SENSE-DATA 29

SENSE-DATA
by means of his senses. The evidence of the senses is, therefore, the
G. E. Moore evidence upon which all our other ways of knowing material objects
seems to be based.
Moore, G. E. (1953) Sense-data. In his Some Main Problems of And what I want first to consider is what sort of a thing this
Philosophy (London: George Allen & Unwin, Ch. II, pp. 28-40). evidence of the senses is; or in other words what it is that happens
Pagination here follows that reference. Also reprinted in Thomas when (as we should say) we see, or feel, a material object, or perceive
Baldwin (ed): G. E. Moore Selected Writings (London: Routledge,
1993), pp. 45-58. one by any other sense. And I propose to take as an instance, for the
sake of simplicity, a single sense only—namely, the sense of sight: I
I have said that I shall now begin discussing the various ways in shall use what happens when we see, as an illustration of what happens
which we know of the existence of material objects—supposing that we in sense-perception generally. All the general principles which I point
do know of their existence. I do not want to assume, to begin with, that out with regard to the sense of seeing, will, I think, be easily
we certainly do know that they exist. I only want to consider what sort transferable, mutatis mutandis, to all the other senses by which we can
of a thing our knowledge of them is, supposing that it is really be said to perceive material objects.
knowledge. I shall afterwards consider whether it is really knowledge. My first question is, then: What exactly is it that happens, when (as
And I said I should begin with the most primitive sort of way in we should say) we see a material object ? And I should explain,
which we commonly suppose that we have knowledge of them— perhaps, to avoid misunderstanding, that the occurrence which I mean
namely, that kind of knowledge, which we should call knowledge by here to analyse is merely the mental occurrence—the act of
means of the senses—the knowledge which we have, for instance, by consciousness—which we call seeing. I do not mean to say anything at
seeing and feeling, as when we feel an object over with our hands. This all about the bodily processes which occur in the eye and the optic
way of knowing material objects, by means of the senses, is, of course, nerves and the brain. I have no doubt, myself, that these bodily pro-
by no means the only way in which we commonly suppose we know of cesses do occur, when we see; and that physiologists really do know a
their existence. For instance, each of us knows of the past existence of great deal about them. But all that I shall mean by 'seeing', and all that I
many material objects by means of memory; we remember the wish to talk about, is the mental occurrence—the act of consciousness
existence of objects which we are no longer perceiving by any of our —which occurs (as is supposed) as a consequence of or
senses. We know of others again, which we ourselves have never accompaniment of these bodily processes. This mental occurrence,
perceived by our senses and cannot therefore remember, by the which I call 'seeing', is known to us in a much more simple and direct
testimony of other persons who have perceived them by their senses. way, than are the complicated physiological processes which go on in
And we know also, we suppose, by means of inference, of others which our eyes and nerves and brains. A man cannot directly observe the
nobody has ever perceived by his senses: we know, for instance, in this minute processes which go on in his own eyes and nerves and brain
way that there is another surface of the moon, different from that which when he sees; but all of us who are not blind can directly observe this
is constantly turned to the earth. All these other ways of knowing mental occurrence, which we mean by seeing. And it is solely with
material objects, I shall have presently to consider, and to contrast them seeing, in this sense—seeing, as an act of consciousness which we can
with sense-perception. But all these other ways do seem, in a sense, to all of us directly observe as happening in our own minds—that I am
be based upon sense-perception, so that it is, in a sense, the most now concerned.
primitive way of knowing material objects: it seems, in fact, to be true, And I wish to illustrate what I have to say about seeing by a direct
that if I had not known of some material objects by means of sense- practical example; because, though I dare say many of you are perfectly
perception, I could never possibly have known of any others in any of familiar with the sort of points I wish to raise, it is, I think, very
these other ways; and this seems to be true universally: no man could important for every one, in these subjects, to consider carefully single
ever know of the existence of any material objects at all, unless he first concrete instances, so that there may be no mistake as to exactly what it
knew of some is that is being talked about. Such mistakes are, I think, very apt to
happen, if one talks merely in generalities;
 30 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 31

and moreover one is apt to overlook important points. I propose, what we mean by a 'sensation' in this phrase, is my seeing of the colour,
therefore, to hold up an envelope in my hand, and to ask you all to look not the colour which I saw: this colour does not seem to be what I mean
at it for a moment; and then to consider with me exactly what it is that to say that I had, when I say I had a sensation of colour. It is very
happens, when you see it: what this occurrence, which we call the unnatural to say that I had the colour, that I had that particular whitish
seeing of it, is. grey or that I had the patch which was of that colour. What I certainly
I hold up this envelope, then: I look at it, and I hope you all will did have is the experience which consisted in my seeing the colour and
look at it. And now I put it down again. Now what has happened? We the patch. And when, therefore, we talk of having sensations, I think
should certainly say (if you have looked at it) that we all saw that what we mean by 'sensations' is the experiences which consist in
envelope, that we all saw it, the same envelope: / saw it, and you all apprehending certain sense-data, not these sense-data themselves. I
8aw it. We all saw the same object. And by the it, which we all saw, we think, then, that the term 'sensation' is liable to be misleading, because
mean an object, which, at any one of the moments when we were it may be used in two different senses, which it is very important to
looking at it, occupied just one of the many places that constitute the distinguish from one another. It may be used either for the colour
whole of space. Even during the short time in which we were locddng which I saw or for the experience which consisted in my seeing it. And
at it, it may have moved—occupied successively several different it is, I think very important, for several reasons, to distinguish these two
places; for the earth, we believe, is constantly going round on its axis, things. I will mention only two of these reasons. In the first place, it is,
and carrying with it all the objects on its surface, so that, even while we I think, quite conceivable (I do not say it is actually true) but
looked at the envelope, it probably moved and changed its position in conceivable that the patch of colour which I saw may have continued to
space, though we did not see it move. But at any one moment, we exist after I saw it: whereas, of course, when I ceased to see it, my
should say, this it, the envelope, which we say we all saw, was at some seeing of it ceased to exist. I will illustrate what I mean, by holding up
one definite place in space. the envelope again, and looking at it. I look at it, and I again see a
But now, what happened to each of us, when we saw that envelope? sense-datum, a patch of a whitish colour. But now I immediately turn
I will begin by describing part of what happened to me. I saw a patch1 away my eyes, and I no longer see that sense-datum: my seeing of it
