THE TREATMENT OF THE PROBLEM Or PPlVACY IN WITTGENSTEIN'S LATER WRITINGS A thesis presented to the University of Cape Town in fulfilment .of the requirements for the degree of Mast e r of Arts in Philosophy. by B. J. HOLIDAY Supervised by B.P. P . Keaney Ph.d. the CO:)vrir:ht c , . · - -Cape Town, 1973. - · ' "" · o, t •<• · t !,e .. ·--. U · . . .. ..> '' SI S is h fd b -1 ·n 1vp~· ~ !·v . . - ··· . _ · e y the I -· _ ... J ... .~ c : t_ :--·:· - ~ ; :) ·.· :n . ' r.. py--- .. -·. ! -,· . ·. ~ .. 1 .. : - • '~ • , :'11 -I m&;' b.:: ,-,. , -· -- · - ' r any part r t f • '''• ~ : ' ·• : ' • • • ·_ \ ' i f "' - •' ,-, .., 1 not- fc ·- n 1 • •· . 1 ,- _,-,N ..::. only and : __ -~~- ~ · ~-~-~::.~ .·. , ~~ ... ~ ......... ,__ . -----~ .... The copyright of this thesis vests in the author. No quotation from it or information derived from it is to be published without full acknowledgement of the source. The thesis is to be used for private study or non-commercial research purposes only. Published by the University of Cape Town (UCT) in terms of the non-exclusive license granted to UCT by the author. (i) TABLE OF CONTENTS Abbrev i ations Acknowledgements Introduction PART I Introduction Chapter I Chapter II Chapter III PART II Introduction Chapter Chapter Chapter IV v VI Chapter VII Chap ter VIII PART III Introduction Ch apter Ch apter Chapter IX X XI Bibliography THE OPENING PHII..DSOPHICAL SECTIONS OF THE INVESTIGATIONS The Atomi stic Concept of Lan gu a ge Philosoph ical Problems and Illusions Following Rules PRIVACY The Concept of a Criterion Private Language Grammatical Li ngu istic Formalism and the Theory of Communication-Intention Sceptical Doubts: The Inner-outer Doubt Sceptic a l Doubts: Wittgenstein's Treatmen t LANGUAGR AS INTENTIONAL ACTION Intention Human Be ing s Language -Games and Forms of Life ( :i -: . 2 :;; ~ _.•.J ''"1 .. : g.: 1 , ~. .. '1 -· ..J. "· ' (ii) ABBREVIATIONS In the footnotes at the bottom of pages, the following abbreviations of the titles of Wittgenstein's works have been used: Tractatus = Tractatus Logico•Philosophicus B.B. F'.M. P.r. o.c. z. = Blue and Br.own Books = Remarks on the Foundations of Mathematics - Philosophical Investigations a On Certainty = Zettel My practice in referring to the Philosophical Investi-gations has been to give section numbers for Part I and page numbers for Part II. (iii) ACKNOWLEDGEMENTS My thanks are due to the Human Sciences Research Council and to the Trustees of the H.B. Webb Scholarship for the financial assistance which has made it possible for me to complete this thesis. I should also like to thank Dr. B.P. Keaney whose encouragement and advice, particularly in matters of reading,have been invaluable. Needless to say the responsibility for the errors in this thesis rest~entirely with me. - 1 -INTRODUCTION In the Tractatus Logico-Philosophicus Wittgenstein wrote: 4.112 "Philosophy is not a.body of doctrine but an activity." This is a thought which to a lesser or greater extent governs all his work. Its full implications only be-came apparent to him after he first enunciated the maxim, and it could be said I think, that his life's work in philosophy consisted in recognizing these implications and in scrupulous-ly and ultimately successfully following them through in his practice. To do this he had to give up the idea that philo-sophy could give us any knowledge about the world which we had not possessed before, and in fact to stress the opposite: that it could only give us insight into what we have always known. Its function was to render the obvious acceptable to the philosopher or rather to enable the philosopher to see the obvious clearly. Such a role for philosophy could not have been as readily palatable as this rather anodyne description makes it sound. The radical nature of his redefinition only begins to strike home when it becomes apparent that there will no longer be any place in philosophy for the:ory or hypothesis, discovery of new facts or denial of the truth of old theories. We may assert nothing "philosophically" and when we do assert I that such and such is the case the assertion does not derive its authority from philosophy. Both the discovery of new racts and explanations or the workings or nature rall outside the competence of philosophy. 110ne might also give the name. 'philosophy• to what is possible·BEFORE all new discoveries . 1. and inventions." Wittgenstein states in the Philosophical Investigations!