Chapter 2

SOUNDS

 

[1] Claiming a special status for one class or category of entities as opposed to others is very common in philosophy. It is the philosophical phenomenon of category-preference. I have been exhibiting category-preference in claiming that material bodies are, in a certain sense, basic in relation to other categories of particulars. But I should like to emphasize the point that there are certain ways in which category-preference may be exhibited, in which I am not exhibiting it. Suppose a s are the favoured type of entity. Then sometimes preference is manifested by the declaration that the word 'exist' has a primary sense or meaning, and that only a s exist in this sense, other things only in a secondary sense; sometimes by the declaration that only a s are real; and sometimes by the declaration that other things are reducible to a s, that to talk about other things is an abbreviated way of talking about a s. I want to emphasize that in saying that material bodies are basic among particulars, at least in our conceptual scheme as it is, I am not saying any of these things. The meaning given to the term 'basic' is strictly in terms of particular-identification. On the other hand, I believe that the facts I have tried to indicate may underlie and explain, if not justify, some of the more striking formulations, which I disavow, of the category-preference which I acknowledge. It seems to me also unobjectionable to use the expression, 'ontologically prior', in such a way that the claim that material bodies are basic particulars in our conceptual scheme is equivalent to the claim that material bodies are ontologically prior, in that scheme, to other types of particular.

These things, I have maintained, are true of our conceptual scheme as it is. The next thing I want to consider is whether, and

[p59]

if so how, it could be otherwise, Could there exist a conceptual scheme which was like ours in that it provided for a system of objective and identifiable particulars, but was unlike ours in that material bodies were not the basic particulars of the system? When I say, 'Could there exist such a scheme?' I mean 'Can we make intelligible to ourselves the idea of such a scheme?'

I have spoken of two sides, or aspects, of identification. They might be called the distinguishing aspect and the reidentifying aspect. The second has not, in the preceding exposition, been at all closely tied to a speech-situation involving a speaker and a hearer. Reidentification may involve merely thinking of a particular encountered on one occasion, or thought of in respect of one occasion, as the same as a particular encountered on another, or thought of in respect of another. Now such thinking clearly involves distinguishing, in thought or observation, one particular from others. So the distinguishing aspect of identification is quite fundamental. But so far in the exposition the idea of distinguishing one particular from others has been closely tied to the situation in which a hearer identifies a particular as the one currently referred to by a speaker. This tie I want now to loosen, while preserving the conclusion that material bodies are, in our actual conceptual scheme, basic to our thinking about particular-identification. I may legitimately do so; for it is not to be supposed that the general structure of such thinking is different when we are concerned to communicate with each other in speech and when we are not. The assertion that material bodies are basic particulars in our actual conceptual scheme, then, is now to be understood as the assertion that, as things are, identifying thought about particulars other than material bodies rests in general on identifying thought about material bodies, but not vice versa; and the question I have just raised, viz. 'Could we conceive of a scheme providing for identifiable particulars in which material bodies were not basic?', must be understood in a correspondingly revised and more general sense. This loosening of the tie with actual speech situations gives more freedom of manoeuvre in the next stage of the inquiry, without prejudicing the possibility that the connexion may ultimately have to be tightened up again.

[p60]

It gives more freedom of manoeuvre in the following way. So long as 'identification' means 'speaker-hearer identification', any question about the general conditions of a scheme providing for identifiable particulars is a question about the general conditions of speaker-hearer identification of particulars. So it is a question which can only arise given that we at least have speakers and hearers communicating with each other. But we can, or at least it seems that we can, raise a similar question without any such prior assumption of speakers and hearers. For each of us can think identifyingly about particulars without talking about them. Now of course it may be that the ability to think identifyingly about particulars is logically dependent on the ability to talk identifyingly to others about particulars. But this, if so, is at least not obviously so. We do not want to prejudge the question whether it is so or not; and we may, without prejudging it, raise a more general question about the conditions of the possibility of identifying thought about particulars.

But how general do we want our question to be? I am going to impose one limit on its generality. In one's own identifying thought, and indeed in one's own identifying talk, about particulars, one can certainly recognize a certain distinction: viz, the distinction between those particular occurrences, processes, states or conditions which are experiences or states of consciousness of one's own, and those particulars which are not experiences or states of consciousness of one's own, or of anyone else's either, though they may be objects of such experiences. Thus, if a tree is struck by lightning, that is one kind of happening; and if I see the tree being struck by lightning, that is another kind of happening. The knife entering my flesh is one kind of event, and my feeling the pain is another. The limit I want to impose on my general question is this: that I intend it as a question about the conditions of the possibility of identifying thought about particulars distinguished by the thinker from himself and from his own experiences or states of mind, and regarded as actual or possible objects of those experiences. I shall henceforth use the phrase, 'objective particulars' as an abbreviation of the entire phrase, 'particulars distinguished by the thinker &c.'. Now it may be that

[p61]

this limit on my question is, in a sense, no limit at all; for it may be that there could be no such thing as identifying thought about particulars which did not involve this distinction. But this, too, is a question which I shall shelve. It is not necessary for me to answer it; and perhaps it cannot be answered.

I can, then, indicate the line of enquiry I have in mind by posing two questions, reminiscent in form and partly in content of Kantian questions: (1) What are the most general statable conditions of knowledge of objective particulars? (a) Do these most general conditions involve the requirement that material bodies should be the basic particulars, or is this simply a special feature of our own scheme for knowledge of objective particulars? Or -- to run the two questions into one -- is the status of material bodies as basic particulars a necessary condition of knowledge of objective particulars?