of a particular whitish colour, having a certain size, and a certain shape, has ceased to exist. But I am by no means sure that the sense-datum—
a shape with rather sharp angles or corners and bounded by fairly that very same patch of whitish colour which I saw—is not still
straight lines. These things: this patch of a whitish colour, and its size existing and still there. I do not say, for certain, that it is: I think very
and shape I did actually see. And I propose to call these things, the likely it is not. But I have a strong inclination to believe that it is. And
colour and size and shape, sense-data* things given or presented by it seems to me at least conceivable that it should be still existing,
the senses—given, in this case, by my sense of sight. Many whereas my seeing of it certainly has ceased to exist. This is one reason
philosophers have called these things which I call sense-data, for distinguishing between the sense-data which I see, and my seeing of
sensations. They would say, for instance, that that particular patch of them. And here is another. It seems to me conceivable—here again I do
colour was a censation. But it seems to me that this term 'sensation' is not say it is true but conceivable—that some sense-data—this whitish
liable to be misleading. We should certainly say that I had a sensation, colour for instance—are in the place in which the material object—the
when I saw that colour. But when we say that I had a sensation, what envelope, is. It seems to me conceivable that this whitish colour is
we mean is, I think, that I had the experience which consisted in my really on the surface of the material envelope. Whereas it does not seem
seeing the colour. That is to say, to me that my seeing of it is in that place. My seeing of it is in another
'I am so extending the use of the word 'patch' that, e.g., the very small black dot place—somewhere within my body. Here, then, are two reasons for
which I directly apprehend when I see a full-stop, or the small black line which I directly distinguishing between the sense-data which I see, and my seeing of
apprehend when I see a hyphen, are, each of them, in the sense in which I am using the them. And it seems to me that both of these two very different things
word, a 'patch of colour'. (1952). are often meant when people talk about 'sensations'. In fact, when you
'I should now make, and have for many years made, a sharp distinction between what are
I have called the 'patch', on the one hand, and the colour, size and shape, of which it is, on
the other; and should call, and have called, only the patch, not its colour, size or shape, a
'sense-datum'. (1952).
 32 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 33

reading any philosopher who is talking about sensations (or about distance from the paper. And so too, with regard to the size of the patch
sense-impressions or ideas either), you need to look very carefully to of colour which you saw: differences in the strength of your eyes and in
see which of the two he is talking about in any particular passage— your distance from the envelope probably made slight differences in the
whether of the sense-data themselves or of our apprehension of them: size of the patch of colour, which you saw. And so again with regard to
you will, I think, almost invariably find that he is talking now of the the shape. Those of you on that side of the room will have seen a
one and now of the other, and very often that he is assuming that what rhomboidal figure, white those in front of me will have seen a figure
is true of the one must also be true of the other—an assumption which more nearly rectangular. Those on my left will have seen a figure more
does not seem to be at all justified. I think, therefore, that the term like this which you in front now see, and which you see is different
'sensation' is liable to be very misleading. And I shall, therefore, never from this which you then saw. And those in front of me will have seen
use it. I shall always talk of sense-data, when what I mean is such a figure like that which you on the left now see, and which, you see, is
things as this colour and size and shape or the patch which is of this different from this, which you saw before. Those directly in front of
colour and size and shape, which I actually see. And when I want to me, may, indeed, have all seen very nearly the same figure—perhaps,
talk of my seeing of them, I shall expressly call this the seeing of sense- even, exactly the same. But we should not say we knew that any two
data; or, if I want a term which will apply equally to all the senses, I did; whereas we should say we did know that we all saw the same
shall speak of the direct apprehension of sense-data. Thus when I see envelope. That you did all see the same envelope, would, indeed, be
this whitish colour, I am directly apprehending this whitish colour: my accepted in ordinary life as a certainty of the strongest kind. Had you
seeing of it, as a mental act, an act of consciousness, just consists in my all seen me commit a murder, as clearly as you all saw this envelope,
direct apprehension of it;—so too when I hear a sound, I directly your evidence would be accepted by any jury as sufficient to hang me.
apprehend the sound; when I feel a tooth-ache I directly apprehend the Such evidence would be accepted in any court of law as quite
ache: and all these things—the whitish colour, the sound and the ache conclusive; we should take such a responsibility as that of hanging a
are sense-data. man, upon it. It would be accepted, that is, that you had all seen me, the
To return, then, to what happened to us, when we all saw the same same man, commit a murder; and not merely that you had all seen some
envelope. Part, at least, of what happened to me, I can now express by man or other, possibly each of you a different man in each case,
saying that I saw certain sense-data: I saw a whitish patch of colour, of commit one. And yet, in this case, as in the case of the envelope, the
a particular size and shape. And I have no doubt whatever that this is sense-data which you had all seen, would have been different sense-
part, at least, of what happened to all of you. You also saw certain data: you could not swear in a court of law that you had all seen exactly
sense-data; and I expect also that the sense-data which you saw were the same sense-data.
more or less similar to those which I saw. You also saw a patch of Now all this seems to me to shew very clearly, that, if we did all see
colour which might be described as whitish, of a size not very different the same envelope, the envelope which we saw was not identical with
from the size of the patch which I saw, and of a shape similar at least in the sense-data which we saw: the envelope cannot be exactly the same
this that it had rather sharp corners and was bounded by fairly straight thing as each of the sets of sense-data, which we each of us saw; for
lines. But now, what I want to emphasize is this. Though we all did (as these were in all probability each of them slightly different from all the
we should say) see the same envelope, no two of us, in all probability, rest, and they cannot, therefore, all be exactly the same thing as the
saw exactly the same sense-data. Each of us, in all probability, saw, to envelope.
begin with, a slightly different shade of colour. AH these colours may But it might be said: Of course, when we say that we all saw the
have been whitish; but each was probably at least slightly different envelope, we do not mean that we all saw the whole of it. I, for
from all the rest, according to the way in which the light fell upon the instance, only saw this side of it, whereas all of you only saw ihat side.
paper, relatively to the different positions you are sitting in; and again And generally, when we talk of seeing an object we only mean seeing
according to differences in the strength of your eye-sight, or your some part of it. There is always more in any object which we see, than
the part of it which we see.
