from childhood onwards we ordinarily assume that when anyone looks at a chair or at his own hand the object which he perceives really has approximately the shape and colour that he sees, and that it existed before and will exist after his perceiving it. in the naive view a rose really is red, a penny round, a stone hard.
difficulties soon confront this theory. is the rose red by night? naive common sense would probably say, "yes, but its redness cannot then be seen." redness needs light to manifest it. if we point out that its manifested colour depends on the colour of the light thrown upon it, common sense will reply that to manifest its "true" or "real" colour it must be bathed in normal, neutral, white light. then what of the green hills which in the distance look blue? the reply might be that they "really" are green, but that the green light from them is modified by the intervening atmosphere. another difficulty lies in the fact that some people are colour-blind. what right have normal people to believe that their coloured view is more true, more objective, than that of the colour-blind man? is he not entitled to say that they "project" upon the external world qualities that are caused by their own sense-organs?
we may feel some doubt about colour, but we confidently assume that, at any rate, things really have the shapes and sizes that we see. but what of the straight walking-stick that looks bent when we dip it into a stream? the visual bentness, we are told, is mere "illusion," produced by the intervening water. the stick still feels straight, though it looks bent. touch is reliable, though vision sometimes misleads. then what of the single nose which, between crossed fingers, feels like two noses? evidently touch itself can mislead. ah! but, says common sense, the fingers are crossed. the situation is abnormal.
in all these examples one particular appearance, in each case, is taken as the "true" or "real " appearance, and the rest are said to be rendered " illusory " by special conditions. yet all the appearances are equally good appearances. if some of them do not really belong to the objects, how can we be sure that any of them do? what special power has white light to reveal the object's real colour, or uncrossed fingers to reveal the real nose?
let us consider the classical example of a penny's appearances. we say that it is round, though in certain views it appears or seems oval. by what right is the circular appearance alone assigned reality?
of course the flat surfaces of the penny can be geometrically defined as circular. each contains a point that is equidistant from every point on the circumference, equidistant, that is, if we look perpendicularly at the surface, but not otherwise. however, if we test vision by touch, we find that the penny can be smoothly rotated between the finger and thumb. this would not happen if it were not circular. once more, then, we refer the inconsistencies of sight to the judgment of touch. but we have seen that touch itself, though more constant, is not really infallible. moreover, whatever the verdict of touch, so far. as vision itself is concerned the oval appearances are just as good appearances of the geometrical circle as the circular appearances. why should the appearance from one particular angle be "real" and the others not? common sense would perhaps say that to secure a true visual appearance of the penny it is necessary to eliminate all distorting factors, all that lies "between" the object and us. we must get as "near" to it as possible. the hills (we are told) are not really blue, as they seem from a distance, but green, as they appear when we walk on them. the tilting of the penny, too, must be regarded as a distorting factor. but why, after all, should the angle of ninety degrees be less of a distorting factor than any other?
this principle of the distorting medium is certainly important, but its application is often arbitrary, and if it is pushed too far it leads to nonsense. it is arbitrary to choose one visual appearance and call it "real" merely because it happens to be the most convenient one to take as a symbol or a counter to represent the whole system of the object's appearances, as in the case of the circular view of the penny. on the other hand, the attempt to penetrate behind intervening media so as to grasp the very object itself may be pressed so far as to lead to results most bewildering to common sense. this is what happens when we consider perception from the point of view of natural science.
science raises two distinct types of problem for the common-sense theory of perception, namely, those problems that spring from physiology and those that spring from physics.
it is fairly well established that sense-perception involves sense-organs and is intimately associated with certain tracts of the cerebral cortex. in seeing, light from the object is focused by the lens of the eye so as to form an image on the retina. in the minute light- sensitive rods and cones of the retina the light sets up chemical changes which in turn set up changes in the nerve fibres. waves of chemical change pass along the fibres into the brain tract at the back of the head, which is the seat of vision. clearly, what actually reaches this visual cortex must be very different from the stimulus which originally affected the retina. if the object really has the colours and shapes which we perceive, how do they manage to be transported by the chemical nerve current?
a further difficulty is raised by the fact that though we perceive such different qualities as colour, sound, pressure, taste, scent, warmth, cold, pain, the nerve current is probably of one and the same quality in every case. how does an identical nerve current transmit such different qualities?
it seems that all our efforts to secure a true appearance of the object, undistorted by intervening media, are bound to be frustrated by the necessary intervention of the nerve current and the sense-organs.
so much for physiology. physics has other difficulties to raise. in its own way it has tried to "get at" the object itself, and has even claimed, in its nineteenth-century phase of optimism, to tell us what the object "really" is. the object, we have been told, is really a system of molecules, which are groups of atoms, which are electro-magnetic systems, wherein (it now seems) there is a ceaseless movement of excessively minute somethings that are not exactly particles and not exactly trains of waves, but rather like both. particles. of what? waves of what? seemingly of the potentiality of doing work, or (as some say) of sheer probability that something will happen.