· nrf one tried to advance THESES in philosophy, it would never be possible to debate them, because everyone would agree to them." 2 • Philosophy is neither scientiric nor super-scientific· and is powerless to impart new inrormation. The seeds of this insight are already to be round in Hume but the rruit which they bear in his writings are a strange 1 ~ P.I. 126 2. P.r. 12a , I J - 2 -inedible hybrid. When Hume advises us to commit to the flames "any volume of divinity or school metaphysics," which does not "contain any reasoning concerning quantity or number", nor "any experimental reasoning concerning matter of fact and· existence,n3 • he unwittingly condemns his own Treatise to the fire. His greatness consists in having seen the connection between the growth of knowledge and the practice of experi-mentation. The tragic flaw in his work is the confusion be-tween science and philosophy - a confusion which Wittgenstein penetrated. ·On the one hand Hume wanted experimentation to provide us with the sort of certainty proper only to logical investigations and Kolakowski argues, correctly I feel, that in opposing the pretensions of 17th. Century metaphysics he un-successfully attempted to show that experiment produces a high-er degree of certainty than did metaphysical speculation. He writes, ·~he destruction of knowledge which Hume•s doctrine was lead· to by its own premise$ is thus accounted for by his striving to endow •true• knowledge with.the very character that .. the seventeenth century-metaphysicians had claimed for it, namely, an absolutely compelling character." 4 • On the other hand the success of the experiments he performed were more or less guaranteed a priori, so undercutting the whole point of any empirical methodology. Jean Piaget has remarked on this score, 11If empiricism thus opened the way to a whole group of fundamental and extremely useful enquiries, it has itself pro-ceeded somewhat apace, remaining satisfied with a minimum of effort. In fact the kind of observations and experiments which it was looking for only started in a methodological fashion in the 19th. Century and it is still for most of the important questions at the phase of a first approximation. The empiricists were themselves contented to proceed more philosophico, i.e. reflecting much and appealing to facts by way of examples and justification: in such cases the facts, 3. Hume, "Inquiry concerning Human Understanding." Section XII. Part III. 4. Kolakowski "Positivist Philosophy". Page 53 - 3 -.• of course, always confirm the hypotheses." 5 • So Hume ad-vanced in the name of philosophy a quasi-empirical psychology, which, though brilliantly and penetratingly formulated, could only result in a welter of conceptual confusion and a mass of assertions poorly supported by insufficient experimental evi-dence. The point that I am trying to make with r~spect to Wittgenstein is that when he says that philosophy is an acti-vity unproductive of theories·he breaks once and for all with metaphysics, Humes postivist metaphysics included, and with I all the debates, nominalist - realist, empiricist - idealist, etc., internal to metaphysics. But what does he leave us with? Is philosphy to be self justifying activity in vacuo? Is it· to contain nothing but great, simple and beautifully expressed statements of the obvious? In Zettel Wittgenstein answers both questions as to the point and structure of philosophy: 11How does it come about that philosophy is so complicated a structure? It surely ought to be completely simple, if it is the ultimate thing, independent of all ex-perience, that you m~e it out to be. Philosophy unties knots in our thinking; hence its results must be simple, but philo-sophising has to be as complicated as the knots it unties." 6 • Strictly speaking its results are. not philosophical- they· are statements of what we have always known but what we come at one time or another to doubt or to misunderstand. 11Philoso-phical 11 is an adjective descriptive of an activity not of a result·. The concept of philosophy has been so transformed by Wittgenstein that 11philosophising" in his sense of the word is an activity which largely concerns itself, not with answer-ing the questions raised by metaphysics but with dissolving the problems and confusions belonging to 11philosophising" in the old, metaphysical sense. These problems are never factual, although metaphysics conceives them to ~e so~ but they are always conceptual, or, as Wittgenstein sometimes prefers to 5. Piaget, ninsights and Illusions in Philosophy", Page 53. 