Now I have suggested earlier that the fact that material bodies are the basic particulars in our scheme can be deduced from the fact that our scheme is of a certain kind, viz, the scheme of a unified spatio-temporal system of one temporal and three spatial dimensions. If this is correct, then to find a scheme in which material bodies were not basic particulars would be, at least, to find a scheme which was not of this kind. This reflection suggests more than one direction in which we might look. But, in particular, it suggests one very simple, though very radical, direction. We might ask, 'Could there be a scheme, providing for a system of objective particulars, which was wholly non-spatial?' This question reminds us once more of Kant. He spoke of two forms of sensibility or intuition, namely Space and Time. Time was the form of all sensible representations, Space only of some. He regarded it as a matter not of absolute necessity, but of very fundamental fact, that we had both, and only, these two forms of sensible intuition. He would probably think it tautological to say that we cannot imagine ourselves possessing other forms, though we could, in some sense, conceive of its possibility. I think he would probably also say that it was in some sense impossible to imagine ourselves not possessing both these forms. 'We cannot represent to ourselves the absence of space.' [1] I do not know quite

 

[p62]

what this means. But whether or not we can 'represent to ourselves the absence of space', I do not see why we should not confine ourselves imaginatively to what is not spatial; and then see what conceptual consequences follow. Kant held that all representations were in inner sense, of which Time was the form; but only some representations were representations of outer sense, of which Space was the form. I suggest that we inquire whether there could be a scheme which provided for objective particulars, while dispensing with outer sense and all its representations. I suggest we explore the No-Space world. It will at least be a world without bodies.

[2] Now, as regards what follows, I must sound a note at once apologetic and cautionary. I shall constantly raise questions in a form in which they may well seem quite unanswerable: especially in the form of asking whether a being whose experience was in certain ways quite unlike ours could or could not have a conceptual scheme with certain general features; or whether a being whose conceptual scheme was in certain general ways quite unlike ours could or could not nevertheless reproduce in it certain features of ours. So presented, these questions may well seem at worst nonsensical, and at best to admit of only the most wildly speculative answers. But, in general, this form of question may be seen as simply a convenient, if perhaps over-dramatic, way of raising more evidently legitimate types of question: questions not about hypothetical beings at all, but rather, for instance, about the extent to which, and the ways in which, we might find it possible to reinterpret, within a part of our experience, some of the most general conceptual elements in our handling of experience as a whole. How far can we map the structure of this whole within a part of itself? Or what structural analogies can we find between some part and the whole of which it is a part? Or, again, how far can we break down the connexions of certain central concepts with each other and with certain types of experience without seeming to destroy those concepts altogether? Questions which belong to these general kinds are no doubt in some sense idle; but appear to be discussable.

[p63]

In the ensuing discussion the emphasis will be found to shift in a certain way. For the selected model of a No-Space world something that was taken for granted in the first chapter comes into question: viz, that distinction made by the user of a conceptual scheme between himself and his own states on the one hand, and other particulars of which he has knowledge or experience on the other. The question, whether the conditions of this distinction could be satisfied in the supposed world, will be found to turn in part, but by no means exclusively, on another question, which echoes some themes of the first chapter, viz, the question, whether the conditions of reidentifiability of particulars could be satisfied in the supposed world. But it is not settled by the discussion of this question; and the further attempt to settle it leads us back, in the third chapter, to a direct consideration of our ordinary world, and of the ways in which the conditions of the distinction in question are in fact satisfied there.

So the present chapter acts in part as a bridge between the first and the third. Some illumination of general features of our actual thinking may perhaps be hoped for by thus inquiring how far such features can be reproduced in phenomenal terms of an artificial simplicity, by observing, so to speak, in what ways we have to shape and model our impoverished material in order to reproduce the structure we know.

[3] What does the suggestion that we explore the No-Space world amount to? What is it to imagine ourselves dispensing with outer sense? Traditionally, five senses are recognized as distinguishable modes of perception of public objects. Of these, taste and smell are strikingly more trivial than the others, and taste in addition has a logical complexity which makes it difficult to handle. It does not seem that to suppose our experience free of gustatory or olfactory elements would, by itself, be to invite a significant conceptual revolution. (One does not see the world differently if one has a cold.) Let us nevertheless, for simplicity's sake, suppose them eliminated. This leaves us with sight, hearing and touch. Which of these shall we have to suppose eliminated in order to eliminate outer sense? It might at first seem that we

[p64]

should have to eliminate them all -- which would bring the inquiry rather swiftly to an end. For while we can certainly discover the spatial characteristics and relations of things by sight and touch, it seems no less certain that we can also discover at least some spatial features of some things by hearing. Sounds seem to come from the right or the left, from above or below, to come nearer and recede. If sounds, the proper objects of hearing, possess in their own right these direction-and-distance characteristics, does it not follow that we shall have failed to eliminate spatial characteristics and concepts even if we adopt the radical hypothesis of a purely auditory experience? This conclusion, however, would, I think, be a mistake, and a fairly obvious one. The fact is that where sense-experience is not only auditory in character, but also at least tactual and kinaesthetic as well -- or, as it is in most cases, tactual and kinaesthetic and visual as well -- we can then sometimes assign spatial predicates on the strength of hearing alone. But from this fact it does not follow that where experience is supposed to be exclusively auditory in character, there would be any place for spatial concepts at all. I think it is obvious that there would be no such place. The only objects of sense-experience would be sounds. Sounds of course have temporal relations to each other, and may vary in character in certain ways: in loudness, pitch and timbre. But they have no intrinsic spatial characteristics: such expressions as 'to the left of', 'spatially above', 'nearer', 'farther' have no intrinsically auditory significance. Let me briefly contrast hearing in this respect with sight and touch. Evidently the visual field is necessarily extended at any moment, and its parts must exhibit spatial relations to each other. The case of touch is less obvious: it is not, e.g., clear what one would mean by a 'tactual field'. But if we combined tactual with kinaesthetic sensations, then at least it is clear that we have the materials for spatial concepts; of the congenitally blind one does not wonder whether they really know what it means to say that one thing is above another, or farther from another than a third thing is. Of a purely visual, or a purely tactual-kinaesthetic, concept of space, one might feel that it was impoverished compared with our own, but not that it was an impossibility. A purely auditory

