There is a double circuit of knowledge connecting man with the cosmos.
It is also to be noted about this sensory knowledge of the cosmos that it is incomplete: for, the Universe being continuous, it must follow that any point is indissociable from its neighbours (however distant these may be); and in this sense there are present at any one point an infinite number of characteristics of the Universe. The human senses, being restricted in number and therefore not infinitely perfect, are unable to apprehend at any one point more than a certain number of these characteristics. This composite of characteristics gathered from the cosmos at a given point constitutes what man calls an 'object' at that point. This idea of an object is likely to be very clear to man, for he is dealing with a concept directly materialised by his senses and which he can therefore express in an immediate fashion. He can invent words, i.e., a language, to designate each of the objects he perceives; he can say, for example, "This is a red wooden sphere three centimetres in diameter".
The next step towards some kind of 'intellectualising' of the concept of an object is to start from this simple notion and then make a number of abstractions from it. To do this, you will need to select only certain characteristics in the object perceived, thus making it possible to have a direct apprehension of the sense of a group of words which need not necessarily correspond to objects actually perceived. For example, in the sentence quoted above, a first step in abstraction would be to do away with the 'red' and read 'a wooden sphere three centimetres in diameter'. This would give a picture of the set of spheres of all possible colours, but of this shape and material. A further abstraction would be to do away with the word 'wooden', and read 'a sphere three centimetres in diameter'. Lastly, in a final abstraction, only the shape would be left — 'a sphere'. Thus we can pass from one abstraction to another, using the objective language which has to do with objects perceived by our senses, and construct a language in which each word has a meaning directly associated with our perception of Nature.
The next stage in improving this objective language consists in using 'relationships' to bring about a correspondence between characteristics belonging to different observations, i.e., to different objects. Thus I can use my senses to consider the object 'Earth'; and I note that it has a characteristic in common with the 'red wooden sphere three centimetres in diameter'. So I state the relationship, 'The Earth has the form of a sphere'.
The Quantum Theory, based entirely on observation, makes use of this objective language in its most elaborate form. Here, the network of objects, abstractions, and relationships reaches a complexity and refinement to which the language of everyday speech is unaccustomed. Nevertheless this language in the nature of the case remains objective; and we must note that it will never be able to express anything else but the Known, that is to say the immediate data of our senses. On the other hand, it will always be relatively easy to 'visualise' the meaning of this language by a direct appeal to the experience of our senses.
But this is where the difficulties begin. Man cannot put into words this 'communion' with the whole cosmos, for it represents an infinite number of perceptions, abstractions, and relationships all lumped together. It is thus not possible to dissect this total intuitive feeling into words. The first word tentatively invented to convey this feeling to someone else would have to be able to express the whole all at once; but it would certainly fail to suggest anything like what a normal 'word' conveys in any known language.
What, then, is to be done? Man's first instinct is to say nothing; but he often has that feeling of silent awareness in front of some marvel of Nature, or when he stands at some sacred spot.
But man wants to communicate to his fellows any experience he may have had of the knowledge of the cosmos, even if this is only intuitive knowledge. He therefore embarks upon a course opposite to that which has led him to work out an objective language.
So he proceeds to do the opposite of abstraction, and begins a process of 'concretisation'. He borrows words from his objective language to symbolise this feeling of instinctive knowledge. This is how all religious language is built up — upon images and symbols. A new step forward in concretisation produces the language of art, which is likewise symbolic inasmuch as it attempts to apprehend a reality hidden beneath the appearances registered by our senses.
The objective language was perfected by 'relationships' between different objects. But since intuition is awareness of the whole, symbolic language will follow the opposite course and progress by means of 'simplifications', reducing the infinite number of relationships between all the points of the Real to a finite number.
Is it possible for a true science to arise from this symbolic language? This is an intensely interesting question, for it is certain that this circuit of direct knowledge brings us fundamental truths which it would be very desirable to incorporate into scientific knowledge.
Could this be true of our symbolic language as applied to the intuitive awareness of the cosmos? It must be admitted that religion and art are two examples of symbolic languages which are not interpersonal, and so not scientific. A work of art — whether in music, painting, sculpture, or poetry — is very far from arousing the same emotional or visual response in everybody.
Why is this so? Surely we should not simply accept this objection without pushing our inquiries a good deal further and discovering why this important circuit of intuitive knowledge could not be the basis of a science.