6. Zettel 452. I \ I I I ' - 4 -put it, grammatical. They never turn on an insufficient testing of data nor on a simple paucity of factual evidence. They arise out of a confusion about the way we use language. Wittgenstein•s recognition of this fact was an essential precursor to .. the development of his philosophical technique. One could say that he made philosophers aware that the very nature of their traditional activity was large-ly a result of conceptual confusion. In other words, not only were there confusions about traditional philosophic problems, like the mind-body problem, but the philosophers concept of the practice whereby he attempted to resolve these problems was itself confused. Hence, in the case of Hume for example: the unsatisfactory amalgamation of logical and empirical investigation leading to a radical scepticism and the loneliness of solipsism. I use Hume as an example because the revolution which Wittgenstein achieved through a reconceptualization of philo-sophy and through a new kind of practice (his "treatment" of philosophic problems) to a large extent took the form of· a break from and a critical engagement with "philosophical psychology" which had received considerabl~ impetus from Hume•s writing. His concern with the other-minds problem, the question of the privacy of sensation, the concept of in-tention and volition is evidence of this. He talks of the bewitchment of our intelligence by means of language and finds this bewitchment.particularly prevalent in philosophies which are involved with concepts pertaining to psychology or to our use of psychological predicates. Chief cause of this puzzle-ment is a tendency to seek out a physical object to which psychological terms are thought to refer. There are many metaphorical and analogical uses of ordinary language which reinforce this tendency. Take for example the instance he gives us in Zettel: 11The soul is said to LEAVE the body. Then, in order to exclude any similarity to the body, any sort of idea that some gaseous thing is meant, the soul is said to I I - 5 -be incorporeal, non-spatial; but with that word 'leave' one has already said it all. Show me how you use the word 'spiritual' and I shall see whether the soul is non-corporeal and what you understand by •spirit•."7 • tent. The tone of this remark is as important as its con-The last sentence especially has the tone of a doctor about to treat a patient, and, as is well known, Wittgenstein conceived of metaphysical problems as symptoms of a kind·of illness - an illness deep-seated in the grammar of our language. But his intention was never to present his investigations as the final cures of all philosophic illnesses. Rather, he wanted those who read them to treat that reading as a sort of apprenticeship during which the skills and techniques requisite for the treatment of problems could be acquired. My attempt in this thesis will therefore be twofold. I shall try to outline the problem of the privacy of sensation as a special case of philosophical scepticism of our having knowledge of other minds. Certain ramifications of this question, particularly the problems of lin~uistic meaning, and intentional action will also be discussed. Secondly I shall try, in discussing Wittgenstein's treatment of this problem, to show that his technique is a satisfactory one, not only for "curing" this problem, but for handling all philosophic prob-lems. In other words the unity of Wittgenstein•s philosophy will be stressed. This unity is not the one which is receiv-ing a great deal of attention at the moment, viz. the unity of the Tractatus and the later.works. But I refer rather to the fittedness of Wittgenstein•s philosophical activity to his domain of interest - conceptual investigations. In these in-vestigations he discovers no new fact, gives no new piece of information. What he does is to prac_tice philosophy in a new way and initiate us into a new form of activity. In this, it seems to me, his work is supremely original and greatly valuable. 7. Zettel 127 - 6 -PART I THE OPENING SECTIONS OF THE "PHILOSOPHICAL INVESTIGATIONS 11 The opening sections of the Investigations contain a number of themes which are central to the later thought of Wittgenstein and which deserve lengthy and detailed treat-ment in their own right. At the same time this first part -up to paragraph 243 - is an indispensible introduction to the discussion of privacy and sensation which immediately follows it. A large part of the section is concerned with breaking the spell of the atomistic conception of language which Wittgenstein had formulated in the Tractatus. The importance of this for the treatment of sensation emerges later when Wittgenstein proposes that sensation - language is expressive and not descriptive and that there can be no such thing as an . h" 1 • b t h h 1.somorp 1.sm e ween a sentence sue as 11 I have a toothac e" and some nstate of affairs." By breaking·with the atomistic· theory Wittgenstein achieves a second result, equally important for the treatment of the problem of sensation: he dispels the force of that Lockean variant of the theory which suggests that the problem of meaning can be overcome by interposing a ') mental object or image between a word and the world. 