[p65]

concept of space, on the other hand, is an impossibility. The fact that, with the variegated types of sense-experience which we in fact have, we can, as we say, 'on the strength of hearing alone' assign directions and distances to sounds, and things that emit or cause them, counts against this not at all. For this fact is sufficiently explained by the existence of correlations between the variations of which sound is intrinsically capable and other non-auditory features of our sense-experience. I do not mean that we first note these correlations and then make inductive inferences on the basis of such observation; nor even that we could on reflection give them as reasons for the assignments of distance and direction that we in fact make on the strength of hearing alone. To maintain either of these views would be to deny the full force of the words 'on the strength of hearing alone'; and I am quite prepared to concede their full force. I am simply maintaining the less extreme because less specific thesis that the de facto existence of such correlations is a necessary condition of our assigning distances and directions as we do on the strength of hearing alone. Whatever it is about the sounds that makes us say such things as 'It sounds as if it comes from somewhere on the left', this would not alone (i.e. if there were no visual, kinaesthetic, tactual phenomena) suffice to generate spatial concepts. I shall take it as not needing further argument that in supposing experience to be purely auditory, we are supposing a No-Space world. I am not, of course, contending that the idea of a purely auditory world is the only possible model for a No-Space world. There are other and more complex possibilities. I select the idea of a purely auditory universe as one that is relatively simple to handle, and yet has a certain formal richness.

The question we are to consider, then, is this: Could a being whose experience was purely auditory have a conceptual scheme which provided for objective particulars? The question is complex and breaks down into a number of others. Consider first the qualification 'objective' in the phrase 'objective particulars'. It might seem at first that this qualification raises no special difficulties. For, as things are at present, different people may certainly be said to hear one and the same particular sound -- not just

[p66]

sounds of the same type, instances of the same kind of sound, but exactly the same particular sound. Sounds may be, and most of those that we hear are, public objects. If, when we talk of a sound, we mean a particular sound, then we may, and usually do, mean an objective particular, a public object. So it might seem obvious that if, in a purely auditory world, we could operate with the concept of a particular at all, then we could operate with the concept of an objective particular. But this is in fact not at all clear. For to call a sound a public object, to say that different people may hear one and the same particular sound, seems to mean at least this: that different normal hearers may roughly simultaneously have roughly similar auditory experiences, or auditory experiences systematically related in statable ways, in roughly the same particular surroundings; and perhaps one should add that, in order to fulfil on any particular occasion the requirement that they are hearing the same sound, the causal source of the relevant auditory experiences should be the same for all of them. We may imagine, for example, that the same piece of music is being simultaneously played in two different concert-halls. We may imagine a moment at which a certain chord is played. Then two different normal hearers, each in a different hall, have, roughly simultaneously, roughly similar auditory experiences. But though in one sense the sound they hear is the same -- it is the same chord for each -- in another sense, the sense we are concerned with, the sounds they hear are distinct. They hear different sound-particulars: for the condition of particular-identity of surroundings, and the condition of particular-identity of causal sources, are not fulfilled. Two listeners in the same concert-hall, however, hear the same sound particulars as each other, as well as hearing the same chord, i.e. the same sound-universal; for in their case the conditions of particular-identity of setting and source are both fulfilled.

I do not want to say that the stated conditions for the identity of sound-particulars heard by different listeners are exhaustive. The case of sounds transmitted through various artificial media, for example, suggests other quite interesting possibilities of different criteria for the identity of sound-particulars. I just choose the

[p67]

conditions stated as being the most obvious set, without ruling out the possibility of others.

Of the stated conditions, the last, concerning causation, may perhaps be neglected for our purely auditory universe. The others present an acute problem. For, it seems, to give the idea of publicity of sounds a meaning in the purely auditory world, we must give meaning, in auditory terms alone, to the idea of other people and to the idea of their being in identical particular surroundings. But to assume that we could give a sense to the idea of identity of particular surroundings in terms of sound alone would be to beg the question. For the sounds in terms of which we were to give sense to this idea would themselves have to be public sounds; otherwise they could not provide particular-identity of surroundings for the different enjoyers of auditory experiences. But it is precisely the possibility of public sounds, in a purely auditory world, that is in question. So we cannot assume a favourable issue here. Indeed the prospect for a favourable issue begins to look unhopeful.

We might, however, raise our hopes by reducing, or trying to reduce, our demands. I glossed 'objective' particulars just now as 'public' particulars, and that involved the ideas of other enjoyers of experience and of shared surroundings. For this gloss there is, as I have already hinted, much to be said, if a certain general line of thought is correct. A summary, which I hope is not too much of a parody, of this line of thought, might run as follows. We could not talk to one another about the private if we could not talk to one another about the public. We could not talk unless we could talk to one another. Above, at any rate, a very rudimentary level, the limits of thought are the limits of language; or 'what we can't say we can't think'. Finally, there is no experience worth the name, certainly no knowledge, without concepts, without thoughts. Applied to the present question, this line of thought yields the conclusion that the whole idea of a purely auditory experience is empty, unless a sense can be given in purely auditory terms to the idea of public auditory objects which are also topics of discourse between beings who hear them.

I shall not, for the moment, try to pronounce on the merits

[p68]

of this line of thought. For I earlier introduced the word 'objective' by giving it what is certainly a more traditional, and possibly a less exacting, sense, in terms of the distinction between oneself and one's states on the one hand, and anything on the other hand which is not either oneself or a state of oneself, but of which one has, or might have, experience. So I shall provisionally interpret the question, 'Can the conditions of knowledge of objective particulars be fulfilled for a purely auditory experience?' as meaning: 'Could a being whose experience was purely auditory, make use of the distinction between himself and his states on the one hand, and something not himself, or a state of himself, of which he had experience on the other?' This question, for the sake of a convenient phrase, I shall reexpress as follows: 'Can the conditions of a non-solipsistic consciousness be fulfilled for a purely auditory experience?' That is to say, I shall mean by a non-solipsistic consciousness, the consciousness of a being who has a use for the distinction between himself and his states on the one hand, and something not himself or a state of himself, of which he has experience, on the other; and by a solipsistic consciousness, the consciousness of a being who has no use for this distinction.

