Space and Sense-Data

Author(s): W. Russell Brain

Source: The British Journal for the Philosophy of Science , Nov., 1960, Vol. 11, No. 43
(Nov., 1960), pp. 177-191
Published by: Oxford University Press on behalf of The British Society for the
Philosophy of Science

Stable URL: https://www.jstor.org/stable/685308

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

and Oxford University Press are collaborating with JSTOR to digitize, preserve and extend
access to The British Journal for the Philosophy of Science

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 The British Journal for the
Philosophy of Science
VOLUME XI NOVEMBER, 1960 NO. 43

SPACE AND SENSE-DATA *

W. RUSSELL BRAIN

I HAVE argued in the past that our knowledge of neurophysiology

necessitates the adoption of some form of the sense-datum theory.
The following are the chief reasons for this. I experience a sensation
only when certain changes occur in the cells which compose my
nervous system. When, by means of such sensations, I perceive an
object outside my body, the sensations are the end-result of a long chain
of physical causes starting at the object and ending in my nervous
system. When what I perceive is in my own body, the same is true,
except that the chain of physical causes begins at the surface of, or
within, my body instead of outside it. In either case, what finally
happens in my nervous system at the moment when I perceive either
an external object or my body is physically quite unlike the initial
change which causes it. There is, for example, no resemblance
between the light-waves which originate in the sun or are reflected
from a table, and the complex electrical changes which occur in my
brain when I see the sun or a table. A sense-datum, therefore, if we
use that term to describe a colour, a sound, a touch, or a smell which
we experience, is caused by the object which we see, hear, feel or
smell, but cannot be either a part of that object or qualitatively like it.
Moreover it is argued that, since in perceiving we are dealing with a
chain of physical causation which takes time, we must always perceive
an object at a time later than that at which the series of physical events
which leads to its perception starts from it. This time is only a fraction
of a second when we are concerned with awareness of events in our
own bodies, or with seeing an object on the surface of the earth, but
it may be appreciable when we are dealing with the sound made by a
rapidly moving object such as an aeroplane, which sounds as if it were
in one place when we can see it ui another, or when light travels from
* Read to the British Society for the Philosophy of Science, 30.x1.59
177
13

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

some celestial body to the eye. In the case of th

it takes about eight minutes for a light-wave to r
argued that the only information which sensation
always be about the past state of the object percei
practical difference in everyday life because the ob
deal change so little from moment to moment,
purposes their present state, at the moment at whic
is identical with that at which the light-wave or so
In the case of astronomical objects, however, th
practical importance, and has to be taken into acco
waves from a distant star take several thousand y
should not know if it had become extinct several
In what sense, therefore, can we be said to be seein
we not conclude from this temporal argument,
sense-data cannot be identical with the object whi
Thls view of the purely causal relationship betwee
data leads to the question, where are sense-dat
scnse-data are situated in a space in which they are rel
and to the body of the observer. Sounds are on th
located; nevertheless sound is in, or comes from,
in most instances, and the directions from which
distinguished from one another. Smells are related
diffuse way, but they also have spatial reference.
feelings are very precisely located, others more dif
are all somewhere.
If we become aware of all these sense-data as the result, and solely
as the result, of changes occurring in the brain, which is inside the skull,
how do they come to be experienced either outside the body when they
are related to external objects, or witlun the body itself, but never
within the brain? Some thinkers, chiefly physiologists and psycho-
logists, have tried to get over thius difficulty by invoking a process,
which they term projection. Sense-data, though depending upon
physiological changes in the brain, are held to be projected to the
spatial position in which they seem to be located. No-one has ever
succeeded in describing how projcction is brought about, and on
examination it scems to be no more than an imaginary process invoked
in an attempt to describe how changes within the brain lead to an
cxpericnce in wlhich a sensory quality is located outside the brain.
Some writers have sought to get over tlus difficulty by supposing
tlhat sense-data or perceptual objects exist in one space and physical
178