 34 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 35

And this, I think, is quite true. Whenever we talk roughly of seeing any colours are parts of the envelope, then we must suppose that all of them
object, it is true that, in another and stricter sense of the word see, we are in the same place. We must suppose that ever so many different
only see a part of it. And it might, therefore, be suggested that why we colours all of them occupy the same surface—this surface of the
say we all saw this envelope, when we each, in fact, saw a different set envelope which you now see. And I think it is certainly difficult to
of sense-data, is because each of these sets of sense-data is, in fact, suppose this, though not absolutely impossible. It is not absolutely
apart of the envelope. impossible, I think, that all the different colours which you see are
But it-seems to me there is a great difficulty even in maintaining really all of them in the same place. But I myself find it difficult to
that the different sense-data we all saw are parts of the envelope. What believe that this is so; and you can understand, I think, why most
do we mean by a part of a material object ? We mean, I think, at least philosophers should have declared it to be impossible. They have
this. What we call a part of a material object must be something which declared, chiefly, I think, on grounds like this, that none of the colours
occupies a part of the volume in space occupied by the whole object. which any of us ever see are ever part3 of material objects: they have
For instance, this envelope occupies a certain volume in space: that is declared that none of them are ever in any part of the places where
to say, it occupies a space which has breadth and thickness as well as material objects (if there are any material objects) are. This conclusion
length. And anything which is a part of the envelope at any moment, does, indeed, go beyond what the premisses justify, even if we accept
must'be in some part of the volume of space occupied by the whole the premiss that several different colours cannot all be in exactly the
envelope at that moment: it must be somewhere within that volume, or same place. For it remains possible that the colour, which some one of
at some point in the surfaces bounding that volume. you sees, is really on the surface of the envelope; whereas the colours
Are, then, any of the sense-data we saw parts of the envelope in this which all the rest of you see are not there. But if so, then we must say
sense ? that though all of you are seeing the same side of the envelope, yet only
The sense-data I mentioned were these three—the colour—the one of you is seeing a sense-datum which is a part of that side: the
whitish colour; the size of this colour; its shape.1 And of these three it is sense-data seen by all the rest are not parts of the envelope. And this
only the colour, which could, in the sense defined, possibly be also, I think, is difficult to believe. It might be, indeed, that those of you
supposed to be apart of the envelope. The colour might be supposed to who are seeing a colour, which is not a part of the envelope, might yet
occupy apart of the volume occupied by the envelope—one of its be seeing a size and a shape which really is the size and shape of one
bounding surfaces,' for instance. But the size and shape could hardly be side of the envelope; and we will go on to consider whether this is so.
said to occupy any part of this volume. What might be true of them is And, first, as to the size. I assumed that the sense-given sizes, which
that the size I saw is the size of one surface of the envelope; and that you see, are all of them probably slightly different from one another.
the shape is the shape of this surface of the envelope. The side of the And, if this be so, then certainly it seems to be absolutely impossible
envelope which I say I saw certainly has some size and some shape; that they should all of them be the size of this side of the envelope.
and the sense-data—the size and shape, which I saw as the size and This side of the envelope can only really have one size; it cannot have
shape of a patch of colour—might possibly be the size and shape of this several different sizes. But it may not seem quite clear, that you all do
side of the envelope. see different sizes; the differences between the different distances at
Let us consider whether these things are so. which you are from the envelope are not so great, but what the patches
And, first, as to the colours. Can these possibly be parts of the of colour you all see might be, at least, of much the same size. So I will
envelope ? What we supposed is that each of you probably saw a give a hypothetical instance to make my point clearer. Suppose this
slightly different colour. And if we are to suppose that all those room were so large that I could carry the envelope two or three hundred
'I had here forgotten that one of the sense-data mentioned was the patch which has yards away from you. The sense-given size which you would then see,
that colour and shape and size—the patch which, I should now aay, is the only 'sense- when I was three hundred yards off, would certainly be appreciably
datum', having to do with the envelope, which I then saw. (1952). smaller than what you see now. And yet you would still be seeing this
*I should now say that any part of the surface of a volume is not a part of the same envelope. It seems
volume, because it is not itself a volume. (1952).