As intimated above, reflection suggests that there is a particular importance in the first word to be coined as an expression of our intuitive knowledge of the cosmos. The intuition must be one, since it is an awareness of the whole; and the first word to be chosen to denote this intuition must define the whole 'substance' on which the rest of the language is to be constructed. We shall then go on to say that this 'substance' is like this or that by building up symbolical imagery from the vocabulary of our objective language.
In order that the symbolic language should have some chance of being scientific, the first word must be scientific too, that is to say, interpersonal, and have one and the same meaning for all of us.
Suppose we try to disconnect ourselves from the whole of our exterior world by withdrawing from all our sense-perception, blocking our ears, and closing our eyes. What instinctive awareness still persists apart from the channel of our senses?
The answer given by religion is 'the soul'. Here is religion's first word: our 'substance' is to be the soul. But this is not a scientific word, for it is on too deep a level; it does not arouse the same feeling in everyone. The whole symbolic language of religion is destined to remain on this deep level and will never, unfortunately, become scientific.
The answer of art is that the 'self' is primary. "A work of art", wrote Emile Zola, "is a corner of creation as seen through the medium of a particular temperament"; and he was quite right. But does the term 'self' express a scientific reality? Here too the level of intuitive knowledge remains too deep: no interpersonal symbolic language could be built up on the 'self'.
What is the verdict of the scientist? He supports the philosopher's opinion, for he holds that the feeling of which man is aware when he has suppressed all the messages from his senses is that of the passing of subjective time, the notion of duration — in fact, the notion of our own ageing.
The miracle has been accomplished! Here at last is our long-sought word, the word to serve as a basis for our scientific symbolic language. Present-day physics, and in particular General Relativity, does in fact tell us that the process of ageing has the following two characteristics:
These two stages in the progress of a symbolic language stand out very clearly in General Relativity, and provide it with its bases and its methods. First, there had to be a concretisation of the intuitive feelings in the form of this concept of space-time, which is the basis of all Relativity. Then there had to be a simplification of the number of relationships provided by space-time, obtained by choosing certain broad principles of Nature (such as the conservation of energy) as postulates on which the whole language would be built up. All this was the work of the a priori or axiomatic method discussed at great length in the previous chapter.
Fig. 1 is a diagram of this double circuit of human knowledge.
As an illustration of this, let us first take the symbolic language of dreams. Imagine that in a dream I have seen myself stretch out my arms and fly through the air. I know that this picture can only be a symbol whose meaning is of necessity something different from what the dream set before me. I can describe my dream to someone else in the form of language, but my hearer and I will be perfectly aware of the symbolic meaning of this language, in spite of the fact that neither of us will have the slightest difficulty in 'visualising' a 'man flying through the air simply by holding his arms outstretched'.
In the same way General Relativity, using the symbolic language of geometry, is able to describe what our Universe looks like quite independently of our sense-perception. Thus General Relativity tells us that an elementary particle can be expressed as 'a certain curvature of space-time'. Now this is an image we can perfectly well 'visualise'. It means that if one were to draw lines representing 'the shortest distance between two points' (the equivalent of our 'straight line' in empty space) in the region of space-time occupied by an elementary particle, we should notice that these lines were 'curved', just as the shortest route between two places with a mountain in between them is necessarily curved. Thus there is no difficulty in understanding what is meant by the 'curvature of space-time'. But — and this is where the symbolism comes in — our senses never reveal an elementary particle in this way; they merely give us a 'corpuscular' image of an elementary particle.
Take, for example, the little luminous impact of an electron striking a cathode ray tube screen: this sensory perception of the electron has no direct relation to a curvature of space-time any more than the sensory image I can conjure up of man is related to 'a man flying through the air'.
Here, then, is an important characteristic of this symbolic language: one can form a clear picture of the images expressed by it, but these images do not correspond to our sensory knowledge of the Universe.
This leads at once to a further question: if this language is not addressed to the images of our 'known' universe, do the logical relationships of the objective language also hold good between the different terms of this symbolic language?
Taking this symbolic language in its widest connotation, that is to say, as including the languages of religion, art, dreams, etc., the general answer to this question must be in the negative. It is possible that the connections between images suggested by the symbolic language may fit in to a logical framework, but this is not a necessary requirement. For the symbolic language may try to express — as it does in art and religion — unconscious and irrational links between man and the whole cosmos. In that case, language works only by way of 'suggestion'; it aims at producing 'feeling' rather than 'knowledge', and for this it is not necessarily bound to have recourse to the rules of logic. Thus one can hardly find any 'logical' relationships between the notes or harmonies of a musical score. The symbolical language of music is therefore not a 'logical' language.