This question, however, is not the only one we have to answer. There is another which turns out to be closely connected with it, viz.: Can we, in purely auditory terms, find room for the concept of identifiable particulars at all? Would there, in the purely auditory world, be a distinction between qualitative and numerical identity? This seems at first to present no particular difficulty. Could not audible continuity or discontinuity be used as a criterion for distinguishing sounds as particulars? That is to say, suppose, first, that during a certain temporal slice of experience, sound of a certain loudness, timbre and pitch began to be heard, continued without interruption and then stopped. Suppose, second, that during such a slice of experience such sound began, stopped, began again and stopped again. In the first case the number of sounds as particulars would be one; in the second case the number of sounds as particulars would be two. In both cases there would be just one and the same sound in the qualitative

[p69]

sense of 'same', i.e. just one sound as universal. Even when sound of some kind is continuous, as, usually, when music is being heard, we can distinguish qualitatively different sounds within the general sound, and hence, by the criterion of interruption, different particular instances of the same qualitative sound. Also, of course, we can, perhaps more easily, distinguish more complex sound-particulars, composed of sets or sequences of the sound-particulars distinguished by the above method. This seem to show that there could be identifiable, in the sense of distinguishable, sound-particulars.

But could there be identifiable, in the sense of reidentifiable, sound-particulars? Unless this question is answered affirmatively, the concept of a particular with which we are working will be, so to speak, a very thin one. Now, of course, sounds could be reidentified, if 'sounds' is taken in the sense of universals or types. A note could be reidentified, or a sequence of notes or a sonata. But what sense could be given to the idea of identifying a particular sound as the same again after an interval during which it is not heard? We cannot turn to the particular-identity of the non-auditory setting of the sounds, to justify our saying, e.g. 'This is the continuation of that same particular sequence of sounds which was heard a while ago'; for, by hypothesis, the sounds have no setting but other sounds. The difficulty can be emphasized by considering what might appear to be a possible exception to it. Suppose a sound-sequence of some complexity -- and here I am speaking of a type or universal -- which has a certain, say, musical unity, and to which I shall refer as M. Suppose within it four 'movements' are distinguishable, A, B, C, D. Suppose an instance of A is heard and then, after a suitable interval, an instance of D is heard. The interval, however, is not occupied by B and C, but by other sounds. May we not suppose that, in this case, when the instance of D is heard, it is identified as a part of the same particular M as that of which the previously heard instance of A was a part; that is, that when D is heard, we have a case of the same particular M, reappearing, so to speak, after an interval; and that thus we have here a case not just of reidentifying a universal, but of identifying a particular as the same again? But, of course, for

[p70]

this suggestion to be of any use, we have to suppose that we have some criterion for distinguishing the case of a reappearance of the same particular M from the case where we just have an instance of A followed after an interval by an instance of D, and where the two are not parts of one and the same particular M. This is reminiscent of the case of the two concert-halls, with the same (type or universal) piece of music being played in both simultaneously. Here the criteria for distinguishing between a later part of the same particular piece and an instance of a later part of the same universal piece were evident enough; they turned, once more, on the non-auditory setting. But in the purely auditory world these criteria are not available; and if no criteria are available for making the distinction, then no sense has been given to the distinction and hence no meaning to the idea of reidentification of auditory particulars. The case is perhaps not quite so poor as this suggests. For a criterion of sorts might be suggested. It might be suggested that where the instances of A and D were fairly soft sounds, while the sounds that filled the interval between them were very loud sounds, then we had a clear case of A and D being parts of the same particular; and when this condition was clearly not fulfilled, then we had a clear case of their not being parts of the same particular. But the reasons for such appeal as this suggestion might have for us are only too evident. It helps us to think of unheard parts of one particular M being drowned or submerged by the stridencies which intervene between the instance of A and the instance of D; and thus to think that they were there to be heard, would have been heard but for these stridencies. But now we have only to think of the reasons, the evidence, we have for thinking something like this in real life -- the visible but inaudible scrapings of the street violinist as the street band marches by -- and then we lose interest in the suggested criterion for the case of the purely auditory world.

Nevertheless certain important and interrelated points emerge from these considerations.

The first is the connexion between the idea of a reidentiflable particular, and the idea of the continued existence of a particular while it is not being observed. This connexion conferred whatever

[p71]

appeal it possessed upon the criterion of reidentification of sound-particulars which I have just considered and dismissed. It was not just that the intervening sounds were loud; it was that they were loud enough for us, from within our familiar world, to think of them as drowning the unheard sounds which linked the earlier and the later parts of the reidentifled particular. But this thought came too evidently from our familiar world, and has no relevance, or has not yet been given any relevance, to our imaginary world. We have yet to show that sense can be given to the idea of continued existence of unobserved particulars in this imaginary world.

This first point leads directly on to the second. The question: Could there be reidentifiable sound-particulars in the purely auditory world? was raised as if it were a further question which had to be considered, over and above another question, viz.: Could a being whose experience was purely auditory make sense of the distinction between himself and his states on the one hand, and something not himself or a state of himself, on the other? But now it seems that these questions are not independent. An affirmative answer to the second entails an affirmative answer to the first. For to have a conceptual scheme in which a distinction is made between oneself or one's states and auditory items which are not states of oneself, is to have a conceptual scheme in which the existence of auditory items is logically independent of the existence of one's states or of oneself. Thus it is to have a conceptual scheme in which it is logically possible that such items should exist whether or not they were being observed, and hence should continue to exist through an interval during which they were not being observed. So it seems that it must be the case that there could be reidentiflable particulars in a purely auditory world if the conditions of a non-solipsistic consciousness could be fulfilled for such a world. Now it might further be said that it makes no sense to say that there logically could be reidentifiable particulars in a purely auditory world, unless criteria for reidentification can be framed or devised in purely auditory terms. And if this is correct, as it seems to be, we have the conclusion that the conditions of a non-solipsistic consciousness can be satisfied in

[p72]

such a world only if we can describe in purely auditory terms criteria for reidentification of sound-particulars.

