Bernard Williams described Wittgenstein as a ‘linguistic idealist’. Kripke called him a sceptic and Russell said that Wittgenstein had given language an untrammelled freedom which it had not enjoyed hitherto. These claims are connected in that they are all directed to something that Wittgenstein denied. The denial in question is that of a reality external to and independent of language on which our use of words is supposed to be based. It is very easy to find this denial objectionable: ‘Surely there must be a reality that is independent of our language and its concepts must correspond to features of such a reality! A denial of this would leave our language hanging in mid-air and turn its logic into something arbitrary! Are we to say that before there were human beings and human language there was nothing – no mountains, no rocks, no water and no rivers? Are we seriously to suggest that the realities with which we engage are a product of our language in the way that, on Berkeley's view, physical reality is a product of our minds, namely ideas – when the ideas in different minds agree in certain ways?’
The objection to Berkeley is that he shares an assumption common to the Sceptic and the Realist, namely that whenever we perceive physical objects we have or are presented with ideas. This means that if we have any knowledge of physical objects this knowledge must be indirect. The sceptic appreciates this and so denies that we can have any such knowledge. To reject this conclusion, Berkeley has to deny that our knowledge of physical objects is indirect. Given the above assumption, however, Berkeley can only do so by reducing physical objects, from the ideas of which alone he assumes we can know them, to these ideas themselves: ‘matter is nothing but ideas’. It is this reduction that is objectionable; it is the crux of his idealism. The assumption which forces him to do so in his determination to oppose scepticism is the original property of philosophical realism. The realist rightly wants to preserve the independent and continuous existence of physical objects in the face of interrupted perceptions of them and so feels forced to claim that our perceptions and knowledge of them must be indirect. Obviously it is the original assumption common to the three philosophical positions – realism, scepticism and idealism – that needs to be rejected. How Wittgenstein does so falls outside the topic of this chapter.
If the linguistic idealism that has been attributed to Wittgenstein is conceived of as parallel to Berkeley's idealism then it must be the claim that reality is the product of human language. Wittgenstein never makes such a claim, but it is easy to thus summarise what one gathers from various remarks he makes in discussing language, logic and mathematics. All the same, it is a caricature of what he says. I shall try to correct this caricature.
‘There is no reality outside the mind and apart from ideas.’ This is Berkeleyan idealism. ‘There is no reality outside language and apart from its grammar – or logical concepts.’ This is linguistic idealism. Their antitheses are Lockean realism and linguistic realism. Indeed, these two forms of idealism are reactions to the two respective forms of realism. Lockean realism thinks of ‘matter’ as a substance, an ‘unknown somewhat’ that supports its various qualities, sensible and causal, or in which these ‘inhere’. Likewise linguistic realism claims that there must be simple objects, bare particulars, devoid of any property, which combined in various ways constitute what we state, describe and speak about in the language we use. These objects constitute ‘the substance of the world’. This is one form of linguistic realism.
I am not suggesting that this is Wittgenstein's conception of the relation between language and reality in the Tractatus. But it bears a certain resemblance to it and some philosophers have read what Wittgenstein says in the Tractatus in this way. On this realist view the simple objects exist independently of language; but on the Tractatus view this is not so. They are ‘instruments of language’; they are the meanings of names. They are not what we talk about or refer to in our use of language; they are what makes reference and the statement of facts possible. Nevertheless, there is an element of realism – linguistic realism – in the Tractatus which may be characterised as ‘Platonic’. For it measures natural languages against language with a capital L which exists independently of these languages and the surroundings of human life in which they are used. The ‘objects’ of the Tractatus thus, while they are internal to language with a capital L, nevertheless exist independently of our actual ways of speaking. Our actual language, our actual use of words, is thus thought to be subservient to something that exists outside of time and space. It is in this sense that we have a form of linguistic realism in the Tractatus that Wittgenstein dismantled bit by little bit in his later writings. This applies equally to his conception of mathematics in the Tractatus.
Thus, while in the Tractatus logic does not need and does not have any metaphysical foundations in an independent reality – ‘logic must look after itself’ – it is itself a metaphysical foundation of natural languages. Actual languages must conform to it; they are the tail which logic with a capital L, as the top dog, wags. In the Investigations, this relation is reversed and the capital L is dropped from both logic and language. We have ‘language-games’ which involve our behaviour; they are organically related in natural languages which are themselves part of human life. They form an important part of the life and culture in which speakers of a language participate. Logic appears in that. It does not have an independent anchor outside or separate from our natural languages and the language-games that form part of such languages. Those in turn are subject to historical change. Thus, logic, though its principles, like the propositions of mathematics, are timelessly true, is not itself rooted in anything timeless.