where now is the smooth, hard, coloured surface of the object? where is sound? where scent, taste, warmth, cold? it begins to look as though they were mere figments of the mind; and as though the object itself were qualitatively unknowable, and in form at once grains and ripples of the unknowable quality.
other difficulties raised by science are connected with the nature of light. light is not instantaneous. it takes time to travel. when we look at a star, we look at something which may have actually ceased to exist thousands of years ago. in what sense, then, can we be said to perceive the star itself? strictly, all physical objects perceived by sight have already ceased to exist as present objects at least a fraction of a second before we see them. in a sense we see into the past.
in yet another manner light raises a difficulty. colour, it seems, involves time. things cannot have colour at an instant. the sensations of colour depend on rhythms of electro-magnetic vibrations. if the redness of the rose is a quality of the object itself, it must somehow be a quality which inheres in a rhythmically changing sequence of events. this seems very odd.
(a) dualism of matter and mind
(b) subjective idealism
(c) phenomenalism
(d) kant's criticism of phenomenalism
bearing in mind all these difficulties, both those of common sense and those raised by science, let us examine some typical solutions offered by philosophers. we will begin by considering the idealist solution, which was the first to be formulated.
the word idealism has many meanings. in ordinary speech it refers to ideals, or goals of action or of hope. in philosophy it seldom has this meaning. broadly, it may be taken to apply to any theory in which "idea" is regarded as a more significant concept than "matter," or "mental" than "material."
even so there are two very different kinds of philosophical idealism. in epistemology (the study of the nature of knowing) "idealism" is the theory that the objects of knowledge (including objects of perception) have no existence apart from the mental states or acts which know them. the contrary view, in which objects exist independently of our knowing them, is "realism."
on the other hand, in metaphysics (the study of the fundamental nature of reality) "idealism" is the theory that, even if the object of knowledge exists independently of our knowing it, its nature is nevertheless in some sense mental. in this chapter we are concerned only with idealism of the epistemological kind.
though the method of this book is not historical, it will be useful at this stage to summarise the growth of epistemological idealism. this famous theme in the history of modem european thought is necessary for the understanding of more recent thought about the external world and the knowing mind.
(a) dualism of matter and mind — starting with the common-sense distinction between matter and mind, john locke formulated a theory according to which "ideas" in the mind represent, or fail to represent, "things", in the external world. some of these "ideas" have qualities which are actually like the qualities of things. others have qualities which are unlike any qualities in things; though apart from quality the ideas may yet for practical purposes represent the things. thus arose the distinction between "primary" and "secondary" qualities. shape, size, number, motion, and solidity were said to be primary qualities belonging to objects themselves, as well as to our ideas. colour, sound, odour, etc. were said to be secondary qualities belonging only to our ideas, but corresponding to powers which a body possesses in virtue of its primary qualities. roughly, the qualities perceived by touch were regarded as primary, because they seemed to be constant. the rest, because variable, were secondary.
accepting the traditional way of thinking, locke regarded things as "substances," and the primary qualities as the "attributes" which inhered in these substances. minds also he regarded as substances, having, among their attributes, ideas corresponding to the primary and secondary qualities of things.
(b) subjective idealism — bishop berkeley showed that the distinction between primary and secondary qualities was unsound, that the theory of ideas as mental representatives of things was very confused, and also that the notion of material substances behind the perceived qualities of things was unnecessary.
our "ideas" of primary qualities are in principle variable, like our "ideas" of secondary qualities. the size and shape of the penny, even its tactual size and shape (perceived by touch), vary with the condition of the observer's body. if his fingers are swollen the penny will seem bigger than usual. size and shape vary also in relation to his previous experience. after handling farthings, pennies seem very large. movement also is relative. when two trains are running abreast at different speeds, the slower seems to passengers in the faster to be moving backwards.
the theory that ideas "represent" material substances distinct from themselves is open to the objection that if we have no direct experience of things, if we never experience them save through the medium of ideas, we have no reason to believe in their existence. also, if things do exist "behind the veil of ideas," we have no means of comparing the ideas with the things, and deciding that some ideas truly represent things and others do not.
berkeley's main positive contribution to philosophy consists in the famous principle that "to be is to be perceived." it seemed to him incredible that any material thing should be able to exist unperceived. accepting the principle, formulated by locke, that all, knowledge was derived from experience, he enquired if anyone ever had experience of any material thing that was not a perceived thing. the obvious answer was, no. therefore we had no right to postulate unperceived things. he held that there were no "material substances" but only "spiritual substances," namely, ourselves and god.
he did not, however, believe that when a thing was not being perceived by any human mind it ceased to exist. he did not, for instance, hold that the foundations of a house ceased to exist as soon as they were buried and out of sight. the humanly unperceived things, he said, were perceived by god, in whose mind they continued to exist.
