Perception and Its Object by P.F. Strawson
Ayer has always given the problem of perception a central peace in his thinking. Reasonably so; far a philosopher’s views on this question are a key both to his theory of knowledge in genera and to his metaphysics. The movement of Ayer’s own thought has been from phenomenalism to what he describes in his Iratest treatment of the topic as ‘a sophisticated Corro of realism’.\ The epithet is doubly apt. No adequate account of the matter can be simple; and Ayer’s account, white distinguished by his accustomed Lucidity and economy of style, is notably and subtly responsive to ait the complexities inherent in the subject itself and to ait the pressures of more or Is persuasive argument which have marked the course of its treatment by philosophers. Yet the form of realism he defends has another kind of sophistication about which it is possible to have reservations and doubts; and, though I am conscious of being far tram clear on the matter myself, I shall try to make some of my own doubts and reservations as clear as I can. I shall take as my text Chapters 4 and 5 of The Central Questions of Philosophy; and I shall also consider a different kind of realism-that advocated by J. L. Mackie in his book on Locke.2 There are points of contact as well as of contrast between Ayer’s and Mackie’s views. A comparison between them will heap to bring out the nature of my reservations about both.
According to Ayer, the starting-point of serious thought on the matter of perception consists in the fact that our normal perceptual judgements always ‘go beyond’ the sensible experience which gives rise to them; far those judgements carry implications which would not be carried by any ‘strict account’ of that experience.3 Ayer sees ordinary perceptual judgements as reflecting or embodying what he calls the common-sense view of the physical world, which is, among other things, a realist view; and he sees that view itself as having the character of ‘a theory with respect to the immediate data of perception’. 4 He devotes some space to an account of how the theory might be seen as capable of being developed by an individual observer on the basis of the data available to biro; though he disavows any intention of giving an actual history of the theory’s development. The purpose of the account is, rather, to bring out those features of sensible experience which make it possible to employ the theory successfully and which, indeed, justify acceptance of it. For it is, he holds, by and large an acceptable theory, even though the discoveries of physical science may require us to modify it in certain respects.
Evidently no infant is delivered into the world already equipped with what Ayer calls the common-sense view of it. That view has to be acquired; and it is open to the psychologist of infant learning to produce at least a speculative account of the stages of its acquisition. Ayer insists, as I have remarked, that his own account of a possible line of development or construction of the common-sense view is not intended as a speculative contribution to the theory of infant learning. It is intended, rather, as an analysis of the nature of mature or adult perceptual experience, an analysis designed to show just how certain features of mature sensible experience vindicate or sustain the common-sense view which is embodied or reflected in mature perceptual judgements. Clearly the two aims here distinguished-the genetic-psychological and the analytic-philosophical-are very different indeed, and it will be of great importance not to confuse them. In particular it will be important to run no risk of characterizing mature sensible experience in terms adequate at best only for the characterization of some stage of infantile experience. It is not clear that Ayer entirely avoids this danger.
What is clear is that if we accept Ayer’s starting-point, if we agree that Our ordinary perceptual judgements carry implications not carried by a ‘strict account’ of the sensible experience which gives rise to them, then we must make absolutely sure that our account of that experience, in the form it takes in our mature lire, is indeed strict-in the sense of strictly correct. Only so can we have any prospect of making a correct estimate of the further doctrines that the commonsense view of the world has the status of a theorywith respect to a type of sensible experience which provides data for the theory; that this experience supplies the evidence on which the theory is based; 5 that the common-sense view can be regarded as inferred or at least inferable from this evidence; and that our ordinary perceptual judgements have the character of interpretations,6 in the light of theory, of what sensible experience actually presents us with.
But can we-and should we-accept Ayer’s starting-point? 1 think that, suitably interpreted, we both can, and should, accept it. Two things will be required of a strict account of our sensible experience or of any particular episode or slice of sensible experience: first, as 1 have just remarked, that it should in no way distort or misrepresent the character of that experience as we actually enjoy it, that is, that it should be a true or faithful account; secondly, that its truth, in any particular case, should be independent of the truth of the associated perceptual judgement, that is, that it should remain true even if the associated perceptual judgement is false. It is the second requirement on which Ayer lays stress when he remarks that those judgements carry implications which would not be carried by any strict account of sensible experience; or, less happily in my opinion, that in making such judgements we take a step beyond what our sensible experience actually presents us with. But it is the first requirement to which 1 now wish to give some attention.
Suppose a non-philosophical observer gazing idly through a window. To him we address the request, ‘Give us a description of your current visual experience’, or ‘How is it with you, visually, at the moment?’ Uncaptioned as to exactly what we want, he might reply in some such terms as these: ‘I see the red light of the setting sun fettering through the black and thickly c1ustered branches of the elms; 1 see the dappled deer grazing in groups on the vivid green grass . . .’ and so on. So we explain to him. We explain that we want him to amend his account so that, without any sacrifice of fidelity to the experience as actually enjoyed, it nevertheless sheds ail that heavy load of commitment to propositions about the world which was carried by the description he gave. We want an account which confines itself strictly within the limits of the subjective episode, an account which would remain true even if he had seen nothing of what he claimed to see, even if he had been subject to total illusion.