I want to give another example of a form of linguistic realism rejected by nominalism which constitutes a form of linguistic idealism. The philosophical question to which they respond concerns the generality inherent in the meanings of our words, which finds expression in our using the same word on different occasions, in our applying one and the same word to different instances of the same thing. ‘Surely,’ one is inclined to think, ‘there must be some consistency in the use of a word, some rhyme or reason for its application to the instances to which it is applied. One does not use the word any old how. There must be something about these instances, something that is true of them all, which makes the word applicable to them. There must be something to which its use is responsible, something independent of this use to which speakers of the language conform so that there is interpersonal agreement in their use of the word.’
This is a form of linguistic realism. It claims that what gives meaning to a word and the generality that belongs to it is ultimately something that exists independently of language – the properties that are common to different things in reality, or the similarities which as a matter of fact we find in them. There is thus on the realist view a correspondence between words and things reminiscent of the correspondence between a lid and the many boxes which it fits.
Thus, in his book Thinking and Experience, H.H. Price says that ‘we cannot help noticing that there is a great deal of recurrence or repetition in the world around us’ and that ‘this perceptual repetition … makes conceptual cognition possible’. He goes on: ‘In a world of incessant novelty … no concepts could ever be acquired, and thinking could never begin’ (Price 1953: 7–8). What, however, he fails to recognise is that the repetition of the same in question is something that we notice as people who already have a mastery of language, of a particular language, and living in a world that is lived with language. Consequently, Price's attempt to base the generality inherent in the meanings of our words on something that exists independently of language fails. He is like someone attempting to lift himself up by his bootstraps. Hume recognised this clearly in his attempt to justify induction when he discovered a vicious circle which such an attempt cannot avoid.
Traditionally, this problem of ‘the one and the many’ is put in terms of general nouns or common names and what the things they name name: what is it they have in common by virtue of which we use the same word to name them? The nominalist answers that they have nothing in common except that they are called by the same name. This is a rejection of the answer of the linguistic realist. However, while it escapes the vicious circle of realism, it makes the use of words and language into a wholly arbitrary matter. Superficially and taken in isolation, there are various remarks in Wittgenstein's writings that are reminiscent of this answer and so various philosophers thought that he was a nominalist.
Rightly wanting to deny that Wittgenstein is a nominalist, Renford Bambrough (Bambrough 1960–1961) turns him into a realist: it is not for nothing that we call the many things we call ‘games’ games. They are related to each other by ‘a complicated network of similarities overlapping and crisscrossing: sometimes overall similarities, sometimes similarities of detail’ (PI 66). But Wittgenstein is concerned to reject his own early idea of a general form of proposition here and is using the example of games – that we call many different things games – as an analogy. ‘Look and see,’ he says, ‘whether there is anything common to all. – For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series of them at that.’ We are looking and seeing these things as language-users. This is being contrasted with what we are inclined to say when we think as a philosopher and are in the grip of a picture which suggests to us what must be the case: the idea of a common essence.
We think that the meaning of a general noun which applies to many different instances of the thing it names is to be found in the common essence in all of them – much in the way that alcohol is to be found in all alcoholic drinks, like beer and wine, gin and whisky (BB 17). And with this idea in mind what actually guides us in applying this general noun goes out of focus. For it is in the actual surroundings in which we use it in a particular case that its meaning stands out clearly. We look past these, however, when we search for the common essence of things to which the word applies – the things that are instances of what the word names. As Wittgenstein puts it: ‘in order to find the real artichoke, we divest it of its leaves’ (PI 164). But it is the leaves that constitute the artichoke, each particular artichoke, and without them there is no artichoke at all. Indeed, ‘the real artichoke’ is a fiction.
It is the same with the thing named and with the meaning of the general noun which names it. Thus – to take one of Wittgenstein's examples – one cannot understand what a smile is by concentrating on instances of the human mouth when it is smiling, trying to discern a common geometrical pattern there. For ‘a smiling mouth smiles only in a human face’ (PI 583). ‘Smiling’, Wittgenstein said, ‘is our name for an expression in a normal play of expressions’ (Z 527). One has to consider the circumstances in which one responds to a smile. I would not be able to respond to a face fixed in a permanent smile. Indeed that would hardly be a smile. By contrast, someone may be looking for something he has dropped. I see it, pick it up and give it to him. He smiles as he says thank you. I smile back. It is in such circumstances that we speak of a smile. Remove them or alter them radically, freeze the smile so that it no longer varies with the circumstances, and you no longer have a smile. Without these varying circumstances, there is no smile, yet we have no way of summing them up by any formula or description.