(c) phenomenalism — the theories of berkeley were favourable to christian orthodoxy, but they contained the seed of heresy. david hume carried berkeley's own principles to their logical and devastating conclusion. he agreed with berkeley's criticisms of locke, but he pointed out that they must be applied not only to the external world but equally to the self and to god., if we had no evidence of the existence of material substances beyond or behind appearances, neither had we evidence of a soul or spiritual substance at the centre of them. what was actually experienced or given in experience was simply a flux of "impressions" and "ideas," or, in more modern language, sensations and mental images. though hume distinguished between "impressions" and "ideas" (sensations and images), he did not suppose that sensations were caused by an external world, and images not. he distinguished between them merely by the greater intensity or vividness of sensations. seeking to discover his "self," hume found that he always came upon some particular experience, never upon that which had the experiences. therefore, he argued, since all knowledge is derived from experience, we have no reason to believe either in physical substances or in the spiritual substance which the self is supposed to be. and if a man has no reason to believe in his own self as a spiritual substance, neither has he reason to believe in other selves, or in god. for intellect unaided by faith there is nothing but the stream of experience.
with hume the philosophical movement initiated by locke seemed to reach its logical conclusion. starting with the dualism of the perceiving mind and the perceived thing, it developed by way of subjective idealism into phenomenalism, the theory that there is nothing whatever but the sequence of experiences, or phenomena.
(d) kant's criticism of phenomenalism — idealist philosophers often seek to retain the essentials of berkeley's doctrine while rejecting the unpleasant consequences suggested by hume.
kant agreed with berkeley at least in believing that the whole knowable form or structure of the perceived world was a product of human experience. but he rejected hume's contention that this world of phenomena was the sole reality. there was for kant another reality behind appearances. with regard to the nature of the mind, he accepted hume's criticism of berkeley's belief in the individual self, but he protested that the concept of a stream of mere "impressions" and "ideas," wholly disconnected with one another, was false. an outstanding characteristic of experience is its unity. all the sensory elements of the present mental state are intimately related in a single experience. they are known together. and the actual character of each is largely determined by its relation to others. the present mental state, moreover, contains within itself knowledge of preceding states. it is what it is in virtue of past experience. to see an apple as an apple, we must have had previous experience of coloured shapes, and of touch and scent. in experience everything penetrates everything else.
to kant it seemed evident also that in perception we were actually affected by something other than ourselves, for our perceptions come to us whether we desire them or not. something, he said, forces them upon us. but the manner or form in which the reality appears to us, he said, is wholly an expression of our own nature. reality is entirely unknown to us save in terms of our own sensibility. the "thing in itself" we can never perceive as it really is. all we can perceive is appearances of it, and the form and quality of these is our own creation.
kant was not thinking merely of the physiology of sensation. he pointed out that everything perceived by us was perceived by means of, or in terms of, our own past experience, and in terms of certain fundamental categories, or fundamental potentialities of experiencing. these, he said, were inherent in mind itself. not only sensory qualities and shapes but space itself, time itself, and causation, he believed to be creatures of our minds, not characters of reality itself. no character whatever, knowable to man, could be assigned to reality, which, in kant's view, was simply the unknowable external cause of our conceiving phenomena.
our experience of ourselves as finite centres of consciousness, as selves, kant regarded as equally false. he agreed with hume that we could not discover in experience anything which might be regarded as distinctively "the self." in criticising hume's theory of the stream of impressions and ideas, kant did not rehabilitate the self, save as an abstract principle of the unity of experience, and the equally abstract principle of subjectivity.
according to kant there were two distinct worlds or spheres of being, one the "noumenal" world, the world of reality, which was wholly unknowable, the other the "phenomenal" world, the world of appearance, the sphere to which our knowledge applied, but in essence a world of illusion. for both physical appearance and the deliverance of introspection were given character by the mind itself.
briefly, the main criticism of kant's theory of perception and knowledge in general seems to be this. his unknowable "thing in itself" is unnecessary. the mere feeling that perception is forced on us by something other than ourselves is as likely to be a product of our own minds as anything else in our experience. consequently we have no reason at all for believing in a "thing in itself," which is different in kind from the actual objects of our experience. and so we are once more back in phenomenalism. all the same, kant's protest against the sheer subjectivity of phenomenalism was a step in the right direction. his mistake lay in setting a gulf between the known phenomena and the unknowable reality, for thus he re-introduced subjectivism. it is incredible that our experience should be all there is; but, further, there is no need to believe that the unexperienced is in principle unknowable.