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

objects in another. (I do not think it necessary

relationship between sense-data and perceptual
not think it has been suggested that they rmght b
spaces.) Since Bertrand Russell has been the c
exponent of this view, I shall illustrate it from h
Naive realism identifies my percepts with physica
that the sun of the astronomers is what I see. This
the spatial relations of my percepts with those
Many people retain this aspect of naive realism
rejected all the rest. But this identificaton
spatial relations of physics hold between electrons
etc., which we do not perceive; the spatial relation
hold between things that we do perceive, and in the
coloured patches. . ... When I say that somethm
there are two different thlungs that I may mea
have a percept which is outside the percept of my
space, or I may mean that there is a physical ob
my body as a physical object m the space of phy
is a rough correspondence between these two. T
outside my body as I see it m perceptual space, and
outside my physical body m physical space. . . . On
whlch have led to confusion was failure to distngu
tual and physical space. Perceptual space con
relations between parts of percepts, whereas ph
of inferred relations between inferred physica
may be outside my percept of my body, but not o
physical thing. Percepts, considered causally, ar
afferent nerves (stimulus) and events in efferent n
location in causal chains is the same as that of certain events in the
brain.1

It would seem to follow from this (and it is a view which I have

myself adopted m the past), that each observer has his own private
perceptual space, whereas there is a physical space common to all
observers. My private perceptual space is the space m which are
located all the sensory characteristics of everything I perceive, that
is objects outside my body, and my body itself, and all that I can per-
ceive as happening within it. Every person is believed to have such a
perceptual space private to him. This idea has both advantages and
disadvantages. One of its advantages is that it provides a habitat for
I Bcrtrand Russell, Human Knowledge: its Scope and Limits, London, 1948, pp.
217-225

179

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

hallucinations, which are private to the observer,

exist in spatial relationship with the world of ext
he perceives. If one observer says that he sees a m
room, in which other people present say they see
difficult to understand if we believe that the sen
objects m the room and their spatial relationships
private to the observer. The hallucinatory appe
though resulting from some abnormality in his n
not to be regarded as an event of a different perce
awareness of the normal occupants of the room. S
who has had a limb amputated experiences a phan
that it feels as though the limb was still there, h
located in a spatial relationship to the rest of his b
to exist in hius private perceptual space though it
longer in physical space.
There are, however, certain disadvantages in th
of us possess a private perceptual space. The first
raises is the relationship of these multiple private
one another and to physical space. Russell clear
private perceptual space is in my brain. It cann
my perceptual bramin, which is situated in my pe
my perceptual head. It must therefore be m my p
that raises the question of the relationship betw
space and physical space. A disadvantage, whiuch
serious, is the complication which it introduce
events of everyday life. In order that I may lift
lips, I must first by sight and touch become aware of
cup as a perceptual object in my perceptual space,
perceptual hand and mouth are situated. I then go
movements in that space guided by my sense-data
and vision, as the result of which I succeed in brin
cup to my perceptual mouth, and I am gratified to
all this has taken place in a perceptual world located
brain, I have succeeded in lifting the physical cup
hand to my physical mouth.
The sense-datum theory, and therefore the c
perceptual spaces based upon it, has to meet criticis
kind, namely that it is self-contradictory. This cr
been expressed by a number of philosophers, is w
Hirst as follows.
18o

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

If we, as minds, never directly perceive material obj

directly aware of images or mental representations a
them, how do we know that there are any material
or what their nature is? We cannot look behind t
to see what their causes are like, if they have an
to notice that his theory is self-refuting, for its co
the prermse assumed m physiology that we do perce
such as sense organs and brains.1

I believe that the facts of physics and physiol

outlined necessitate some form of the sense-datum
come to think that it is possible to retain this with
believing in a multiplicity of private perceptual sp
revised sense-datum theory, m the form in which
it, may go some way towards reconciling the repr
of perception with the views of some at least of th
In an earlier discussion of this question I said:
If at least what the philosophers call secondary quali
sounds, colours and so on are quite unlike the phy
give rise to them, then we must regard them as s
reality and say that the receptive function of the ce
provide us with a symbolical representation of the wh
world, not only distinguishng objects by their
conveying to us the spatial relationships which ex
and at the same time giving us similar symbolical
our own bodies and their relationship with the e
this mformanon, of course, is given us not merely f
awareness or contemplation, though that may s
product of it, but m order that we may act; hence i
with which we are familiar, with the motor activiti
And Russell said:

Percepts, considered causally, are between events

(stimulus) and events m efferent nerves (reaction
causal chains is the same as that of certain events in t