Similarities too, as I said, strike us as language users and in particular circumstances, as do outstanding features which Wittgenstein illustrates beautifully in his discussion of what makes a friendly face friendly (BB 145–6). When we are in doubt, these are what we consider so as to be clear that a general noun which we are inclined to use in a particular case does apply. These are the considerations we refer to in justifying its application. The philosopher, however, here as in other cases – for instance as in the case where, like Hume, he seeks for the justification of induction itself, the ultimate justification of any inductive conclusion – wants to know the considerations that govern the application of any general term. He thinks that if these vary from case to case, there must be behind them, something general that is the same in all cases in which we apply a general term supporting them – the core of the artichoke. He thinks that the considerations to which we resort in any particular case constitute a justification only because of what lies behind them all – namely what we take for granted in our use of general terms. He thinks that what we normally call a justification in a particular case is not a justification unless what we thus take for granted is itself justified: the ultimate justification for the application of a general term.1
The philosophical theses of realism and nominalism are responses to this quest for an ultimate justification. They are answers to the question ‘what is the ultimate justification for the application of the general term?’ taken at face value. They raise the insuperable objections that I have pointed out – the vicious circle in the case of realism and the appearance of an arbitrariness in our use of general terms, and that means of all terms, in the case of nominalism. Wittgenstein did not take that question at face value; he rejected realism and he answered the charge of arbitrariness. In his answering of that charge, he rejected nominalism, and so the kind of idealism which it constitutes.
Renford Bambrough appreciates that Wittgenstein is not a nominalist; but unclear about the way Wittgenstein avoids leaving our use of language hanging in midair, he turns Wittgenstein into a realist here. It is his own realism that gets in the way of an appreciation of the subtleties of Wittgenstein's contribution here. Basing himself on what Wittgenstein says about ‘games and family resemblances’, he claims that in the end all classification rests on the similarities and differences with which nature presents us independently of our systems of classification. He calls them ‘objective similarities and differences’ (Bambrough 1960–1961: 221). Ultimately, he says, ‘there are only similarities and differences from which we may choose according to our purposes and interests’ (ibid.: 222).
There can, however, only be similarities and differences between things that we identify as this or that kind of thing. In other words, where there are similarities there are already classifications. That is, similarities and differences cannot come logically before classes and, therefore, they cannot provide an ultimate basis for all classification – for the existence of kinds and our naming of them. The claim thus can be seen to involve a vicious circle, like the one discovered by Hume in his quest to find an ultimate justification of all inductive conclusions. Wittgenstein makes this point in connection with ostensive definition which Augustine sees as the basis from which language develops. He points out that giving an ostensive definition presupposes the existence of a language and that to understand it one must have already come a long way. As he puts it: ‘the ostensive definition explains the use – the meaning – of the word when the overall role of the word in language is clear’ (PI 30).
As I put it elsewhere (Dilman 1981: 182–3), classification necessarily presupposes objects, situations, actions, etc. that we can refer to, name and characterise prior to and, therefore, independently of the classification. The similarities and differences that it draws on go with that; they belong with our name for or characterisation of what we classify. We do not, as it were, classify bare particulars; we do not start in a grammatical vacuum. And where we have objects to classify, we have already names for them, and so a whole range of similarities and differences from which we can select in classifying them in different ways. That is, we must already have come a long way before we can classify things; we must already have things to classify. If we are to explain what we mean by, say, ‘elm tree’, we shall point to similarities and differences between trees we call by that name and others we call by other names. But then we would have to go on to explain what makes a plant a tree as opposed to, say, a shrub, and so on. And what we shall come down to in the end will not be, as it were, bare particulars which we call by the same name because of the way they resemble each other, but the grammar in which we carry out the comparisons in question. So what we come down to in the end are not, and indeed cannot be, similarities and differences exhibited by nature.
It is just at this point that Wittgenstein raised the question whether grammar, in this case the grammar within which we make comparisons, is arbitrary, whether it is not responsible to anything. It is at this point that he asks whether nature has nothing to say about what we call ‘the same’, ‘similar’ and ‘different’ in various connections of our lives (Z 364). He does not deny that she has something to say; only, he says, she speaks in another way than Bambrough imagines. In Zettel, Wittgenstein asks: ‘If I say “there is a particular similarity among the primary colours” – whence do I derive the idea of this similarity?’ (Z 331). He responds: ‘Just as the idea “primary colour” is nothing else but “blue or red or green or yellow” – is not the idea of that similarity too given simply by the four colours?’ Here, we can imagine Bambrough protesting: ‘Is then our concept of primary colours arbitrary – like my concept of alphas as opposed to the South Sea Islander's classification of trees?2 “Then might one also take red, green and circular together?”’. Wittgenstein's response is: ‘Why not?’