(a) criticism of idealism
(b) the essence of realism
(c) difficulties in realism
(d) what we perceive
(e) an analysis of experience
(f) some types of realist theory
(a) criticism of idealism — "realism " like "idealism" has many senses. we are here concerned with realism as a theory of knowledge, and in particular of perception. in this sense realism starts with a denial of the first principle of idealism, namely, that to be is to be perceived.
the defence of idealism's central principle, it will be remembered, runs as follows. we never perceive anything that is not a perceived thing. therefore we have no reason to suppose that unperceived things exist.
this argument, says the realist, must be made more precise. i never experience anything that is not experienced by me. from this it might be argued that i have no reason to believe that anything exists that is not experienced by me. this theory, that my experience is all there is, is called solipsism, and is very different from berkeley's theory. berkeley's argument points further than he was prepared to go. he clung to the belief in many things other than his own perceptions. he believed in things unperceived by him, though perceived by other selves, or by god. this was irrational. in effect berkeley said: "everything that i experience is a thing experienced by me. therefore everything must be experienced by someone." the argument is illogical.
there is another fallacy in berkeley's argument. even if in my experience two things, namely, "what is perceived" and "my perceiving," always do, as a matter of fact, occur together, it does not follow that they are necessarily connected and could not occur apart. in my experience, of course, the perceived objects themselves and my activity of perceiving do always occur together. it does not follow that the objects involve the perceivings. of course, this argument does not prove that the objects do not involve the perceivings, or that they can exist apart. but if for other reasons we incline to believe that they exist apart, this argument enables us to do so in spite of berkeley. and we have other reasons, as berkeley himself might have realised. otherwise he would not have needed to bring in god to support the things that no man perceived.
the world of immediate experience, of sounds, smells, coloured shapes, tactual shapes, and so on is most intelligibly conceived as part of a much greater world with which. it is all. of a piece. suppose that i plant a seed in my garden. it is no longer to be seen. according to the account of the phenomenalist or the solipsist, the whole truth is that certain events have occurred in the world constituted by my immediate experiences, my phenomenal world. after some days there follow the experiences known as "the appearance of a seedling." this may be followed at intervals (whenever i go and look) by a series of experiences known as "the growth of a cabbage." so far as immediate experience is concerned there is no reason whatever for these fragmentary and disconnected experiences. but if we correlate them with the whole system of other people's experiences, comprising agriculture and botanical research, and if we conceive that in my absence unseen events were happening which might have been experienced, though actually they were not, we arrive at the concept of a germinating seed and a growing cabbage. thus, though we do not prove the existence of a world independent of experience, we do give a rather more intelligible account of my fragmentary experience than the phenomenalist's or the solipsist's account.
it is important to realise that what we are postulating is not an unexperienceable substance which has experienceable attributes, or an unknowable "thing in itself" upon which the mind mysteriously "projects" characters created by itself. we are postulating simply a sequence or tissue of objective events of the same kind as those which we do actually experience. we are not even postulating any sort of causation more inward or necessary than the observable regularity of the sequences of events. we are merely completing the fragmentary observed sequence by means of an unobserved but coherent system of events of the same order.
at this point it is perhaps well to mention an argument against realism which has been derived from modem physics. it is said to be impossible, in principle, to discover both the position and the momentum of an electron in its orbit. for in trying to observe the electron's momentum the observer inevitably alters its position, and vice versa. the answer to this argument is that it is not his observing that affects the situation, but simply the impact of the "light-waves" to which it must be subjected if it is to be observed at all.
(b) the essence of realism is the conviction that the objects of perception are not created or altered simply by being perceived. in the realist view experience is essentially a relation between the experiencer and something other than the experiencer; though, as we shall see, the special difficulties of realism have often, forced realists to give very strange accounts both of the relation and of its terms. this central principle of realism agrees with the common-sense theory of perception, and is therefore subject to many of the criticisms that we have already encountered in the common-sense theory. somehow it must meet these criticisms more effectively than common sense can do.
(c) difficulties in realism — the central difficulty of all realist theories is this. if it is true that the object of experience is not created by, or even changed by, being experienced, how comes it that apparently we sometimes experience things that do not exist?
the walking-stick dipped in the stream "looks" bent and "is" straight. where is the bent stick? in a dark lane we see a shadowy cow, which presently turns out to be no cow at all but a bush. the toper sees pink snakes that "do not exist." dreams are "sheer figments of the mind." imagination creates purely mythical beasts, like the centaur. we cannot escape the difficulty by saying that these objects are simply "ideas," which do not "correspond" with anything in the physical world. to do so would be to deny the essential principle of realism. we should be setting up a veil of mental stuff between the perceiver and the world. if we postulate purely mental objects in illusion, we must allow them also in true perception, since there is no discoverable inherent difference between the objects of illusion and the objects of perception. but to allow that the objects of perception are mental is to be forced, step by step, through the positions of locke and berkeley into hume's sheer phenomenalism.