But if a symbolic language is also aiming at being scientific, it must of necessity also be logical. This is the case with the symbolic language of General Relativity. It uses the language of geometry and mathematics, which is the logical language par excellence.
We have therefore reached the conclusion that a symbolic language is not necessarily logical except when it also claims to be scientific.
This language, particularly in the form occurring in a Unitary Theory, allows us to draw up a map of the Real, that is to say, to describe what the Universe is independently of man's sensory observations. This map does not appear to have any 'gaps' in it, for it gives an account of the look of the Universe not only at every point but also in its totality. So we have a right to ask the question: Is this map without 'gaps' a complete map? That is to say, can it give an answer to all the questions we could ask about the Universe?
This is an important problem: for if the answer is in the affirmative, it will mean that once this map has been made there will be nothing more to discover — at least in principle. All that ever could be known about the Universe would be as it were 'foreseen' in this map; and anything that man could observe would find a place and an explanation in it.
Very well then. Let us by way of example take the map of the Real proposed by the Unitary Theory, and let us see, by examining it on the smallest and on the largest scale, if we are not led to put at least one question that the map cannot answer.
At the small end of the scale, all seems to be well: the distinction between the Real and the Known disposes of the wave/particle difficulty, for the real is continuous and undulatory; discontinuity appears only in the Known and arises from the limitations of our senses.
The map of the Real then gives a place to all observed phenomena, in particular the physical fields — nuclear, electromagnetic, and gravitational.
But there remains the problem of the Universe as a whole. This map shows space-time to be curved and bending back on itself like the surface of a sphere; so that if one set out on a journey through the cosmos 'following one's nose', one would in the end come back to one's starting-point — as happens on this Earth.
We can 'visualise' this situation without any difficulty. We have only to realise that a 'straight line' drawn in space-time is never the straight line of Euclidean geometry, but is of necessity slightly curved, so that if produced far enough it becomes a circle and closes in on itself.
But this raises a 'simple' question which, however, on further examination appears to be not quite so simple: if the Universe is analogous to the surface of a sphere, what is there 'inside' or 'outside' the sphere?
Is this a question that can be answered? "Yes", says the physicist, on first thoughts. "I have just shown that the entire Universe consists of the surface of this sphere. You therefore have no right to ask me 'what there is' outside this sphere, for if there had been something it would have been part of the Universe (since the Universe is the sum total of all things), and I shouldn't have told you that the Universe consisted of nothing but the surface of the sphere. Your question is therefore meaningless; I can no more answer it than if you asked: 'What is there inside a shadow?' A shadow has no inside; and the Universe has no inside or outside."
But if the questioner persists and is a bit of a geometrician (and probably persists because he is a geometrician!), he may comment as follows:
"I cannot imagine any geometrical figure without giving it some sort of outline. And so I cannot imagine a straight line actually drawn in space without considering it as delimiting two regions of a surface, that is to say, a geometrical figure having one more dimension than my straight line. In the same way, I cannot understand what a plane surface is unless I endow it with the quality of separating two regions of a volume, that is to say, a figure with one more dimension. Thus I cannot imagine the surface of the sphere which symbolises the whole of our Universe without at the same time imagining the volume in which this surface is immersed. Without hesitation of fear of error, I can extrapolate these considerations to the space-time which constitutes the reality of our Universe. If this space-time is comparable to a four-dimensional geometrical figure, I am forced to admit that this figure in 'immersed' in a space of one more dimension, that is, five; moreover, if I admit that this fifth dimension is describable in geometrical language, I could continue this reasoning ad infinitum, and see the necessity for a sixth, and then a seventh, dimension, etc., etc. So I prefer to say that this fifth dimension does exist, but that the space that constitutes it is not describable in any language at all. This conclusion is different from the one that maintained there was nothing outside our Universe describable in four dimensions; for I have just shown that this 'outside' exists, though it cannot be described."