What Wittgenstein says about ‘similarity’ (BB 133) and ‘simplicity’, namely that it makes no sense to speak absolutely of the simple parts of anything, that is ‘outside a particular language-game’ (PI 47) applies equally to ‘arbitrariness’ and also to ‘reality’. Grouping red, green and circular together makes an arbitrary collection for us – that is within our language and culture. Wittgenstein asks: what have red and green on the one hand and yellow and blue on the other in common? (BB 134). We would say, ‘Nothing’. Wittgenstein then imagines ‘a use of language (a culture)’ in which there is a common name for the first two colours and another one for the other two. In the way he sets up this example, he makes it easy for us to see that, for the people he imagines, there would be nothing arbitrary about the classification on which these common names are based.
Wittgenstein gives a great many examples. I shall mention only one other of them. When we hear the diatonic scale, we are inclined to say that after every seven the same note recurs, and, asked why we call it the same note again one might well answer ‘Well, it's a C again’ (BB 140–1). But Wittgenstein points out that this is not the answer we want. For we want to know what makes it a C again. The truth is that we agree in what we hear when we listen to the diatonic scale and we have been taught to use the word ‘the same’ of notes at intervals of an octave on the diatonic scale. Wittgenstein imagines someone who heard the same note after every four or three notes on the scale or – to put it differently – to describe what he hears as ‘hearing the same note again’. The important point here is that we could all have heard the notes on the diatonic scale the way he does and then we would have said that every third or fourth note is the same note. This is the same point as the one Wittgenstein argues in the Investigations with his example of the arithmetical series ‘n + 2’ (PI 143 and §185).
We may speak of Wittgenstein's pupil in §185 as stupid. We speak of people who do not hear the differences we hear between notes as ‘tone-deaf’ – i.e. deaf to the aspects of the sounds that we hear as tones of music. But the reality of this aspect depends on this reaction being shared and on much else in our life that surrounds the activity of music making and listening to it. Thus, if we can describe a person like Wittgenstein's pupil as ‘stupid’ or in the other case as ‘deaf to music’, thus implying the existence or reality of something to which he is blind or deaf, that is only because of the overwhelming agreement of reactions from which he is separated.
So if we had a society where the reactions of those we call ‘tone-deaf’ or ‘stupid’ or ‘feeble-minded’ were the prominent ones, i.e. those that are shared, then those aspects that are important to us in our life would not exist within that society. They would not be part of the reality of those belonging to that society. As Wittgenstein puts it: ‘One imagines the feeble-minded under the aspect of the degenerate, the essentially incomplete. … And so under that of disorder instead of a more primitive order (which would be a far more fruitful way of looking at them)’ (Z 372).
Nominalism, I said, is a form of linguistic idealism, for in rejecting realism it leaves the use of language hanging in midair. It claims that the generality that the meanings of words have depends on language without being able to relate the use of language to anything at all. Wittgenstein, we have seen, does reject realism without embracing any such idealism. To put it more generally, yes, for Wittgenstein ‘reality’ is a relative term, it is internally related to language, internal to the grammar of particular modes of discourse which form part of a natural language. But a natural language is rooted in the life of the people who speak it. Indeed we are the kind of creatures we are in the life we live with language; and all three – our language, our life and we are in constant interaction with each other. They are seemlessly related. Essences, which Wittgenstein says are expressed by grammar (PI 371), are a product of this interaction in the course of human history.
Normally we would say that physics and astronomy are empirical disciplines whereas, by contrast, mathematics and formal logic are not; they are a priori, formal or reflective disciplines. What this means in the case of physics is that, however we develop theories in it, the results that we obtain by means of experiments, thought out and devised in the light of these theories, play a crucial role in our consideration of the acceptability of these theories. More simply, it means that physics ultimately deals with observable phenomena and is based in the end on our observations of these phenomena – however we make such observations.
In philosophy, empiricism starts innocently with a concern regarding the bases of the statements we make in what we observe – in our ‘experience’ as it is often put – and moves onto the question of what these statements must mean if they are to be based on ‘experience’. That is, there is a move here from truth to meaning. We are familiar with the way that Locke and Hume tried to derive the meanings of the terms of our language from ‘experience’, including those of what they called ‘general terms’, to analyse these into ‘ideas’. Later empiricists tried to do the same with the sense of statements, or ‘propositions’ as they often called them. More recently, Quine, or at any rate early Quine, turned his attention to the relation between language as a whole, considered as a system of propositions, and ‘experience’. Even in his pragmatism, he remained faithful to empiricism: ‘the myth of physical objects is a device for working a manageable structure into the flux of experience’ (Quine 1961: 44).