(d) what we perceive — before trying to deal with this grave difficulty of error and illusion let us seek a clearer view of what we perceive when we suppose ourselves to be perceiving truly. we must also relate our findings to our scientific knowledge of physical objects.
i perceive a pear. i see it not merely as an area in my visual field, a pattern mottled in green, yellow, and red. i see it as a smooth-surfaced three-dimensional volume of a certain shape and at a certain distance. i vaguely perceive it also as fragrant and as internally wet and sweet. i can, if i will, distinguish between the characters which are given in sensation and those which are "imagery" recovered from my past experience.
if i approach the pear it appears larger. if i move round it, its perceived shape changes. evidently, if we cling to the principle of realism, we must believe that the perceivable shape of the pear itself consists of an infinite number of possible views from an infinite number of possible view-points. this is true equally of the pear's visual and of its tactual shape, save that for "views" we must substitute "touchings" and "graspings." no one of these visual and tactual " appearances" can logically have priority over the others. of course we ordinarily say that the pear itself has one particular geometrical shape in the physical world; but what is this shape really? it is an abstraction, a mathematical formula, derived from the actually perceived host of visual and tactual "views." it is a formula by means of which other such "views," perceivable from other view-points, can be calculated.
the world presented to my sight and touch is thus a world of "views," or (in bertrand russell's phrase) of "perspectives." the actual views that i perceive can be correlated to form a vast but fragmentary system of views all centred on my particular view-point, which can be located in the abstract geometrical space. this fragmentary system i can complete by correlating it with a host of views which i am not having, but which could be had from an infinite number of possible viewpoints.
the physical world, according to this theory, is an immense system of these views or aspects, which exist whether anyone is perceiving them or not. in this world a physical object, say a pear, is a particular sub-system of aspects which are all correlated with a certain volume in the abstract geometrical space. this abstract volume is that which common sense regards as the location of "the object itself." but strictly, what is concrete is not the geometrical volume but the aspects. the geometrical volume is an abstraction. we must certainly avoid the theory that the sensory characters which make up the object are simply on the surface of its geometrical volume. the blue of a distant mountain is not on its surface. neither is the greenness that appears in a closer view. even its hardness is not strictly confined to its volume, for hardness is simply resistance to pressure, and presupposes something other than the unpenetrated volume itself. as professor whitehead has said (but many realists reject his theory), sensory characters are not "simply located." they do not simply inhere in a particular place. they are essentially "in a place from a place." not only the mountain's distant blue, but also its nearer green, and even the touched hardness of its rock, are essentially "there from here." this being so, the more we press forward to reach the very object itself, located in its geometrical volume, the more of the object we lose on the way, until at last we find we have nothing whatever but abstract volume.
if we penetrate within the volume all that we find is more views, more aspects, more appearances, not of the object as a whole but of its parts. and these, in the last analysis possible to us, are such ghostly appearances as electrons, protons, positrons, neutrons, and electromagnetic vibrations. such abstruse objects we cannot adequately conceive in terms of our ordinary experience. in trying to conceive them we hopelessly misconceive them. as we have seen, they are thought of as neither particles nor wave-trains, yet something like both. but particles or wave-trains of what? we may try to conceive them as somehow consisting of sheer "pushfulness." nothing else is discoverable in them. indeed, even pushfulness ought not to be attributed to them. they are just movement. but movement of what? of something altogether inconceivable. they are completely abstract, mere factors in a complex mathematical formula. then what of the actual, concrete, characterful objects of experience? what of the mountain, the pear, the rose? and what of the limbs and heart and brain of the beloved? how can these be mere factors in a mathematical formula?
clearly, in our attempt to strip off all that does not belong actually to the object itself, we have completely missed all that is actual in the object. it seems as though the object were everywhere but in the volume where it is said to exist. take, for instance, one of the ultimate physical units studied by science, say an electron in one atom of the pear. it is not a minute thing in a particular region of the atom. it is a determination or influence pervading the whole atom. it is a factor in, and an expression of, a system. and what is the atom itself without its environment? it is essentially a member in a larger system. without an environment it can neither express itself nor even be itself. the case of the whole pear or mountain or beloved is the same. each is a factor in the world, a colourfulness pervading the world, and in turn a particular effect of, or expression of, the world's complex nature.
in case it is still thought that we are really misconceiving the whole matter, and that all these mysteries can be explained in terms of the substance confined within a certain concrete volume, and radiating effects into other such "located" substances, let us consider space itself. the very volume that the object is said to occupy, and from which it is said to scatter its effects, is something pervading the whole of space, is a system of possible views from every possible view-point, inside and outside the abstract volume. even a volume is "in a place from a place." it is a hole in space. and you cannot have a hole without an environment that is not the hole. space itself is nothing but the abstract system of all possible volumes, and each volume is a determination of the whole system.
(e) an analysis of experience — such, very roughly, is the world as it is perceived and scientifically studied. what is to be said of its relation to the perceiving and studying mind? let us first catalogue the various factors which philosophers have sometimes supposed to be involved in experience. some of these factors are denied by some philosophers, but let us begin by enumerating all the factors that have been suggested. we may summarise them as follows:
according to locke, the experient experiences only the mental content, not the material world. the mental content is composed of mental stuff, of "ideas"; and these "represent" the real objects. for subjective idealism there are no material objects; and the content is again mental. for phenomenalism also there is no physical thing behind phenomena; but also there is no experient and no experiencing. there is only mental content, or the stream of experience, which is not to be analysed into subject, act, and object.
