 |
 |
 |
 |
 |
|
|
|
|
|
SENSE DATA AND THE PERCEPT THEORY RODERICK FIRTH 1949-50  |
3. THE PERCEPT THEORY The thesis that ordinary perception is, as James puts it, "one state of mind or nothing", has been systematically defended by advocates of the Percept Theory. They have tried to show that perceptual consciousness is not a twofold state by proving that all those things which are in any way present to consciousness in perception are present in exactly the same way. Or, to describe their method more precisely, they have tried to show that it is impossible to discover within perception the two types of consciousness which are essential not only for the truth but for the very meaningfulness of the Sense-datum Theory. The Gestalt psychologists, in particular, deserve credit for the patient and methodical manner in which they have presented evidence against this distinction; they have begun by showing how artificial it is and have ended by presenting arguments to destroy it completely. The Sense-datum Theory has been made particularly vulnerable to such criticism by a shift of opinion among its supporters concerning a certain phenomenological question -- the question namely, whether or not we are ever directly aware of depth in visual perception. There are still a few contemporary epistemologists who seem to believe, along with Locke and Berkeley, that depth is a conceptual or interpretational element in visual perception; they regularly speak of visual sense-data as "patches" of colour and describe their shapes in the language of plane geometry. But among contemporary psychologists who are investigating perceptual phenomena, there seems to be general agreement that there is no phenomenological justification for making this traditional distinction between visual depth and the other two spatial dimensions. And Price, an epistemologist whose analysis of perceptual consciousness is extremely acute, says specifically that our sense-datum when we look at a tomato has "a certain visual depth"29 From the point of view of those who defend the Percept Theory, the important fact about this shift of opinion within the older school of thought is not the manner in which the area of direct awareness is now delimited within perceptual consciousness. The important fact is that this shift of opinion represents a first step towards the recognition that in perception we are conscious of many qualities and relations which do not differ in their phenomenological status from those few which have traditionally been attributed to sense-data. Thus it is but one small step, as the Gestalt psychologists have shown, to the recognition that such qualities as simplicity, regularity, harmoniousness, clumsiness, gracefulness, and all the innumerable so-called "shape qualities" can also have the same phenomenological status as colour and shape. 30 And it is but one small additional step from this to the recognition that the same holds true of qualities fittingly described by such adjectives as "reptilian", "feline", "ethereal", "substantial", and perhaps most of the adjectives in the dictionary. And this, of course, finally forces the admission that the qualities belonging to objects of direct awareness cannot be thought of as limited in the manner traditionally assumed, by the use of one or another particular organ of sense. For it may sometimes, indeed, be quite correct to say that the experience of a man looking at a distant mountain from a warm room comprises both whiteness and coldness, each in precisely the same manner, and neither in any other manner. John Dewey has discussed some of these phenomenological facts in his Art as Experience and has pointed out that they do not clash in any way with our knowledge of physiology. The organic processes which condition perceptual experience are not limited to processes in a particular sense organ; the eye or the ear, as Dewey puts it, is "only the channel through which the total response takes place". Hence it should not be very surprising to discover that the so-called "visual qualities" do not always occupy a unique or central place in visual perception. "When we perceive, by means of the eyes as causal aids, the liquidity of water, the coldness of ice, the solidity of rocks, the bareness of trees in winter, it is certain that other qualities than those of the eye are conspicuous and controlling in perception. And it is as certain as anything can be that optical qualities do not stand by themselves with tactual and emotive qualities clinging to their skirts."31 Dewey's primary objective in the discussion
from which these sentences are quoted is
to refute what I have called the first point
of agreement among philosophers and psychologists
who accept the Sense- datum Theory; his primary
objective, in other words, is to show that
we cannot find within ordinary perceptual
consciousness a limited set of qualities
having the unique phenomenological status
which has been thought to distinguish the
objects of direct awareness. But Dewey's
description of perceptual experience also
contains an implicit criticism of the second
point of agreement among those who accept
the Sense-datum Theory, and thus serves as
an introduction to the final, and perhaps
most important step, in the development of