Our observer is quick on the uptake. He does not start talking about lights and colours, patches and patterns. For he sees that to do so would be to falsify the character of the experience he actually enjoyed. He says, instead, ‘I understand. l’ve got to cut out of my report ail commitment to propositions about independently existing objects. Well, the simplest way to do this, white remaining faithful to the character of the experience as actually enjoyed is to put my previous report in inverted commas or oratio obliqua and describe my visual experience as such as it would have been natural to describe in these terms, had I not received this additional instruction. Thus: “I had a visual experience such as it would have been natural to describe by saying that I saw, etc. . . . [or, to describe in these words, ‘I saw . . . etc.’] was it not for the obligation to exclude commitment to propositions about independently existing objects.” In this way [continues the observer] I use the perceptual claim-the claim it was natural to make in the circumstances-in order to characterize my experience, without actually making the claim. I render the perceptual judgement internal to the characterization of the experience without actually asserting the content of the judgement. And this is really the best possible way of characterizing the experience. There are perhaps alternative locutions which might serve the purpose, so long as they are understood as being to the same effect-on the whole, the more artificial the better, since their artificiality will help to make it clearer just to what effect they are intended to be. Thus we might have: “It sensibly seemed to me just as if I were seeing such-and-such a scene” or “My visual experience can be characterized by saying that I saw what I saw, supposing I saw anything, as a scene of the following character . . .'”
If my observer is right in this-and I think he is-then certain genera conclusions folio. Our perceptual judgements, as Ayer remarks, embody or reflect a certain view of the world, as containing objects, variously propertied, located in a common space and continuing in their existence independently of our interrupted and relatively fleeting perceptions of them. Our making of such judgment implies our possession and application of concepts of such Objects. But now it appears that we cannot give a veridical characterization even of the sensible experience which these judgements, as Ayer expresses it, ‘go beyond’, without reference to those judgements themselves; that our sensible experience itself is thoroughly permeated with those concepts of objects which figure in such judgements. This does not mean, that is, it does not follow directly from this feature of sensible experience, that the general view of the world which those judgements reflect must be true. That would be too short a way with scepticism. But it does follow, I think, that our sensible experience could not have the character it does have unless-at least before philosophical reflection sets in- we unquestioningly took that general view of the world to be true. The concepts of the objective which we see to be indispensable to the veridical characterization of sensible experience simply would not be in this way indispensable unless those whose experience it was initially and unreflectively took such concepts to have application in the world.
This has a further consequence: the consequence that it is quite inappropriate to represent the general, realist view of the world which is reflected in our ordinary perceptual judgements as having the status of a theory with respect to sensible experience; that it is inappropriate to represent that experience as supplying the data for such a theory or the evidence on which it is based or from which it is inferred or inferrable; that it is inappropriate to speak of our ordinary perceptual judgements as having the character of an interpretation, in the light of theory, of the content of our sensible experience. The reason for this is simple. In order for some belief or set of beliefs to be correctly described as a theory in respect of certain data, it must be possible to describe the data on the basis of which the theory is held in terms which do not presuppose the acceptance of the theory on the part of those for whom the data ore data. But this is just the condition we have seen not to be satisfied in the case where the so-called data are the contents of sensible experience and the so-called theory is a general realist view of the world. The ‘data’ are laden with the ‘theory’. Sensible experience is permeated by concepts unreflective acceptance of the general applicability of which is a condition of its being so permeated, a condition of that experience being what it is; and these concepts are of realistically conceived objects.
I must make it quite dear what I am saying and what I am not saying here. I am talking of the ordinary non-philosophical man. I am talking of us ail before we felt, if ever we did feel, any indignation to respond to the solicitations of a general scepticism, to regard it as raising a problem. I am saying that it follows from the character of sensible experience as we ail actually enjoy it that a common-sense realist view of the world does not in general have the status of a theory in respect of that experience; white Ayer, as I understand biro, holds that it does. But I am not denying that to one who has seen, or thinks he has seen, that sensible experience might have the character it does have and yet a realist view or the world be false, to him the idea may well present itself that the best way of accounting for sensible experience as having that character is to accept the common realist view of the world or some variant of it. He might be said to adopt, as a theory, the doctrine that the common realist view of the world is, at least in some basic essentials, true. But this will be a philosopher’s theory, designed to deal with a philosopher’s problem. (I shall not here discuss its merits as such.) What I am concerned to dispute is the doctrine that a realist view of the world has, for any man, the status of a theory in relation to his sensible experience, a theory in the light of which he interprets that experience in making his perceptual judgements.
