Content uploaded by Nicholas Maxwell
Author content
All content in this area was uploaded by Nicholas Maxwell on Oct 27, 2014
Content may be subject to copyright.
PHYSICS AND COMMON SENSE*
NICHOLASMAXWELL
I
Physicalism
andCommonSense
THE
basic prob
lem to be discussed is this: To
what extent, in
what
sense, may the discoveries of physics legitimately conflict
with our
ordinary common sense views about the worl
d based on
our ordinary
experience? In order to discuss this question, I
consider a particular
extreme view about the nature of the world
—
a view I call Physicalism
—
which may seem to be supported by
the findings of physics, and
which asserts at least that:
(a)
The world is made up entirely of only a few different sorts of
things
—
the fundamental physical entities.
(b)
Precise, exceptionless laws govern the way in which these
entities
change.
(c) Human perception is almost entirely deceptive: almost all
perce
p
tual qualities, e.g. qualities such as colours, sounds and
smells, have
noreal, no objective existence.
According to physicalism, only those qualities which apply to
the
fundamental entities, or to aggregates of fundamental entities,
such as
mass, posit
ion, electric charge, etc., really exist.
Many eminent physicists appear to have believed in some form
of
physicalism. Thus Planck has written that certain '. . .
considerations
... and not any logical argument. .. compel us to
assume the existence
of anot
her world of reality behind the world
of the senses; a world
which has existence independent of man,
and which can only be per
ceived directly through the medium of
the world of the senses, and by means of certain symbols which
our senses allow us to appre
hend. It
is as though we were
compelled to contemplate a certain object in
which we are
interested through spectacles of whose optical properties
we were
entirely ignorant.'
1
Again, Einstein has written: ' Physics is
an
attempt conceptually to grasp realit
y as it is thought independently
* Received I5.vii.65
1
M. Planck,
The Universe in the Light of Modem Physics,
Alien
&Unwin, 1931,
PP
-
8
-
9
295
NICHOLAS
MAXWELL
of its being observed. In this sense one speaks of "physical reality"’
1
,
and:
'The be
lief in an external world independent of the
perceiving
subject is the basis of all natural science. Since
however, sense per
ception only gives information of this external
world or of "physical
reality" indirectly, we can only grasp the
latter by specula
tive means.'
2
The findings of physics do not of course oblige us to adopt
some version of physicalism. We may argue that scientific
investigation
involves an extension, elaboration and refinement of
common sense, of
our ordinary experience, and therefore
that
scientific discoveries cannot
contradict in any fundamental way the
tenets of common sense that are
based on ordinary experience.
According to this second view
—
which
I call the ' common sense'
theory
—
scientific discoveries can only
undermine those of
our
ordinary views about the world that are based
on inadequate or
distorted observation.
Thus scientists may find some means to extend the range of
our
experience, to that which is very distant perhaps, or very
small, or to
that which occurred long ago. A
s a result of
obtaining this new,
direct or indirect, observational evidence, the
scientist is in a position
to correct previous assumptions made
about the nature of that part of
the universe in question.
Again, it is the physicist's fundamental aim to dis
cover, in so
far as
this is possible, those very general regularities between
phenomena to
which there are no exceptions; to discover, in other
words, the ' laws
of nature '. On the whole physicists have been
extraordinarily success
ful in this search. But
from our immediate,
uninformed experience it is not at all obvious that the nature of the
world is such as to be amen
able to this kind of investigation. Thus
from ordinary human exper
ience, not altogether unreasonable
conclusions may be reached about
th
e nature of things in the
world, which may seem to be rendered
highly implausible by the
subsequent apparent discovery of universal
laws of nature
—
by
the discovery that it is possible to formulate laws
which
(a)
are
not refuted by experience (perhaps withi
n certain limits)
and
(b)
make accurate predictions possible.
But none of this implies that there is anything inadequate or
dis
torted about our observation of the familiar ' furniture of the
earth '.
Thus, according to this second view, as long as we are
not
mad,
drugged, blind, deaf or dreaming, our ordinary experience
provides us
with knowledge about the visual, auditory, tactile,
etc., properties of
1
P.Schilpp(ed.),
AlbertEinstein:Philosopher
-
Scientist,
Tudor,New
York,1957
,
p.81
2
A.
Einstein,
TheWorldasIseeIt,
BodleyHead,1935,pp.156
-
157
.
296
PHYSICS AND COMMON SENSE
the things on this earth which no legitimate scientific theory
could ever refute. Roses really are red, dogs really do growl,
etc.
These two theories, the 'common s
ense' theory and
physicalism, have of course been formulated with no very great
precision. Any number of different versions of each theory may
be developed. Enough has been said however, I hope, to make
quite clear the follow
ing crucial point: we have her
e two very
different theories about the nature of the world, for either of
which the claim may be made that it is scientific, that it is
supported by science, and takes into account possible future
scientific discoveries.