the Percept Theory. It must not be inferred, however, that James and others who deny that perception is a twofold state would not admit that there are certain types of disposition of the perceiving organism, or possibly even certain types of conscious experience, which invariably accompany our perceptual consciousness of physical objects. Such possibilities must be considered in determining the epistemological implications of the Percept Theory and will be briefly discussed in Part II of this paper, but they are irrelevant to the central thesis of the Percept Theory. For the duality denied by those who accept this theory is a duality of what we may perhaps call the sensuous aspect of consciousness at the moment of ordinary perception. At that moment, they maintain, we are conscious in a certain manner of a physical object which is somehow presented to us completely clothed in sensuous qualities. These qualities are presented as qualities of the object; indeed they are in no sense abstracted or otherwise distinguished from the presented object; and they are not limited to the qualities which have traditionally been mentioned in descriptions of sense-data. And finally, according to those who accept the Percept Theory, the sensuously clothed object is the only sensuous content of consciousness during ordinary perception. Sense-data, of course, if they exist as the objects of pure states of direct awareness, may properly be described as having sensuous qualities; but sense-data do not occur as constituents of perceptual consciousness. There is, of course, considerable disagreement about the proper way to express this conclusion. The traditional Sense-datum Theory is based on a supposed distinction between the direct awareness of sense-data and the mediated consciousness or "perception" of physical objects; those who reject this distinction, therefore, are rejecting not only the traditional concept of sense-datum but also the theory that ordinary perception is mediated in some manner or other, 34 by the presence of sense-data. Since both aspects of the traditional distinction must stand or fall together, there is no unambiguous way in which the traditional terminology can be used to express the positive conclusions of those who accept the Percept Theory. On the one hand it is probably misleading for advocates of the Percept Theory to assert bluntly, as they frequently do, that in perception we are directly aware of physical objects. For in addition to suggesting the very distinction which the Percept Theory rejects, the word "directly", when used in such a context, may have certain epistemological connotations which are not relevant to the phenomenological issue. 35 On the other hand it would create an unnecessary paradox if advocates of the Percept Theory were to assert that in ordinary perceptual experience we are not directly aware of anything; and this too might have confusing epistemological connotations. Consequently, even though either of these two modes of expression would be adequate if it were carefully explained, I shall avoid both of them in the following pages. William James sometimes used the word "percept" to refer to the content of consciousness during perception; it is this fact which has made the name "Percept Theory" seem to me appropriate for the particular theory of perceptual consciousness which he himself supported. If we were to adapt this terminology to satisfy our present need we could say that according to the Percept Theory we are presented in ordinary perception not with a sense-datum but with "object-percept" only, and we could speak more specifically, when necessary, of "cat-percepts", "mountain percepts", etc. For the problems to be discussed in the following pages, however, the terminology used by Price is even more convenient. Price uses the term "ostensible material object" to refer to that part of the content of perceptual consciousness which is not a sense-datum; it is thus possible to express the fact that in perception we are conscious in a certain manner of a physical object, by saying that we are presented with an ostensible physical object. Price himself does not accept the Percept Theory, but those who do may describe their position by saying that in ordinary perception we are presented with ostensible physical objects but not with sense-data. 36 The following passage from Price, as a matter of fact, provides a fitting conclusion for this section of the discussion, for it aptly describes the unmediated character of perceptual consciousness to which supporters of the Percept Theory have tried to draw attention. That Price could write a passage like this and still accept the Sense-datum Theory, is a mystery of the kind which the next section is intended to solve "Somehow it is the whole thing, and not just a jejune extract from it, which is before the mind from the first. From the first it is the complete material thing, with back, sides, and insides as well as front, that we 'accept', that 'ostends itself' to us, and nothing less; a thing, too, persistng through time both before and after . . . and possessed of various causal characteristics. . . . Already in this single act, even in a momentary glance, we take all these elements of the object to be there, all of them, as Mr. Joseph has said in another connection, we must not suppose that because there is only a little definite before the mind, therefore there is only a definite little."37. Notes 4. METHODOLOGICAL PROBLEMS In the pages immediately following, therefore, I shall discuss a number of possible errors which might account, at least in part, for the popularity of the Sense-datum Theory. Such a procedure seems likely to be more fruitful, considering the history of this issue, than further efforts to review the phenomenological evidence for the Percept Theory or to seek for more felicitous ways of describing perceptual consciousness.