To put the point summarily, whereas Ayer says we take a step beyond our sensible experience in making our perceptual judgements, I say rather that we take a step back (in general) from our perceptual judgements in framing accounts of our sensible experience; far we have (in general) to include a reference to the former in framing a veridical description of the latter.
It may seem, on a superficial reading, which Ayer had anticipated and answered this objection. He introduces, as necessary for the characterization of our sensible experience, certain concepts of types of pattern, the names far which are borrowed from the names of ordinary physical objects. Thus he speaks of visual leaf patterns, chair patterns, cat patterns, and so on.7 At the same lime, he is careful, if I read him rightly, to guard against the impression that the use of this terminology commits him to the view that the employment of the corresponding physical-object concepts themselves is necessary to the characterization of our sensible experience.8 The terminology is appropriate (he holds) simply because those features of sensible experience to which the terminology is applied are the features which govern our identifications of the physical objects we think we see. They are the features, ‘implicitly noticed’, 9 which provide the main clues on which our everyday Judgements of perception are based.
This is ingenious, but I do not think it will do. This we can see more clearly if we use an invented, rather than a derived
terminology for these supposed features and then draws up a table of explicit correlations between the invented names and the physical object names. Each artificial feature Dame is set against the Dame of a type of physical object: our perceptual identifications of seen objects as of that type are held to be governed by implicit noticing of that feature. The nature and significance of the feature names is now quite clearly explained and we have to ask ourselves whether it is these rather than the associated physical-object terms that we ought to use if we are to give a quite strict and faithful account of our sensible experience. I think it is clear that this is not so; that the idea of our ordinary perceptual judgements as being invariably based upon, or invariably issuing from, awareness of such features is a myth. The situation is rather, as I have already argued that the employment of our ordinary, full-blooded concepts of physical objects is indispensable to a strict, and strictly veridical, account of our sensible experience.
Once again, I must make it clear what I am, and what I am not, saying. I have been speaking of the typical or standard case of mature sensible and perceptual experience. I have no interest at ail in denying the thesis that there also occur cases of sensible experience such that the employment of full-blooded concepts of physical objects would not be indispensable, and may be inappropriate, to giving a strict account of the experience. Such cases are of different types, and there is Ode in particular which is of interest in the present connection. An observer, gazing through his window, may perhaps, by an effort of will, bring himself to see, or even witlessly find himself seeing, what he knows to be the branches of the trees no longer as branches at all, but as an intricate pattern of dark lines of complex directions and shapes and various sizes against a background of varying shades of grey. The frame of mind in which we enjoy;-if-we ever do enjoy, this kind of experience is a rare and sophisticated, not a standard or normal, frame of mind. Perhaps the fact, if it is a fact, that we can bring ourselves into this frame of mind when we choose may be held to give a sense to the idea of our ‘implicitly noticing’ such patterns even when we are not in this frame of mind. If so, it is a sense very far removed from that which Ayer’s thesis requires. For that thesis requires not simply the possibility, but the actual occurrence, in all cases of perception, of sensible experience of this kind. Ode line of retreat may seem to lie open at this point: a retreat to the position of saying that the occurrence of such experiences may be inferred, even though we do not, in the hurry of lire, generally notice or recall their occurrence. But such a retreat would be the final irony. The items in question would have changed their status radically: instead of data for a common-sense theory of the world, they would appear as consequences of a sophisticated theory of the mind.
This concludes the first stage of my argument. I have argued that mature sensible experience (in general) presents itself as, in Kantian phrase, an immediate consciousness of the existence of things outside us. (Immediate, of course, does not mean infallible.) Hence, the common realist conception of the world does not have the character of a ‘theory’ in relation to the ‘data of sense’. I have not claimed that this fact is of itself sufficient to ‘refute’ scepticism or to provide a philosophical ‘demonstration’ of the truth of some form of realism; though I think it does provide the right starting point far reflection upon these enterprises. But that is another story and I shall not try to tell it here. My point so far is that the ordinary human commitment to a conceptual scheme of a realist character is not properly described, even in a stretched sense of the words, as a theoretical commitment. It is, rather, something given with the given.
But we are philosophers as well as men; and so must examine more closely the nature of the realist scheme to which we are pretheoretically committed and then consider whether we are not rationally constrained, as Locke and Mackie would maintain we are, to modify it quite radically in the light of our knowledge of physics and physiology. Should we not also, as philosophers, consider the question of whether we can rationally maintain any form of realism at all? Perhaps we should; but, as already remarked, that is a question I shall not consider here. My main object, in the present section, is to get a clear view of the main features of our pre-theoretical scheme before considering whether It IS defensible, as it stands, or not, I go in a somewhat roundabout Way to work.