I wish to stress that these are th
eories about the nature of the
world, theories about things, and not primarily theories about the
nature of science or the theories of physics (although each theory
may, incident
ally as it were, imply different things about the
nature of science). Thus, a
ccording to physicalism, the world is
made up solely of a few different kinds of fundamental physical
entities, nothing really being coloured, etc.; according to the '
common sense ' theory, the world is at least made up of all the
different kinds of thing
s that we ordinarily experience, many of
which really are coloured, etc.
We have here surely an extraordinary state of affairs. Physics
appears to be a particularly precise discipline; most physicists at
any one time seem to agree as to which theories are
acceptable,
which un
acceptable. Yet we have just seen that wholly divergent
views may be held about what physics can tell us about the
nature of the world.
The reason for this, briefly, is that precisely what we consider
physics can tell us about the natu
re of the world depends to a
consider
able extent on the kind of interpretation we think can
legitimately be given to the mathematical formalism of a
physical theory. And on this point there is no general agreement.
Let us suppose a wholly successful, all
-
embracing, fundamental
physical theory has been found which suits ideally the
physicalist's purpo
ses. We may suppose that this '
ideal' theory is
such
that
given
any isolated
system,
and
given
the
so
-
called
'initial conditions' corres
ponding to som
e state of that system,
then the theory will in principle enable us to predict all
subsequent states of the system. Further we may suppose that the
theory at least ostensibly postulates the existence of just a few
different kinds of fundamental entities.
T
he physicalist would interpret such a physical theory in the
follow
ing manner. Given any isolated system, in order to
describe what really exists at any instant we require (a) the
fundamental theory, which
297
NICHOLAS MAXWELL
states in general terms
what kind of things the fundamental entities
are,
(b)
the initial conditions, which specify the precise values at the
moment
in question of such variable attributes of the
fundamental entities as
position, momentum. According to the
physicalist, a descript
ion of
what really exists at a given moment
is only
complete
if it enables us to
deduce descriptions of
subsequent states of the system. By hypothesis,
the above ' ideal'
theory does provide such a ' complete ' description.
The physical theory might howeve
r be given a quite different,
non
-
physicalist interpretation. Briefly, it might be argued that the
theory
would not
really
postulate the existence of fundamental
entities. The
theory would
really
only describe a
model,
a
conceptual tool invented by
the phy
sicist solely in order to facilitate
the prediction of observations.
Thus the fact a successful physical
theory apparently postulates and describes fundamental physical
entities in no way implies that such
entities
really
exist.
1
Defending such different i
nterpretations to one and the same
mathe
matical formalism clearly amounts to defending different
theories
about the nature of the world. The question arises: Does a
successful
mathematical formalism, given a physicalist
interpretation, constitute a
possib
le theory of physics, as opposed
perhaps to a theory of meta
physics?
I accept here without argument Popper's solution to the
problem of
demarcation between physics and metaphysics: a
theory, in order to
belong to physics, must at least be
experimentally f
alsifiable.
2
From this requirement it follows that the kind of
interpretation of
the mathematical formalism of a physical theory,
demanded by physical
ism, which we may call
'
tentative realism
', is
legitimate. In practice
of course no isolated system can be
observed; but experiments can
nonetheless be devised to test
deductive consequences of theories in
tended to apply to isolated
systems.
It should be noted that even if one day the kind of ideal
funda
mental physical theory described above is found, the
ph
ysicalist will
still never know with certainty that entities precisely
like those described
by the theory do really exist, since it will
never be possible to know
1
Variants of this view have been defe
nded by Berkeley, Mach,
Poincar
é
, Bridg
man, Heisenberg,
Bohr, and by the logical positivists.
For an exposition and criticism
of this view, under the heading '
Ins
trumentalism', see K. Popper, '
Three Views
concerning Human
Knowledge', in
Conjectures and Refutations,
Routledge and Kegan
Paul,
London,1963.
2
K.
Popper,
TheLogicofScientificDiscovery,
Hutchinson,London,
1959
298
PHYSICS
AND COMMON
SENSE
with certainty that the theory is true. Nonetheless the theory
might
be completely true, and hence entities precisely like those
postulated by
the
theory
might
really exist. The fact that the
theory must be open
to experimental refutation ensures that it is
meaningful to call the theory
false, which in turn ensures surely that
it must be at least
meaningful
to
callthetheory
true.