 36 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 37

quite impossible that these two very different sizes should both of them the shape of the big square is the same in quality—qualitatively
be the size of the envelope. So that here the only possibility is that the identical—with that of the small square, but is not numerically the
size which you see at some one definite distance or set of distances, same—not numerically identical: the shape of the big square is
should be the envelope's real size, if you ever see its real size at all. numerically different from that of the small, in the sense that they are
This may be so: it may be that some one of the sense-given sizes which two shapes, and not one only, of which we are talking, though both are
we see is the envelope's real size. But it seems also possible that none the same in quality: both are squares, but the one is one square and the
of them are; and in any case we all see the envelope, just the same, other is another square. There is, then, a difference between two
whether we see its real size or not. different kinds of identity: qualitative identity and numerical identity;
And now for the shape. Here again it seems quite impossible that and we are all perfectly familiar with the difference between the two,
all the shapes we see can be the envelope's real shape. This side of the though the names may sound strange. I shall in future use these names:
envelope can have but one shape: it cannot be both rhomboidal, as is qualitative identity and numerical identity. And now to return to the
the shape which you on the left see, and also rectangular, as is the case of the envelope. Even supposing that the sense-given shape which
shape seen by those in front; the angles at its corners cannot be both you in front see is rectangular, and that the real shape of the envelope is
right angles and also very far from right angles. Certainly, therefore, also rectangular, and that both are rectangles of exactly the same shape;
the sense-given shape which some of you see is not the shape of this it still does not follow that the sense-given shape which you see is the
side of the envelope. But here it may be said, it is plain enough that one shape of the envelope. The sense-given shape and the shape of the
of the sense-given shapes seen is its real shape. You may say: The envelope, even if they are qualitatively the same, must still be two
shape seen by those in front is its real shape; the envelope is different shapes, numerically different, unless they are of the same size;
rectangular. And I quite admit that this is so: I think we do know, in just as the shape of a large square must be numerically different from
fact, that the envelope really is roughly rectangular. But here I want to the shape of a smaller one. And we saw before how difficult it was to
introduce a distinction. There are two different senses in which we may be sure that any of the sizes which you saw were the real size of the
talk of the shape of anything. A rectangle of the size of this envelope, envelope. And even if the sense-given size which some one of you sees
and a rectangle of the size of this blackboard, may both, in a sense, is the real size of the envelope, it still does not follow that the sense-
have exactly the same shape. They may have the same shape in the given shape which you see is numerically the same as the shape of the
sense, that all the angles of both are right angles, and that the envelope. The two may be numerically different, just as in the case of
proportions between the sides of the one, and those between the sides two different squares, side by side, of the same shape and size, the
of the other, are the same. They may, in fact, have the same shape, in shape of the one is not the shape of the other; they are two numerically
the sense in which a big square always has the same shape as a small different shapes. We may say, then, that if those of you who see
square, however big the one may be and however small the other. But rectangular shapes, do see rectangular shapes of different sizes, only
there is another sense in which the shape of a big square is obviously one of these can possibly be the shape of the envelope: all the others
not the same as that of a small square. We may mean by the shape of a may be of the same shape—the same in quality—but they cannot be the
big square the actual lines bounding it; and if we mean this, the shape shape of the envelope. And even if some one of you does see a shape,
of a big square cannot possibly be the same as the shape of a smaller which is of the same size as the shape of the envelope, as well as being
one. The lines bounding the two cannot possibly be the same lines. And of the same shape (and it is very doubtful whether any of you does) it
the same thing may be true, even when there is no difference in size would yet be by no means certain that this sense-given shape which
between two shapes. Imagine two squares, of the same size, side by you saw was the shape of the envelope. It might be a shape numerically
side. The lines bounding the one are not the same lines as those different from the shape of the envelope, although exactly similar both
bounding the other: though each is both of the same shape and of the in shape and size. And finally there is some reason to suppose that none
same size as the other. The difference between these two senses in of the sense-given shapes which any of you see are exactly the same,
which we may talk of the shape of anything, may be expressed by even in quality, as the shape of the envelope.
saying that
 38 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 39

The envelope itself probably has a more or less irregular edge; there are it. This area was also a sense-datum. And in this area we can dis-
probably ups and downs in the line bounding its side, which you at that tinguish parts—this part, and this part, and this. And it might be
distance cannot see. thought with regard to parts, at least, of this area, that two things are
Of the three kinds of sense-data,1 then, which you all of you saw, true. Firstly, that part at least of the sense-given area which each of you
when I held up the envelope, namely, the whitish colour, its size, and saw, is really numerically identical with some part of that seen by all
its shape, the following things seem to be true. First, as regards the the rest. And secondly, that this part, which you all saw, is also a part of
colour, no one of you can be sure that the exact colour which you saw the area occupied by the real envelope. In other words, you might
was really a part of the envelope—was really in any part of the space, comfort yourselves by supposing, that even if the colour presented by
which the real envelope (if there was a real envelope) occupied. Then your senses is not a part of the real envelope, and even if the shape and
as regards the size, no one of you can be sure that the size which you size presented by your senses are not the shape and size of the real
saw was the real size of the envelope. And finally as regards the shape, envelope, yet at least there is presented by your senses a part of the
no one of you can be sure that the shape which you saw was really of space occupied by the real envelope. And against this supposition I
exactly the same shape as that of the envelope; still less can you be sure confess I cannot find any argument, which seems to me very strong.
that it was the shape of the envelope, that the bounding lines which We are all, I think, very strongly tempted to suppose that this is so.
composed it were numerically the same bounding lines as those which That, for instance, this space which I touch is really seen by all of you
enclosed the envelope. And not only can none of you be sure of these —this very same place—and that it also is part of the space which the
things. As regards the sizes and shapes which you saw, it seems quite real envelope occupies. The best argument I can think of against this
certain that some of you saw sizes and shapes which were not the real supposition is the following; and I think it is enough to render the
size and shape of the enve~ lope; because it seems quite certain that supposition doubtful. If we are to say that part of this sense-given area
some of you saw sizes and shapes different from those seen by others, which I see is really numerically the same with part of those which you
and that these different sizes and shapes cannot possibly all be the size see, and that it is also numerically the same as part of the area occupied
and shape of the envelope. And as regards the colours it seems fairly by the real envelope, then we must either again accept the hypothesis
certain, that the colours which you saw cannot all have been in the that all the different colours which we see as occupying the area are
envelope; since it seems fairly certain that you all saw slightly different really in the same place and in the same place as the real envelope, or
colours, and it is difficult to believe, though not absolutely impossible, else we must say that the colours only seem to be in this sense-given
that all these different colours were really in the same place at the same area and are not really there. But there is the former objection to
time. supposing that several different colours are all really in the same place.
