to belong

to or inhere the same physical surface simultaneously, which seems
unlikely. On the other hand, if we still believe that this view is
true, and also believe that all these different colours papppp_inhere
the same area of the surface at the same time, then the only alternative
left is that we declare the accepted view regarding the nature of sense ‘.“
data in this respect (viz., that different people actually perceive, for
instance, different colours simultaneously), although they are all per-

ceiving the same physical surface, as false; but this view has not been

 

proven false; hence, the view that we are directly perceiving the part
of the surface of a hand in the same sense in which we directly perceive
sense data is at least questionable.

A second objection against this view is that if this view is
true then in the cases of seeing double, where we certainly have two
sense data each of which is pf_the surface seen, we would have to say
that pgtp_sense data are identical with the surface seens, which again
seems unlikely since the surface seen is one where as the sense data
seen are two, and both cannot be identical with one and the same
surface.

(2) According to Moore the second type of answer to the
question of the relation of sense data to material objects has several
versions. Of these different versions, the only one which Moore thinks
to be plausible, is that according to which the relation R that holds

between the sense datum and the surface in question is an ultimate and

 

unanalyzable relation,

168

which might be expressed by saying that "ny" means the
same as "y is an appearance or manifestation of x." I.e.,
the analysis which this answer would give of “This is part
of the surface of a human hand" would be "There is one and
only one thing which it is true both that it is part of the
surface of a human hand, and that this sense datum is an
appearance or manifestation of it.12

The main objection against this view, according to Moore, is

3‘;

that how we can possibly know with regard to any of our sense data that

we experience in a particular case, that there is one thing and one

ugh-17$: HI." ’1. 1".

thing only which has to them (i.e., to the sense data) such a supposed

...—-. ‘fl'fi.

ultimate, unanalyzable relation. Those who claim that there is such a

 

relation have not given any grounds for their claim.
(3) The third alternative answer that Moore considers is

Mill's view according to which material things are "permanent possi-
bilities of sensation." On such a view when one knows that "This is
part of the surface of a human hand" what one knows with regard to the
sense datum, is not that the sense datum in question is itself part of
the surface of a human hand; nor does one know with regard to any
relation, that the thing which has that relation to the sense datum,
is part of the surface of a human hand; but

a whole set of hypothetical facts each of which is a fact

of the form. "If these conditions are fulfilled, I should

have been perceiving a sense datum intrinsically related to

this sense datum in this way," "If these (other) conditions

had been fulfilled, I should have been perceiving a sense

datum intrinsically related to this sense datum in this
(other) way,” etc., etc.13

 

 

For instance, on this view, a proposition of the form, "I am seeing
a disc," has to be interpreted as something like this: If I were to

move my body in certain ways, I should be directly apprehending certain

 

169

sense data (other than the one(s) that I am apprehending now), e.g.,
tactual ones, which I should not be apprehending as result of these
movements of my body, if my present visual experience were a mere
hallucinatory or imaginary experience.

From Moore's reference to and discussion of this view in his

""1

different works,'“ it seems that he is inclined to prefer some such
view to the rest of the views, so far as the analysis of the perceptual

statements about material objects is concerned. But he has also pointed

 

out serious difficulties with this view.

Dr.”
One of the main difficulties with this view is that none of the

type of interpretations of perceptual statements suggested above (viz.,
"I am seeing a disc" is equivalent to “If I were to move my body . . ."
etc.) is in its ultimate form; i.e., each of them so far is indefinite
and incomplete. The reason for this is that all Of these interpreta-
tions contain some physical object terms, like 'body' or 'coin' or
'disc,‘ etc.; in other words, these interpretations themselves involve
reference to material objects, and in order to make the interpretation
complete, these terms are to be interpreted in terms of hypothetical
statements, and so on until all such terms are so interpreted. Sec-
ondly, it is doubtful if there is any intrinsic relation between the
sense datum that one is apprehending at the moment and the sense data
thatziane would apprehend when certain other conditions are fulfilled;
and even if there is any such intrinsic relation we do not know so far
lunv to explain either the ontological nature of that relation or our

knowledge of it. Finally, if this view were true then the interpre-

taticn1 of such terms as 'round' or 'square,' etc., would be very

170

different from what they are ordinarly taken to mean. Ordinarily we
take these qualities as qualities of material things; since this view
conceives of matter or material things in a completely different way,
the interpretation of terms like 'circular,‘ etc., as applied to
material things have to be very different from their common usage.
The same problem arises regarding the interpretation of statements
asserting the past existence of material things or the existence of
unperceived material things. According to Moore, they all have been
interpreted, on this view, in a Pickwickian sense.15