4 • Wittgenstein continually points out the fruitlessness of this device in explaining the meaning of words, and begins to point to the danger~ of the mental picture being tJeated as a private objec1; .. a metaphysical peg on ~hich to hang a variety of philc,sophic prob:J.ems including the problems. t):f sensation. Two other important topics are dealt with. He gives a valuable insight into what he considers to be the character of a philosophic problem : the impression it gives of having an unplumbable depth and the apparent conflict between the aim of a philosophic inquiry and the nature of the object of the 1. Erik Stenius, 1'Wittgenstein•s 1 Tractatus 1 11 , pp.91-96 2. Locke, "An Essay Concerning·_:Hwnan Underst.apding 11 , Bk. ·III, Ch. I. - 7 -inquiry, the latter seeming to resist the attempts of the former to penetrate its essence. His discussion of this topic is not only important as a statement of his position with respect to philosophy, it also givesus the atmosphere of a philosophic "disquietude" - such as can lead to a man striking his breast and crying ·~ut surely another person can't have THIS pain". 3 • (Both Wittgenstein and Wisdom place great store on· capturing the psychological atmosphere o( a philosophical perplexity.) He also suggests here that the solution to the 11Problemtt of sensation may well be grammatical and not empirical. Thirdly he discusses what is involved in ufollow-ing a rule 11 , and this is extremely important because it pre-pares the ground for his refutation of the private-language argument which forms one of the underpinnings of the theory that sensations are private. The discussion shows once again how the understanding as a generator of private mental images is fruitlessly employed as a device whereby philoso-phers try to explain how we learn rules and how we fol.low them. We find that his discussion recurs at a number of points in the Investigations and is pursued in the "Founda• tion of Mathematics" (F.M. 1•5) and I shall treat it at some length. My method of dealing with the first section will be to stick as closely to Wittgenstein•s text as possible, giv-ing something in the nature of a running commentary on it. 3. P.I. 253 ,, -a CHAPTER I THE ATOMISTIC CONCEPT OF LANGUAGE "4.01 A proposition is a picture of reality A proposition is a model of reality as we imagine it 4.0311 One name stands for one thing, another for another thing, and they are combined with one another, In this way the whole group .. like a tableau vivant • presents a state of affairs. 4.032 It is only in so far as a proposition is logically segmented th~t it is a picture of a situation,u 1. Wittgenstein begins the Investigationsby trying to discourage certain rigid and artificial ways of treating langu-age. He shows that language always transcends the confines drawn for it by an atomistic t~eory. T~e philosophical idea that language is made up of words which name objects can, he argues, be made a definition pf language only if we are prepared to accept that a great deal of what in the ordinary way we call language, will be excluded by ~he definition. And even if we invent a language which conforms to t~is'definition w~ can see that a great deal more is involved in:teaching, learning and , I using this language than is suggested by a theory of simple name-object correlation. Let us take for example the language-game ·.described in Philosophical Investigations 2 and 8. Here a builder A calls out words "block" npillarn, "slab 11 and 11beam 11 and his ' ' assistant is trained to bring him the appropriate object. In addition, the language contains numerals, {the letters of the alphabet); also the words nthis" and "there" used with a point-ing gesture; and a chart of colour samples.· A gives an order like: nd - slab - there." At the same time he shows the assistant a colour sample, and when he says "there" he points 1. Tractatus. .- 9 -to a place on the building site. From the stock of slabs B takes one for each letter of the alphabet up to "d", of the same colour as the sample, and brings them to the place indicated by A ..... 