Both of these accounts recogmse the importanc

motor functions and, by implication at least, the d
latter upon the former. Russell goes further m lo
regions of the brain regarded causally as between af
1 R. J. Hirst, ' Perception, Science and Commonsense ', M
2 W. Russell Brain, Mind, Perception and Science, London, 1
3 Russell, op. cit., p. 225
181

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W RUSSELL BRAIN

pathways. Percepts are thus treated, both as

respect of their physiological substrata, as isolab
accompamments or sequelae.
No doubt we habitually consciously isolate the
in an experience from our actual or potential ac
them, and this has led to the belief that percept
events, and the neurophysiological basis of which
the termmnation of the relevant afferent sensory p
to the neural substratum of any motor activities to
rise. I suggest that both sense-data and perc
which, however convenent for many practic
leading when we come to consider the nature
epistemologically or physiologically; that expe
nothing but sensory charactenstics, do not exist
when we talk about sense-data and percepts is so
and that, whatever it is, it is not to be found st
between sensory and motor nerve-pathways m t
If we adopt the sense-datum theory in the
Russell, the whole of the space which I perceive
brain. Hence when I lift a teacup to my lips, th
of my hand, the pressure arid weight of the han
thumb, the sensation of the movement of my ar
pressure of the cup upon my lips, together with
of these vanous experiences, are all located w
success in performing the manoeuvre depends u
various bram events which cause me to have the
represent corresponding events in the physical t
bouiy m such a way that when I experience this
these physical events are taking place. This leads
of the view being put forward m this paper.
Whatever view different schools of philoso
sense-data, no-one, as far as I know, has dem
opposed to hallucinatory perception, such physi
place. Hence when I lift a cup to my lips I know
is, and where my lips are in physical space. If
follows that our sensc-data may give us inform
relationships of external objects both to one ano
bodies. This information guides, and is validat
tivities in relation to such external objects. This
with the idea of a multiplicity of pnrivate perceptu
182

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

observer. As far as the physical world is concern

space, in which physical objects, amongst which
exist. The spatial elements in perception give u
the spatial relationship of bodies im physical s
and subjective about perception is that this infor
to each one of us, and presents itself in the form of
another of Russell's phrases, centred upon our ow
Thls view of perception is not inconsistent w
the sensory qualities of objects, colours, sounds,
are the products of the activity of our nervous syste
unlike the objects which are their remote and ini
spatial characteristics of the objects I perceive, suc
and relative position, the spatial characteristics o
the spatial relationships between external objec
the charactenstics and relationships of those objec
Unlike other sensory qualities, they are not crea
system, the function of which in this respect is
information, that is, to make me aware of them.
How on this view are we to deal with the crticism derived from
the sense-datum theory, that the time factor involved in perception
means that what we are aware of can never be the physical object
perceived? The answer to this is that if being aware of objects is
identical with receiving information about them, the time interval
involved in perception does not impair the accuracy of the information
as far as normal perception of most terrestrinal objects is concerned.
When I see an object, even a moving object, in spite of the speed of
light and the time taken for the transrmssion of a nerve impulse from
my retina to the visual area of my cerebral cortex, the object really is
where I see it. The same is true of an object held in my hand; for
the time taken by the sensory nerve-impulses to reach my brain does
not render the mformation about the size, shape and position of the
object inaccurate. The test of the motor response can be applied in
each case. The mformation is reliable as a guide to movement in
respect of both the distant object and the object grasped. When we
come to astrononucal objects, the time factor involved m the conduc-
tion of light waves may be great enough to make a practical difference.
Now, however, we are dealing with mformation about the object's
past, and this becomes inaccurate only if we wrongly suppose that it
relates to the object's present situation. Even m such cases, however,
we may have to introduce a qualification. When I speak about shading
183

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

my eyes from the sun, I do not normally mean t

which is by now in a slightly different position f
see. For practical purposes the sun, as the object o
light and heat, is equivalent to the light and heat w
part of their passage through the atmosphere, an
mation about the sun as an object in thius sense is
when I look at a star which may be situated many mil
away, unless I am astronomically sophisticated I sh
the existence of the astronomers' star, but I shall p
physical object composed of the light waves from
last part of their course before they reach my ey
again the accuracy of the information is to be mea
motor response, i.e. shading the eyes from the sun
star.