I have spoken of our pre-theoretical scheme as realist in character. philosophers who treat of these questions commonly distinguish Different forms of realism So do both Ayer and Mackie. They both mention, at One extreme, a form of realism which Mackie calls ‘naive’ and even ‘very naive’, but which might more appropriately be called ‘confused realism’. A sufferer from confused realism fails to draw any distinction between sensible experiences (or ‘perceptions’) and independently existing things (or ‘objects perceived’) but is said (by Mackie expounding Hume) to credit the former with persistent unobserved existence. IO It should be remarked that, if this is an accurate way of describing the naive realist’s conception of the matter, he must be very confused indeed, since the expression ‘unobserved’ already implies the distinction which he is said to fail! to make. Speaking in his own person, Mackie gives no positive account of the naive realist’s view of things, but simply says that there is, historically, in the thought of each of us, a phase in which we fail to make the distinction in question.l1 It may indeed be so. The point is Ode to be referred to the experts on infantile development. But in any case the matter is not hereof any consequence. For we are concerned with mature perceptual experience and with the character of the scheme to which those who enjoy such experience are pre-theoretically committed. And it seems to me as certain as anything can be that, as an integral part of that scheme, we distinguish, naturally and unreflectively, between our seeing and hearings and feelings-our perceiving-of objects and the objects we see and hear and feel; and hence quite consistently accept both the interestedness of the former and the continuance in existence, unobserved, of the latter.
At the opposite extreme from naive realism stands what may be called scientific or Lockian realism This form of realism credits physical objects only with those of their properties which are mentioned in physical theory and physical explanation, including the causal explanation of our enjoyment of the kind of perceptual experience we in fact enjoy. It has the consequence that we do not, and indeed cannot, perceive objects as they reality are. It might be said that this consequence does not hold in an unqualified form. For we perceive (or seem to perceive) objects as having shape, size, and position; and they reality do have shape, size, and position and more or less such shape, size, and position as we seem to perceive them as having. But this reply misconstrues the intended force of the alleged consequence. We cannot in sense perception-the point is an old one-become aware of the shape, size and position of physical objects except by way of awareness of boundaries defined in some sensory mode-for example, by visual and tactile qualities such as scientific realism denies to the objects themselves; and no change in, or addition to, our sensory equipment could alter this fact. To perceive physical objects as, according to scientific realism, they really are would be to perceive them as lacking any such qualities. But this notion is self-contradictory. So it is a necessary consequence of this form of realism that we do not perceive objects as they really are. Indeed, in the sense of the pre-theoretical notion of perceiving-that is, of immediate awareness of things outside us-we do not, on the scientific-realist view, perceive physical objects at all. We are, rather, the victims of a systematic illusion which obstinately clings to us even if we embrace scientific realism For we continue to enjoy experience as of physical objects in space, objects of which the spatial characteristics and relations are defined by the sensible qualities we perceive them as having; but there are no such physical objects as these. The only true physical objects are items systematically correlated with and causally responsible for that experience; and the only sense in which we con be said to perceive them is just that they cause us to enjoy that experience.
In between the ‘naive’ and the ‘scientific’ varieties, Ayer and Mackie each recognize another Corro of realism, which they each ascribe to ‘common sense’. But there is a difference between Ayer’s version of common-sense realism and Mackie’s. For Mackie’s version, unlike Ayer’s, shares Ode crucial feature with scientific realism
The theory of perception associated with scientific or Lockian realism is commonly and reasonably described as a representative theory. Each of us seems to himself to be perceptually aware of objects of a certain kind: objects in space outside us with visual and tactile qualities. There are in fact, on this view, no such objects; but these object appearances can in a broad sense be said to be representative of those actual objects in space outside us which are systematically correlated with the appearances and causally responsible for them. The interesting feature of Mackie’s version of common-sense realism is that the theory of perception associated With it is no less a representative theory than that associated with Lockian realism The difference is simply that common sense according to Mackie, views object appearances as more faithful representatives of actual physical objects than the Lockian allows: in that common sense, gratuitously by scientific standards, credits actual objects in space outside us with visual and tactile as well as primary qualities As Mackie puts it, common sense allows ‘colours-as-we-see-them to be resemblances of qualities actually in the things’ 012 On both views, sensible experience has its Own, sensible objects; but the common-sense view, according to Mackie, allows a kind of resemblance between sensible and physical objects which the scientific view does not.
I hope it is already clear that this version of common-sense realism is quite different from what I have called our pre-theoretical scheme What we ordinarily take ourselves to be aware of in perception are not resemblances of physical things themselves This does not mean, as already remarked, that we have any difficulty in distinguishing between our experiences of seeing, hearing and feeling objects and the objects themselves. That distinction is as firmly a part of our pre-theoretical scheme as is our taking ourselves, in general, to be immediately aware of those objects. Nor does it mean that we take ourselves to be immune from illusion, hallucination, or mistake. We can, and do, perfect1y adequately describe such cases without what is, from the point of view of the pre-theoretical scheme, the quite gratuitous introduction of sensible objects interposed between us and the actual physical objects they are supposed to represent.