In all this, that is, in empiricism in philosophy or philosophical empiricism, language – that is the meanings of its terms and their generality, the sense of its propositions – are thought to be founded on what we experience independently of language. What we are so supposed to experience is thought of as ‘empirical reality’ with which all reality, or reality with a capital R, is equated, and our contact with it is considered to be unmediated by language. That is, that contact, usually referred to as ‘sense perception’, is thought of as a natural or quasi-natural process, in the way that images are formed in a mirror. We may have to learn to identify what we see, to name it, but we see what we see independently of what we learn. Or at any rate, this is considered to be so of the items that go to constitute what we see – our ‘ideas’ or ‘sense data’, so-called. Thus, empiricism in philosophy is a form of linguistic realism. Insofar as it equates all reality with empirical reality, that is with what we experience, the object of human experience, it is faced with the question whether in what we experience there is a residue that lies beyond such experience. Here a division is formed between them, such as we have between Locke and Berkeley.
Kant pointed out that ‘intuitions’ (what is supposed to be given or supplied by means of the senses) are ‘blind’ without concepts, that is apart from language – and, let me add, apart from the forms of behaviour which language extends and in which it is rooted. He also said that ‘concepts without intuitions are empty’. In other words, human experience, including of course ‘perceptual experience’, cannot be divorced from human language, and mutadis mutandis human language cannot be divorced from human experience. As Wittgenstein would put it, mathematics, for instance, which he described as a ‘phraseology’, cannot be divorced from its application in ‘civil life’, and hence from the life outside the pure mathematician's study where it has many different applications.
Kant argued elaborately in his Critique of Pure Reason, against empiricists, that human experience is not constituted of ‘ideas’ or ‘sense data’ which are caused by an independent reality, or, in the case of hallucinations, by our brain, so that possibly the whole of what we consider to be real may be a dream. It is essentially directed to an object. This undercuts philosophical scepticism. For when I have a hallucination, my consciousness is object-directed; what I see is a hallucination of something, for example a dagger. This object-directed experience in the case of a hallucination is parasitic on the veridical perceptual experience. Thus, unless I know what a dagger is I cannot have a hallucination of a dagger; whatever hallucination I have, it cannot be that of a dagger. Hence unless I am capable of perceiving things, I cannot have hallucinations at all. This is how Kant puts it in a letter to Hertz:
If I had the mentality of a subhuman animal, I might have intuitions [i.e. sensations] but I should not be able to know that I have them, and they would therefore be for me, as a cognitive being, absolutely nothing. They might still exist in me … exercising influence upon feeling and desire … without my thereby acquiring the least cognition of anything, not even of these my own states.3
When Kant says ‘they might still exist in me exercising influence upon feeling and desire’ he means, I think, that if I were, let us say, a dog, I might go after a bone, want to eat it, perhaps bury it and remember it the next day, go and dig it out and start chewing it. So, clearly, as a dog I would be able to recognise a bone when I see and smell one. I would be able to distinguish a real bone from a leather or plastic imitation one. This is something that would show in my behaviour. What I would not be able to do is voluntarily turn my thoughts to the bone, bring it to mind, consider it in its absence. However, even if having ‘the mentality of a subhuman animal’ enables me to be capable of only this much, I would still say that physical reality does figure in the life and behaviour of a creature with such a mentality. Such a creature's behaviour is clearly directed to physical reality; undoubtedly it engages with physical things. In human life, such behaviour and engagement is, of course, extended in the use of language in various directions; they enter into forms of behaviour which do not exist in a dog, for instance – forms of behaviour interwoven with the use of language; in other words, what Wittgenstein calls ‘language-games’.
To return to empiricism, what makes an empirical proposition true or false is something, facts, that we establish by observation and experiment. What then is it that makes it true? What we observe. The empiricist speaks here of the proposition confronting experience, of our comparing it with what it thus confronts. But what it confronts is not experience, of course, but what we experience, what we observe, in a particular situation of our life. This is a life lived with language and the comparison takes place within a particular grammar. The criteria that we employ in our comparisons belong to and come from our language. What we compare it with, what makes it true or false, cannot be described or identified without repeating the proposition or some other proposition equivalent to it. Much of what we experience or observe, the results of our experiments if the proposition is a hypothesis in physics, is mediated by the language we speak, and, in the case of the hypothesis, by the language of physics. This is something not recognised by empiricist philosophy. It is something that Kant began to bring out.