It has been frequently suggested, first of all, that some of the philosophers and psychologists who accept the Sense datum Theory have committed what is sometimes called "the physiological fallacy".39 The physiological fallacy consists in assuming, a priori, some particular type of relationship between physiological facts and phenomenological facts. It is empirically demonstrable, for example, that the nature of particular states of perceptual consciousness is determined partly by the direct physiological effects of the stimulus, and partly by the past experience and present interests of the perceiver. But to conclude from these facts alone that there must be at least two constituents in every state of perceptual consciousness, one of them (the sense-datum) corresponding in some simple fashion to the direct physiological effects of the stimulus, would be to commit to the physiological fallacy and obscure the actual character of perceptual consciousness. In reply to this fallacious form of argument,
if made explicit, it would be sufficient
to point out that the only way to decide
a question of this sort is by direct inspection
of perceptual consciousness itself. But Dewey
and others have also shown that so far as
our present knowledge of physiology is concerned,
the Sense-datum Theory is not even favoured
by considerations of elegance. 40 Some philosophers
have complicated the matter by actually defining
a sense-datum as that constituent of a perception
which is caused by the physical stimulus.
Russell does this in his Philosophy although
elsewhere he accepts the conventional definition
in terms of direct awareness. A sense-datum,
he says in Philosophy, is "the core,
in a perception, which is solely due to the
stimulus and the sense-organ, not to past
experience."41 But even if we were to
accept this definition, it would still remain
an open question whether there are sense-data
as so defined, and it is still true that
this can be decided only by direct irispection
of perceptual consciousness. But the Sense-datum Theory could no longer be distinguished from the Percept Theory, of course, if the assertion that perceptual consciousness includes direct awareness of a sense-datum were interpreted to mean merely that perceptual consciousness involves an intrinsic point of view. Almost all the philosophers who accept the Sense-datum Theory, moreover, have made statements which are incompatible with the proposition that visual sense-data are the front surfaces of ostensible physical objects. Thus to clarify the relationship between sense-data and the interpretational or conceptual element in perception, they say that the latter may vary with our past experience and present attitudes even though the former remain unchanged, and that the former may change even though the latter do not vary at all. 43 As naive children, for example, our awareness of a purple sense-datum when we look at a distant mountain might be accompanied by the perception of a purple mountain, whereas an exactly similar sense-datum, occurring at a later age, might be accompanied by the perception of a green mountain. And for similar reasons the sense-data produced by tomatoes in a dark cellar might vary from light grey to dark grey with changes in the illumination, although the ostensible tomatoes might at the same time remain uniformly red. According to those who accept the Sense-datum Theory, in short, the qualities of sense-data and the qualities of ostensible physical objects can vary independently to some extent, and this implies, of course, that sense-data cannot be identified with the front surfaces of ostensible objects. If the Sense-datum Theory were true, indeed, it would rarely it ever, he correct to apply the same determinate adjectives both to an ostensible physical object and to the sense-datum which is presented along with it. Thus we should have to maintain that whenever we are perceiving a physical object with a surface which is ostensibly red and circular, we are also directly aware of a sense-datum which is probably elliptical in shape and which may very well be orange or purple or grey in color. We should have to maintain, in short, that even when we look at a single physical object we are almost always