The odd thing about Mackie’s presentation is that at One point he shows himself to be perfect1y well aware of this feature of the real realism of common sense; far he writes, ‘What we seem to see, feel, hear and so on o . o ore seen as real things without us-that is, outside us. We just see things as being simply there, of such-and-such sorts, in such-and-such relations. . ‘ 13 He goes OD, of course, to say that ‘our seeing them so is logically distinct from their being so’, that we might be, and indeed are, wrong But he would scarcely dispute that what is thus seen as real and outside us is alg_ seen as coloured, as possessing visual qualities; that what is felt as a real thing outside us is also felt as hard or soft, smooth or rough-surfaced-as possessing tactile qualities The real realism of common sense, then, does indeed credit physical things with visual and tactile properties; but it does so not in the spirit of a notion of representative perception, but in the spirit of a notion of direct or immediate perception.
Mackie’s version of common-sense realism is, then, I maintain, a distortion of the actual pre-theoretical realism of common sense, a distortion which wrongly assimilates it, in a fundamental respect, 10 the Lockian realism he espouses. I do not find any comparable distortion in Ayer’s version. Re aptly describes the physical objects we seem to ourselves, and take ourselves, to perceive as ‘visuotactual continuants’. The scheme as he presents it allows for the distinction between these items and the experiences of perceiving them and far the causal dependence of the latter on the former; and does so, as far as I can see, without introducing the alien features I have discerned in Mackie’s account. It is perhaps debatable whether Ayer can consistently maintain the scheme’s freedom from such alien elements white continuing to represent it as having the status of a ‘theory’ in relation to the ‘data’ of sensible experience. But, having already set out my objections to that doctrine, I shall not pursue the point. .
Something more must be said, however, about the position, in the common-sense scheme, of the causaI relation between physical object and the experience of perceiving it. Although Ayer admits the relation to a place in the scheme, he seems to regard it as a somewhat sophisticated addition to the latter, a latecomer, as it was, far which room has to be made in an already settled arrangement.14 This seems to me wrong. The idea of the presence of the thing as accounting far, or being responsible far, our perceptual awareness of it is implicit in the pre-theoretical scheme from the very start. For we think of perception as a way, indeed the basic way, of informing ourselves about the world of independently existing things: we assume, that is to say, the general reliability of Our perceptual experiences; and that assumption is the same as the assumption of a genera causal dependence of our perceptual experiences on the independently existing things we take them to be of. The thought of my fleeting perception as a perception of a continuously and independently existing thing implicitly contains the thought that if the thing had not been there, I should not even have seemed to perceive it. It really should be obvious that with the distinction between independently existing objects and perceptual awareness of objects we already have the general notion of causal dependence of the latter on the former, even if this is not a matter to which we give much reflective attention in our pre-theoretical days.
Two things seem to have impeded recognition of this point. One is the fact that the correctness of the description of a perceptual experience as the perception of a certain physical thing logical/y requires the existence of that thing; and the logical is thought to exclude the causai connection, since only 10gicalIy distinct existences can be causally related. This is not a serious difficulty. The situation has many parallels. Gibbon would not be the historian of the decline and fall of the Roman Empire unless there had occurred some actual sequence of events more or less Corresponding to his narrative. But it is not enough, far him to merit that description, that such a sequence of events should have occurred and he should have written the sentences he did write. For him to qualify as the historian of these events there must be a causal chain connecting them with the writing of the sentences. Similarly, the memory of an event’s occurrence does not count as such unless it has its causal origin in that event. And the recently much canvassed ‘causal theory of reference’ merely calls attention to another instance of the causal link which obtains between thought and independently (and interiorly) existing thing when the former is rightly said to have the latter as its object.
The second impediment is slightly more subtle. We are philosophically accustomed-it is a Human legacy-to thinking of the simplest and mega obvious kind of causal relation as holding between types of item such that items of both types are observable or experienceable and such that observation or experience of either term of the relation is distinct from observation or experience of the other: that is, the causally related items are not only distinct existences, but alg_ the objects of distinct observations or experiences. We may then come to think of these conditions as constituting a requirement on ail primitive belief in causal relations, a requirement which could be modified or abandoned only in the interests of theory. Since we obviously cannot distinguish the observation of a physical object from the experience of observing it-for they are the same thing-we shall then be led to conclude that the idea of the causal dependence of perceptual experience on the perceived object cannot be even an implicit part of our pretheoretical scheme. but must be at best an essentially theoretical addition to it.
But the difficulty is spurious. By directing our attention to causal relations between objects of perception, we have simply been led to overlook the special character of perception itself. Of course, the requirement holds far causal relations between distinct objects of perception; but not far the relation between perception and its object. When x is a physical object and y is a perception of x, then x is observed and y is enjoyed. And in taking the enjoyment of y to be a perception of x, we ore implicitly taking it to be caused by x.