In giving a tentat
ive realist interpretation to physical theories we
are
not obliged to assert that entities postulated by
all
acceptable
physical
theories really exist. Suppose a theory, by postulating
certain en
tities, is able to make accurate predictions within
certain
conditions
or limits, but outside these, goes wildly astray,
and is, in other words, refuted. Clearly, in this case, not all of the
entities postulated by the
theory really exist. Nonetheless the
theory might be retained, perhaps
because it is the only the
ory
available, or perhaps because of its relative
simplicity.
The kind of tentative realist interpretation of physical
theories indicated here is often opposed by a view which
combines a non
-
realist interpretation of theories that apply to the
so
-
called mi
cro
-
level,
with a realist interpretation of theories that
apply to the macro
-
level.
Such a view is prompted partly by the
belief that scientific theories
must be observationally verifiable
(the kind of belief that is presupposed
by Hume's polemic against
c
ausation as ' necessary connections' be
tween events), partly by
the fact that such a view enables one to side
step the apparent
wave/particle paradoxes of micro
-
phenomena. Un
fortunately there
is insufficient space for a discussion of such matters
here.
1
The view that physicalism might conceivably be
formulatable as
a physical theory raises the problem: Why would
not such a theory be
refuted immediately by our ordinary
experiences? Is not the theory
that the world is in reality
colourless, soundless, odour
less, etc., refuted
by the fact that we do
see colours, hear sounds, smell smells?
I wish to argue that physicalism is only a defensible theory if
inter
preted in such a way that it does
not
imply that there are no
colours, etc.,
in the world. The problem
here is to interpret
physicalism in such a
way that it does not contradict an
acceptable version of the ' common
1
For a criticism of the principle of complementarity (and thus of
phenomenalist interpretation of quantum mechanics) see: M. Bunge,
' Strife a
bout Complementar
ity ', this
journal 6,
21; for a suggestion
of how apparent wave/particle paradoxes of
micro
-
phenom
ena may
be resolved see A. Land
é
,
'
Why do Quantum Theorists ignore the
QuantumTheory? '
this Journal,
15,
60.
299
NICHOLAS MAXWELL
sense' theory. But before indicating how the ' common sense'
theory and physicalism may be reconciled, I wish to discuss two
ver
sions of physicalism that reject outright the ' common sense '
theory.
The thoroughgoing
physicalist
clearly
req
uires a t
heory of
perception
which explains why our ordinary experience does not
refute
physicalism. From Popper's demarcation requirement, it
follows that
if physicalism is to be formulatable as a physical
theory, then from the
theory, and a description of the
constitution
of the hum an brain and
sense organs, it must be possible to deduce
a theory of perception which:
t. Explains why most of our
experience is delusive.
2. Indicates precisely what aspect of our
experience is
not
delusive, so
that predictions of
the physicalist
theory may be tested experiment
ally.
2
TwoPhysicalist Theories Rejected
The two physicalist theories of perception I wish to discuss are
(a)
dualism,
(b)
the discriminatory response/brain process theory,
defended
recently byProfessorS
mart.
1
For the purposes of this discussion I shall call those qualities
which,
according to physicalism do really, objectively exist,
physicalist,
and
those perceptual qualities which, according to
physicalism, do not really,
objectively exist,
phenomenal.
This
leaves open the possibility that
some, but clearly not all,
physicalist qualities, such as for example
shape, may also be
perceptual.
(a) Dualism.
This doctrine accounts for the alleged
deceptiveness
of human perception as follows. We know that if I
perceive, let us say,
a red rose, then a causal sequence of events
takes place between the
surface of the rose and my brain via my
eyes and optic nerve. Accord
ing to the dualist, it follows from this
that my visual sensation of the
rose must be an entity
caused by, or
in some way associated with, some
event in my brain, but
nonetheless distinct from any event in my brain, and open only to
my own inspection. Hence, in perceiving the rose, I
really only
perceive my visual sensation of the rose. I can therefo
re have no
reason to believe that the perceptual quality, redness, applies
to the
physical object I have called a rose: redness is a quality that can
apply only to visual sensations that occur in the mind. Of course in
practice if a physical object causes
us to have the visual sensation
of
redness, then we will tend to call that physical object' red '; but
strictly
it must always be
false
to ascribe a perceptual quality to a
physical
1
J. J. C. Smart,
Philosophy and
Scientific Realism,
Routledge &
KeganPaul
, Lon
don, 1963, Chaps. 4,5
300
PHYSICS AND COMMON SENSE
object. In general, physicalist qualities apply only to physical
objects,
perceptual qualities only to sensations or sense impressions,
which exist
exclusively in our minds. Dualism thus involve
s a
serious addition to
physicalism, since according to dualism the
world is made up of the two
entirely different sorts of things: (i)
fundamental physical entities, (ii)
sense impressions, or, more
generally, ideas. Ideas exist only in
dimensionless' bub
bles',
called ' minds ', which are associated in some
strange way with
certain clusters of fundamental physical entities called
' brains '.