2 • Even in a primitive language as simple as this one a diversity of operations is to be found. For instance it can easily be seen that the meaning of the names can only be discovered by seeing how they are used and how they are· taught. "Block" and 11slab 11 are neither used nor taught in the same way as· the names of the numbers of the stones "a", "b", "c", "d" are. The context of linguistic usage and· the type of· training employed to teach the use of words, determines the meaning of the word. This is what Wittgenstein is driving at when he employs one of his favourite illustrations: '"I set the break up by connecting. up rod and lever' ... Yes, given the whole of the rest of the mechanism. Only in con-junction with that is it a break-lever; it may be anything or nothing." 3 • .. At this stage his position seems to be stated rather than argued, but its force comes out if we consider whether the command "Slab1 11 is a word or a sentence. The expression can function· in either way - which way will be determined by how it is used. We might say that it was a degenerate sentence but if this tempts us into saying that it must be a contraction of our sentence "Bring me a slab1 11 then we should remember that there is no reason why the latter expression should not be considered an expansion of the former. When we say that "Slabl" means 11Bring me a slabl" then all we are asserting· is that it can sometimes be used in the same way and applied in the same context. The expressions differ when "Bring me a slabl" is used •.•as four words" in order to dis-tinguish it from ... oth.er combinations such as, "Bring him a slabl" etc. But here too lurks a temptation to think that when we 2~ P.I. 8 3. P.I. 6 I t I I I ' ! - 10 -use it by contrast, the other sentences float before our minds, as do the birds in Plato's aviary in the Theaetetus,4• and that we "mean" this one by singling out in thought and taking hold of it with the immaterial hands of the understanding. In fact the contrast can be made because of the context of "Bring me a slabn in our language. Because we have a mastery of English the other contrasting combinations exist as possi-bilities in our language. Someone lacking this mastery and so being unaware of the possibility of using the expression as a contrast to others might always utter "Bring me a slab1 a as one word. His do~ng this NEED not involve his having any different mental concomitant from ours - need not inv,olve any-thing mental happening at all. The same sense and different senses depend on the use in the context of the language. ' The focus of his discussion has changed somewhat and he is now concerned to ~ettle accounts with the idea that psychological processes are a n~cessary and sufficient hypothesis for the explanation of how words mean. He is aiming over.tly at Frege•s theory t~at every sentence contains an assumption which is asserted. Frege states his position as follows: "An interrogative sentence and an indicative one ~ontain the same thought; but the indicative con-tains something else as well, namely, the assertion ••• Therefore two things must be dis-tinguished in an indicative sentence: the content which it has in common with the corresponding sentence question, and the assertion. The former is the thought, or at least contains the thought. So it is possible to express the thought without laying it down as true ••• Consequently we must distinguish: 1. the apprehension of a thought - thinking 2. the recognition of the truth of a thought - judgement 3. the manifestation of this judgement - assertion". 5 .. This implies that something takes place in the mind which is expressed by the assertion sign (a vertical bar) 6 • in addition to the word.s: "such and such is the case." Wi ttgenstein 4. Theaetetus 199. s. Frege, nThe Thought: A Logical Enquiry~' Mind Vol. LXV No. 259 6. Frege, _nBegrif:fsschrifti'S2 trans. P.T. Geach. I - 11 • replies: 11And if' I write not 'It is asserted that ••• 1 , but 1 It is asserted: such and such is the case,' the words •It . is asserted simply become superf'luous. 11 7 • So the game of' asserting something does not require a· mental state- the "asserting staten - if' it is to be played, anymore than a particular set of mental pictures are necessary to playing a game of' chess. • And here Wittgenstein introduces the concept of' the language-game in order, nto bring into prominence the f'act that the speaking of' language is part of' an activity or of' a f'orm of' lire." a. Just as a f'orm of' lire contains a diver-sity of' practices, so with language. The primitive language game which we described earlier, belies a fairly complex set of' practices which are necessary in order that the language may be used and another set in order that it may be taught. Wittgenstein considers the teaching practice an important linguistic feature to be described in an investigation into meaning. ni used at one time to say that in order to get clear how a· certain sentence is used, it was a good idea to ask oneself' the question, 'How could one verify such an assertion.