The point that Moore seems to emphasize is that the question
regarding the relation of sense data to material objects can be
definitively answered only if a correct analysis of the perceptual
statements about material objects is available. But so far we do not
have any analysis of these statements which is free from obvious dif-
ficulties. Any such analysis, in its turn, depends upon definite
answers to certain basic questions regarding the ontological status
of sense data, for instance, questions like: Do sense data, at least
some of them, exist unperceived (i.e., in the same sense in which we
know that physical objects do)? And, are sense data physical in the
sense that they occupy some public space in the sense in which physical
objects do? I have discussed Moore's views about such problems earlier
in Chapter 111. To recapitulate summarily, his final answers to these

questions seem to be inconclusive; in other words, it may be said that

 

he has not given any final answers to these questions. On the one hand,

he has claimed that he is strongly inclined to believe that at least

171

eppe_sense data do exist unperceived16 and do occupy the same space17
which is occupied by the corresponding surfaces of physical objects;

he has claimed that there are good reasons (though not conclusive ones)
to think that at least some sense data do so exist and occupy space,
and as such, there are good reasons to believe that some sense data
may be identical with the surfaces of physical objects that we see.

But on the other hand, he also recognizes that there are some good
reasons (though not conclusive ones) to doubt that this is the case,

i.e., to doubt whether they are so identical with physical surfaces.

 

Although Moore has not given any definite answers to questions regard-
ing whether sense data are mental, or physical, i.e., whether or not
they occupy areas of space publicly observable (e.g., areas of surfaces
of physical objects), it seems from his latest writings (although even
here his views are not completely free from hesitations), that he is

more inclined to hold that sense data do not exist unperceived and they

 

do not occupy any public space, and they are not identical with any
area of the surface of a physical object. He says at one place of
his "Reply."

I am inclined to think that it is impossible that anything
which has the sensible quality "blue" . . . more generally,
anything which is ever directly apprehended, any sense
datum . . . should exist unperceived, as it is that a
headache should exist unfelt.18

In another passage of the same work, he writes"

I know perfectly well that, if my present situation had
been different from what it is only in the respect that
I had been wearing blue spectacles of plain glass, then
the part of the surface of my hand which I am now seeing
would have looked to me of a somewhat different colour

172

from that which it now looks to me. . . . But, if so, the
directly seen object, which would then have “corresponded"
to that piece of surface cannot possibly be identical with
the piece of surface in question. And if lp_is not, then
certainly nothing else which I am directly seeing is
identical with that piece of physical surface.19

It is on the basis of passages like these that I have said

above that he is more inclined to hold the views regarding sense data :—

that I mentioned there; but these passages should not be taken without

qualifications since there are also other passages there which exhibit

the

tions; one such very interesting passage is the following: 4

And

fact that these views of his are not completely free from hesita- 5‘

 

I am now seeing part of the surface of my hand; and I do
now not only feel sure but know, with regard to this object
I am seeing which i§_part of the surface of my hand, that
it is part of the surface of my hand. And also I do pep,
at the very same time, feel some doubt as to whether a
certain object, which I am directly seeing, is identical
with the object which I am seeing which is part of the
surface of my hand. But to say that I feel doubt as to
this, is to say that it is possible that it l§_identical.
And if_it is identical, then I am both feeling sure of and
doubting the very same proposition at the very same time.
I do not say, of course, that I am_doing this. I only say
that, so far as I can see, I don't know that I'm not.2°

 

yet at another place in the same context, he says,

And this is the truth. I am strongly inclined to take
both of these incompatible views. I am completely puzzled
about the matter, and only wish I could see any way of
settling it.21

173

Section IV. Moore's Final Views About the
Relation Between the Two

 

Moore has persistently continued his search to discover what
the relation between the two should be until his very last work. In
his last paper, "Visual Sense Data," he has explicitly said that the
relation between a sense datum and the corresponding physical surface
papppp_possibly be one of identity.22 His final observations as to what
this relation can be if it is not one of identity, are very much based

upon certain distinctions that he has made in this connection.