Such a dualist theory of perception does provide an
explanation
of how human perception might be deceptive, of how
t
hings might not
really be as they seem. It is a mistake, however, to
suppose that dual
ism can be derived from the existence of the
causal sequence of events
involved in any perception. We may
stipulate that a necessary condi
tion for perceiving an object
is that a
certain kind of causal sequence of
events should take place
between the object and the brain via the eyes and optic nerve. The
visual sensation (as opposed to the perception)
we may define as
the last event in the above sequence, leaving open the
question of
whether this is a brain process or some peculiarly ' mental'
occurrence. From such an explication of' perception ' it follows
that
far from
only
being able to perceive our own visual
sensations, in
practice we
never
perceive our visual sensati
ons.
A more serious objection to dualism as a physicalist theory of
per
ception is however that it does not meet the second of the
above two
requirements for such a theory. According to dualism,
we can have
no reason to believe that any experience is non
-
d
elusive, if our exper
ience is interpreted as providing information
about the physical world.
This is because, according to dualism,
we can
only
perceive our own
sensations, it being impossible to
perceive any physical object.
1
In
other words, the assumpti
on that
dualism is true implies that there can
be no evidence in favour of
the theory. (More precisely, dualism
implies that no singular
existential proposition about the physical world
can be verified, and
hencethat no physicalist theorycan be experimen
t
allyrefutable.)
(b) The Discriminatory Response/Brain Process
theory of
Professor
Smart. Briefly, Smart's thesis is that phenomenal
qualities must be
understood not as intrinsic, unanalysable,
objective properties of things,
1
This, basically, is the po
int behind Berkeley's polemic against
Lockean substance,
see ' Three Dialogues' in G. Berkeley,
A New
Theory of Vision and other writings,
Everyman, London, 1910.
301
NICHOLAS MAXWELL
but in terms of the discriminatory responses of human beings.
Colour,
for example, is to be analysed in the following manner.
First we explicate the phrase ' discriminate with respect to colour
' without
introducing colour as an undefined primitive. A person
discriminates
between objects with respect
to colour if
the
discri
mination can be made when the objects are illuminated
with daylight but cannot be
made when the objects are
illuminated with monochromatic light. A
normal human
percipient with respect to colour is then defined as a
person who
is able to make all the colou
r discriminations that anyone
else can
make. We then define objects of the same colour as objects
which a normal human percipient would not be able to dis
-
criminate between with respect t
o colour. Thus Smart declares:
‘
We might say at a first shot, tha
t
'
this is red' means roughly that
a
normal human percipient would not easily pick this thing out of a
heap
of geranium petals, though he would easily pick it out of a
heap of
lettuceleaves.'
1
Smart acknowledges that this account of colour has one slight
fl
aw: we can imagine that colours of all objects undergo a
systematic change,
so that for example lettuce leaves become red,
geranium petals green.
According to the above account, this
radical change would not be detectable, since human beings
would continue
to make colour dis
criminations just as before. In
order to meet this difficulty, Smart
admits that we must take into
account the inner experiences that accom
pany discriminatory
responses. But this does not involve admitting
that unanalysable
mental or p
sychic entities exist, since these inner
experiences
are, according to Smart, nothing more than complicated
neurophysiological events.
This last point
—
which we may call, following Smart, the '
brain process theory of inner experiences'
—
does not, I think,
r
epresent a
weak point in Smart's position. The theory is based on
the argument
that in having an inner experience I am presented
with no evidence for
the existence of entities over and above
brain processes, in the sense
required by dualism. Thus, if I
exp
erience, for example, the visual sensation of a red patch, then
I have undeniable evidence only that something exists which, in
some unknown respect, resembles that
which exists when I
perceive a red patch. Hence all available evidence
supports the
hypothe
sis that the ' something ' in question is a particular
brain
process. This does not imply however that' brain process' and
'
inner experience ' have the same meaning.
1
J.J.C.Smart,op.cit.
,p.
79
302
PHYSICS AND COMMON SENSE
The brain process the
ory is often considered as an implausible
last
step in the physicalist's reductive programme. I shall assume
here
however, without further discussion, that this theory, if
formulated and defended entirely independently of physicalism, is
an entirely sensib
le,
unobjectionable view, with considerable
ordinary empirical evidence
in its favour. I assume that it is in
full accordance with our ordinary
experience, and by no means
implies that men are merely mechanisms,
or that life can have no
significance or val
ue. In other words, I assume
that the brain process
theory, formulated independently of physicalism,
is in full
accordance with the ' common sense ' theory. It is, I think, in part
just because the brain process theory is thought to presuppose
physicalism
that it is usually considered to be so implausible.