conscious, though in different ways, of two colours and two shapes. To those who support the Percept Theory it seems so clear that ordinary perception is not characterised by any such duality, that they may perhaps be excused for suspecting that their opponents, when they actually examine a state of perceptual consciousness, contradict their own theory by identifying visual sense-data and the front surfaces of ostensible physical objects. If this fallacy is committed by any of the philosophers and psychologists who accept the Sense-datum Theory, it is probably committed chiefly by those whose conception of the perceptual consciousness of physical objects is over intellectualised. Such philosophers and psychologists are likely to conceive of perceptual consciousness as a twofold state consisting of direct awareness of a sense-datum and a purely intellectual or purely conceptual "interpretation" of this sense-datum. Our consciousness of a particular physical object in perception is consequently thought of as nothing more than a state of belief that there exists a physical object of a particular kind, and the special sensuous character of this mode of consciousness is completely overlooked. It is easy to see that such a conception could blind one to the very phenomenological facts which would correct it, and lead to the fallacy of identifying sense-data and the front surfaces of ostensible physical objects. For even if we should decide that it is appropriate to describe our perceptual consciousness of physical objects as a kind of belief, 44 it is surely as very special kind of belief -- the kind, namely, that is characterised by the presence of an ostensible physical object. But an ostensible physical object, as supporters of the Percept Theory have tried to point out, is presented, or appears, or "ostends itself" fully clothed in sensuous qualities. If, therefore, a philosopher or psychologist were to suppose, because he accepted a priori an over-intellectualised conception of perceptual consciousness, that only the traditional objects of direct awareness can have sensuous qualities, he could very easily fall into the error of believing that the front surface of an ostensible physical object is a sense-datum. It is difficult to believe, however, that this error could explain the acceptance of the Sense-datum Theory by those philosophers and psychologists who show quite clearly that they are fully aware of the sensuous character of ostensible physical objects. Price, for example, has made a characteristically acute analysis of what he calls the "pseudo-intuitive" features of our perceptual consciousness of physical objects. 45 He criticizes those whom he calls "Rational Idealists" for their over-intellectualised conception of this mode of consciousness, and points out that it is actually very similar to direct awareness of sense-data. He quots with approval Husserl's statement that the object of perception is "leibhaft gegeben", and adds that it "just comes, along with the sense-datum: it just dawns upon us, of itself".46 In fact Price's only reason for refusing to say that our consciousness of physical objects is intuitive, appears to be epistemological rather than phenomenological. 47 There is little doubt, therefore, that Price is fully aware of the sensuous character of ostensible physical objects. He seems to recognise that a ripe tomato hanging on a vine in the sun is "leibhaft gegeben" with all its sensuous qualities of redness, and smoothness, and warmth, and sweetness. Yet he would also maintain that when we look at the tomato we are ordinarily presented with another entity, a sense-datum, which may have qualities quite unlike those of the ostensible tomato. For those who support the Percept Theory it is difficult to see how there could be room, so to speak, for such conflicting sets of sensuous qualities in one and the same state of perceptual consciousness, and even Price says that in ordinary perception we "fail to distinguish" between the sense-datum and the ostensible object. 48 He does not seem to doubt, nevertheless, that there is a sense-datum in every perception, and a sense-datum which can be distinguished from the front surface of an ostensible physical object.