• But this is just one way among others of' getting clear about the use of' a word or a sentence. For example another question which is of'ten very useful to ask oneself' is, •How is th~s word iearned? How would people set about teach-ing a child this word? :1 n9• One type of' teaching which the builder's assistant might need will involve the. giving of' ostensive definitions and even the ostensive definition game presupposes a knowledge of' the meaning of certain words on the part of' the pupil. e.g. If' I define •two• by pointing to two objects then this might be taken as·re:t:erring to just these particular objects, or to their colour. I can then say "This number is called • two • n, but then • number' must be understood or .it too needs 7. P.I. 22 a. P.I. 23 9. From, "Ludwig Wittgenstein 11 by D.A.T.G. and A.C.J., The Australasian Jou~nal of' Philosophy XXIX, 2, P.79, quoted by Hartnack. - 12 -to be defined. The series of definitions will vary in length according to whether the pupil "gets" a definition or requires a further one which will depend· on what one might call his linguistic experience. 11So one might say: the ostensive definition explains the use - the meaning - of the word when the overall role of the word in language is clear". 10• If it is argued that such prior knowledge is unnecessary and that the meaning can be learnt by seeing or guessing what is being pointed to, then descriptionsmust be given of what, for example, ''Pointing to the shape" or "pointing to the colour 11 consists in. One might answer·by saying, "I fix my attention on the colour", but this can involve doing very different things and the 11mental act of attention" is not sufficient for us to be able to say that someone is attending. The c~iteria of someone•s attending will be his actions within a certain context -moves in a game. Similarly we can say that some<>ne is 11solving 1 a chess problem" not by hypothesising about his inward state I but by watching· what he does and referring that to the context of his actions. Further, if "meaning the colouru always in-volves my doing just these things (moving my eyes· in this way, gesturing like that) and having just these sensations, it is still quite conceivable that the person whom I am teaching may observe my movements, feel my sensations, and yet interpret my ostensive definition in a way I had not foreseen. And here Wittgenstein enunciates in the clearest possible terms one of the great philosophic temptations to nonsense •••• 11because we cannot specify any.one bodily action which we call pointing to the shape (as opposed for example to the colour) we say that a spiritual (mental, intellectual) activity corresponds to these words. Where our language suggests a body and there is none: h h ld lik t ;s a . "t n 11. t ere, we s ou e o say • sp~r~ • Closely connected with the giving of ostensive definitions is the use and teaching of names and the temptations to oversimplify and to spiritualise are as powerful, here as there. A name may be related to a thing in various ways. 10. 11. P.L. 30 P.I. 35 B.B. p.l2 I ~ I I I I - 13 -The relation comes out in the use, and is not "some remarkable act or the mind, as it were, a baptism or the object". This view makes us think that this state of mind is expressed by the words •this• and •that• since they are always used in nam-ing ("This ·is called •. •") ·and so are thought to signify a necessary mental correlate - a nomenative state of mind one might say. We should remember Wittgenstein argues that 'this• and •that• are not used as names -we don•t say 11That is called • this • • n ·· The other form or over simplification - and the more persuasive one, I think • is that a name signifies a simple to which it corresponds, or that ir it signifies a structure, and the elements of this structure become detached £rom each other, then the name stands for these simple elements - this one, this one, and that one. This view is one which I take to be held by Russell 12 • and which is illustrated by his following statements on the topic: 11When I say that an unspecified member. or a class occurred, my statement is significant provided I know what class is meant; but in the case or a true proper name, the name is meaningless unless it names something, and ir it names something, that something must occur. This may seem reminiscent or the ontological argument, b~t