(f) some types of realist theory — the only contention common to all forms of realism seems to be that, although mental content is not all there is, there must be no fundamental distinction between mental content and the "real" object that is experienced. content, however false, must be part of the objective world, not simply a creature of the mind.
the most radical type of realism is that which has been advocated by the american new realists and by bertrand russell. it is in many ways akin to phenomenalism, but it allows that there is more in the universe than experience. theories of this kind, which russell calls "neutral monism," describe the whole universe in terms of fleeting sensory elements (particular coloured shapes, noises, touches, and so on), called sense data, sensa (singular sensum), sensibilia (singular sensibile) or sensibles. these are said to be the only constituents both of the physical world and of the mind. these " neutral" elements or sensibles are said to be capable of being related together as two -distinct kinds of system, namely physical objects and minds. thus when i perceive a pear, a certain coloured shape enters into the system which is my mind; but it is also a member of the physical system which constitutes the pear. someone else, perceiving the same pear, picks up certain other visual components of the physical system of sensibles which constitutes the physical pear.
my perception of the pear includes much more than the characters now present to my senses. though i am not touching the pear, i perceive it as smooth, cool, softish. these characters i have sensed on other occasions, in association with other visual appearances of pears. on this occasion those past objects of experience are brought into mental relation with the present sensibles which i receive from this particular pear. i vaguely apprehend those past "cools," "smooths," "softs" again now, without noticing their location in the past.
but we must beware of the word "i." in this type of theory the perceiver, the "experient," is simply the whole system of " my" past experience, which. in perceiving this pear now, is assimilating new material to itself. "experiencing" is the relation between the whole system of "my" experienced objects and this newly-assimilated object.
in this type of theory, in spite of much that is of permanent importance, there are many serious difficulties. we are left wondering how a mere system of sensibles can, so to speak, "know itself together" as a single mind; and further, how this very strange sort of mind, which is entirely composed of objects and has no subjectivity, is capable of striving and feeling. i shall not here describe the ingenious ways in which supporters of theories of this type meet these difficulties. instead, i shall try to show that the theory, in its extreme form, breaks down over the central realist problem, namely error and illusion.
let us take the case of the bush which in the dark lane appears as a cow. the vaguely cow-like and bush-like shape reinstates in my present consciousness certain past sensibles belonging not to bushes but to cows. metaphorically, i see the past sensibles now, and fail to note their pastness. my error consists in ignorance. to believe in the illusory cow is to fail to see that the cow-like shape is not a complete cow.
this account of illusion is certainly correct up to a point, but is it the whole truth? is error simply ignorance, or is it something positive, based on ignorance? does not the error consist, not merely in failing to distinguish between the present and past sensibles, but in jumping to an unjustifiable conclusion? for the deluded observer does not merely say to himself (in effect), " here are certain cow-like characters, along with certain bush-like characters." he says, "that is a cow." if the mind is entirely passive, is simply a "cross-section of the universe" (as some say), and nothing but a fragmentary bit of the world, how can it commit error at all? a mind that was purely receptive would be simply a faithful recorder, a camera. and a camera, we are told, cannot lie. of course, in a sense a camera often does lie, and flagrantly; but the lying is really the work of the observer's mind when he falsely interprets the photograph. similarly in the bush-cow incident, the mind does, of course, passively receive a sort of "composite photograph" of past and present; but this photograph itself does not lie. nor does the error consist in passively accepting the photograph. the error consists in actively misinterpreting the data to mean much more than is actually given, namely to mean a present cow.
impressed by these considerations, some realist philosophers, for instance the critical realists of america, have reintroduced in one form or another the concept of mental activity. for them, experiencing is not simply a passive relationship but a positive act of assertion or judgment, which may fall into positive error. in doing this, they have also to reintroduce, though in a would-be realist manner, the distinction between the mental content and the physical thing.
in this view the bush-cow illusion must be described somewhat as follows. the mind passively receives certain vague bush-like but also cow-like shapes, and these recall past experience of cows. the mind thereupon actively asserts this physical object to be a whole cow. the mental content, then, consists of the universal character "cow," which is asserted to belong to this present physical object.
this account raises the whole question of the status of universal characters, such as redness, cowishness, justice. are there any such things? or are they sheer figments? let us for the present assume that there are such things as universal characters. then according to the theory under discussion, error consists in asserting that a certain system of universal characters belongs to a certain object when in fact it does not. the theory is clearly in danger of falling into the old dualism of ideas and things, and thence into phenomenalism. if in the case of illusion we distinguish between mental content and the thing, we must also do so in the case of true perception. this looks like abandoning realism. can this defeat be avoided?