Here we have the answer to the objection to the sense-datum

theory made by Hirst and others. As I have said elsewhere,

If we start our description of perceptual knowledge with a subject

and an object, we shall rapidly get mto difficulties from which there is
no escape. The fact with which we must begin is the fact of knowledge,
experience, or mformation, if we use the last word, as I think we must,
to imply a receiver as well as something received. If we start with
knowledge or experience we start with the subject-object relationship
already given. We do not need to ask how we become aware of things
outside ourselves because it is with that awareness that we begin. It is
easy to fall into the error of supposing that the sense-data generated by
the brain must be entirely unlike the physical objects which, by acting
upon the brain from the outside world, produce them. But, with some
important exceptions, thls view is minustaken. . . . One of the most
important functions of the brain is to provide us with an accurate
representation of the spatial structure of the external world as well as
of our own bodies. An important part of this structure is the ever-
changing relationshlp between our bodies and their environment:
awareness of the externality of what is outside our bodies is therefore
given m all ordinary acts of sense perception.-

Hence while there may be forms of the sense-datum theory which

imply that, in Hirst's words, 'we, as minds, never directly perceive
material objects but are only directly aware of images or mental
representations allegedly caused by them ', and in respect of which it is
legitimate to ask, 'how do we know that there are any material or
1 W. Russell Brain, The Nature of Experience, 1959, pp. 31-32
184

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

physical causes, or what their nature is?' this is not t

the theory now proposed. In that we do have di
external objects, and of our own bodies. We do
things such as sense organs and brains. We can m
infer from our observations the speed of light a
impulse, and learn from physical and physiological
the structure and activity of the nervous system, t
elements are the products of the nervous system
no mformation concerning material objects, we
that material objects were the causes of our sensatio
we have sufficient information concerning the stru
objects, there is nothing illogical m inferring that
our perceptions are subjective. All that is neede
inference is the power of spatial and tempora
Whether we discriminate objects such as pointer-read
depend upon black and white contrast or contras
be quite irrelevant to the accuracy of the observano
It may be said that this does not meet Hirst's
our knowledge of the structure of the external wor
sense-data which are generated by our brains, a
subjective, we cannot know that there are such t
objects to cause our sense-data. It is necessary, ther
the relationship between structure and sense-da
structure is independent of some sense-data is an eve
We recogmse a photograph although it is a flat
representation ofa coloured three-dimensional object
it as a photograph of the same object whether it
Similarly we recognise the structure of a word, wh
in black on white or blue on red. The distant hills look small and
blue, and as we approach them they become large and green, but we
recognse them as the same hills. In order that we may recogmse
a spatial structure visually, we must be able to discriminate perceptual
differences, but within a wide range it does not matter what shades
are used to differentiate them, provided the shape of the boundary
between them remains the same. Hence, provided an area of the
body surface, such as the retina, possesses receptors which are stimulated
from different points of space, and the nervous system can discriminate
the nervous impulses excited by such stimuli, we possess the basis of
visual discrimination of spatial structure in the external world, and the
fact that this discrimination reaches consciousness literally coloured by
I8s

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

sense-data, which themselves do not exist m the

not mean that the information they mediate is n
Most mammals do not possess colour-vision, w
insects, birds, the higher primates, and man. Thi
orgamsms which do possess colour-vision hav
system capable of responding differentially to dif
length of light. Those without colour-vision c
objects by vision, but those with colour-vision po
of discrimination.
Thus we arrive at the view that m perception w
ledge of the size, shape, and spatial disposition
What we know about them in these respects is, s
errors, genuine knowledge. They are where we
m a public physical space. Our knowledge of t
private, but that does not mean that we create
space m which our perceptions of external obje
true broadly of terrestnal objects, with which
When we come to celestial objects, we must eithe
meaning the last phase of the energy they propag
our bodies, or we must say that the spatial charac
have present knowledge are those of their past, a
remote past.
To understand the part played by sense-data
necessary to look at the problem created for the
the physical stimuli reaching it from external ob
develop a capacity to respond to them. The ph
extremely diverse, compnsing as they do light-w
changes, sound-waves, the presence of particular
or m food substances, and the physical changes d
the pressure of objects on the body, the moveme
gravity. The organism must develop specific r
selectively responsive to each variety of physic
such receptors can do when stimulated is to ex
wluch consist of electncal changes m the nerve
receptor to the central nervous system. These
differ somewhat in frequency and rate of conduc
comparatively small range. There is no evidence t
differences m the nerve-impulses from particular
to the very wide range of differences between th
All our awareness of perceptual difference ther
186