1
Objections to Smart's physicalism lie elsewhere, in connection
with his' discriminatory response/brain process' analysis of qualities
such as
colours. (It should be noted that acceptance of a version of
the brain process theory does not automatically implicate us in an
acceptance of
the role which this theory plays in Smart's analysis of
colours.) Crude
ly, the obvious objection to Smart's physicalism is
surely this: the
theory just does not account for t
he existence of
qualities like
redness.
Whenever we see a red object surely we just
are indubitably aware of
the existence, somewhere in the world, of a
unique, unanalysable quality
which we may call' redness'. Yet
according to Smart it is just this
kind o
f quality that does not exist
at all, either in things or in our minds.
Unfortunately, in order to give a precise formulation to this
apparently simple objection, it seems it is necessary to raise the
following
somewhat intricate considerations.
Smart does
not always make it quite clear which of the two
follow
ing positions he is defending:
(i) The physicalist analysis of
sentences such as' This is red ' explicates
satisfactorily what we
ordinarily mean by this sentence,
(ii) ' This is red ', as ordinarily
understood, is always
false;
hence all perceptions of colours are
delusive, and colours are only pheno
menalqualities.
Clearly Smart must be defending one or other of these
positions,
but not both. It would seem that the following
objection to the first
p
osition is decisive: an analysis of' This is red
' in terms of discrimin
atory responses of certain biological
mechanisms misses the most
important part of what is ordinarily
meant bythis sentence, namely that
1
ForadefenceofthesepointsseeN.Maxwel
l,
PhysicsandCommon
Sense, a critique
of Physicalism,
M.A. thesis, 1965, Manchester
University, pp.72
-
126.
303
NICHOLAS MAXWELL
unanalysable part which can only be understood by actually
seeing
a
red object, or, at the very least, by experiencing
the
visual sensation of
redness. To this Smart might reply ' But this
experience can in turn
be analysed in terms of some brain process,
which clearly could be fully
described without mentioning any
such quality as " redness " '. The
answer to this (acce
pting the
brain process theory as formulated here) is
that in order fully to
understand ' This is red ' it is necessary to have the
brain process in
question occur in one's own brain; it is wholly irrelev
ant whether
or not one is in possession of a descri
ption of this brain
process.
In having this brain process occur in one's own brain for the
first
time something wholly new is learnt; one discovers what red
things are like, just that which the congenitally blind never know.
To this the following qualif
ication must be added. It is just
possible, as far as our knowledge goes at present, that the
public
meaning
of' This is red ' (i.e. that meaning which is common to
all the rather
different meanings which people
may
give to ' This is
red ') can be com
plet
ely analysed in Smartian terms. This would
be the case if, from
person to person, the inner experience that
accompanied the perception
of a red object was different. Suppose
the brain process that accom
panies my perception of a red object
is A, while the
brain process that accompanies your perception of
a red object is B; suppose further that
if B is induced by some
means in my brain I experience a visual sensa
tion which I regard
as quite different from ' experiencing the visual sensation of
redness'. In
this case I would say ' What I mean by
" This is red "
is quite different from what you mean '. Nonetheless a
part of what
I meant by ' This is red ' would be the same as what you
meant
by this sentence. This ' public ' part would be explicated
entirely
by
a Smartian analysis.
However, given these circumstances (which may very well not
be the case) the
private
meaning which I give to 'This is red
'
would still
be a perfectly genuine meaning. For another person
to understand
this private meaning it would be n
ecessary only for
that person to have
a brain process which I judged to be
sufficiently like A in the relevant
respect in order to constitute for
that person ' experiencing the visual
sensation of redness'. Thus
arguments such as Wittgenstein's against the
possibility of a
private language,
1
do not apply here, since
a definite procedure
exists for determining whether two people have
similar inner
experiences.
To sum up, we must reject the contention that Smart's
analysis of
1
L.Wittgenstein,
PhilosophicalI
nvestigations,
Blackwell,Oxford,
1958,§258,etc.
304
PHYSICS
AND COMMON
SENSE
colours explicates all that we ordinarily mean by such sentences as '
This
is red '.
But no doubt Smart does not wish to make the above claim for
his
analysis of colours.
Smart would maintain I think that a sentence
such
as' This is red ', if given its ordinary meaning, is always false.
In other words Smart is defending a physicalist theory of
perception, according to which all perceptions of colours (as
ordinarily underst
ood) are false,
i.e. according to which colours (as
ordinarily understood) are pheno
menal qualities only.
The question arises: Does this theory satisfactorily explain why
all
perceptions of colours are delusive, why colours are
phenomenal qualities only?
It is important to realise that Smart's
discriminatory response/brain process theory does not in itself
explain why certain
perceptual qualities are only phenomenal.