Does the entailment hold in the other direction too? That is, does the existence of the idea of a reidentifiable particular, and hence the idea of a particular which continues to exist while not being observed, entail the existence of the distinction between oneself and states of oneself on the one hand and what is not oneself or a state of oneself on the other? The answer to this question I shall postpone for the moment. Later, I shall suggest a technique for answering it and all similar questions, i.e. all questions about whether something is or is not a sufficient condition for the existence of a non-solipsistic consciousness. Let us merely notice in passing, and dismiss, one temptation to give an affirmative answer to this question on mistaken grounds. One might be tempted to answer affirmatively simply as a result of confusing two different ideas: that of the being with the solipsistic consciousness and that of the philosophical solipsist. But the being with the solipsistic consciousness, whom, for short, I might call the true solipsist, would not think of himself as such; nor as a philosophical solipsist; nor as anything else. He certainly would not think that everything particular which existed was himself or a state of himself. One who claimed to think this might indeed have some difficulty, not necessarily insuperable, in reconciling his doctrine with the idea of a number of particulars which continue to exist unobserved. But the true solipsist is rather one who simply has no use for the distinction between himself and what is not himself. It remains to be seen whether a conceptual scheme which allows for reidentifiable particulars must necessarily also make room for this distinction.

Meanwhile let us pass to the next point. Let us inquire how, in our familiar world, the requirements just established are fulfilled. That is to say, with what feature or complex of features of our familiar world is the idea of reidentifiable particulars, existing continuously while unobserved, most intimately, naturally and generally connected? I think the answer is simple and obvious, though the detailed description of the feature in question would be of great complexity. Roughly speaking, the crucial idea for us

[p73]

is that of a spatial system of objects, through which oneself, another object, moves, but which extends beyond the limits of one's observation at any moment, or, more generally, is never fully revealed to observation at any moment. This idea obviously supplies the necessary non-temporal dimension for, so to speak, the housing of the objects which are held to exist continuously, though unobserved; it supplies this dimension for objects which are not themselves intrinsically spatial, such as sounds, as well as for objects that are. Thus the most familiar and easily understood sense in which there exist sounds that I do not now hear is this:

that there are places at which those sounds are audible, but these are places at which I am not now stationed. There are of course other senses which can be given to the idea of unheard sounds. But many of them turn on correlations between auditory phenomena and phenomena of other kinds (e.g., non-auditory phenomena causally associated with auditory phenomena) and on the extrapolation of these correlations beyond the general limits of human auditory discrimination. So these do not help us here. Alternatively they turn on such an idea as that of failing sensory powers. But why do we think of our powers failing rather than the world fading? This choice cannot be used to explain a conception it presupposes.

Let us return, then, to the most familiar sense in which we think of sounds which exist now but are unheard by us, and to its relation to the idea of places. We have already seen that the idea of place, and with it that of a spatial system of objects, cannot be given a meaning in purely auditory terms. Yet it seems we must have a dimension other than the temporal in which to house the at present unheard sensory particulars, if we are to give a satisfactory sense to the idea of their existing now unperceived, and hence to the idea of reidentification of particulars in a purely auditory world and hence, perhaps, to the idea of a non-solipsistic consciousness in a purely auditory world. So our question becomes this. Since we cannot give any literal, even though impoverished, interpretation of spatial concepts in purely auditory terms, can we at any rate find some sort of variable feature in auditory items which will provide what we might call an analogy of

[p74]

Space? And of course -- whatever this turns out to mean -- a sufficiently close analogy for our purposes?

But how close is sufficiently close? We want the analogy of space to provide for the unperceived, but existing, particular. Roughly, we want it to provide for something like the idea of absence and presence -- but not just of absence and presence in the most utterly general sense these words could bear, but absence or presence in a sense which would allow us to speak of something being to a greater or lesser degree removed from, or separated from, the point at which we are. In other words, we want an analogy of distance -- of nearer to and further away from -- for only, at least, under this condition would we have anything like the idea of a dimension other than the temporal in which unperceived particulars could be thought of as simultaneously existing in some kind of systematic relation to each other, and to perceived particulars. Of course the spatial phenomena with which we are seeking an analogy are infinitely more complex than this. Remote particulars are located, not just in one dimension of distance, but in three; particulars may be unperceived, not because they are too remote, but because they are hidden by others or because, of all the directions we may be looking or feeling in, we are not looking or feeling in theirs. But we may well despair of reproducing analogies for all this complexity in auditory terms. Looking for the simplest feature for which we might find an analogy, it seems that this of distance is the easiest. For the loomings-up and dwindlings and obliterations of perspective we might find an analogy.

It is customary to distinguish three dimensions of sound: timbre, pitch and loudness. Timbre we may discount; for differences of timbre do not seem to admit of any systematic serial ordering. Pitch seems much more hopeful. Indeed, we customarily speak of differences of pitch on analogy with a spatial dimension -- we speak of higher and lower notes -- and moreover we customarily represent these differences by spatial intervals. If the analogy holds in one direction, may it not also hold in the other? Suppose we imagine that the purely auditory experience we are considering has the following characteristics. A sound of a certain distinctive