This brings us finally to what is probably the most interesting explanation for the resistance which contemporary epistemologists have shown to the Percept Theory. For although it seems unlikely that careful phenomenologists like Price have committed either of the two fallacies so far discussed, Price's comment that in ordinary perception we "fail to distinguish" between the sense-datum and the ostensible physical object, does suggest that he, and perhaps others, are accepting a methodological presupposition which supporters of the Percept Theory would wish to reject. Our state of mind in perception, Price says, "is, as it were, a dreamy half-awake state, in which we are unaware of a difference between the sense-datum and the ostensible physical object".49 And this naturally raises the question: How can anyone claim to know this particular fact about perceptual consciousness and at the same time believe that there is evidence to support the Sense-datum Theory? If it be admitted, in other words, that in perception we are not aware of any difference between the sense-datum and the ostensible physical object, what possible evidence could there be that both of them are present to consciousness during perception? 50 The importance of this question is also indicated by certain passages in Broad's discussion of perceptual consciousness in Scientific Thought. 51 To illustrate what I have called "the Sensory Core Theory", Broad draws an analogy between sense-data and printed words. In reading a familiar language, he says, "what interests us as a rule is the meaning of the printed words, not the peculiarities of the print. We do not explicitly notice the latter unless there be something markedly wrong with it, such as a letter upside down. . . . In exactly the same way", he explains, "we are not as a rule interested in sensa". We ordinarily notice them only when they are queer, as when we see double, though "even in a normal case, we generally can detect the properties of sensa . . . provided that we make a special effort of attention". These statements raise the same methodological question: If it be admitted that in perception we ordinarily do not "explicitly notice" sensa, what possible evidence could there be that we are actually conscious of them? Now it is quite possible that statements like these just quoted from Broad and Price are the results of careful phenomenology; it is possible that these philosophers can actually discover within a single perception the two levels of consciousness suggested by their statements -- a level of "explicit awareness" and a level of "implicit awareness". In that case the evidence for the fact that we are implicitly aware of sense-data in perception can be obtained by direct inspection of perceptual consciousness, in the very same manner, indeed, in which we would proceed to obtain evidence for the fact that we do not "explicitly notice" such sense-data or that we "fail to distinguish" them. And in that case it might be argued that the difference between the Sense-datum Theory and the Percept Theory is little more than a difference about the meaning of such expressions as "present to consciousness"; for perhaps in the limited sense designated by the word "explicitly", most of the philosophers and psychologists who accept the Sense-datum Theory would be quite willing to admit that sense-data are not ordinarily present to consciousness during perception.
Most supporters of the Percept Theory, however, have made it quite clear that their opposition to the Sense-datum Theory could not be mitigated by rephrasing the issue in terms of any such verbal distinction. They have said that direct inspection of perceptual consciousness convinces them that sense-data as traditionally described are not present at all, thus implying that perceptual consciousness simply does not manifest the levels suggested by the use of such expressions as "explicltly notice". There are, to be sure, various interpretations of "explicit" and "implicit" which would make these words applicable to the content of perceptual consciousness; indeed it would be surprising if there were not in view of the complexity of perceptual phenomena. But those who support the Percept Theory seem to be unanimous in their belief that direct inspection does not reveal the presence of sense-data, either explicitly or implicitly, within the ordinary perceptual consciousness. It is understandable therefore, that supporters of the Percept Theory should look elsewhere for an explanation of the fact that some philosophers and psychologists still cling to the Sense-datum Theory even though they seem to be admitting that we are not actually aware of sense-data at the moment of perception. And the explanation which naturally suggests itself is that these philosophers and psychologists would in fact admit the truth of the Percept Theory if they limited themselves to direct inspection of perceptual consciousness, but that they also employ another method, wittingly or unwittingly, in which they have greater faith. This possibility is easily illustrated by referring to the statements I have quoted from Broad. Broad asserts, as an empirical fact, that in perception we do not explicitly notice our sense-data, just as in reading a book we do not explicitly notice the print. If this is an empirical fact, however, then Broad presumably discovered it by direct inspection of perceptual consciousness; he inspected his