 

The first distinction is between the two referents of the
demonstrative 'this' or 'that,‘ and the second distinction between
two different senses or modes of seeing (which also holds relative
to the cases of other senses with necessary changes) corresponding
to the two referents of 'this' or 'that.‘ I have briefly mentioned
the first distinction in Chapter III while discussing how the analysis
of perceptual statements leads to the introduction of sense data; in
this section I shall discuss this distinction in detail emphasizing
the role it plays in explaining the ontological relation between a
sense datum and the corresponding physical surface. I have discussed
and explained the second distinction in Chapter III while discussing
how the analysis of sense perception as an act of consciousness leads
to the introduction of the notion of sense data, and to a certain
extent while discussing the epistemological status of sense data and
the epistemological relation between a sense datum and a physical

surface; in this section I shall refer to this distinction only to

 

174

the extent it overlaps the discussion of the ontological relation
between the two.

In the paper "Visual Sense Data," Moore says that the demon-
strative 'This' or 'That' in perceptual statements like 'This is a
penny,‘ or 'I am seeing this,‘ or 'That is a hand,’ etc., refers to
or denotes two different objects at the same time and not only one, :%1
as it is ordinarily supposed to. On the one hand, the demonstrative i
'this' or 'that,‘ in such statements, is a short for a phrase of the

kind that Russell has called a definite description; in such a usage,

 

the word 'this' or 'that,’ in Moore's view, refers to or denotes the ;*
part of the surface of a physical object that we are seeing; and yet,
on the other hand, it also denotes or refers, at the same time, to
another object which we are directly seeing in apprehending or recog-
nizing ppp_the surface of the physical object which also we are geejpg_
at the same moment, lggk§_to us, and in this case the object referred
to is sense datum. The definite description which 'this' or 'that' is
a short phrase for is, according to Moore, “the object of which tpj§_is
part of the surface"; and, in this sense (i.e., in the sense of a def-
inite description), "you could be said to 'know' the penny 'only by
description,‘ although you can also correctly be said to be seeing

the penny. . . . Perhaps we might say that the penny in such a case

is only 'seen by description'!"23 The point that Moore wants to
emphasize is that if the demonstrative 'This' or 'That' in the state-
ment 'This (or that) is a penny' is short for a definite description,

and as such refers to the physical surface which satisfies that

175

description, and at the same time, refers to a sense datum corresponding

 

to the physical surface, then the proposition 'This (or that) is a
penny' is a proposition apppt_two objects at once, and not about only
one. At this point Moore points out that although in some of his
earlier works he has maintained that the ultimate of the principal
subject of any perceptual statement is always a sense datum, that does
not conflict with his later claim, viz., that such a judgment is always
about two objects at once. Since, he says, when he has said that a
sense datum is the ultimate or real subject of a perceptual statement,
he has not meant that it (the sense datum) necessarily has to be al§p_
the pply_subject of the statement; rather the claim “implies that it
[i.e., the sense datum] is ppp_the only subject."

In order to explain the relation between the two referents
of 'This' (or 'That'), he relates this distinction with that between
'seeing' (or 'seen') and directly seeing (or directly 'seen'). When
the word 'This' refers to an object which is directly seen (in an
ordinary case of visual perception), the referent of 'This' is a sense
datum; when the word 'This' is a short for a definite description, it
refers to an object, answering the description, which is seen; in
this case, the referent of 'This' is a physical surface. But the
important point is that in such a situation both the referents are
present at once and both the modes of seeing corresponding to the two
referents occur at once or simultaneously. And this is a fact which,
according to Moore, we do not usually recognize in everyday perceptual

experiences; we do not recognize that whenever we are seeing a physical

 

‘5“ '....'- vb"- -

' ‘3

I
'3‘“: '0

3"

176

object or any part of its surface, we are also directly seeing some

 

object which is directly given to the sense, namely, a sense datum and
about which we are not sure whether this directly seen object is identi-
cal with the surface of the physical object that we are seeing. Moore
mentions these two points as follows:

(l) that the function "x is seeing a physical object" entails
the function "x is seeing a sense datum” (= "x is seeing some
object directly"), or, in other words, that the sense in
which we "see" when we say that we see a thing which is a
physical object, is such that the seeing of a physical object
necessarily involves the seeing of a sense datum, and also
(2) that there is some reason to think that . . . no sense
datum which we are seeing is ever identical with any physical
surface which we are seeing; or in other words, that though e~i
the seeing of a physical object necessarily involves directly
seeing some object, yet there are good reasons . . . for
thinking that no physical surface is ever directly seen, and
that therefore the seeing of a physical object necessarily
involves the direct seeing of an object which is not a
physical reality at all.2

 

 

He explains and illustrates this relation with a standard case
of visual perception of a physical surface under ordinary and normal
conditions, i.e,, where one is not seeing the object double, etc.,
and is seeing the surface of the object in continuity (i.e., without
any of its parts being hidden by some other object); for instance,
when one is looking at a white wall. The gpjte_wall which one sees
without any coloured glasses on, would look different if one looks at
it with some coloured glasses on. If one looks at the same white wall
with blue glasses on, the wall--which is white and not bluish white--

looks bluish white to one. According to Moore, this is

177

merely another way of saying that I am directly seeing an
expanse which really is of a bluish white colour, and which
at the same time has to the surface which is ppp bluish-
white a specific relation . . . 'R'--a relation which
entitles me to assert that, in directly seeing that bluish-
white expanse, I am seeing the surface of the wall which is
ppp bluish-white.26

 

(The same type of phenomenon is obtained when, for instance, the same
surface of the penny looks circular and elliptical from two different
angles.) It is obvious that in a situation like this although it will
be correct to say that one sees a white wall, it will not be correct

to say that one sees a bluish-white wall in the same sense of 'seeing.‘

 

1 $.1th .i'..-'IK 61-. -'~ 3: is? In {Exiting . A
z .,

When the white wall looks bluish white to a person, he is seeing some-
thing directly which he is not seeing when he is seeing the surface of
the same wall which is white. But in any such case a person also knows
that the object that he is directly seeing is definitely related with
the physical surface that he is seeing. Moore writes,

If I am ppp_directly seeing a bluish-white expanse which

pap some such relation to a wall which is pgp bluish white,

how can I possibly know that that wall l§_looking bluish

white to me? It seems . . . quite plain that I cannot 'see'

in the common sense any physical object whatever without

its 'looking' somehow to me, and, therefore, without my

directly seeing some entity which has R to the object I am

said to see. . . . And . . . that entity i§_a visual sense

datum.27

Moore's final observation on this issue is that the questions

regarding the relation of a physical surface to its corresponding sense
datum can be answered definitively only if we know the correct analysis
of the perceptual statements, which we do not know so far (although we
do know with certainty that such statements are true, and also know

with certainty that the sense data which we directly see, in any

178

instance of ordinary cases of visual perception, are related to the
corresponding physical objects, or their surfaces).

It may be said that, at least so far as the relation between
a physical surface and a sense datum is concerned, there are some
elements of doubt in Moore's views; but the element of doubt lies
ppp_in not being sure as to whether such statements are kpppp_to be
true, but in not being sure as to whether the analyses of such state-
ments given so far are correct. While replying to some of his critics,
Moore has made some positive comments on his own on this issue:

Mr. Bouwsma goes on to say . . . that my doubt . . . cannot
be resolved; that there is pp way of settling the question
whether the directly seen object . . . is or is not identical
with that part of the surface of my hand which I am seeing.

. . This question, of which Mr. Bouwsma asserts so dogmat-
ically that there is no way of settling it, that “there is
nothing to do but to go on doubting," is the very same one
about which Mr. Marhenke asserts that he "is sure no philos-
opher will ever find the answer to it until we know what a
correct analysis is." Mr. Marhenke, then, thinks that it is
possible that it should be settled; and as to this I think he
’is clearly right as against Mr. Bouwsma. l, of course, do
not know how this particular philosophic question is to be
settled. . . . But that ways of settling this and other
philosophic questions will not some day be discovered, I
certainly do not know; and Mr. Bouwsma certainly does not
know it either. There is certainly something else to do
besides going on doubting; and that is to go on thinking
about it.2

 

Later again while talking about the role of analysis relative to the
two languages, viz., sense data-language and common sense-language,
in the context of commenting on Ayer's criticism that the question
regarding the relation of sense data to physical objects cannot be

settled by the analysis of any standard language, Moore writes:

 