[p75]

timbre is heard continuously, at a constant loudness, though with varying pitch. This sound is unique in its continuity. We may call it the master-sound. It may be compared with the persistent whistle, of varying pitch, which, in a wireless set in need of repair, sometimes accompanies the programmes we listen to. In addition to the master-sound, other sounds or sequences of sound of various degrees of complexity are heard. Some of these sequences may be supposed to have the kind of unity which pieces of music have. They recur and are recognized. They are highly complex universals with particular instances. One can imagine that transitions up and down the pitch-range of the master-sound sometimes occur quite fast; while at other times the pitch of the master-sound remains invariant for quite considerable periods. One may imagine, finally, that variations in the pitch of the master-sound are correlated with variations in the other sounds that are heard, in a way very similar to that in which variations in the position of the tuning-knob of a wireless set are correlated with variations in the sounds that one hears on the wireless. Thus suppose a particular instance of one of the unitary sound-sequences I mentioned is being heard. A gradual change in the pitch of the master-sound is accompanied by a gradual decrease, or a gradual increase followed by a gradual decrease, in the loudness of the unitary sound-sequence in question until it is no longer heard. If the gradual change in pitch of the master-sound continues in the same direction, a different unitary sound-sequence is heard with gradually increasing loudness. If it is reversed, the whole accompanying process is reversed too. Here the comparison is with gradually tuning-out one station and tuning-in another -- and back again for the reversal. Only of course instead of a tuning-knob being gradually turned, we have the gradual alteration in the pitch of the master-sound. If, on the other hand, the pitch of the master-sound changes very rapidly, the change is accompanied by that kind of cacophonous succession which one gets by twirling the tuning-knob around at top speed. And if the pitch of the master-sound remains constant, then one recognizable unitary sequence of sounds duly completes itself and another begins.

In these circumstances, one might feel, the analogy would be

[p76]

close enough to yield the picture of a sound-world which allowed for reidentifiable particulars. The pitch of the master-sound at any moment would determine the auditory analogue of position in the sound-world at that moment. The sound-world is then conceived of as containing many particulars, unheard at any moment, but audible at other positions than the one occupied at that moment. There is a clear criterion for distinguishing the case of hearing a later part of a particular unitary sound-sequence of which the earlier part has been heard previously, from the more general case of merely hearing the later part of the same universal unitary sound-sequence of which an earlier part has been heard previously. Suppose for instance a certain unitary sound-sequence, to which we may refer as M (M being the name of a universal) is being heard at a certain pitch-level of the master-sound -- say at level L. Then suppose the master-sound changes fairly rapidly in pitch to level L' and back again to L; and then M is heard once more, a few bars having been missed. Then the sound-particular now being heard is reidentified as the same particular instance of M. If, during the same time, the master-sound had changed not from L to L' and back again to L, but from L to L', then, even though M may be heard once more, a few bars having been missed, it is not the same particular instance of M that is now heard, but a different instance. Once again, the wireless supplies the easy comparison: one can tune out a station, and tune it in again while the same piece is being played; or, instead, one might tune in a different station where the same piece is being simultaneously played by a different orchestra.

But, of course, though the analogy, and hence the resultant conceptual scheme which allows for re-identifiable particulars, may be fairly persuasive, fairly attractive, it is not compelling. We could adopt a different scheme of description which allowed for reidentifiable universals but not for reidentifiable particulars. What we cannot consistently do is, as it were, to appear to accept a scheme which allows for reidentification of sound-particulars and then to say that, of course, particular-identity would always be in doubt; that there would be no possibility of certainty about

[p77]

it. This would be the position of philosophical scepticism about the identity of sound particulars, and ultimately, about the independent reality of the sound-world. It would involve that kind of inconsistency which I commented on earlier -- the simultaneous acceptance and rejection of a certain conceptual scheme for reality. Alternatively it could be construed as a kind of muddled advocacy of a different scheme: in this case, of one which either did not allow for reidentification of particulars, or which envisaged more stringent or more complex criteria of reidentification than those which I have described.

Let us briefly pause to compare the situation in the auditory, and in the ordinary, worlds. In describing a possible scheme which allows for reidentification of sound-particulars in the auditory world, I have, obviously, described a scheme which allows for their reidentification without any kind of reference to particulars of any other type than their own; for no other type of particular comes into consideration. In the auditory and in the ordinary worlds alike, the possibility of reidentification of particulars depends on the idea of a dimension in which unperceived particulars may be housed, which they may be thought of as occupying. But, for our ordinary world, the word 'housed' is barely a metaphor and the word 'occupying' is not a metaphor at all. For in our ordinary world that 'dimension' is, precisely, three-dimensional space. Now it is the general character of this dimension which, for any conceptual scheme, determines the types of particular which can be reidentified without dependence on particulars of other types. So, in our actual scheme, the particulars which can be thus independently reidentified must at least be intrinsically spatial things, occupiers of space; and sound-particulars, not being of this character, are not independently reidentifiable. But in the imagined scheme we are now considering, the dimension in question is supplied by variations in purely auditory phenomena. The dimension is, so to speak, the pitch-range of the master-sound. So independently reidentifiable particulars may, in this scheme, themselves be purely auditory.

To return to the auditory analogy of distance, whereby we

[p78]

tried to allow for a conceptual scheme providing for reidentifiable particulars. I said that the analogy might be fairly persuasive, but was not compelling. Some might find it less persuasive than others. I can imagine one who is not disposed to be at all persuaded by it arguing like this: You have referred to the three characteristic kinds of variation of which sound is capable, viz, loudness, pitch and timbre, and have tried to make them, and in particular pitch, yield between them the analogue of spatial distance. A quite essential element in the construction is the device of the master-sound; whatever was accomplished, was done with the aid of this trick. If we now compare sound with colour -- something intrinsically spatial -- we see how weak the analogy really is. For colour, like sound, exhibits three characteristic modes of variation -- brightness, saturation and hue -- of which the first two, like pitch and loudness, admit of serial ordering in respect of degree, while the last, perhaps, like timbre, does not. In the case of a visual scene, we may be presented with coloured areas, exhibiting between them variations of all three kinds simultaneously; and so far there is an analogy with sound. But when we are presented with such a scene, we are also and necessarily presented with something which simultaneously exhibits a further principle of ordering of its parts. Suppose we break up the scene, as it were, into its uniform elements, i.e. into elements no one of which at any moment exhibits variations, but each of which is of a definite hue, brightness and saturation. Then these simultaneously presented elements, besides being related to one another in these three respects, are also simultaneously presented as being related in another respect: viz, in a respect which leads us to characterize one as being above or below or to the left or to the right of another, or, if there is difficulty over these words at the phenomenal level, which at any rate leads us to characterize one as being further away from another in a certain direction than a third is & c. The point is that relations between elements in respect of the spatial dimension are presented simultaneously, all at once,we need no changing master-patch to give us the idea of this dimension. But relations between elements in respect of the auditory analogue of the spatial dimension cannot be presented simultaneously, all at