consciousness while reading and found that he was not explicitly noticing the print, and he inspected his consciousness during perception and found that he was not explicitly noticing sense-data. Nevertheless, Broad explains, we can detect the properties of the printed words by attending to the print "as in proof-reading", and we can similarly detect the properties of sensa by making "a special effort of attention". Now it is clear that the special act of "attending" by means of which we explicitly notice the print and the sense-data must be at least somewhat different from the act of direct inspection which, Broad admits, does not reveal the presence to consciousness of either the print or the sense-data. Any other conclusion would be self-contradictory, for if the act of "attending" were identical with this act of direct inspection, it would obviously be impossible for any one to discover that we do not explicitly notice either the print or the sense-data. And this suggests the possibility that some of the opposition to the Percept Theory can be explained on methodological grounds. It suggests that some of the contemporary philosophers and psychologists who accept the Sense-datum Theory may believe that there is a better method of discovering phenomenological facts than the method of examining the phenomenon directly. A good deal has been written on this subject, but it is still interesting and important, and deserves further attention. The method of settling phenomenological questions which has sometimes been regarded as better than the method of direct inspection, involves a unique operation which I shall call "perceptual reduction". This operation is familiar to everyone who has participated in discussions of the traditional problems of perception, because in such a context there is a certain use of the expression "really see" such that what we really see can be determined only by performing this operation, and not by direct inspection alone. Thus if I were asked simply what I see right now, I should probably reply: "A sheet of white paper"; but if I were asked what I really see, especially in the context of psychological or epistemological discussion, I should probably answer; "A patch of pale yellow". In the latter case I should assume that I was being asked to perform the operation of perceptual reduction first and then to describe my state of consciousness as revealed by direct inspection. And since this paper happens at the moment to be illuminated by artificial light, the answers to the two questions would, for physical and physiological reasons, be different. The operation of perceptual reduction has two rather distinct effects when it is performed on a state of perceptual consciousness. The first of these two effects is to make the ostensible physical object progressively less and less determinate. If I were to perform the operation while looking at a tomato, for example, the ostensible tomato which is present to consciousness would, so to speak, become less specifically distinguished as an individual. Starting as a tomato with worm holes it might be reduced to a tomato with "some sort of holes" in it, and then to a tomato with spots on its surface, and so on. It might eventually become "some sort of globular object", or even just "some sort of physical thing". But when this last stage is reached, or perhaps even before, there is a second effect: a radical cue takes place and a new object of consciousness appears and grows more and more determinate. Our state of consciousness is approaching a pure state of sense-datum awareness, and this new object is therefore not an ostensible physical object at all but the kind of thing which is corriectly called a sense-datum; and it is not until this second stage in the process has begun that we are able to describe what we "really see" and to report, for example, that we are presented with "a red patch of a round and somewhat bulgy shape". In fact the properties which we attribute to this new object of consciousness are usually incompatible with those which characterised the original ostensible physical object. This description of the effects of the operation of perceptual reduction is undoubtedly over-simplified. Some psychologists and philosophers would probably insist that in the final stage, when we become aware of a sense-datum, we are also conscious of an extremely indeterminate physical object -- that there are, in other words, no pure states of direct awareness. Others might maintain that the process by which the final stage is reached varies considerably from one occasion to another. But so far as the present issue is concerned, the only relevant fact is that the operation of perceptual reduction destroys the state of perceptual consciousness on which it is performed; it is an operation, to be precise, which has the effect of replacing a state of perceptual consciousness by a state in which we are aware of sense-data. It would be difficult to exaggerate the importance of the reduction throughout the history of modern philosophy and modern psychology. This operation reached the height of its importance in the psychological methods of Wund, Titchener, and their followers, who declared it to be the very essence of the experimental technique of introspective psychology. They interpreted it, however, as a procedure for cleansing perceptual consciousness of its nonsensory constituents. They believed, as Köhler says in criticizing them, that "as psychologists our task is to separate . . . 