179

It is in this last assertion of his, that the correct
answer to my question cannot be discovered by the analysis
of any standard usage, that he seems to me to have gone
h0pelessly wrong. His only reason for saying so seems to
be that there is no standard usage either of the philo-
sophical term "sense datum," or of the terms ”directly
apprehend" or "directly see"; and he is perfectly right
that there is no standard usage of these terms. What he
has failed to see is that my question can be put without
any use of these expressions . . . and that the answer to
it does depend upon the analysis of expressions which
undoubtedly have a standard usage in ordinary life. The
expressions I mean are those which consist in saying such
words as "This is a penny" or "That is a penny" together
with some standard gesture which seems to explain what
object we are referring to by the words "this" or "that";
and if we could only discover the right analysis of what
is meant by such expressions, my question would be answered.
It is by "going on thinking" about the analysis of such
expressions as these that I hope my question will some day
be answered--perhaps, has already been answered in some
work which I have not read.29

Thus, so far as our knowledge of the ontological relation
between sense data and physical surfaces is concerned, and our knowledge
of the ontological status of the sense data is concerned, it may perhaps
be said that there is some element of skepticism, but only in a very
qualified sense. That is, so far as the ontological relation between
sense data and physical object is concerned, we do know that there 1;

a definite relation between the two; what we are not certain about §p_

 

fap, is the exact nature of the relation between the two. If there is
any element of skepticism then it is there only in the sense in which
Moore has made this acknowledgment, and only to the extent to which it
is implied by this acknowledgment. But if this acknowledgment of
Moore's is at all indicative of an element of skepticism, it is very

weakly indicative of skepticism. This is so for this reason: from

 

180

this acknowledgment it does not follow (a) that we do not know the
nature of this relationship at all; (b) that we cannot ever possibly
know the nature of the relationship. What follows at the most from
this acknowledgment is only this: that so far we have not arrived
at a complete and definite knowledge of the epapp,nature of this P?
relationship (since we have not so far arrived at a complete analysis ‘-
of the perceptual statements of common sense). It is quite possible

that the discovery of a single intermediary step, not so far arrived

at in the process of analysis, would make it complete and definite.

 

And Moore has never claimed either that the analysis so far arrived
at is final or that it is the only possible analysis. Rather, he
has shown a strong optimism towards the possibility of finding out
a correct analysis of the same.

Moore has always made a distinction between knowing a
proposition to be true and knowing the correct analysis of a
proposition; i.e., on his view it is quite possible that one knows
a particular proposition to be true and yet may not know the correct
analysis of the same proposition. In his entire theory of empirical
knowledge Moore has repeatedly emphasized the point that we do know
for certain that material objects (animate or inanimate) exist, and
we also know for certain that sense data exist; and we further know
that sense data are related to material objects. His position is thus
clearly in opposition with that of the skeptics.

The problem in Moore's case arises only insofar as the analyses

of such propositions and the terms contained in them are concerned; and

 

181

even regarding the analyses of these propositions he has never said
either that they papppp_be analyzed or we do not know their analyses
appall, all that he has said that we do not know their complete
analyses §p_fap, But the point is that even if we cannot know the
correct analysis of such statements, this fact would not, by any means,
affect his views regarding the knowledge of the truth of perceptual
statements asserting the existence of material objects. In other
words, so far as the question as to whether or not we know such
propositions to be true, Moore has given a positive answer and has
justified his claim. His views about the unavailability of a correct
analysis of such propositions at the moment does not undermine his
claims regarding the knowability of such propositions. The point he
has wanted to make throughout his theory of empirical knowledge is that
such propositions are and can be known to be true, as opposed to his
opponents' claim that such propositions cannot be known to be true.

The final stand of Moore's makes it quite clear that it would
be incorrect to attribute to his views any of the following character-
izations (as some of his critics seem to have suggested), namely, (a)
that Moore's philosophy of common sense and his philosophy of sense data
are in conflict with each other, or, (b) that Moore himself has not
ultimately been able to avoid skepticism in his own views regarding
empirical knowledge, or, (c) that his views regarding common sense and
sense data constitute two completely different and unrelated philos-
ophies running parallel as two separate alternative ways of dealing

with the problem of empirical knowledge.

 

FOOTNOTES

Chapter II

 

1G. E. Moore, The Refutation of Idealism, in Philosophical [I
Studies (Totowa, N.J.: Littlefield, Adams & Co., 1968), p. 2l.