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

differences m the spatial and temporal configura

changes reaching the brain. When such changes o
the appropriate sensation, and it is arguable that
electrical changes is the experiencing of the sensa
It is inevitable therefore that a percept should
elements, and, as we have seen, it is these wh
discrimmation possible. External objects, ther
through the nervous system, and such of their ph
can be distinguished by our sensory receptors
distinguishable sensory qualities of sense-data. W
the fact that the sensory characteristics of object
medium through which light-waves pass on their
us. The colour of distant objects is affected by t
phere, and if I look at the world through rose
objects will have a rosy appearance. This process
objects with sensory qualities m order to discrim
properties would be useless if the label were not a
and this is what happens m perception. We h
mformation concerning the external world, as
which it is perceived with sense-data which sy
physical properties. The same is true of our ow
Thus, although it is sometimes convenient to ta
data, and the localisation of sensory pathways an
vous system, this is an abstraction from a more
totality. Quite early m infancy the visual app
stimulates the young child to make movements w
direction of the object. If it succeeds in grasping i
by sensations of touch, and knowledge of the di
becomes linked m an elementary way with kn
ments required to grasp it. So perception is fr
motor, and sense-data through experience may
other information related to other potential sen
weight, etc.
Much of this may not be obvious even on introspection, but it has
at least ongmated m conscious experience. Physiology shows us that
perception is profoundly influenced by factors which do not as such
enter consciousness at all. The main evidence for the part played by
these unconscious afferent impulses is the profound disorgamsation of
perception which occurs when they are themselves disorgamsed or cut
off. For example, our whole perceptual onrientation m space depends
187

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

upon the continuous reception of nervous impulses f

yet we are completely unaware of these as s
perceptual information is conscious information m
are aware of it as constituting an individual elemen
ceive, like a colour. Nevertheless it contributes t
alone illustrates the inadequacy of any philosophic
ception which is limited to the untutored observatio
however introspective.
Finally there is a question which is fundament
standing of perception, and indeed of consciousnes
wrote:

Let us suppose that we are watching an observer looking

There is a sense-datum or a group of sense-data which ar
Light waves, which in their grouping mantain a circular rela
travel from the object to the eyes of the observer. On a circ
on each of ins two retinae disturbances are set up which exc
impulses which travel through his optic nerves, tracts and r
to the visual areas of the cerebral cortex. Only when th
impulses reach the cortex does the observer see the circle
logical idealism is true we rmght expect to finmd that there is
circular about the events at the cerebral cortex, for it is the
told, which are 'projected' on to the outside world when w
a circle. Nothiung of the sort is true. The area of cortica
which exists when we perceive a circle is divided into two hal
each cerebral hemisphere. Pathways connecting them exist, b
appear to play no part m our perception of the two halves o
as one whole, for this still occurs when the connecting pa
corpus callosum) is divided.' Neither half is sermcircular; it i
the shape of D, the closed end lying m front and the open e
The right half of the circle is represented m the left cerebral
and vice versa, and the lower quadrant is represented above t
There is another complication. We saw that a circular ne
turbance occurred in each retina; thus each half of the c
double representation in the nervous system, and, though th
for the two corresponding halves-that is, the two right halv
two left halves--come to be close together behind the opu
and are represented in the same area of the cerebral cortex,
anatomical point at which they fuse, as Sherrmington2 demon

'1A. J. Akelaints, Arch. Neurol. Psychiat., Chiucago, 1941, 45, 288

2 Sherrngton, The Integrative Action of the Nervous System, Lond
354 sq.
188

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

his experiments with flicker. Finally, since the tw

are represented m cortical areas lying parallel to eac
disturbance is three-dimensional.
Thus when we perceive a two-dimensional circle
of an activity m the brain which is halved, redu
inverted, distorted, and three-dimensional. If p
is to be really physiological it must admit that its
breaks down because the circle which is said to be
cerebral cortex never existed there at all.x