'meanings' from the seen material as such, the manifold of simple sensations". They admitted that "it may be a difficult task to effect this separation and to behold the net sensations which are the actual data"; but the ability to do so, they maintained, "is precisely the special talent which transforms the layman into a psychologist".52 Now whether or not it is misleading to say that the operation of perceptual reduction is a technique for "cleansing" perceptual consciousness, it is quite certain, I believe, that this operation differs from direct inspection. And it is also quite certain that it destroys the state of perceptual consciousness on which it is performed, for when we reach the final stage in which we are aware of a sense-datum, we are no longer presented with the fairly determinate ostensible physical object which originally existed. If, therefore, any of the philosophers and psychologists who accept the Sense-datum Theory have simply failed to notice the difference between direct inspection and perceptual reduction, it is fair to say that they have committed a very serious fallacy, and one which might explain the resistance which they offered to the Percept Theory. This fallacy is one particular form of what James called the "psychological fallacy par excellence" -- the fallacy of reading into a state of consciousness the characteristics of something(in this case another state of consciousness) which is externally related to it. 53
But perhaps there are some philosophers and psychologists who believe that perceptual reduction is a legitimate method for discovering the content of perceptual consciousness, and who are nevertheless fully aware of the difference between this method and the method of direct inspection. If they recognise this difference and the fact that the two methods yield incompatible conclusions about the nature of perceptual consciousness, and if they sincerely believe that perceptual reduction is the more trustworthy method, then it is perhaps not accurate to say that they have committed a fallacy. They are, however, accepting a debatable hypothesis which ought to be carefully formulated and examined. I shall call this hypothesis the "Exposure Hypothesis". According to the Exposure Hypothesis, the operation of perceptual reduction does not produce a state of consciousness which is simply other than the original state of perception on which it is performed. It produces, on the contrary, a state of direct awareness which was contained in the original perception. To put the case very simply, indeed, we might say that according to this hypothesis the only difference between the two states is that the sense-datum of which we are aware is obscured in the earlier one by the presence of an ostensible physical object. The method of perceptual reduction, therefore, is a method designed to expose the sense-data which are presumed to be contained in ordinary states of perceptual consciousness. This exposure is achieved, according to the hypothesis, by destroying the consciousness of physical objects which accompanies and obscures the sense-data, so that the bare sense-data themselves become accessible to subsequent acts of direct inspection. Now to grasp the full import of the Exposure Hypothesis, it should be recognised that it grants a unique and privileged epistemological status to the particular attitude (the "reducing attitude") which we adopt in order to initiate the process of perceptual reduction. This attitude of "doubt" or "questioning" is, to be sure, an attitude of special importance to the psychologist and perhaps to the artist, but it is, on the other hand, only one among a seemingly infinite number of attitudes which we can adopt in the presence of an ostensible physical object. There are mercenary attitudes and pedagogical attitudes and martial attitudes and so on indefinitely, and each one is capable of affecting the content of perceptual consciousness; thus even though the visual stimuli are similar, for example, the qualities of a tomato as seen by a hungry child will surely be very different from the qualities of a tomato as seen by someone looking for a missile to throw at a candidate for political office. And sometimes, moreover, such changes in attitude are consciously solicited -- as for example when we revisit some favourite childhood scene and try to recapture something of its former meaning. Yet it would scarcely occur to anyone to suggest that by forcing such changes of attitude we can find, in the resulting state of perceptual consciousness, the real but previously unobservable content of the original state. The two states would be regarded as related to one another, to be sure, by the fact that they are caused by the same external stimulus, but the one would scarcely be takers to be a constituent of the other. Those who accept the Exposure Hypothesis, therefore, have singled out one particular attitude from among the multitude which we can adopt in the presence of an ostensible physical object, and have attributed to this attitude the rare epistemological power of exposing otherwise unobservable characteristics of perceptual consciousness. And there does not appear to be the slightest empirical justification for this. If the truth of the Sense-datum Theory were assumed a priori, and it were also assumed that there is some procedure for discovering the sense-data