let us begin by temporarily agreeing that the mind does first apprehend universal characters only in particular events; but let us allow it the power of reverting to past instances of such universals. we must not suppose that it detaches the universal from the particular situations in the past and turns it somehow into "idea." we must conceive that the mind somehow reverts to the past situation, and apprehends again the character of that situation, or rather the character common to many situations of a certain type. but in saying this we raise very serious questions about the nature of "the mind" and about the status of universals. granting that there are such entities as universals, and that particular instances of them are somehow discoverable in particular situations, what sort of being have they? how can a universal character be common to, or identical in, a number of instances? further, if the mind apprehends patterns of universals, which may be either true of the "thing " or false, are we not once more putting a veil between the mind and the actual world?
evidently this less radical kind of realism, though it avoids the difficulties of the more radical kind, has difficulties of its own, and is in danger of collapsing into dualism and phenomenalism. evidently, also, we cannot properly judge it till we have formed a clear opinion about universals. this in due course we shall attempt. but first we must consider another kind of theory of the external world, a theory which is in some respect akin to realism, but is said by its advocates to avoid realism's pitfalls, along with all other pitfalls of epistemology. this it does by the simple denial that any kind of epistemology is possible.
the theory of logical positivism is one which i discuss with great diffidence, for i came upon it late in life, and have not the equipment to judge it authoritatively. this, perhaps, is enough to account for my deep, though perhaps ill-founded, conviction that it is one of those many brilliant achievements of specialism which, though immensely fertile, are weakened by inadequate basic assumptions.
as we have seen, the theory accepts actual sense-experience, and refuses to accept any entities that cannot in theory be verified in sense-experience. like phenomenalism, it regards sensations as simple events not analysable into a knower, a mental act, and a known. it agrees with berkeley and hume in holding that sensations (or "sense-contents," as it sometimes calls them) cannot exist unperceived. or rather, it says that the proposition "sense-contents exist unperceived" is meaningless, since it can never be verified in sense-experience. thus logical positivism denies one of the principles of realism. on the other hand, and surprisingly, logical positivism agrees in a manner with locke, since it allows that material things do (in a very peculiar sense) exist unperceived. material things, it says, are not composed of sensory characters, or of sense-contents; they are said to be "logical constructions out of" sense-contents. this does not mean that they. are mere mental figments. they are what they are, independently of any particular mind's perceiving. and so, in a sense, they exist unperceived. but what they are is of the nature not of actualities but of possibilities. the logical positivist agrees with john stuart mill's dictum that physical objects are "permanent possibilities of sensation." in calling a thing a pear we name a complex and systematic possibility of sensations, past and future. the word " pear" is a symbol definable in terms of all the possible sensations commonly said to be sensations "of a pear." this possibility exists whether anyone sees the pear or not. but it is not to be interpreted as a hidden "something" behind the sensations. it is simply the systematic manner in which sensations of a certain class have occurred and may be expected to recur. the definition of "a pear" is a formula which is true of this class of sensations.
in exactly the same way the definitions of the objects of science, such as atoms and electrons, are formulae descriptive of possible sensations, namely the sensations resulting from the hosts of experiments on which atomic and electronic theory is based. the status of atoms and of pears is identical. both are objective possibilities of sensation. they are not hidden causes of sensation.
the logical positivists refuse to tell us what causes these possibilities. causation is for them simply the observed invariable regularity of sense-experience. about any reality behind sense-experience it is impossible, in their view, to say anything that is not nonsense. our knowledge is strictly confined to the world of sense-experience. to have any genuine meaning at all, a statement must refer to this world. questions and statements about non-sensible causes of sensation are mere verbiage.
in considering any theory we must seek clear consciousness of its particular emotional bias. in this case the bias is in favour of what may be called "intellectual puritanism," or the longing to be independent of all irrational beliefs. intellectual puritanism is horrifying to the mind that craves metaphysical illumination, but it is seductive to the intellectual puritan himself. indeed the passion for scepticism may sometimes induce an irrational conviction that a theory must be true simply because it is intellectually puritanical. we must try to avoid this prejudice no less than its opposite. the only relevant question is this. is the theory that perceived objects and scientific objects are merely "logical constructions" really an adequate account of our experience of them?
before accepting the theory we should have to be given a precise idea as to what a "logical construction " really is. but i shall not pursue that question. it is sufficient that a "logical construction" is a formula derived from experience, and useful for the prediction of future experience of the same type. if this is correct, certain very awkward consequences seem to me to follow from the theory that physical objects are logical constructions.