[p79]

once. They turn essentially on change. Roughly, two visual elements can be seen all at once as at a certain visual distance from one another; whereas two auditory elements cannot be heard all at once as at a certain auditory distance from one another. Or, to put it in another way: the momentary states of the colour-patches of the visual scene visibly exhibit spatial relations to each other at a moment; whereas the momentary states of the sound-patches of the auditory scene do not audibly exhibit the auditory analogue of spatial relations to each other at a moment. Not in their momentary states, but only in swelling or fading as the pitch of the master-sound varies in time, could the particular sounds exhibit such relations. But surely the idea of simultaneous existence of the perceived and the unperceived is linked with this idea of the simultaneous presentation of elements, each of a definite character, but simultaneously exhibiting a system of relations over and above those which arise from the definite character of each. Surely the former idea is necessarily an extension of the latter, is just the idea of such a system of relations extending beyond the limits of observation. So the objector might argue. (So arguing I think he would, at least in the last sentence, over-reach himself, by ignoring the importance, for his own doctrine about the extension of the idea of such a system of relations, of the notion of movement of the scene and the observer relative to each other, and hence of change. But he might meet this point by saying that he was stating only a necessary and not a sufficient condition of such an extension.) If the objector argues so, there is a sense in which we cannot meet his objections. That is to say, though we might complicate our auditory world-picture in many ways, we cannot, while keeping it an auditory picture, incorporate just that feature which he seems to be insisting on as a condition of regarding the analogy of space as close enough to satisfy him. Indeed, nothing but a system of spatial relations, and possibly nothing but spatial relations as visually perceived, can be conceived by us which would satisfy this condition. If this is so, then the objector is not simply criticizing our method of analogy-seeking, but rejecting the whole idea of any such analogy. This might be a reasonable thing to do if the ground for doing so were that there just

[p80]

are no formal parallels worth considering between the sensibly spatial and the auditory. But this view would be simply false. One should remember here not only the spatial analogies implicit in our ordinary talk about sounds, but the persistent and in no way irrational tendency of critics of music and the plastic arts to discuss the formal properties of the works they are criticizing in terms which, in their literal application, belong to each other's vocabularies.

The necessary incompleteness of the analogy, then, is not a decisive objection. There remains a doubt about the meaning of saying that we have here a possible reinterpretation of the idea of an unperceived and hence of a reidentifiable particular. What are the tests for whether it is a possible reinterpretation or not? I do not think there is any test beyond what we find it satisfactory to say. One can certainly influence the finding by pointing to respects in which the parallel holds or fails to hold -- and can also suggest improvements. But no more.

The question, whether we could find room in the purely auditory world for the concept of a reidentifiable particular, was not, however, the only question we set ourselves. There was also the question, whether the conditions of a non-solipsistic consciousness could be satisfied in such a world. An affirmative answer to the first question appeared as at least a necessary condition of an affirmative answer to the second. Whether it was also a sufficient condition, was a point I left undecided. It might appear obvious that it was a sufficient condition. For the concept of a reidentifiable particular was held to entail that of a particular's existing while unobserved and hence, in general, the distinction between being observed and being unobserved, or at least some closely analogous distinction. But how can this distinction exist without the idea of an observer? How, therefore, can the being with the auditory experience make use of any such distinction without the idea of himself as an observer? Moreover, when we were preparing to construct our auditory analogue of space, we spoke of ordinary observers as thinking of themselves as being at different places at different times. Must not the being with the purely auditory experience similarly think of himself as 'at' different places in

[p81]

auditory space? This reasoning is attractive. [1] But, since the entire object of this speculation is to put the maximum pressure on the normal associations of our concepts, it will be consonant with our general programme to resist this attractive reasoning if we can. I think we can resist it. The question essentially is whether a distinction parallel in other respects to the ordinary 'observed-unobserved' distinction can be drawn without the need for any idea such as we ordinarily express by the first person singular pronoun and associated forms. Why should it not be? Let us consider a possible technique for answering such questions. We are to imagine ourselves, our ordinary selves, with all our ordinary conceptual and linguistic apparatus at our disposal, writing reports on a special part of our experience. The part is defined by the description given of the purely auditory world, But the writing of our reports is governed by an important rule. The rule is that we are not, in writing our reports, to make use of any concepts which derive their function from the fact that this special part of our experience is in fact integrated with our experience at large, forms part of a wider whole. All the concepts or expressions we employ must find their justification within the part of our experience in question. They must all be concepts or expressions of which we find the use essential or convenient merely in order to do justice to the internal features of this part of our experience. For example, supposing that the description of the purely auditory world is as we have so far given it, then if, in writing our reports, we write the sentence 'I heard M after N at L' (for the purposes of this example it does not matter whether 'M' and 'N' are names of universals or not), we should have broken this important rule. The verb 'to hear' is one we must not use. It is redundant, since the description of the universe of discourse in question specifies that it contains no sensory items other than sounds. And as far as the description so far given is concerned, the personal pronoun appears equally superfluous. The sentence in the report should read simply 'N was observed at L followed

[p82]

by M'. That is to say, for the description of the universe so far given, we do not appear, if we follow the rule, to need to make any use of the distinction between oneself and what is not oneself.