within ordinary perceptual consciousness, then, indeed, it might be inferred that the reducing attitude must have the unique epistemological power attributed to it by the Exposure Hypothesis. But in that case, of course, the Exposure Hypothesis could not in turn be used as part of an argument to support the Sense-datum Theory. Whatever the empirically distinguishable features of the reducing attitude may be, they do not indicate that the operation of perceptual reduction is anything more than one method among many of substituting one state of consciousness for another. The only argument for the Exposure Hypothesis, so far as I know, which might have some appeal to an empiricist, is the argument that to deny this hypothesis is to cast suspicion on all intellectual analysis. It is possible that Price is employing this argument, for example, when he criticizes the view " that just as dissection destroys a living organism, so intellectual analysis destroys that which is analysed, and substitutes something else in its place". According if to this view, he continues, "since all thought may be regarded as analysis, we are forbidden to think"54 Lewis suggests the same argument when he writes: "The given is in, not before experience. But the condemnation of abstrction is the condemnation of thought itself. Nothing that thought can ever comprise is other than some abstraction which cannot exist in isolation."55 To infer, however, that the rejection of the Exposure Hypothesis casts suspicion on intellectual analysis, or that it implicitly denies the possibility of thought, is to overlook the very distinction which criticism of the Exposure Theory is intended to clarify -- the distinction, namely, between introspective reduction and direct inspection. For there is, of course, no inconsistency at all in asserting that introspective reduction is merely a process of substitution, and at the same time maintaining that there is another process -- i. e., direct inspection -- which is quite compatible with genuine intellectual analysis and which does not destroy the very thing which is to be analysed. "In intellectual analysis", says Price, "I do not do anything to the object before me. I find relations with it. I discover that it possesses various characteristics . . . . But those relations and characteristics were there before I discovered them. The only change that has occurred is a change in myself. I was ignorant and now I know."56 And surely there is no reason why one who rejects the Exposure Hypothesis must deny the possibility of such a process as this. As a matter of fact the supporters of the Percept Theory have been especially interested in describing the nature of perceptual consciousness, and such description requires analytical thought. The very assertion that perceptual consciousness is not a twofold state is itself the result of a kind of analytical process, and so is the more specific assertion that I am presented with an ostensible tomato clothed with certain sensuous qualities. But these assertions describe perceptual consciousness itself, as revealed by direct inspection, and not the substitute provided by the operation of perceptual reduction. Those who accept the Sense-datum Theory have made similar distinctions in explaining their own position. They have pointed out that we cannot learn more about a particular sense-datum by changing the physical conditions of observation, for we merely frustrate ourselves if we try by such means to "get a better look" at a sense-datum. By moving our bodies, by putting on spectacles, and by turning on the light, we may indeed learn more about the stimulus-object, but if we perform these operations in order to dissect our sense-data, then indeed "we murder to dissect". And those who reject the Exposure Hypothesis are arguing analogously that we frustrate ourselves if we perform the operation of perceptual reduction in order to describe perceptual consciousness. In neither case does the argument imply that analysis is impossible. But if the rejection of the Exposure Hypothesis does not imply the impossibility of analysis, it is difficult to see what argument could possibly be advanced to support it. Yet the hypothesis has apparently been accepted by many philosophers and psychologists in the past, and the history of the conflict between the Sense- datum Theory and the Percept Theory suggests that it will continue to be accepted for some time to come. There may, of course, be unexpressed arguments which have not been uncovered; and it is even possible that the three errors discussed in this section cannot account for more than a small part of what Köhler calls the "natural tendency" to favour the Sense-datum Theory. But if this is the case then those who support the Percept Theory will naturally hope that their opponents may soon provide them with a full explanation of the phenomenological or epistemological basis of the Sense-datum Theory. The next and final section of this paper is devoted to an examination of some of the epistemological implications of the Percept Theory. For many philosophers it is only the possible implications of this Theory which can give importance to the phenomenological issues which we have been discussing; and it is not impossible that the revolutionary nature of some of the supposed implications of the Percept Theory can account in part for the resistance which it has encountered among epistemologists. Notes |