To-day I should qualify and extend this sta

cortex is a necessary condition of vision, for its d
subject blind, but the same is true of the optic n
we attach importance to the cortex? Because it is
point on the afferent visual pathway of which th
not follow that the function of vision is locat
inaccurate to speak of its location in any particula
system. If the brain largely acts as a whole, visio
of the whole of it, in which case it is represented
in a different way from its representation in th
latter then becomes only the last point at which
concerned with vision exist as an isolable grou
true of other sensory pathways, and of the mot
pathways too. If so, the fact that there is nothin
activity of the visual cortex when we see a circle
larger question: is there anything circular abo
activities in these circumstances? Is neuropsycho
a necessary concept, or indeed a meaningful one?
a circle there is always something circular in the
discussing tilted circles). If I see a circle there is a
on my retinae. If I feel a penny pressed into
circular disturbance in the skin of my hand. T
cerning the circularity of the visual or tactile obj
the relevant nerves in a complex pattern dependin
nerve-fibres excited, the rate and frequency with
up them, and the temporal pattern of their indi
other words it is coded. Up to and including the
is a detectable correlation between a nerve fibre
retina or skin excited, as shown by the sensation a
it and the defect caused by its destruction. B
1 W. Russell Bran, Philosophy, 1946, 21, 133
189

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 W. RUSSELL BRAIN

no such correlation detectable. Somehow this

information in a spatio-temporal pattern of fre
into a static representation of a spatial pattern.
How this is done goes to the heart of the problem
relationship. We do not yet know enough to des
but we can see at least the outline of the process. Co
facts. I ask someone to close his eyes and stretch o
moving his extended forefinger to describe a cir
keeping the forefinger still, but moving the wrist,
circle, by wluch I mean a circle m the same place: an
both the finger and wrist still, to do it again by mo
shoulder. No one has any difficulty m doing this,
doubts that the three circles are, approximately at least
What does this involve m psychological and ne
Psychologically there is initially the decision to ca
ment, which in some people will be accompanied
of the circle to be described. Neurologically, a s
impulses will cause contractions of the right muscle
to produce the desired effect. For each of the thre
muscles will be different, and the parts of the motor c
excited will differ correspondingly. Psychologicall
ment is occurring there will be awareness of its ci
(neurologically) from sensory receptors chiefly m an
which again will be different in each case. The
impulses will reach different regions of the sensor
will be nothing circular about their disposition
system, then, must have the means of extracting fr
diverse sources the conscious information that the
describing a circle m the same place, just as on the
to use different groups of muscles to the same end
passing that to describe one circle involves the ret
of the earlier part of the movement as a guide to
memory is also required to enable the second and th
the first. But the process is much more complex t
three successive movements would not describe circles m the same
place unless the body had meanwhile remained stationary, so all the
sensory and motor pathways and muscles concerned m the maintenance
of the posture of the body have played an unconscious part m the
procedure. And 'the same place' has meaning only m relation to
the rest of the space in wuhich the body and external objects are situated,
19go

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms
 SPACE AND SENSE-DATA

so the conscious and unconscious factors contribu

space are also involved. (No one, for example,
procedure accurately if as the result of an attack
appeared to him to be going round.) Viewed i
sorimotor substrata of the three movements are diff
information is correlated with that from the who
' body-image ') the result is awareness that the mo
tip is the same m each case. Hence, awareness of
movement is the product of a very complex neu
of mformation m space and time.
Patzll points out that as in mathematics a prope
the same in spite of its transfer from one co-ordin
is termed an mvariant, so the cerebral activity, w
ness of a single space out of nervous impulses
sources, does so by creating mvariants. A similar
our recognition of the identity of visual objects
relation system in which they are at any momen
might add, in spite of the perceptual changes pr
their distance from the observer and perspective.
invariants is the function of the schemas, whose
speech I have discussed.2
So when we see a circle, the fact that the patt
m the visual cortex is not itself circular is of no f
All they need to do is to transrmt mformation m
form of electrical impulses for mtegration at
nervous system. We do not at present know how
us aware, either that we are seeing something,
seeing is circular.

1 0. P6tzl, Die Aphasialehre von Standpunkte der klinischen

optisch-agnostischen Stiirungen, 1928
2 W Russell Brain, 'The Concept of the Schema in Neuro
Perspectives in Neuropsychiatry, ed. by D. Richter, 1950,
and Agnosa ', m Neurology by S. A. Kinmer Wilson, ed. b
3, 1413.

191

This content downloaded from

138.232.236.29 on Sun, 12 May 2024 14:14:08 +00:00
All use subject to https://about.jstor.org/terms