 

 

2Ibid., p. 22.
3Ibid., p. 25.

“I. Jun-1.“ .. .-

“Ibid., p. 25.

 

5A. R. White, G. E. Moore, A Critical Exposition (Oxford:
Basil Blackwell, T969), p. ll.

61bid., p. 12.

7G. E. Moore, Is Time Real? in Some Main Problems of Philosophy
(New York: Collier Book, T966), pp. 222-223.

 

 

8G. E. Moore, Hume's Philosophy, in Philosephical Studies
(Totowa, N.J.: Littlefield, Adams & Co., T968), p. T64.

 

96. E. Moore, A Defence of Common Sense, in Philosophical
Papers (New York: Collier Books, T966), p. 44.

 

1°G. E. Moore, Material Things, in Some Main Problems of
Philosophy (New York: Collier Books, 1966), pp. l42-l43.

 

 

11Moore, Disjunctive and Other Properties, in Some Main Problems
of Philosophy (New York: Collier Books, T966), p. 372.

 

'2Moore, What is Philosophy? in Some Main Problems of Philosophy
(New York: Collier Bpoks, T966), pp. l5-l6.

 

13Ibid., p. 41.

1"Moore, Material Things, p. T48.

 

15Ibid., p. 145.
16Ibid., p. 146.

182

183

”Ibid., p. 147.
18Ipici., p. 155.

19Moore, What is Philosophy? p. 20

 

20Moore, Material Things, p. 147.

216. E. Moore, Are Material Things Real? in Lectures on
Philosophy, ed. by C. Lewy (New York: Humanities Press Inc., 1966), F1.
p. 14; cf. G. E. Moore, Proof of an External World, in Philosophical
Papers (New York: Collier Books, 1966), p. 136.

 

 

Chapter III

 

 

1G. E. Moore, Some Judgments of Perception, in Philosophical
Studies (Totowa, N.J.: Littlefield, Adams & Co., 1968), p. 221.

 

2Ibid., p. 228.
3Ibid., p. 229.
I’Ibid., p. 232.

5G. E. Moore, A Defence of Common Sense, in Philosophical Papers
(New York: Collier Books, T966), p. 53.

 

6Ibid., p. 53.

7G. E. Moore, Sense-Data, in Some Main Problems of Philosophy
(New York: Collier Books, T966), p. 42.

8Ibid., p. 44.

9Ibid., p. 46.

1°Moore, A Defence of Common Sense, pp. 53-54.
11G. E. Moore, A Reply to My Critics, in The Philosophy of

G. E. Moore, ed. by P. A. Schilpp (New York: Tudor Publishing Company,
1952), p. 644.

 

12Ibid., p. 631.
13Ihid., p. 639.

1"G. E. Moore, Sense-Data, p. 62.

184

15G. E. Moore, Ways of Knowing, in Some Main Problems of
Philosophy (New York: Collier Books, 1966), p. 91.

 

16Moore, Sense-Data, p. 62.

 

17G. E. Moore, Propositions, in Some Main Problems of Philosophy
(New York: Collier Books, 1966), pp. 83-84.

 

18Moore, Sense-Data, p. 45.

 

196. E. Moore, The Status of Sense-Data, in Philosophical
Studies (Totowa, N.J.: Littlefield, Adams & Co., 1968), pp. 169-170.

 

 

20Ibid., p. 171.

21Moore, Sense-Data, pp. 53-54.

 

22Ibid., p. 57.
23Ibid., p. 57.
2"Ibid., p. 60.
25Ibid., p. 60.
26Ibid., p. 58.
27Ibid., p. 63.

286. E. Moore, The Nature and Reality of the Objects of
Perception, in Philosophical Studies (Totowa, N.J.: Littlefield,
Adams & Co., 1968), p. 73.

 

29Ibid., pp. 73-74.

3°Ib1°d., pp. 74-75.

31Moore, The Status of Sense Data, p. 181.
32Ibid., p. 184.

33Ibid., pp. 184-185.

3"Ibid., p. 181.

35Ibid., p. 181.

36Moore, Sense-Data, p. 58.

 

 

 

185

Chapter IV

 

1Moore, What Is Philosophy? p. 32.