When Wittgenstein spoke of the ‘limits of empiricism’ he meant, I think, that empiricist philosophers were taking the kind of empiricism we find in our physics and common sense, that is in verifying a proposition, checking its truth, beyond the limits within which it applies. They did so by confusing what makes a proposition true with the reality we take for granted in referring to what makes it true. Thus, in a spirit of philosophical empiricism, one may come to be tempted to say that what makes a claim such as ‘the teddy bear from my childhood still exists’ true ultimately is physical reality. For the teddy bear is part of physical reality: it is a physical object. So a philosopher may say: if the teddy bear exists then physical objects exist – very much as G.E. Moore argued. But ‘physical object’ or ‘physical reality’: what is that? Again the philosopher: if my teddy bear is a physical object, and physical objects exist independently of us and of our pronouncements in language, then mustn't physical reality exist independently of our language?
This is once more the idea behind the philosopher's quest for an ultimate justification. The empiricist philosopher goes beyond the bounds of non-philosophical empiricism in seeking such an ultimate justification of what it must be possible to attain empirically. He mislocates the concept of physical reality – a formal concept – by treating it as an ordinary, empirical concept, signifying a class of objects such as trees, mountains, houses, etc. A whale is a mammal, a mammal is an animal, an animal is a living thing, a living thing is an organism, an organism is a physical thing or object: an infinite class. This is exactly how Russell reasoned. But at the limits of what can be ascertained, verified and justified empirically, lies grammar, or what belongs to grammar, and that cannot be justified empirically.
We could sum this up: our language is not founded on an empirical reality with which we are in contact through sense perception. Rather, our language determines the kind of contact we have with such a reality and our conception of it. This is Wittgenstein's Copernican Revolution. I make no distinction between the form of that reality and our conception of it. There is no distinction between them as there is between, say, a table and our concept of a table. It is not what we experience that determines the kind of language we speak; but, rather, the kind of language we have developed, or, better, the kind of language that has developed in a society or community, determines the kind of experience that we have, as members of that community, and so the internal objects of those experiences – the form of the objects to which those experiences are directed.
The empiricist philosopher thus needs to be made to recognise that much of what is first in the senses presupposes the categories of our everyday language. For he holds the opposite of this, namely that there is nothing in language and thought that was not first in the senses. The senses, however, enter the foundations of human language and knowledge in a very different way, namely through their peculiar role in the life of those who speak our language. Both the use of the senses and the use of language are part of our natural history and they are intertwined. On the one hand, the ways in which we use our senses and the activities in which the senses play an important part, are inconceivable apart from language. On the other hand, the language we speak is one that has developed in the context of activities in all of which the use of the senses plays an important role.
The weakness of empiricism is that it gives epistemological and logical priority to the senses. When, in criticism of this, Kant said that ‘intuitions without concepts are blind’ (Kant 1961 (1781/1787] A51/B75) he meant that intuitions or sensations without concepts cannot be directed to objects, they cannot amount to perceptions. Our sense organs are not ‘windows to the outside world’ so to speak. We use our eyes to see and that is something we learn. The use of the senses and the use of language are interdependent. Certain natural reactions are interwoven with both of them – with sight in animals for instance. A dog reacts in certain ways to sights and smells. One such reaction is that it may sniff what comes before its eyes – just like a child may reach for it with his hands. In the absence of such reactions we cannot attribute sight – object-directed sight – to either the dog or the child. The eyes may let in the light, images may form on the retina, and the optic nerve may conduct the electrical impulses thus generated to the brain. But still, in the absence of such reactions, the dog and the child do not see anything; they do not have anything to see. They are living organisms, but with respect to sight they are not very different from a camera that takes pictures of objects to which it is directed.
I stressed that we use our senses – just as we, as seeing and thinking creatures, use the camera. This is something we learn in conjunction with a great many other things. We use our senses in different situations of the life we live and within various frameworks. We look at a vase and see it, see the material of which it is made, see its shape, its colour, etc. Each of these things presuppose different frameworks within which we have learned to operate. As Wittgenstein puts it: ‘Do not believe that you have the concept of colour within you because you look at a coloured object – however you look’ (Z 332).