This seems to be the state of things with regard to these sense-data And as to the only remaining possibility, namely, that they only seem to
—the colour, the size and the shape. They seem, in a sense, to have had be in this sense-given area; it may be objected that so far as the sense-
very little to do with the real envelope, if there was a real envelope. It given area is concerned, the colours we see really do occupy it—that
seems very probable that none of the colours seen was really a part of they not only seem to be but really are there—that there can be no
the envelope; and that none of the sizes and shapes seen were the size doubt about this. If we are talking of the area really presented by the
or the shape of the real envelope. senses as occupied by the colours, this area, it may be said, undoubt-
But now I wish to mention one other sense-datum, of a kind that we edly if occupied by the colours: it is nothing but the space over which
all saw, which might be thought to have more to do with the real the colour is spread. So that, if the area, which I see, really is
envelope. Besides the patch of colour and its shape and size, we did, in numerically the same as those which you see, then it will follow that all
a sense, all see the space which this patch of colour occupied. The the different colours we see really are in the same place. This argument,
patch of colour seemed to occupy a certain area; and we can by I say, does not seem to me to be absolutely conclusive. It does seem to
abstraction distinguish this area from the patch of colour occupying me possible that the colour I see only seems to be in the sense-given
area, which I see. But it is, I think, sufficient to
'The patch itself, which has that colour and shape and size, again forgotten (1952)-
 40 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 41

suggest a doubt whether any part of this sense-given area seen by me it is apprehending it. And they would say this not only with regard to
really is numerically the same as any part of any of those seen by you. sense-data like colours, sounds, hardness, smoothness, heat, cold,
Well now: Chiefly, I think, for reasons of the sort which I have aches, which seem to us to occupy space—to be localised. They would
given you, an overwhelming majority of philosophers have adopted the say it also with regards to the sense-given spaces which these things
following views. Reasons of the sort which I have given are not the seem to occupy. For instance, the sense-given area, occupied by this
only ones which have been alleged as reasons for holding these views, patch of colour: I see it now, and while I see it, it is: that particular area
but they are, I think, the ones which have really had most influence in is one among the contents of the Universe. But now that I turn my head
getting them adopted, and they are, it seems to me, by far the strongest away, it, that particular area I saw, has entirely ceased to exist. With my
reasons for adopting them. However that may be, whatever the reasons, seeing of it, it also has ceased to be. I may indeed be still seeing an area
an overwhelming majority of philosophers have, I think, adopted the exactly like it: this area for instance, which I now see, seems to be
following views; and I wish you to realise these views as clearly as exactly like, and only distinguishable by the fact that it is occupied by a
possible. different colour. But these two areas, they would say, though perhaps
They have held with regard to absolutely all the sense-data1 and exactly like, are not the same. They are no more the same than is this
every part of any sense-datum, which we ever apprehend by any of our part of the total area which I now see the same as that part. The
senses, the following things. particular sense-given area which I just now saw has entirely ceased to
They have held (i) that absolutely no part of the sense-data, which I be.
ever apprehend, exists at all except at the moment when I am This, then, is one view, which an overwhelming majority of philo-
apprehending it. They have held, that is to say, that except at the sophers have held with regard to sense-data. They have held that every
moment when I am apprehending it, there simply is not in the Universe sense-datum, of every kind, and every part of every sense-datum, is
any particular sense-datum which I ever apprehend. If, for instance, I something which only is or exists, so long as the person apprehending it
look at this envelope again and now turn away my eyes for a moment, is apprehending it.
then while I saw that particular patch of whitish colour, there was that (2) And they have held too this second view. Namely, that no two
particular patch of colour in the Universe: there certainly was, for I saw of us ever apprehend exactly the same sense-datum. They would allow
it. But now that I no longer see it, that particular patch of colour has that we might, perhaps, apprehend sense-data exactly alike; but they
ceased to exist. It no longer is in the Universe, any more than my would say that even though exactly alike—the same in quality—they
seeing of it is. They are both of them, both the colour and my seeing of cannot ever be numerically the same. That this is so with regard to
it, things which were, but which are no longer: both of them equally sense-data which exist at different times, would, indeed, follow from
and in the same se nse have completely ceased to be. These the first view. If this particular patch of colour which I see now, has
philosophers would not deny, indeed, that there may still be in the now, when I turn away my head, entirely ceased to be, it follows that
Universe a patch of colour exactly like that which I saw. For instance, nobody can be seeing it now. But it is worth while to emphasize that
some one else might at this moment be seeing a patch of colour exactly this is the view actually held by most philosophers. It is held, for
like it. But this other patch of colour, though exactly like, they would instance, that if somebody were to come and look at this envelope,
say, is certainly not the same: they may be exactly the same in quality, immediately after I had looked at it, standing at exactly the same
but they are not numerically the same. The patch of colour which I saw distance from it and in the same direction, having exactly the same
cannot be now existing even though another exactly like it may be. And power of eye-sight, and the light also not having changed at all, so that
they would say this with regard to absolutely all the sense-data, which he saw a patch of colour exactly similar to that which I had just seen;
any of us ever apprehends. Each of them only is, so long as the person nevertheless the patch of colour which he saw would not be the same as
apprehending that which I had just seen. It would be numerically different from it, in
'These three propositions about what philosophers have held are only true if the word the same sense, in which, supposing you see two spots of colour, of
'sense-datum' be understood in the sense explained in footnote 2 on p. 30, i.e. in such a exactly the same size and shape side by side, the one spot, though
sense that 'patches' are sense-data, but their colour, size and shape, are not. (1952). exactly like the other, is yet not the
 42 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 43

same, is numerically different from it. And it is held too, that no two every sense-datum that an/ person ever directly apprehends exists only
persons can see the same sense-datum, or any part of the same sense- so long as he apprehends it, (2) that no sense-datum which any one
datum, even at the same time: a point which does not follow from the person directly apprehends ever is directly apprehended by any other
last view. For though it might be true that all the sense-data, which any person, and (3) that no sense-datum that is directly apprehended by one
of you now sees in looking at this envelope ceased to exist the moment person can be in the same space with any sense-datum apprehended by
you ceased to see it; yet it might be true that, while you were seeing and any other person—that no sense-datum that is seen or heard or felt by
while, therefore, it exists, some other of you might be seeing at least a me can possibly be either in the same place with or at any distance
part of one of them too. But this is what is denied by this second view. from any that is seen or heard or felt by any one else. These three
It is denied that any two of you are at this moment seeing, even in part, things are, I think, the chief things that are meant, when it is said that
the same sense-data. It is asserted that every part of every sense-datum all sense-data exist only in the mind of the person who apprehends
which any one of you sees now, is numerically different from any part them; and it is certainly the common view in philosophy that all sense-
of any sense-datum seen by any other of you. data do only exist in our minds. I do not think myself that this is a good
And the third view, which is held by an overwhelming majority of way of expressing what is meant. Even if all these three things are true