This concludes the second phase of my argument. I have tried to bring out some main features of the real realism of common sense and of the associated notion of perception. From the standpoint of common-sense realism we take ourselves to be immediately aware of real, enduring physical things in space, things endowed with visual and tactile properties; and we take it for granted that these enduring things are causally responsible for our interrupted perceptions of them. The immediacy which common sense attributes to perceptual awareness is in no way inconsistent either with the distinction between perceptual experience and thing perceived or with the causal dependence of the former on the latter or the existence of other causally necessary conditions of its occurrence. Neither is it inconsistent with the occurrence of perceptual mistake or illusion-a point, like so many others of importance, which is explicitly made by Kant. 15 Both Ayer and Mackie, explicitly or implicitly, acknowledge that the common-sense scheme includes this assumption of immediacy-Mackie in a passage I have quoted, Ayer in his description of the common-sense scheme. Unfortunately, Mackie’s acknowledgement of the fact is belied by his describing common-sense realism as representative in character and Ayer’s acknowledgement of it is put in doubt by his describing the common-sense scheme as having the status of a theory in relation to sensible experience.
It is One thing to describe the scheme of common sense; it is another to subject it to critical examination. This is the third and most difficult part of my task. The main question to be considered, as already indicated, is whether we are rationally bound to abandon, or radically to modify, the scheme in the light of scientific knowledge.
Before addressing ourselves directly to this question, it is worth stressing-indeed, it is essential to stress-the grip that common-sense non-representative realism has on our ordinary thinking. It is a view of the world which so thoroughly permeates our consciousness that even those who are intellectually convinced of its falsity remain subject to its power. Mackie admits as much, saying that even when we are trying to entertain a Lockian or scientific realism’ ‘our language and our natural ways of thinking keep pulling us back’ to a more primitive view. 16 Consider the character of those ordinary concepts of objects on the employment of which our lives, our transactions with each other and the world, depend: our concepts of cabbages, roads, tweed coats, horses, the lips and hair of the beloved. In using these terms we certainly intend to be talking of independent existences and we certainly intend to be talking of immediately perceptible things, bearers of phenomenal (visuo-tactite) properties. If scientific or Lockian realism is correct, we cannot be doing both at once; it is confusion or illusion to suppose we can If the things we talk of really have phenomenal properties, then they cannot, on this view, be physical things continuously existing in physical space. Nothing perceptible-I here drop the qualification ‘immediately’, far my use of it should now be clear-is a physically real, independent existence. No two persons can ever, in this sense, perceive the same item: nothing at all is publicly perceptible.
But how deep the confusion or the illusion must go! How radically it infects our concepts! Surely we mean by a cabbage a kind of thing of which most of the specimens we have encountered have a characteristic range of colours and visual shapes and felt textures; and not something unobservable, mentally represented by a complex of sensible experiences which it causes. The common consciousness is not to be fobbed off with the concession that, after all, the physical thing has-in a way-a shape. The way in which scientific realism concedes a shape is altogether the wrong way for the common consciousness. The lover who admires the curve of his mistress’s lips or the lover of architecture who admires the lines of a building takes himself to be admiring features of those very objects themselves; but it is the visual shape, the visually defined shape, that he admires. Mackie suggests that there is a genuine resemblance between subjective representation and objective reality as far as shape is concerned; 17 but this suggestion is quite unacceptable. It makes no sense to speak of a phenomenal property as esemb/ing a non-phenomenal, abstract property such as physical shape is conceived to be by scientific realism The property of looking square or round can no more resemble the property, so conceived, of being physicality square or round that the property of looking intelligent or looking it can resemble the property of being intelligent or being it. If it seems to make sense to speak of a resemblance between phenomenal properties and physical properties, so conceived, it is only because we give ourselves pictures -phenomenal pictures-of the latter. The resemblance is with the picture, not the pictured.
So, then, the common consciousness lives, or has the illusion of living, in a phenomenally propertied world of perceptible things in space. We might Cali it the lived world. It is also the public world, accessible to observation by ail: the world, in which Ode man, following another’s pointing finger, can see the very thing that the other sees. (Even in our philosophical moments we habitually contrast the colours and visual shapes of things, as being publicly observable, with the subjective contents of consciousness, private to each of us, though not thereby unknowable to others.)
Such a reminder of the depth and reality of our habitual commitment to the common-sense scheme does not, by itself, amount to a demonstration of that scheme’s immunity from philosophical criticism. The scientific realist, though no Kantian, may be ready, by way of making his maximum concession, with a reply modelled on Kant’s combination of empirical realism with transcendental idealism He may distinguish between the uncritical standpoint of ordinary living and the critical standpoint of philosophy informed by science. We are humanly, or naturally-he may say-constrained to ‘see the world’ in Ode way (that is, to think of it as we seem to perceive it) and rationally, or critically, constrained to think of it in quite another. The first way (being itself a causal product of physical reality) has a kind of validity at its own level; but it is criticality and rationally speaking, an inferior level. The second way really is a correction of the first.