It must be admitted that this kind of argument seems perfectly justifiable. If I describe the Universe as a map of space-time (and therefore four-dimensional), it looks as though I must at the same time admit, according to the foregoing reasoning, that this space-time is immersed in a space with at least one more dimension (that is, at least five). Now, as I postulated at the start that space-time constituted the whole of the Universe, I have reached a contradiction, since I am showing that there must necessarily be something 'outside' this Universe. Thus the question 'what is there outside the surface of the sphere representing our Universe?' has shown up a flaw in the 'logic' of the map we had made of the Real. Not only is the map unable to answer this question; but the very framing of the question exposes a contradiction in the map itself.
This seems to be a serious objection which needs deeper analysis. After all, we built up all this geometrical language of symbolism from a certain number of postulates that had been well borne out by Nature; and from these postulates we reached the map of the Real by means of simple logical deductions, 'theorems' which are therefore as 'true' as the postulates themselves. Where then has the contradiction crept in?
Gödel succeeded in demonstrating quite conclusively the following fact: any logical language, such as that of mathematics or of geometry, is built up from a certain number of definitions and postulates on the basis of which this whole language is constructed by a series of logical steps. Within this logical language it is thus not possible to find any contradictions. But anyone has the right to choose a new set of postulates and definitions and, on them, to construct another language. It may then be possible for a 'theorem' deduced from this second language to be in contradiction to a 'theorem' deduced from the first language, in spite of the fact that the postulates from which both these languages set out were not themselves contradictory.
On reflection, one can understand the significance, and even the physical explanation, of this important result arrived at by Gödel. The Universe is continuous and, as we have already demonstrated, each point is indissoluble from the whole, so that at every point there are an infinite number of relationships with the cosmos, appearing as an infinity of aspects. A language of any kind attempts to organise this infinite number of relationships according to a 'system' defined by postulates and definitions laid down as a basis for this language. Thus we make a sort of 'cross-section' of Nature. The language reduces the infinite number of relationships at any point to a finite number; it is therefore always incomplete. But on the other hand it introduces the possibility of a logical system, which is never possible when dealing with an infinite number of relationships in a continuous and unorganised milieu.
This throws light on the fact that any aspect of the Universe described by a particular language is never an 'absolute' aspect, but is essentially dependent on the premises selected for the construction of the language in question. Since these aspects of the cosmos are infinite in number before language sets about organising them, they could just as well be described as 'black' or 'white' according to the way in which it was decided to 'organise' the infinite number of universal relationships at any given point. This is not to say that the Universe could not be described as both 'black' and 'white' in the same language; it simply means that one language will see as 'black' what another sees as 'white'. There is no contradiction in the 'territory' — that is to say, in 'what is'; but there may be contradictions between the 'maps' made by men by means of language to 'describe' Nature.
A little while ago, I intentionally mixed two different languages to show how this can lead to a contradiction. Without warning the reader, I did in fact mix the objective language of the known with the symbolic language of the real. For the symbolic language had shown us that the Universe taken all together was like the surface of a sphere, and only the surface (without any talk of 'inside' or 'outside'). But then another question arose, coming — though we did not specifically mention the fact, nor was it very obvious — from the language of our senses (and not from the symbolic language).
We had stated, it will be recalled, that it is not possible to imagine any figure other than as immersed in a space of at least one more dimension. Well, in that statement we had left the framework of our symbolic language, which had said to us: "I am talking of the surface of a sphere; I know exactly what that is — there is no ambiguity about it, for I can give you the equation for that surface. But mine is a symbolic language: don't try to see Nature through the medium of your senses as the surface of a sphere. To do this will probably land you in impossibilities or contradictions, just as it was impossible for you to 'see' the 'flying man' in Nature whom you described in the symbolic language of dreams."
Yet the basic postulates of the 'objective-known' and the 'symbolic-geometric' languages are not contradictory: they are simply different, and in a certain sense complementary. Nevertheless their 'theorems' do lead to contradictions, for the Universe cannot be simultaneously four-dimensional (as the symbolic language assures us) and have at least five dimensions (as the objective language suggests). The apparent contradiction between the 'wave' and 'particle' theories was in fact also the result of mixing these two languages; it disappears as soon as a distinction is drawn between the language of the Known and the symbolic language of geometry.
Kurt Gödel's great merit was to demonstrate mathematically that if one took two different languages (and more specifically the languages of mathematics and metamathematics — this last being a language constructed for discussing the language of mathematics itself), it was always possible to find a meeting-point between these two languages where there would be an apparent contradiction. An aspect described as 'white' by one language would be described as 'black' by the other. We have just seen why and how this is physically possible.