Peter Winch writes: ‘God's reality is independent of what any man may care to think. But it is within the religious use of language that the conception of God's reality has its place’ (Winch 1972: 12). In other words, if the religious use of language had not developed among men, if men were totally practical creatures, lived somewhat like ants, worked incessantly in order to have food and survive, co-operated instinctively, but showed no appreciation for anything, no gratitude for what they benefited from, etc. – in short if they lived a soulless existence – then, whatever language they had developed, it would not have a religious use. Such people would be recognisably different from us, they would lack much of the behaviour and many of the feelings we have. Their world, the world in which they live, would be different from ours.
Here one may ask: is it their world which is different or do they simply fail to apprehend things that we apprehend? The latter way of speaking presupposes that we and they share a common world, a common language. It is in that language that we may say to someone speaking our language: ‘Look, you thought that what you said did not offend him; but he was offended.’ Here offence and being offended exist in the world of the person we are talking to; he has a conception of it, only he has missed it in this particular case. What if he is totally insensitive, always misses it? Well, then, that is precisely what we say: he is insensitive. This is a ‘disorder’ in him, like blindness. Could we not say that he has no conception of what others feel when hurt? Yes, perhaps he is totally egocentric: he does not have a conception which is available to him in the life and culture of the people with whom he lives. He is oblivious to something that exists or has reality in the world of his language and culture. At least there is a sense – a perfectly ordinary sense – in which he understands the meaning of the word, ‘so-and-so has been offended’. Certainly he is himself capable of being offended. Offence and being offended belong to the world of his language, to its universe of discourse, but he is limited in his sensitivity to others.
Not everything that is alive has a life, of course. Trees are alive, but they do not have a life. For that to be possible, some sort of action and behaviour is needed: going after certain things, reacting to them. Such a life may of course be rich or limited. Life, in this sense, gives a living creature a world. We can thus speak of the world of cats, the domestic cat, for instance. A cat can be upset by the way you treat it, but it cannot be offended. It sits by a hole for hours on end, plays with a mouse before killing it, stalks birds, eats food, rubs itself against things in anticipation of something it wants, purrs when stroked, copulates. These are the things at the centre of its life, the things it knows, does and wants. They make up its world. It does not live in the same world as we do – to a large extent anyway. Many of the things with which we engage and that are, therefore, real to us, have no reality for the cat. They do not exist in or form part of its world. The cat is not merely oblivious of them. For that to be the case, they must be available to its apprehension, they must exist in the world in which cats live.
This is not like saying of a cat that it does not live in my house or garden. It does not even live in my house anyway, in the sense in which I live there. It lives there in an attenuated sense of the phrase. My house is the place it sleeps in and comes for its food. It is part of its territory. What exists in my house may be removed; I may buy some new furniture. But what exists in my world, the world I share with others, what forms part of it, cannot be removed. It may disappear, perhaps, in the course of time; but then in that respect my world, the world I share with those of my generation, would have changed.
Does that mean that there was a time when God did not exist? And does this mean that God could not have created the world as Christians believe? What the second question forgets is that, like most worlds, especially human worlds, the world to which the Christian belief in question belongs is multi-dimensional. One of its dimensions is time and the past. That is, it encompasses a time which stretches in two directions endlessly: the past and the future. So there is no problem about locating creation in the past for as far back as you wish – in our world. As for the creation of the world itself, that is yet another matter. It is obviously not like the creation of an artefact by an artisan, a work of art by an artist. Those are events in the world. The creation of the world, therefore, cannot be like that. But this is not the place to discuss what is meant by God's creation of the world. Certainly part of what it means is that we stand in a creaturely relation to God and that everything that faces us in our life is the will of God. This raises the further question of what it means for us to stand in a creaturely relation to God – what it means for the believer to see his relation to others in that light. It raises the question of what it means to regard events in life as the will of God. But these further questions, difficult as they may be, do not raise any new problems for the relation of our life and language to our world and its dimensions of reality.
What I have argued in this section of the present chapter is that the human world with its many dimensions of reality belongs to our life and language as does our mode of existence as human beings. This world, our life and language, and our mode of existence, must have evolved together and are inseparable. I have expressed this in my own way, but I take it all from Wittgenstein. We have seen that in Zettel §372 he asks us to imagine a society of the feeble-minded, and comments on our inclination to think of its people as ‘essentially incomplete’ – that is as living in a world many aspects of which are a closed book to them.
But is a cat ‘essentially incomplete’ because, unlike a dog, it is aloof from those who take care of it? ‘Why can't a cat be more like a dog!’, its owner may say. ‘Why can't it show some appreciation and gratitude; why can't it be giving; why must it be all for itself!’ What in such words one fails to recognise is that a cat is a cat and lives in its own world. That world only partly coincides with a dog's world and even less, much less, with a human world. The term ‘the human world’, of course, is itself a generic term and covers many differences. To think of the world in which a society of the feeble-mind, such as Wittgenstein asks us to imagine, live ‘under the aspect of disorder instead of a more primitive order’ is like thinking of a cat as an ungrateful and narcissistic creature. As I have explained, I do not see any linguistic idealism in what Wittgenstein asks us to think in the case of his imaginary society of the feeble-minded.