 

2A. E. Murphy, Moore's "Defence of Common Sense," in lpe_
Philosophy of G. E. Moore, ed. by P. A. Schilpp (New York: Tudor
Publishing Company, 1952), p. 3T6.

3C. A. Mace, On How We Know Material Things Exist, in lpe_
Philosophy of G. E. Moore, ed. by P. A. Schilpp (New York: Tudor
Publishing Company, T952), pp. 295-298.

“R. McKeon, Propositions and Perceptions in the World of G. E.
Moore, in The Philosophy of G. E. Moore, ed. by P. A. Schilpp (New York:
Tudor Publishing Company, 1952), pp. 455, 467, 470-471, 479.

 

5Murphy, Moore's "Defence of Common Sense," pp. 316-317.

6Moore, Sense-Data, p. 42.

 

7G. E. Moore, Do We Know That Material Things Are Real? in
Lectures on Philosophy, ed. by C. Lewy (New York: Humanities Press
Inc., 1966), p. 49; also see G. E. Moore, Certainty, in Philoegphical
Papers (New York: Collier Books, 1966), pp. 223-246.

 

Chapter V

1G. E. Moore, Propositions, in Some Main Problems of Philosophy
(New York: Collier Books, 1966), p. 71.

 

2Ibid., p. 73.
3Ibid., p. 76.
I'Ibid., p. 84.
5Ibid., pp. 86-87.
6Ibid., p. 87.
7Ibid., p. 79.
8Ibid., p. 81.
9Ibid., p. 82.

 

186

1°Moore, A Reply_to My_Critics, pp. 634-635.

11Moore, Sense-Data, p. 62.

 

12Ipid., p. 62.

13Moore, Material Things, p. 146.

 

l"Moore, Sense-Data, p. 64.

 

15Moore, Prepositions, pp. 83-84.

 

16Moore, Sense-Data, p. 62.

 

17G. E. Moore, Ways of Knowing, in Some Main Problems of
Philosophy (New York: Collier Books, 1966), p. 92.

1albid., p. 96.

'96. E. Moore, Do We Know That Material Things Are Real? in
Lectures on Philosophy, ed. by C. Lewy (New York: Humanities Press
Inc., 1966), p. 49.

2°G. E. Moore, Certainty, in Philosophical Papers (New York:
Collier Books, 1966), p. 239.

21Ibid., pp. 223-224.

22Ibid., p. 231.

23Ipid., p. 235.

2"Ibid., p. 239ff; cf. G. E. Moore, Hume's Philosophy, in

Philosophical Studies (Totowa, N.J.: Littlefield, Adams & Co., 1968),
p. 161.

 

25Moore, Hume's Philosophy, p. 161.
26Moore, Do We Know That Material Things Are Real? p. 48.
27Moore, A Defence of Common Sense, p. 43.

28G. E. Moore, Hume's Theory Examined, in Some Main Problems of

 

Philosophy (New York: Collier Books, 1966), pp. 140-141.
29Ibid., p. 141.

30Moore, Sense-Data, pp. 41-42.

 

31Moore, Ways of Knowing, p. 99.

 

fi""'-' «3 a ’\ '1 J' V -1“ ‘~'

I"

.r’

v_
‘1’;. ‘

 

187

32Ibid.. pp. 100-101.

33Ibid.,

p. 100.

3“Ibid., p. 102.

35G. E. Moore, Hume's Theory, in Some Main Problems of

Philosophy (New York:

 

36Moore,
37Moore,
38Ibid.,
39Ibid.,
“01bid.s
“IIde.,
“ZIbId.,
“31bid.,
l"'Ib1'd.,
“SIde.,
“GIde.s
l"Ibidu.
“81bid.s
l"’Moore,
soMoore,
51Moore,
52Moore,
53Moore,
Sl'Moore,
55Ibid.,
561616.,

 

 

Sense-Data, p. 63.

 

Hume's Theory Examined, pp. 125-126. i

p. 136.

pp. 135-136.
p. 136.

p. 137.

p. 138.

p. 138.

p. 139f.

pp. 139-140.
pp. 142-143.
p.'l42.

p. 142.

Hume's Philosophy, p. 165.

Material Things, p. 157.

Hume's Philosophy, p. 156.

Material Things, p. 158.

Hume's Philosophy, p. 155.

Material Things, p. 159.
p. l60.
p. 160.

Collier Books, 1966), p.

117.