The authentically Kantian combination is open to objection in many ways; but, by reason of its very extravagance, it escapes Ode specific Corm of difficulty to which the scientific realist’s soberer variant remains exposed. Kant uncompromisingly declares that space is in us; that it is ‘solely from the human standpoint that we can speak of space, of extended things etc.’, 18 that things as they
are in themselves are not spatial at all. This will not do far the scientific realist. The phenomenally propertied items which we take ourselves to perceive and the apparent relations between which yield (or contribute vitally to yielding) our notion of space, are indeed declared to have no independent reality; but, when they are banished from the realm of the real, they are supposed to leave behind them-as occupants, so to speak, of the evacuated territory -those spatially related items which, though necessarily unobservable, nevertheless constitute the whole of physical reality. Ayer refers in several places to this consequence; and questions its coherence.19 He writes, for example, ‘I doubt whether the notion of a spatial system of which none of the elements can be observed is even intelligible.’
It is not clear that this difficulty is insuperable. The scientific realist will claim to be able to abstract the notion of a position in physical space from the phenomenal integuments with which it is originally and deceptively associated; and it is hard to think of a conclusive reason for denying him this power. He will say that the places where the phenomenally propertied things we seem to perceive seem to be are, often enough, places at which the correlated physically real items really are. Such a claim may make us uneasy; but it is not obvious nonsense.
Still, to say that a difficulty is not clearly insuperable is not to say that it is clearly not insuperable. It would be better to avoid it if we can We cannot avoid it if we embrace unadulterated scientific realism and incidentally announce ourselves thereby as the sufferers from persistent illusion, however natural. We can avoid it, perhaps, if we can succeed in combining elements of the scientific story with our common-sense scheme without downgrading the latter. This is the course that Ayer recommends, 20 and, 1 suspect, the course that most of us semi-reflectively follow. The question is whether it is a consistent or coherent course. And at bottom this question is Ode of identity. Can we coherently identify the phenomenally propertied, immediately perceptible things which common sense supposes to occupy physical space with the configurations of unobservable ultimate particulars by which an unqualified scientific realism purports to replace them?
1 approach the question indirectly, by considering once again Mackie’s version of common-sense realism According to this version, it will be remembered, physical things, though not directly perceived, reality possess visual and tactile qualities which resemble those we seem to perceive them as possessing; so that if, per impossibite, the veil of perception were drawn aside and we saw things in their true colours, these would turn out to be colours indeed and, on the whole, just the colours with which we were naively inclined to credit them. Mackie does not represent this view as absurd or incoherent. He just thinks that it is, as a matter of fact, false. Things could really be coloured; but, since there is no scientific reason for supposing they are, it is gratuitous to make any such supposition.
Mackie is surely too lenient to his version of common-sense realism That version effects a complete logical divorce between things’s being red and its being red-Looking. Although it is a part of the theory that a thing which is, in itself, red has the power to cause us to seem to see a red thing, the logical divorce between these two properties is absolute. And, as far as I can see, that divorce really produces nonsense. The ascription of colours to things becomes not merely gratuitous, but senseless. Whatever may be the case with shape and position, colours are visibitia or they are nothing. I have already pointed out that this version of common-sense realism is not the real realism of common sense: that realism affects no logical divorce between being red and being red-Looking; for it is a perceptually direct and not a perceptually representative realism The things seen as coloured are the things themselves. There is no ‘veil past which we cannot see’; for there is no veil.
But this does not mean that a thing which is red, that is, red looking, has to look red ail the lime and in all circumstances and to ail observers. There is an irreducible relativity, relativity to what in the broadest sense may be called the perceptual point of view, built in to our ascriptions of particular visual properties to things. The mountains are red-looking at this distance in this Light; blue looking at that distance at that light; and, when we are clambering up them, perhaps neither. Such-and-such a surface looks pink and smooth from a distance; mottled and grainy when closely examined; different again, perhaps, under the microscope.
We absorb this relativity easily enough for ordinary purposes in our ordinary talk, tacitly taking some range of perceptual conditions, some perceptual point of view (in the broad sense) as standard or normal, and introducing an explicit acknowledgement of relativity only in cases which deviate from the standard. ‘It looks purple in this light,’ we say, ‘but take it to the door and you will see that it’s really green.’ But sometimes we do something else. We shift the standard. Magnified, the fabric appears as printed’ with tiny blue and yellow dots. So those are the colours it really is. Does this ascription contradict ‘it’s really green’? No; far the standard has shifted. Looking at photographs, in journals of popular science, of patches of human skin, vastly magnified, we say, ‘How fantastically uneven and ridge it really is.’ We study a sample of blood through a microscope and say, ‘It’s mostly colourless.’ But skin can still be smooth and blood be red; far in another context We shift our standard back. Such shifts do not convict us of volatility or condemn us to internal conflict. The appearance of both volatility and conflict vanishes when we acknowledge the relativity of Our ‘reallys’.