The Universe is like the surface of an ocean. In the language of the air, above its surface, a wave would be a 'plenum'; but in the language of the water, beneath the surface, the selfsame wave is a 'hollow'. In short, each of these languages represents a different 'cross-section' of Nature, and so each of them gives us some additional information about Nature by showing a multiplicity of facets, without ever exhausting their infinite total number.
The symbolic languages of religion, art, psychoanalysis, and General Relativity are thus 'cross-sections of Nature', at levels of varying depth; they are complementary to one another in their attempt to restore that direct intuitive liaison which exists between man and the cosmos as a whole. Objective language, which describes the sensory links between man and Nature, is yet another 'cross-section' on a different plane.
Although all these languages help man to a better understanding of his Universe, it is nevertheless necessary to be very careful when putting questions in one language about the description given in another language. We have just stressed the contradictions that can arise, unless great care is taken, through the use of two languages at the same time — the symbolic language of geometry and the objective language of the senses. We must now go on to note that this sensory language, the language of our ordinary daily relationships with one another, is itself composed of a multitude of languages; and Kurt Gödel is there to warn us of these contradictions and the resulting difficulties which may arise unless the situation is appraised with the greatest possible insight.
There is no doubt that language, which is so important as a medium of communication between men, is itself of fundamental importance in each of our lives. But what exactly is its role, and how far is it really fundamental?
If we are to give the right answer to this question, a very careful examination of it will be necessary.
One's first reaction is to say that language is important, but that it is after all only a 'tool'. The primordial element in human communication is the idea and the personal character belonging to it, since the idea varies freely from one person to another. Language, on the other hand, is only a means, and is common to everybody; besides, in a certain sense, it is not free, for it is very generally applied to objects and phenomena which 'exist', and on which language is quite content to model itself. In other words, language appears to be conditioned by the idea. When two individuals do not agree with one another in the course of a discussion, the blame for the disagreement must be put upon their respective 'personalities' (that is to say, their ideas as a whole) with regard to the precise point under discussion. Language plays only a secondary part in the affair — an important part, but secondary. The same is true of two gladiators engaged in combat. Their weapons are important, yet secondary: the prime consideration is the strength and skill of each combatant. In the same way, language is a weapon; but it is the man behind it who really counts.
Now, is this a true account of the position? How free are we to express our ideas? Is it not true that these ideas, as they come to us, are conditioned by the language we have at our disposal? We surely have a right to ask ourselves this question, after having seen in the preceding pages what a language really is.
Each of us builds up his own language on the basis of a certain number of postulates and definitions which are certainly very closely linked with man himself. One man will choose certain postulates according to his temperament, his intellectual aptitudes, and the circle in which he has been brought up; another man will make use of other elements in the reality that constitutes his interior and exterior world to construct his own specific postulates. These two men's postulates are not, generally speaking, contradictory with one another: they are simply different. The possibility of choice in the postulates from which a language sets out arises from the fact that everyday language is based upon the objects perceived by the senses. Each of us, by associating objects with one another by means of certain relationships, and by making use only of some characteristics of these objects to form abstractions, arrives at certain concepts (which are in fact the postulates of language) that are specific to himself. Thus the word 'table' is a concept that does not hold exactly the same meaning for each of us. One person will conjure up the picture of a square four-legged table; another, living in a place where tables are usually different, will picture a round table with three legs. So there can be two 'postulates' in the language dealing with the concept of a 'table'; and they are not contradictory but simply different.
Thus all language, setting out from postulates, is built up by logical steps — at least in the case of those who are considered 'sane'; and we are concerned only with them for the time being. So language is built up like geometrical theorems by way of logical deduction from a finite number of postulates.
A man who is well versed in Euclidean geometry is considering certain figures to which Euclidean language applies. Let us suppose that he is commenting for the benefit of his audience on a triangular figure drawn on a sheet of paper. He can choose from among a certain number of ideas and judgments referring to this triangle. He can affirm — amongst other things — that a particular side of the triangle is the greatest, or that a particular side is the shortest. All his audience agree with him. Then at a certain moment the geometrician states that the sum of the angles of this triangle is exactly equal to 180 degrees. At this point, one of his hearers expresses disagreement, and intervenes with a statement that the sum of the angles of a triangle is never exactly equal to 180 degrees, but slightly less.