After all, to any perceptual
quality whatsoever there will
correspond certain discriminato
ry res
ponses and brain processes of
human beings. Hence merely from the
existence of such
discriminatory responses and brain processes we can
not deduce that
the quality in question is only phenomenal, and does not
really, or
objectively, exist. Thus the
theory is no more than a device
whereby the term for a quality which has been shown to be
phenomenal
can be given a physicalist interpretation.
Why then does Smart maintain that colours as ordinarily
understood
are phenomenal? Basically Smart distinguishes
physicalist from
phenomenal qualities as follows: Physicalist
qualities are just those
which physicists will ascribe to the
fundamental physical entities and
to all possible assemblies of
fundamental entities, once (or if) they form
ulate a satisfactory,
basic,
all
-
embracing physical theory. Phenomenal
qualities are then those
qualities which we do ordinarily attribute to
objects and appear to
perceive, but whose existence cannot be predicted
from a full
knowledge of the properties of the fundamental entit
ies.
But the
mere fact that a physical theory does not predict the existence
of a
perceptual quality cannot in itself be a sufficient reason for con
-
cluding that that quality does not really exist. In fact perception of
a
quality not predicted by a physica
l theory would
refute
that theory
un
less it could be shown that either (i) our perception of the
quality is
always delusive, or (2) the theory does not imply that the
quality does
not exist. Hence nothing that has been mentioned so
far supports the
hypoth
esis that all perceptions of colours are
delusive.
Smart does provide certain
ad hoc
arguments in support of the
hypo
thesis that qualities such as colours do not really, objectively
exist, and
305
NICHOLAS MAXWELL
therefore that our perceptions of colou
rs are always delusive.
For
example he argues that the view that colours have an objective
existence
is rendered extremely implausible by the fact that the
physical property
which corresponds to any colour is usually
complex and, from the point of view of
physics, wholly
arbitrary.
1
Thus, to light of any
specific colour, there corresponds
infinitely many mixtures of light of
different wavelengths and
intensities. But these considerations show
only that from the point
of view of physics, classification in te
rms of colours seems
arbitrary; it does not follow that colours do not really,
objectively
exist. From the point of view of the person who perceives
colours,
classification in terms of colours is very far from arbitrary.
It is true that doubt may be raised
as to what is to count as a
veridical
experience of colour, particularly by the findings
reported by E. H. Land.
2
But any such doubt must presuppose in
the end that
some
experience of colouris veridical.
3
An Acceptable Physicalism
I wish to turn now to
an examination of our original problem:
How can physicalism be interpreted so as to render it consistent
with
an acceptable version of the ' common sense ' theory?
In essence the solution I offer to this problem is very simple.
Both theories provide compre
hensive descriptions of the world.
These two
descriptions are however compatible, since they are not
both the same
kind
of description. Thus each description applies to
almost all there is,
and yet does not tell us all that there is to know
about that to w
hich it applies. Our task in what follows will be to:
(a)
Explicate the precise requirements which each description
must
fulfil.
(b)
Show that an ideal, fundamental physical theory would
support
physicalism given this interpretation.
(c)
Show that the two
kinds of description are compatible.
(d)
Refute such claims as that onlyphysicalism describes the world
as it
really, essentially
or
objectively
is.
(a)
In saying what exists at any moment we can only say what
that
which exists is
like,
in some respect o
r other. A number of very
differ
ent
kinds
of resemblances can be found between things. The
physical
ist and the '
common sense' descriptions classify things in
terms of
different
kinds
of resemblances between things.
1
J. J. C. Smart, op. cit. pp. 69
-
72
.
2
S
cientificAmerican,
May1959,200(5),84
-
99.
306
PHYSICS AND COMMON SENSE
I suggest that physicalism is to be interpreted as providing
the
following kind of classification of things:
(i) That which exists at any instant is classified, in the simplest
possible
way (i.e. with the smallest possible basic vocabulary), in
terms of
causal consequences, i.e. is described in such a say that
descriptions
of what exists subsequently can be deduced. (In the
terminology
of physics, given an isolated system, that w
hich exists
at any instant
is described in such a way that descriptions of
subsequent states of
thesystemcan be deduced.)
(ii) Things are classified only in terms of those resemblances which
any
intelligent being, however its sensory equipment may be
con
struc
ted, can discern, discover, become aware of.
It is assumed
that these two requirements are compatible.
It should perhaps be
emphasised that the first requirement makes
the following
presupposition: ' It is at least possible, or conceivable,
that from
a
true description of what exists at one instant, true descrip
tions of
what exists subsequently can be deduced.' (It is precisely this
which is presupposed by a tentative realist interpretation of the
kind of
' ideal' fundamental physical theory descr
ibed on p. 297.)