My examples are banal. But perhaps they suggest a way of resolving the apparent conflict between scientific and commonsense realism We can shift our point of view within the general framework of perception, whether aided or unaided by artificial means; and the different sensible-quality ascriptions we then make to the same object are not seen as conflicting once their relativity is recognized. Can we not see the adoption of the viewpoint of scientific realism as simply a more radical shift-a shift to a viewpoint from which no characteristics are to be ascribed to things except those which figure in the physical theories of science and in ‘explanation of what goes on in the physical world in the processes which lead to our having the sensations and perceptions that we have’?21 We can say that this is how things really are so long as the relativity of this ‘really’ is recognized as well; and, when it is recognized, the scientific account with no more conflict with the ascription to things of visual and tactile qualities than the assertion that blood is really a mainly colourless fluid conflicts with the assertion that it is bright red in colour. Of course, the scientific point of view is not, in One sense, a point of view at all. It is an intellectual, not a perceptual, standpoint. We could not occupy it at all, did we not first occupy the other. But we can perfectly well occupy both at once, so long as we realize what we are doing.
This method of reconciling scientific and common-sense realism requires us to recognize certain relativity in our conception of the real properties of physical objects. Relative to the human perceptual standpoint the grosser physical objects are visuo-tactite continuants (and within that standpoint the phenomenal properties they possess are relative to particular perceptual viewpoints, taken as standard). Relative to the scientific standpoint, they have no properties but those which figure in the physical theories of science.
Such a relativistic conception will not please the absolute-minded. Ayer recommends a different procedure. He suggests that we should conceive of perceptible objects (that is, objects perceptible in the sense of the common-sense scheme) as being literally composed of the ultimate particles of physical theory, the latter being imperceptible, not in principle, but only empirically, as a consequence of their being so minute. 22 I doubt, however, whether this proposal, which Ayer rightly describes as an attempt to blend the two schemes, can be regarded as satisfactory. If the impossibility of perceiving the ultimate components is to be viewed as merely empirical, we can sensibly ask what the conceptual consequences would be of supposing that impossibility not to exist. The answer is clear. Even if there were something which we counted as perceiving the ultimate particles, this would still not, from the point of view of scientific realism, count as perceiving them as they really are. And nothing could so count; far no phenomenal properties we seemed to perceive them as having would figure in the physical explanation of the causal mechanisms of our success But, so long as we stay at this point of view, what goes far the parts goes far any wholes they compose. However gross those holes; they remain, from this point of view, imperceptible in the sense of common sense.
Ayer attempts to form one viewpoint out of two discrepant viewpoints: to form a single, unified description of physical reality by blending features of two discrepant descriptions, each valid from its own viewpoint. He can seem to succeed only by doing violence to one of the two viewpoints, the scientifico I acknowledge the discrepancy of the two descriptions, but claim that, once we recognize the relativity in our conception of the real, they need not be seen as in contradiction with each other. Those very things which from one standpoint we conceive as phenomenally propertied we conceive from another as constituted in a way which can only be described in what are, from the phenomenal point of View, abstract terms. ‘This smooth, green, leather table-top’, we say, ‘is, considered scientifically, nothing but a congeries of electric charges widely separated and in rapid motion.’ Thus we combine the two standpoints in a single sentence. The standpoint of common-sense realism, not explicit1y signalled as such, is reflected in the sentence’s grammatical subject phrase, of which the words are employed in no esoteric sense. The standpoint of physical science, explicit1y signalled as such, is reflected in the predicate. Once relativity of description to standpoint is recognized, the sentence is seen to contain no contradiction; and, if it contains no contradiction, the problem of identification is solved.
I recognize that this position is unlikely to satisfy the determined scientific realist. If he is only moderately determined, he may be partially satisfied, and may content himself with saying that the scientific viewpoint is superior to that of common sense. Re will then simply be expressing a preference, which he will not expect the artist, for example, to share. But, if he is a hard-liner, he will insist that the common-sense view is wholly undermined by science; that it is shown to be false; that the visual and tactile properties we ascribe to things are nowhere but in our minds; that we do not live in a world of perceptible objects, as understood by common sense, at ail. He must then accept the consequence that each of us is a sufferer from a persistent and inescapable illusion and that it is fortunate that this is so, since, if it were not, we should be unable to pursue the scientific enterprise itself. Without the illusion of perceiving objects as bearers of sensible qualities, we should not have the illusion of perceiving them as space-occupiers at all; and without that we should have no concept of space and no power to pursue our researches into the nature of its occupants. Science is not only the offspring of common sense; it remains its dependant. For this reason, and for others touched on earlier, the scientific realist must, however ruefully, admit that the ascription to objects of sensible qualities, the standard of correctness of such ascription being (what we take to be) intersubjective agreement, is something quite securely rooted in our conceptual scheme. If this means, as he must maintain it does, that our thought is condemned to incoherence, then we can only conclude that incoherence is something we can perfect1y well live with and could not perfect1y well live without.