Here then is a problem on which these two geometricians are divided. It is clear that the sum of the angles of a triangle cannot simultaneously be equal to and less than 180 degrees.
One way of treating this problem would be for the Euclidean geometrician to treat his interrupter as mad and, if he persists in his objection, throw the geometry book at his head.
A second and wiser course would be to ask the interrupter to justify his statement that the sum of the angles of a triangle is less than 180 degrees. The interrupter would then reply that his geometrical postulates were not taken from Euclid, but from Riemann. Now in Riemann's geometry, there is a possibility of space being 'curved'. This curvature is confirmed by physics, but it is ignored in Euclid's 'ideal' geometry. Thus in a triangular figure drawn in our physical space, the sum of the angles is never exactly equal to 180 degrees, but always slightly less.
Now let us put the following two questions:
First, were these two geometricians free in the choice of ideas and judgments they made on the subject of the triangular figure?
The answer must be no — they were not free at the actual moment of making their judgments. They had been free beforehand to choose the postulates belonging to their language, always supposing that society had given them any possibility of choice. But once the postulates were chosen, one of them was obliged, if he wished to remain logical, to assert that the sum of the angles of a triangle was equal to 180 degrees, whilst the other could continue to be logical only by asserting that it was less. In other words, this example clearly shows that ideas and judgments are conditioned by language, and not the other way round. It also shows with equal clarity that two languages built up on non-contradictory postulates (Riemann's geometry being simply more general that Euclid's) may lead to contradictory statements; for the sum of the angles of a triangle cannot at the same time be equal to and less than 180 degrees.
And now our second question: can we say which of these two geometricians was right — the supporter of Euclid or of Riemann?
Our first answer might be that both were in a sense right, for they had both argued logically on the basis of non-contradictory postulates. But the disciple of Euclid was speaking of a triangle in an uncurved space, that is, an ideal triangle — for our physical space is in fact always curved. Euclidean language is thus seen to be an ideal language. But the disciple of Riemann was thinking of a triangle in our physical space, and so was using a physical language and physical postulates. The confusion between these two languages led to a contradiction.
Another answer might be that the problem of knowing which of the two was right no longer arises in practice as soon as each of the two observers realises that the same words belonged to two different languages. This realisation constitutes an enrichment of their knowledge. They now admit that both their statements have a right to be made, and it will certainly never enter their heads in the future to accuse each other of bad faith, or treat each other as mad.
This was what Alfred Korzybski had been insisting upon since 1930 in his creation of the science of language and human behaviour usually known as General Semantics. He had seen the necessity for this realisation of what is truly implied by a language, and the importance of language as conditioning our judgments. Many of his views now appear out of date; but he had nevertheless caught a glimpse of the fundamental part played by language in the humane sciences, and even in sciences pure and simple.
As already stressed, language is of prime importance in science, for it decides what 'cross-section' we shall choose to consider in Nature. We have already seen that dilemmas as crucial as that of continuity-versus-discontinuity, which has divided scientists for thousands of years, can be finally resolved by realising that it is a question of two different languages.
On the other hand, once this fact has been realised, a new window has in fact been opened on to man and on to Nature. Its effect is to induce in us a certain tolerance which might be called 'an enlightened tolerance'. It must not be confused with a systematic and undiscriminating acceptance of any point of view, no matter what it may be. It is a tolerance based upon what we learn from our present sources of knowledge, not accepting everything, but willing to look at everything. It is a question of choosing our references on the scale of the Universe as a whole, instead of being content with that part of reality which is orientated towards us personally and individually, seeing that the individual can never detach himself from the postulates that have conditioned his thought in the material and social environment in which he has been brought up. It is a question of calmly assessing how another person talks and therefore thinks differently from us about the world. In short, it means looking for more than we have already seen.
Surely it is time now to think of revising our language and launching an attack on the knowledge of man himself. Surely the mass of accumulated observations on the subject of man, from the biological as well as the psychological point of view, is a justification for choosing some broader postulates for the description and subsequent study of man. We have just seen the central importance, for the very ideas we form, of the language in which we express them. This broadening of our horizon through a change in language has already takem place in physics: should not an attempt be made to carry out a similar programme on the subject of man? For the problems connected with life, or even with psychism, will hardly be solved without a change in the postulates of the languages of biology and psychoanalysis.
We shall attempt to examine these points in the next two chapters, which will deal respectively with biological man and psychic man.