I spoke of ‘our world’, ‘our reality’ and said that ‘reality’ is a relative term. That sounds odd because the possessive pronoun is totally redundant in the use of our language, indeed in the use of any language. For we speak to people who share our language. But when we do philosophy and our questions lead us to consider ‘what using language is’ and so to consider what we take for granted in our use of language, and to imagine different languages, different uses of language, the situation changes. If we can make ourselves at home in this new situation, we may come to appreciate that there is no reality independent of the speakers' shared reactions, engagements, life and language. Reality, if I may put it this way, is what we respond to, engage with and live in for those engagements. It is the dimensions to which what we refer to and describe in our language belongs. It is the internal or ‘intentional’ object of our responses, engagements and speech.
As I argued, the world in which we live, the human world, with its different dimensions of reality, has developed hand-in-glove with our life and language. They are all of a piece, and so is our mode of being in that world, what makes us human beings. They cannot be prised apart. This is very different from the claim that reality is created by language. That would be linguistic idealism. As for the existence of things that precede human existence and human language, what is in question are historical, or rather prehistorical, facts, and it takes human language to state them. The truth of what we thus may state, for instance that there were dinosaurs before there were any human beings, is independent of our stating it. It is, however, the possibility of such facts that is dependent on language and that possibility is part of what characterises our world.
I said that our conception of reality is internal to language and that the conceptions of reality internal to different languages do not entirely coincide. There are difficult questions here which Wittgenstein discusses in On Certainty, which would make the topic of another paper. But I simply want to touch very briefly on the question of whether what I have just stated is ‘cultural relativism’. Cultural relativism is the claim that what is true in one culture may be false in another. In other words, it claims that the same belief may be true in one culture or for one people and false in another culture or for another people. I dissociate myself and Wittgenstein totally from such a claim. What is problematic here turns on the identification of the same belief in different cultures. In so far as the same belief is held in two different cultures, this is a respect in which the two cultures overlap. In such a case, if one of the two cultures regards it as true and the other as false, one or the other must be mistaken. The same belief cannot be both true and false at the same time – true for one people and false for another. But people belonging to different cultures may hold some very different beliefs from each other; people belonging to a different culture from ours may have criteria of truth incommensurable with ours. However, then the beliefs measured by such criteria would be different to what they appear to us to be, the words in which they are stated when translated to our language would be misleading. They would not mean what we may take them to mean – that is if we understand them at all.
What Wittgenstein said about someone who believes in the Last Judgement or the Resurrection and someone who does not (APR 53–5) applies here too. They may not hold opposite or contradictory beliefs to each other. What one person rejects may not be what the other believes; but this may not be apparent to them at all. The rejecter may be far from understanding what the believer believes. They may be speaking at cross purposes to each other, reasoning within very different grammars. The rejecter may take what is in question as a scientific or quasi-scientific belief; for instance: ‘Particles will rejoin in a thousand years, and there will be a Resurrection of you.’ He may fail to see that it takes a religious use of language to express what the believer believes; he may even lack all conception of a religious use of language. The situation is worse in the case of beliefs held by people belonging to an alien culture.
It is easy to confuse a claim which attributes relativity to truth and one which speaks of the relativity of criteria of truth belonging to different ‘language-games’. Then, in rightly wishing to reject the first, one may come to deny the second. Wittgenstein certainly upheld the second, but only the second, view; it is of a piece with the way he talked of different language-games and also different world-views.
Let me return to linguistic idealism and repeat that what it claims is a simple determination of reality by language. In the case of nominalism, this determination of reality by language is expressed in the claim that what makes different individual things the same kind of thing is simply that we call them by the same name. In contrast, what Wittgenstein brings out is that while our conception of reality is internal to our language, that language itself is rooted in a life which the speakers share. It has come to have the form it has through a process of historical development. The forms of language within which we speak and reason are conditioned by facts of our natural history and environment. This is very different indeed from linguistic idealism.
1 See Dilman 2002 Ch. 3 section 5.
2 ‘Alphas’ are Bambrough's example of a supreme kind of arbitrariness, a word which is supposed to name each of a random collection of objects. See Bambrough 1960–1961: 199–200.
3 Quoted in Bennett 1966: 104–5.