Hume of
course rejected the above proposition. Hence in
accepting the first
requirement for a physicalist description (and,
incidentally, in accepting
a realist interpretation to physical
theories), we are committed to deny
ing the va
lidity of Hume's
defence of the proposition ' It is not possible,
not conceivable, that
from a true description of what exists at one instant,
true
descriptions of what exists subsequently can be deduced '.
1
The ways in which ' common sense ' descriptions
classify things
are
less easy to specify precisely. In general, things are classified
solely in
terms of resemblances easily discernible to human
beings, in terms of
resemblances associated with the experiences,
interests, and purposes of
human beings.
A ' common sense '
description of an object may
classify that object in terms of
(amongst other things):
(i) What the object looks lik
e, feels like, etc., to a human
being,
(ii) The object's past (e.g. how the object was made, where
it came
fr
om).
(iii) What the object is use
d for.
(iv) The object's causal p
roperties.
Often these, and other, kinds of classifications are combined
in
any one description, although of course one kind may
predominate.
1
For a refutation of Hume on this point see N.
Maxwell, op.
cit. pp.57
-
69
30
7
NICHOLAS MAXWELL
(b)
It is clear enough, from our previous discussion, that a
certain
kind of legitimate fundamental physical theory would, if
given a tenta
tive realist interpretation, support physicalism in the
abov
e sense. Whether or not contemporary physics supports such
a version of
physicalism is, as I have already indicated, a
controversial issue.
(c)
It is further quite clear that it is possible for physicalism and
the
' common sense' theory, given the above in
terpretations, to
be com
patible. If physicalism satisfies the above two
requirements for a
physicalist description, then qualities that are,
for example, discernible
only to beings with sensory equipment
similar to those of human
beings will fall wholly o
utside the
province of the theory. Physicalism will neither imply that such
qualities exist, nor imply that such qualities
do not exist. At the
most the theory would imply, given a physicalist description of a
human being, that that human being will make c
ertain
kinds of
discriminations in certain situations, and will have certain brain
processes. Since physicalism does not imply the non
-
existence of
the
quality I discern in seeing a tomato say and call ' redness', it
must be
compatible with physicalism to
say of an object ' This is
red ' in this
sense.
It should be noted that the requirements for the physicalist
des
cription do not necessarily imply an altogether sharp
distinction be
tween physicalist and non
-
physicalist qualities.
Thus it might be
maintain
ed that even such apparently typical non
-
physicalist qualities as
colours are physicalist qualities, since they
are (i) not altogether causally
inefficacious, (ii) perhaps
perceivable by any intelligent being if we
permit sufficiently
drastic brain surgery
.
(d)
We turn now to a refutation of the claims that only the
physicalist
description is
really
true, that only physicalism describes
the world as it
really,essentially
or
objectively
is.
The question ' Is the world
really
as described by the ''
common
se
nse " theory or as described by physicalism?' only
makes sense if the
two descriptions are incompatible, if only one
can be true. But as we
have already indicated, a comparison of the
requirements for each des
cription makes it quite clear that it is
perfe
ctly possible for both des
criptions to be true. Hence the
above question makes no more sense
than: ' Is this rod
really
one
foot or twelve inches long?' Both
colours and electrons
really
exist, although they are very different sorts
of things.
It may be a
rgued that fundamental particles, and all aggregates
of
such particles, are
essentially
colourless,
essentially
without
perceptual
308
PHYSICS
AND
COMMON
SENSE
properties, precisely the reverse being true of anything to which
the
'
common sense'
description refers. Hence the two kinds of
descrip
tion are incompatible: only one can be true.
Bu
t what is meant bythe phrase '
the
essential
properties of a thing'?
The following two related definitions may be given. The essential
properties of an elect
ron, for example, are
(a)
thosewhich we consider
to be the most important,or (b
)thosewhich a thingmustpossessif it is
to be called an electron. Clearly in giving the essential properties of
theelectron, accordingtothesecond definition, we are gi
vingnomore
than an explanation of what we mean by the term ' electron '. There
will be a tendency to define 'electron' in terms of the properties which
we consider to be the most important. For this reason the two defini
-
tions ofessential properties wil
l tend to beequivalent.
From this explication of the notion of' essence ', it follows that
we
may grant that the essential natures of electrons and stones say are
differ
ent without thereby being committed to maintaining that
the two
kinds of descriptions
are incompatible.
It is truethat from acompletephysicalist descriptionaloneitwould
beimpossibletodeduce
theperceptual
qualitiesofthings,butthisisdue,
not to the fact that things do not really possess perceptual qualities,
but to the fact t
hat the physicalist description is
incomplete:
it does
not tell
us all that there is to know about the world. It does not tell us
what it
is like to be a human being alive and experiencing in the
world. In
particular, it tells us little about that aspect o
f objects which
wediscoverindirect perception.