It might appear that we should introduce the need for this distinction by modifying the description in the following way. So far we have supposed that movement up and down the range of the master-sound merely occurs. We have introduced no distinction between moving and being moved. Suppose we introduce such a distinction. Suppose, that is to say, that the being whose experience is purely auditory sometimes just suffers change of position -- change just occurs -- and sometimes initiates it. (If anyone asks how this is to be understood in terms of movement along an auditory scale, I refer him to differences in the way he anticipates what he is going to do and what is going to happen to him -- differences in the kinds of knowledge he has of these two things.) It might seem that the introduction into our universe of this distinction -- the distinction, roughly speaking, between changes that are brought about, and changes that merely occur -- would necessitate the introduction of the idea of that which brings about the deliberate changes, and hence of the idea of the distinction between oneself and what is not oneself. Surely, one might say, in a Locke-like phrase, the idea of oneself as an agent forms a great part of the idea of oneself. Indeed, I think that it does so, and perhaps a necessary part. Yet the suggested modification of the imagined universe may be insufficient to necessitate the problematic distinction. Suppose our 'reports' are to be composed with an eye to the future as well as the present and the past. Then we shall need some way, in playing our report-writing game for the revised universe, of marking the distinction between what, in terms of our ordinary conceptual apparatus, we might call announcements of intention on the one hand, and predictions on the other. But this distinction can very well be marked without the use of the first person. We shall need something, perhaps, like a grammatical distinction of voice (such a movement will occur; such a movement will be executed). But there is so far no reason why we should admit also the grammatical distinction of person. We need to distinguish what happens by agency from what does

[p83]

not. But we do not need to distinguish agents. The same applies to the reports proper, that is, those which refer to the present and the past. The impersonal form of scientific papers, in which a distinction is nevertheless made between what was done, and what was found to happen, will be perfectly adequate for the reports. More exactly, we shall not need, in the language of the reports, a distinction between a personal and an impersonal form.

If, then, this modification of the purely auditory world will not, on the test suggested, suffice to yield the conditions of a nonsolipsistic consciousness, what further or alternative modifications are required to do so? Would any, indeed, be sufficient? These questions start echoes of many others in philosophy which are, in one way or another, connected with the issue of solipsism. Think for a moment of our ordinary conceptions of ourselves, of the kinds of ways in which we talk of ourselves. We do not only attribute to ourselves sense-perception of things other than ourselves, and action and intention. We attribute to ourselves physical characteristics of a kind shared by other basic particulars of our actual conceptual scheme; that is to say, we have material bodies. We attribute to ourselves thoughts and feelings, and pains and pleasures, which we also attribute to others; and we think of ourselves as having transactions with others, as influencing and being influenced by them. It is not obvious which of these features are essential to a non-solipsistic scheme, and which, therefore, we must try to reproduce or find analogues for in the deliberately restricted sensory terms of the auditory world. Could we reproduce all of these features, while not extending the range of sensory experience beyond the auditory? It seems unlikely, but it is perhaps not impossible. We may, for example, suppose our inhabitant of the auditory world to be able not only to initiate movement along the pitch-range of the master-sound, but also to initiate sounds of a different character from those not initiated by him -- endow him, so to speak, with a voice. The problem of equipping him with a persistent audible body may perhaps be solved by means of the master-sound itself. It is audible to him all the time, and we may suppose that for each inhabitant of the auditory world, there is a master-sound of a different timbre,

[p84]

though no one hears another's except when it is at the same pitch-level or nearly the same pitch-level as his own. Two hearers are then in the same auditory place. We still seem to be short of what is required; as is evident if we re-think this description in terms of the auditory experience of a single such being. What we have introduced, in introducing different 'voices', is different sets of auditory items of which it may be supposed (a) that they are like the sounds initiated by a single such being in a general way; (b) that they are unlike other sounds not initiated by him; (c) that every such set differs from every other in characteristic ways; (d) that each characteristically differing set is associated constantly with a sound which is like his own master-sound in a certain general way, and is never heard by him at a different, or at all widely different, pitch from the pitch of his master-sound at any moment. The most favourable further direction is probably that of simultaneously supposing that (a) of the sounds not initiated by a particular being, those which are like those so initiated may be indirectly influenced in certain standard ways by sounds which are initiated by him; and (b) that sounds of this character tend to stimulate (provide 'reasons' or 'motives' for) initiated changes either in position or sound-initiation. This seems to open the door to something like communication. We might even further suppose that the ability to initiate movement is a development for the single being, not an original capacity, and follows, as it were, a period of subordination to another master-sound. Evidently, in making such suppositions, one would be trying to produce as close an analogy as possible of the actual human condition. But the fantasy, besides being tedious, would be difficult, to elaborate. For it is too little clear exactly what general features we should try to reproduce, and why. It might be better at this point to abandon the auditory world, and face the issues raised by solipsism in closer connexion with the ordinary world. This task will occupy us in the next chapter.

Before leaving the auditory world altogether, I should consider a possible objection to the whole procedure of this chapter. I raised the question, whether we could make intelligible to ourselves the idea of a conceptual scheme which provided for objective

[p85]

particulars, but in which material bodies were not basic to particular-identification; and I selected the model of the auditory world as one from which bodies were altogether absent. I claimed that some of the conditions of such a scheme could be fulfilled in the terms of the model; but concluded that in order to satisfy ourselves that they would all be fulfilled, we should have to reproduce, in the restricted sensory terms available, more and more general features of the actual human situation. At intermediate stages in the elaboration of the model of a purely auditory experience, I spoke of it as satisfying the conditions for a conceptual scheme which included such-and-such features of our own and excluded such-and-such others. But by what right do I assume the possibility of such types of experience, and of such schemes? By what right, in particular, do I assume that there could be such a thing as a solipsistic consciousness?

The objection is one which I hope I have already anticipated. I make no such assumptions as are here questioned. My real concern is with our own scheme, and the models of this chapter are not constructed for the purpose of speculation about what would really happen in certain remote contingencies. Their object is different. They are models against which to test and strengthen our own reflective understanding of our own conceptual structure. Thus we may suppose such-and-such conditions; we may discuss what conceptual possibilities and requirements they can be seen by us as creating; we can argue that they fall short, in such-and-such ways, of being conditions for a conceptual structure such as our own. In all this we need no more claim to be supposing real possibilities than one who, in stricter spheres of reasoning, supposes something self-contradictory and argues validly from it. Indeed we may, if we wish, think of each stretch of argument as preceded by a saving hypothetical clause, by such words as 'If such a being, or such a type of experience, were possible ... '.

[p86]

[end]