philosophers about sense-data is this. of all the sense-data which I ever directly apprehend; it does not seem
They hold, namely (3) that none of the sense-data apprehended by to me to follow that they exist only in my mind, or indeed are in my
any one person can ever be situated either in the same place with, or at mind in any sense at all except that they are apprehended by me. They
any distance in any direction from, those apprehended by any other are, so far as I can see, not in my mind in the sense in which my
person. In other words, they hold that any sense-datum apprehended by apprehension of them is in my mind: for instance, this whitish colour,
me cannot possibly be in the same place as any sense-datum even if it does only exist while I see it, and cannot be seen by any one
apprehended by any one of you : and that this is true of any pair of else, does not seem to me to be in my mind in the sense in which my
persons you like to take. That is to say, this patch of colour seen by me seeing of it is in my mind. My seeing of it is, it seems to me, related to
is neither in the same place with, nor at any distance in any direction my mind in a way in which this which I see is not related to it: and I
from, any that is seen by any of you: the two simply have no spatial should prefer to confine the phrase 'in the mind' to those things which
relations of any kind to one another. With regard to the different sense- are related to my mind, in the way in which my seeing of this colour,
data seen by me at any one moment, they would indeed admit that these and my other acts of consciousness are related to it. But whether they
have, in a sense, spatial relations to one another. This corner of the could be properly said to be in my mind or not, certainly all the sense-
patch of colour which I see really is at a certain distance, in a certain data, which I ever directly apprehend, are, if these three things are true
direction, from this corner; and at another distance in another direction of them, dependent upon my mind in a most intimate sense. If it is
from this other corner. But they would say that all the different sense- really true of all of them that they exist only while I am conscious of
data within my field of vision at any one time have distance and them, that nobody else ever is directly conscious of them, and that they
direction from one another only within a private space of my own. That are situated only in a private space of my own, which also exists only
is to say, no point in this private space of mine is either identical with, while I am conscious of it, and of which no one else is ever directly
nor at any distance from, any point within the field of vision of any conscious—then certainly nothing could well be more thoroughly
other person. The sense-given field of vision of each of us, at any dependent on my mind than they are. Most philosophers have, I think,
moment, constitutes a private space of that person's own;—no two certainly held that all sense-data are dependent on our minds in this
points in any two of these spaces, can be related to one another in any sense. This has been held both by philosophers who believe that there
of the ways in which two points in any one of them are related. are material objects and that we know of their existence, and by those
These three views have, I think, been held by an overwhelming who believe that there are no such things as material objects, or, that, if
majority of philosophers. They have held, that is (1) that absolutely there are, we do not know it. It has, in fact, an overwhelming weight of
authority in its
 44 SOME MAIN PROBLES OF PHILOSOPHY SENSE-DATA 45

favour. And I am going to call it for the moment the accepted view. same space; and whether, when the space we are talking of is the sense-
And as regards the question whether this accepted view is true or given space presented with the colours, it can be true that these colours
not, I confess I cannot make up my mind. I think it may very likely be only seem to occupy this sense-given space, and do not really occupy
true. But I have never seen any arguments in its favour which seem to it. And no arguments of this kind seem to me to be perfectly
me to be absolutely conclusive. The strongest arguments in its favour, conclusive, though they do seem to me to have weight. And on the
as I said, seem to me to be arguments of the sort which I have given other side, in favour of the contrary view, there seems to me the fact
you. This one, for instance : That if we are to say that any portion of the that we all have a very strong tendency to believe it. I find it very
sense-given spaces apprehended by each of us at the same time, really difficult not to believe that when I look at this, and turn away my head,
is numerically the same portion of space, then we must hold either that the colour which I just saw is not still existing; that the space in which I
the very same portion of space may be occupied at the same time by saw it is not still existing too; and that the colour is not still in that
several different colours or that it only really is occupied by the colour space. And so too, I find it very difficult to believe that this space,
which one of us sees and only seems to be so by those which the rest of which I see—this very same portion of space —is not also seen by all
us see or that it only seems to be and is not really occupied by any of of you. I point at it; and what I point at seems to be a part of the sense-
the colours which any of us see. There do seem to me objections to given space which I see; and I cannot believe that by pointing at it I do
saying any of these three things; but, on the other hand, the objection to not make plain to you also, which portion of space I am pointing at. We
none of them seems to me perfectly conclusive: it seems to me possible all constantly assume that pointing at a thing is of some use; that if I
that any one of them might be the truth. One argument which has been point at a thing, that serves to show you which thing I am talking about;
urged by some philosophers as being conclusive seems to me to have that you will see the same thing, which I see, and will thus know what
absolutely no weight at all. It has been urged, namely, that we can see it is that I see. And it certainly seems as if the thing at which I am
directly, without the need of any argument, if we will but think of it, pointing now is part of the sense-given space which I see; and that,
that all sense-data are a sort of thing which can only exist while the therefore, if you see what I am pointing at, some portion of the sense-
person perceiving them is perceiving them: it is urged that this is a self- given space which each of us sees must be the same. But on the other
evident truth like the truth that 2+2=4. This argument seems to me to hand, I can imagine that I am mistaken about this. I can imagine that
have no weight at all. It seems to me that it is simply false that what it what I am sure that you see is not a part of my sense-given space; and
says is self-evident. I can perfectly well conceive that the very same that what you see, when you see the place I am pointing at, is not a part
sense-data, which I see at one time, should exist even when I am not of your sense-given space either: and that the supposition that some
seeing them: and I cannot, by merely considering the possibility, portion of our sense-given spaces must be identical, arises from our
determine whether it is true or not. And moreover, I think, that the confusion of sense-given space with the real space, which we do really
apparent strength of this argument has been largely due to the all of us see—but see in another sense. I can, therefore, not find any
confusion I spoke of above—the confusion between the sense-data arguments, either, which seem to me conclusive against the accepted
which I see and my seeing of them. Many philosophers have, as I said, view: the view that all the sense-data I see, including every portion of
not only called both of these two very different things 'sensations', but my sense-given space, are private sense-data of my own, which exist
have treated them as if they were the same thing. And, of course, when only while I directly apprehend them, and no part of which can be
I cease to see a given sense-datum, I do cease to see it: my seeing of it directly apprehended by any one of you. And what I wish to do in the
certainly does cease to exist. They have, then, argued, treating the rest of this lecture is this. I wish for the moment to suppose that this
sense-datum as if it were the same thing as my seeing of it, that the accepted view is true; to suppose that absolutely all the sense-data of
sense-datum ceases to exist too. But this is surely mere confusion. We each of us are private to that person, in the sense I have explained; and
are, then, I think, if we are to find conclusive arguments in favour of then to consider what, supposing this view is true, can be the nature of
this accepted view, thrown back upon such questions as whether many our knowledge of material objects by means of the senses, if we nave
different colours can all occupy the such knowledge at all.