 

188

57Moore, A Defence of Common Sense, p. 40.

581bid., pp. 40-41.

59Moore, Do We Know That Material Things Are Real? p. 46.
6°Ibld., p. 48.

61G. E. Moore, Proof of an External World, in Phil050phical
Papers (New York: Collier Books, 1966), p. 142.

 

62Ibid., p. 144.

63Ibid., p. 145.

6"Moore, Do We Know That Material Things Are Real? p. 49.
65Moore, Certainty, p. 239.

66Ib‘id., p. 239.

67Ibid., p. 242.

68Ibid., p. 243.

69Ibid., p. 244.

70Moore, Do We Know That Material Things Are Real? pp. 49-50.
71Ib‘id., p. 52.

Chapter VI

 

1Moore, What Is Philoepphy? p. 38.

2G. E. Moore, Other Philosophical Questions, in Lectures on
Philosophy, ed. by C. Lewy (New York: Humanities Press Inc., 1966),
pp. 180, 190.

 

3G. E. Moore, The Justification of Analysis, in Lectures on

Philosophy, ed. by C. Lewy (New York: Humanities Press Inc., 1966),
p. 166.

 

l'G. E. Moore, Principia Ethica (Cambridge: University Press,
1960). p. 7.

5Moore, A Defence of Common Sense, p. 53.

 

of Sense-Data, p. 189.

189

6Moore, The Justification of Analysis, pp. 168-169.
7Moore, A Defence of Common Sense, p. 52.

8Ibid., p. 37.

9Ibid., p. 55.

10Moore, A Reply to My Critics, p. 629.

11Moore, A Defence of Common Sense, p. 55.

m1
1.
a:

12Ibid., pp. 56-57.
13Ibid., p. 57.

1"Moore, Some Judgments of Perception, p. 224; cf. The Status

 

 

15Moore, The Status of Sense-Data, p. 191.

 

16Ibid., pp. 182-183.

17Moore, The Nature and Reality of Objects of Perception, p. 93.
18Moore, A Reply to My Critics, p. 658.

19Ibid., p. 642.

2°Ibid., pp. 636-637.

21Ibid., p. 659.

226. E. Moore, Visual Sense Data, in British Philosophy in the

Mid-Century, ed. by C. A. Mace (New York: The Macmillan Company,

1957),

Moore,
p. 629.

p. 210.

23Ibid., p. 206.

zl'Ib'id., p. 210.

25Moore, A Reply_to My Critics, pp. 644-645.
26Moore, Visual Sense-Data, p. 208.

27Ibid., pp. 208-209.

28Moore, A Reply to My Critics, pp. 637-638.

296. E. Moore, Addendum to My Reply, in The Philosophy_pf G. E.
ed. by P. A. Schilpp TNeerork: Tudor Publishing Company, 1952),

 

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Klemke, E. D. The Epistemology of G. E. Moore. Evanston, 111.:

Malcolm,

Moore, G.

Rollins,

Schilpp,

White,A.

'58 (1949).

Northwestern University Press,ii969.

Norman. "Certainty and Empirical Statements." Mind 51 i
1942).

"Defending Common Sense." The Philosophical Review

E. "Are Materials of Sense Affections of the Mind?‘I
Proceedings of the Aristotelian Society 17 (1916-17).

 

"Indirect Knowledge." Proceedings of Aristotelian Society,
Supplement, 9 (1929).

Lectures on Philosophy. Edited by C. Lewy. New York:
Humanities Press Inc., 1966.

Philosophical Papers. New York: Collier Books, 1966.

Philosophical Studies. Totowa, N.J.: Littlefield, Adams
& Co., 1968.

Principia Ethica. Cambridge, Mass.: Harvard University

.Press, 1960.

Some Main Problems of Philosophy. New York: Collier Books,

.1966.

Visual Sense Data. British Phi1050phy in the Mid-Century,

.Edited by C. A. Mace. New York: The Macmillan Company, 1957.

C. D. ”Certainty." The Encyclopedia of Philosophy. Edited by
P. Edwards, Vols. 1 and 2. New York: Macmillan Publishing Co.,
Inc. and The Free Press, 1972.

P. A., ed. The Philosophy of G. E. Moore. New York: Tudor
Publishing Company, 1952.

R. G. E. Moore--A Critical Exposition. Oxford: Basil
Blackwell, T969.

190