Misunderstanding ofthispoint may to some extent be responsible
for the view, defended for example byR. Harre,
1
that stones and elec
-
trons belong to different ontological classes, that both stones and
ele
ctrons exist, but in different senses of exist'. Certainly the ' essen
-
tial ' properties of stones and electrons are very different. Further, the
epistemological status of these properties is in each case very
different. Propositions that attribute percept
ual properties to stones
may be veri
fied in a straightforward way by observation:
propositions which attribute physicalist properties to fundamental
physical entities are
never verified; at most they survive all attempts
to refute them. None
of this impli
es however that stones and
electrons exist in different
senses of exist'. Once we have made
quite clear what sort of thing a
stone oran electron is, then it canonly
be either true or false that stones
exist, either true or false that
electrons exist.
1
R
.Harr
é
,
Theories
andThings,
SheedandWard, London,
1961,p.85
309
NICHOLAS MAXWELL
Again, it may be argued that the physicalist kind of
description is
truer than other kinds because it is more objective.
Thus Smart
argues '. . . our ordinary mann
er of talking about the
world is suffused
with concepts which relate the things in the
world to our human con
cerns and interests, and which depend, in
often unnoticed ways, on our
human physiology and our particular
station in space
-
time. . . . If the
ant
hropocentricity inherent in
these concepts is not brought out into the open we can have a
misleading picture of the world. We think,
for example, of
objective colour
qualia
or of an objective
now,
much as the
Hebrews looked up at the dome of the sky and th
ought that this was
a solid half
-
spherical shell, or firmament, and did not realise that
the
apparently solid object was an illusion of their own
perspective.'
1
We may agree with Smart that '
Our ordinary manner of
talking
about the world is suffused with c
oncepts which relate the
things in the
world to our human concerns and interests, and
which depend
...
on
our human physiology'. As much is clear
from the conditions for the
'common sense
' description. Whether
or not it follows that such a 'manner of talki
ng about the world' is
subjective cannot be decided until the distinction between
subjectivity and objectivity has been
defined (see below). But
whatever
our
decision
on
this
point,
from
the
fact
that
the
'common sense' description is' suffused
with concepts
which relate
the things in the world to our human concerns and inter
ests, and
which depend
. . .
on our human physiology', it does
not
follow
that such a description is in any way false, or that the corres
-
ponding qualities do not exist. Descr
iptions such as ' That is
—
a
car,
a Picasso, green, a waste
-
paper
-
basket, etc.', relate the things
described
to human concerns and interests, and can only fully be
understood and verified by human beings, but are not, on that
account, false. The fact
that y
ou must possess a human
physiology in order to perceive the
greenness of things does not
imply that grass is not really green.
We may grant that if typically ' common sense ' descriptions
are
employed as if they meet both requirements for the physicalist
d
escrip
tion, then all such descriptions will be false. If colours, for
example,
are thought of as physicalist qualities, then it would be
false to attribute
colours to things. But we are not obliged to give
such an interpreta
tion to the concept of colour.
We are not
obliged to interpret the word ' red ' so that ' This is red ' is bound
to be false.
Finally, a few words about the distinction between objectivity
and
subjectivity. I wish to suggest that the traditional manner in
which
1
J.J.C.Smart,op.cit
.p.49
.
310
PHYSICS AND COMMON SENSE
this distinction is drawn makes an implicit appeal to Cartesian
dualism,
and must be rejected if we reject Cartesian dualism. It
might be stated
thus: a quality is objective if it exists
independently of the obse
rver, subjective if its existence depends
on the existence of the observer.
This has a clear enough meaning
if we accept a dualist theory of per
ception: physicalist qualities are
objective, while phenomenal quali
ties, qualities associated with the
experi
ences, concerns or emotions of
human beings, are
subjective. But if we reject dualism, the above
distinction
becomes wholly unclear.
I suggest that by objective we mean
inter
-
personal,
and by
subjective
we mean
personal,
that which is not inter
-
personal.
T
hus colours, relative to a group of non
-
colour
-
blind people, are
objective, but relative to all human beings, or, even more
generally, relative to all rational beings, are subjective. It should
be noted that in calling a
quality either objective or subject
ive we
are making an implicit appeal
to a group of people. It should also
be noted that in calling a
quality
subjective we do not imply that
the quality does not really exist (or does
not exist ' independently of
the observer '), and in calling a
descripti
on
subjective we do not
imply that the description is not really true. We
imply only that
not all members of the group of people in question can
perceive the
quality, fully understand the description.
Given this definition of objectivity, we may grant the
physicalist
that his is the only description that is objective
relative to all rational
beings.
UniversityofManchester
311