 46 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 47

I return, then, to my original question: What happens, when we all are more apt to call 'sensations', such as that peculiar something we call
sec this envelope ? 'heat', and which we directly apprehend when we put our hands close to
I began, you may remember, by saying that a part of what happened a fire, and those peculiar somethings which we call hardness or
to me was that I saw certain sense-data—a particular whitish patch of smoothness, or the pain of a toothache. And in all these cases, so far as
colour, of a certain size and shape, and also the area which this patch of I can see, what I mean by 'direct apprehension', namely, the act of
colour did, or seemed to, occupy. This, the seeing of certain sense-data, consciousness, is exactly the same in quality: that is to say, the actual
was also a part at least of what happened to you. But now, having for seeing of a colour, considered as an act of consciousness, differs in no
the moment accepted the philosophical view that all the sense-data seen respect at all from the actual hearing of a sound, or the actual smelling
by any one of us are seen by that person alone, we have got this far of a smell. They differ only in respect of the fact, that whereas the one
further: namely, that, if we do in fact all see the same envelope, this is the direct apprehension of one kind of sense-datum, the other is the
seeing of the envelope cannot possibly consist merely in our seeing of direct apprehension of another kind: the one, for instance, of a colour,
those sense-data; this seeing of sense-data, which I declared at first to the other of a sound. And what they are is perhaps best realised by
be at least a part of what happens when we see the envelope, must, we considering the difference between what is happening when you are
now find, be a mere part of what happens; it cannot possibly be the directly apprehending a given sense-datum and what happens when you
whole, if we all do really see the same envelope: for we do not, cease to apprehend it. For instance, you look at this envelope, and you
according to the accepted philosophical view, see the same sense-data; actually see a particular colour: you directly apprehend that particular
the sense-data which we see are not, even as regards the least part, the colour. But, then, if you turn away your eyes, you no longer directly
same. It remains, |! j then, to enquire, what else beside the seeing apprehend it: you no longer actually see the colour which you saw. But
of sense-data, can have happened when we saw the envelope. But you may still be thinking of it—thinking of just that colour which you
before we go on to consider this I want to insist upon one point, with saw a moment ago: you may, therefore, in a sense still be conscious of
regard to this first part of what happened—namely this, which I have it, though you are no longer directly apprehending it. Here, therefore, is
called the seeing of certain sense-data. I said before, that, if I wanted to one way of having before the mind, which is not direct apprehension:
use a term which would apply not only to the sense of sight but to any the way which we call 'thinking of or remembering. That is to say, you
other sense, I should use the term 'direct apprehension' of sense-data. may still be thinking of the colour which you saw, and therefore having
And the point I wish now to insist on is what exactly this way of it before your mind in a sense, although you are no longer directly
perceiving1 things, which I call direct apprehension, is. It is certainly apprehending it. No doubt, when you think of it, you are still directly
one of the most important ways we have of perceiving things. And I apprehending something: you may, for instance, be directly
want, in future, to be able to refer to it by the name 'direct appre- apprehending an image of it—one of those faint copies of sense-data,
hension' and therefore I want you to realise as clearly as possible what which are called images. But you are no longer directly apprehending
sort of a thing this way of perceiving which I call 'direct apprehension' the coloured patch which you saw; the image which you are directly
is. It is, as I said, that which happens when you actually see any colour, apprehending, though it may be like, is not the same; and the relation
when you actually hear any sound, when you actually feel the so-called which you now have to the image is obviously different from that
'sensation' of heat, as when you put your hand close to a fire; when you which you have now to the sense-datum, which you saw but do not now
actually smell a smell; when you feel the so-called sensation of see; while this relation which you now have to the image, is the same as
hardness, in pressing against a table; or when you feel the pain of a that which you had to the sense-datum, just now when you actually saw
toothache, etc., etc. In all these cases you directly apprehend the sense- it. You directly apprehend the image now in exactly the same sense as
datum in question—the particular Colour, or sound, or smell: or those you just now directly apprehended the sense-datum, of which it is an
peculiar sense-data, which we image: but you are no longer now directly apprehending the sense-
•There is another, very different, use of 'perceive', in which we are said to perceive datum which you were directly apprehending a moment ago.
that so and so is the case, i.e. to perceive, not a 'thing', but a fact or truth (see below, p.
77, footnote). I can be said both to see a man, and also to see that, e.g., he has a beard.