if physical objects are logical constructions out of sense-experience, so are minds. the logical defence of my belief in other minds must be derived from sense-experience of their bodies. logical positivism unhesitatingly agrees that minds are logical constructions, not only "other minds" but "my mind." from certain factors in sense-experience i logically construct "physical objects," from other factors i construct "other minds," and from yet other factors i construct "my mind." those parts of sense-experience which are the ground of the logical construction of "my mind" are those (whatever they be) in virtue of which i talk about my feelings, my desiring, my thinking, and so on. now according to the logical positivist i have precisely the same kind of sensory ground for the logical construction of "other minds" as for the logical construction of" my mind." according to him "other minds" are logically constructed solely out of my perception of the behaviour of other human bodies, and "my mind" is constructed solely out of my perception of the behaviour of my own body. but is this ingenious theory strictly true? i do not "construct" "my mind" out of public data available both to me and to other people observing my behaviour. i "construct" it out of a very special kind of data. my knowing-feeling-striving is matter of immediate acquaintance, for me, but not for others. they may infer it from my observable behaviour; but they do so only by analogy, from acquaintance with their own knowing-feeling-striving. similarly i, for my part, may infer "other minds" from observation of the behaviour of other bodies and the analogy with my own behaviour and my own knowing-feeling-striving. but from observation alone i cannot construct other minds, but only robots. "other minds," in fact, cannot conceivably be verified in sense-experience. the proposition that there are "other minds" is therefore meaningless. "other minds" are mere "metaphysical entities," or pseudo-entities. in fact, the logical outcome of logical positivism is solipsism. i am aware that logical positivists deny this; but i cannot see that they do so with reason. they claim that "the only distinction between a conscious man and an unconscious machine resolves itself into a distinction between different types of perceptible behaviour" (ayer). but this is not true. by saying that another person is conscious i do not mean merely that he observably behaves purposefully. i mean that he experiences, that he, like myself, knows-feels-strives. i have no access to his experience; my own is constantly with me. no doubt it is very difficult to say precisely what this introspective "awareness of my awareness " really is, but to deny its existence, and to say that i know my own experiencing in the same manner as i know the experiencing of others is not to clarify. but to confuse thought.
if it is true that " other minds " are something more than logical constructions out of sense-experience, more than formulae, then physical objects may also be so. the fact that we have no idea what character they have apart from sense-experience is no logical ground for asserting that they are merely possibilities of sense-experience. they are complexes of experienceable characters lying both within and without our experience. this, so far as i can see, is the core of the matter. in some sense, which we do not pretend to be able to state clearly, physical objects are to be thought of as actualities (not mere possibilities) extending beyond the system of human experience. this being so, the realist account of them seems to me on the whole the most plausible.
the same argument applies to scientific objects, such as electrons. but since the scope for error is immensely greater in respect of these than in respect of perceivable objects, we can never (in the present state of science) be sure that the official account of scientific objects is more true than false. moreover, scientific objects cannot be conceived in terms of the characters of sense-experience. they have not colour, sound, smell, taste. even shape and hardness belong to them, if at all, in a very metaphorical sense. consequently, though they are actualities, we are almost wholly ignorant of their nature.
this is not the place to pursue the problem of perception into all its ramifications. here we are concerned only with the bearing of philosophy on life, and we must be content merely to gain some realisation of the confusion of our thought about the nature of perception, and to note certain tentative conclusions which seem to be justified.
the upshot of this chapter seems to be briefly as follows. in the first place we may, and indeed we must, cling to the central principle of realism, and believe that there is a world that exists whether anyone perceives it or not. this world is not in principle unknowable, though, of course, all that we know of it is very fragmentary and slight. when we perceive, we really do perceive something of this objective world. this something turns out to be very different from the common-sense account of it, and even more different from the scientific account of it. but we have reason to believe that, whatever the precise logical status of the perceived world, it is not sheer illusion. it is all of a piece with the vast unknown, not a veil between us and reality. we may be confident that reality is not the quality-less abstraction, the fog of mere numbers, the haze of probabilities, that is the upshot of physics. it is at least as rich and colourful as the shreds of it that we perceive. and it contains, amongst other things, the living and sometimes lovely forms of our fellow human beings. no doubt, the perceivable aspects of physical objects in the external world are inextricably tangled up with our own sense-organs and nervous systems. no doubt, what we perceive is partly an expression of our particular dispositions and past experience. but what we perceive is not simply fictitious. moreover, our own sense-organs and brains are themselves physical objects, part of the great tissue of physical existence. and though in our simplicity we may perceive things all jumbled together that are in important manners distinct, so that we form a very grotesque view of reality, still when, for instance, a man perceives his beloved, the form that he perceives really is a character of the objective world, though its relations to other factors in the world may be very difficult and perhaps impossible to state clearly. similarly, though he himself, the perceiving mind, is not the simple thing that he seemed, and is in fact inextricably tangled up with the rest of the world, he is not a mere phantom, or a mere abstraction from the unity of his experience. at a much later stage of our enquiry we shall try to discover in more detail what kind of a thing he is.
had space permitted, this chapter would have included a discussion of the pragmatists' account of sense-perception. i append a note as a starting-point for study.
the pragmatists' main contention is that knowledge affords, not insight into the objective world, but merely formulae useful for action. thus the atomic theory is true simply in that it leads to successful practical operations. in respect of perception, pragmatists seem to divide into two camps. professor f. c. s. schiller holds that a perception is true or false simply as a sign-post leading to success or failure in our enterprises. it does not reveal the actual character of the objective world. this view leads to subjective idealism and, i believe, to solipsism. william james, on the other hand, admits that perception begins with the reception of something which is objective to the perceiving mind. his position is therefore at bottom realist.