 48 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 49

, I hope, then, you understand what I mean by 'direct apprehension.' that there might possibly be something other than your own sense-data
And one reason I have insisted on the point is this. I think many and images, certainly does not merely consist in directly apprehending
philosophers have assumed more or less unconsciously that this way of a certain number of sense-data or images or both.
having things before the mind, which I call 'direct apprehension' is the There must, therefore, be some other ways of knowing of the
only way in which we ever have anything before our minds. They have existence of things besides the mere direct apprehension of sense-data
assumed, that is to say, that the only thing which ever happens, in our and images. And, in fact, it seems to me quite certain that sense-data
minds, whenever we have anything before our minds, merely consists and images are not the only kinds "of things which we directly
in the fact that we directly apprehend certain sense-data or certain apprehend. For instance, suppose I look at this envelope again, and
images, or both at the same time. And there is, I think, obviously a directly apprehend the whitish colour; it seems to me that if I try to
certain excuse for this assumption. For this kind of having before the observe what is happening in my mind, I can also directly apprehend
mind—the direct apprehension of sense-data and images—is certainly not only the whitish colour but also my own direct apprehension of it:
far easier to observe and to understand the exact nature of than any that is to say, that just as my seeing of the colour consists in my direct
other. If you try to observe what is going on in your mind at any apprehension of it, the colour, so, if I happen to observe my seeing of it,
moment, it is easy to see that you are directly apprehending certain this observation consists in the direct apprehension of my seeing of it—
sense-data, or certain images, or both; but it is not by any means easy to of something, that is to say, which is neither a sense-datum, nor an
see that anything else is happening in your mind at all. At least that is image, but the direct apprehension of a sense-datum. I think, therefore,
what I find. And even if you are convinced, as you may I think be we certainly sometimes directly apprehend not only sense-data and
convinced, that something else is in fact happening, it is very difficult images, but also our own acts of consciousness: and we may, I think,
to see exactly what the nature of this something else is: far more directly apprehend \ other things also.
difficult than to see what the direct apprehension of sense-data or of But there are, I think, certainly other ways of knowing, which do
images is. not merely consist in the direct apprehension of anything. And if we do
It is, therefore, I think, very natural to suppose that all knowledge ever know of the existence of material objects by means of our senses,
consists merely in the direct apprehension of sense-data and images; our knowledge of their existence, on the accepted view with regard to
and many philosophers have, I think, constantly assumed this. But now sense-data, must, I think, partly consist in one of these other ways of
observe what results if we combine this view with that view with knowing. But it is very difficult to analyse exactly what these other
regard to sense-data which I have called the accepted view— a view ways of knowing are; and that, I think, is one chief reason why many
which is, of course extended to all images, and which does, in fact, philosophers have supposed that we do not know of their existence at
much more obviously apply to images than to sense-data. It then all.
follows that no one does in fact ever have before his mind anything at In order to shew quite clearly that there are ways of knowledge
all except certain sense-data and images, which are quite private to other than direct apprehension, and also, in at least one instance, as
himself, and which can never be before anyone else's mind. And the clearly as I can what sort of a thing such knowledge is, I will return to
question then arises how any one of us can possibly know that there is an instance which I mentioned just above, the instance of memory.
anything else at all in the Universe except his own private sense-data I look at the envelope again and I see the whitish colour. I turn my
and images; how he can possibly know, for instance, that there are in head away, and I no longer see it. But I remember that / did see it a
the Universe, either the minds of other people, or material objects, or moment ago. I know that I did see it. There is nothing that I know more
the sense-data and images of other people. And obviously, on these certainly than this. Moreover I know that that whitish colour was: that
hypotheses, these are questions which must be answered in the there was such a thing in the Universe. I know, therefore, now of the
negative. On these hypotheses, nobody can possibly know of the past existence of that whitish colour; and yet I am certainly not directly
existence of anything at all except his own sense-data and images. But, apprehending it now. I may, indeed, possibly be apprehending now an
then, on the same hypotheses, nobody can even think that there might image more or less like it. And,
be anything else: for to think
 50 SOME MAIN PROBLEMS OF PHILOSOPHY SENSE-DATA 51

according to the view that all knowledge consists merely in the direct The seeing of a material object—or the perceiving one by any other
apprehension of sense-data and images, it is very natural to suppose sense—would therefore, on this view, be something quite different
that my memory of what I just now saw consists merely in my direct from the seeing of sense-data. The seeing of sense-data consists in
apprehension of an image of it now. But if you consider a moment, I directly apprehending them. But the seeing of a material object does
think you can easily see that this cannot possibly be the case. If it were not consist in directly apprehending it. It consists, partly in directly
the case, I could not possibly know that the image which I now see was apprehending certain sense-data, but partly also in knowing, besides
at all different from the colour which I saw a moment ago. And yet this and at the same time, that there exists something other than these sense-
is just what we all constantly do know whenever we remember data. And so, too, if we ever see that a material object is round or
anything. We know that there was something in the past different, in square, or in a particular position in space ; this also would consist, not
some respects, from anything which we are directly apprehending now. in directly apprehending these things, but in knowing, when we do
Memory, in fact, always carries with it the possibility of our knowing directly apprehend certain sense-data, certain things about something
this: that there was something which we are not now directly quite c her than these sense-data.
apprehending and different in some respects from anything which we I will now try, first of all, to describe more clearly exactly what sort
are now directly apprehending. of a thing I take this perception of material objects to be. And will then
And it seems to me that, on the view we have accepted with regard go on to consider what sort of reasons we may have for supposing that
to sense-data, our knowledge of the existence of material objects by this sort of perception really is knowledge: for supposing, that is, that
means of the senses must be analogous to memory at least in this: it there really does exist something other than the sense-data, which we
must consist in our knowing that there exists something different from directly apprehend, and that this something has certain properties and is
any sense-datum or image which we are directly apprehending at the a material object.
moment. This would seem to be the minimum which we must know, if
we are to know of the existence of any material object by means of the
senses. We must know, when we directly apprehend certain sense-data,
that there exists also something other than these sense-data—
something which we do not directly apprehend. And there seems no
sort of reason why we should not know at least this, once we have
dismissed the prejudice that we cannot know of the existence of
anything except what we directly apprehend. Of course, merely to know
this, would be to know very little. If the something, whose existence we
know of really is, in fact, a material object, we might be said to know
of the existence of a material object, even if we did not know that it
was a material object. But, we must know much more than this, if we
are to know also that this something is a material object. And
moreover, if we are to know that we all saw the same envelope, we
must know that the something, of whose existence we each of us know,
is the same something. But there seems no reason again why we should
not know many things of this kind. In the case of memory, we certainly
do know, with the utmost certainty, a very great many things, about the
something which we remember, beyond the mere fact that it wa$ and
was different from anything which we are now directly apprehending.