i t::is really only part of the derini tion or "name". A proper name names something or which there are not a plurality or instances, and names it by a convention ad hoc, not by a description composed of words with previously assigned meanings. Unless, therefore, the name names something, it is an empty noise, not a word.n 13. This position, says Wittgenstein, confounds the mean• ing or a name with the bearer or a name. "When Mr. N.N. dies one says.that the bearer or the name dies, not that the mean-ing dies. And it would be nonsensical to say that, for ir the name ceased to have meaning it would not make sense to say •Mr N N ·s d d 1 n 14 • In very th · of • • • ~ ea • many cases e mean~ng · 12. Stenius, op. cit. p.ll9 13. Russell~ "An Enquiry into Meaning and Truth", Pelican, p.30 14. P.I. 40. - 14 -a word is its use, and at most we can say that the meaning of a name is sometimes taught by pointing to its bearer. The role of naming is such that it makes the language-game of descrip-tion possible - but the actual naming is not a move in the language-game but a preparation for the game. When one names an element one makes it a paradigm which serves as a means of representation in a language, rather like any standard of measurement does in measuring (the standard metre rule in Paris). The element itself has no extraordinary properties such as its being eternal and indestructible, rather one might say that it has grammatical properties assigned to it in order that it may serve as a representative paradigm. "And to say, 'If it did not exist, it could have no name• is to say as much -·and as little as: if this thing did not exist we could not use it in our language game ••••" l5. So a man can die and his name still mean something; therefore the so-called •indestructible elements• must be con-strued as paradigms used·in the language-game without which the name would have no meaning because it would have no use. This conclusion however, can tempt one into thinking that if no paradigm exists as part of the language - (as a patch on a colour chart, say) - then it flashes before the mind the same every time, and is in this sense indestructible. But since we often remember incorrectly, or .alternatively, since memories fade, the mental picture is no more permanent and indestructible than the physical picture. · The "solution" to the "problem" of finding an indestructible paradigm is attempted on the basis of a misconception. When we say that one cannot say "Red exists" because if there were no red it could not be spoken at all, we think we are talking about a particular property of red - its eternal and indestructible character. In fact we are making what Wittgenstein will later call a grammatical pro-position about the way the word is used in the language-game. What he has shown so far is that (a) language involves far more than a combination of names (b) that 11naming 11 and 15. P.I. 50 I I ! I i t f I I - 15 -ostensively defining are different but related practices, (c) that names do not "stand for 11 indestructible simples, (individuals, objects)·but that these objects are paradigms and there paradigmatic role is indicated by the way the names of the objects are used, and by the grammatical nature of the propositions which we make about paradigms such as nred 11 or 11blue11 • The material we have dealt with so far has'been collected in order to launch an attack at one of the strongest citadels of logical atomism, namely the contention that analysis ' always yields a mor~ fundamental form of sentence. The atomist says that the names in sentences correspond to indestructible elements of reality. Wittgenstein points out how odd this idea is since we have no notion of what such an element would be like. All we.have seen are components of structures. e.g. the legs, arms and back of a chair. In the Tractatus he had written - "Every statement about complexes can be re-solved into a statement about their constituents and into-pro-positions that describe the complexes completely (2.0201). Objects make up the ~ubstance of the world. That is why they cannot be composite (2.021). Objects are what is unalterable and subsistent; their configuration is what is changing and unstable" (2.0271) and a little further on 11A proposition has one and only one complete analysis" (3.25).· He now calls this theory into quest ions with the famous nbroom in the corner" example, which has aroused the ire of Professor Marcuse 16·• whose force of attack on it seems to me to be in inverse pro-portion to his understanding the point he is attacking. We can analyse the sentence, "My broom is in the corner" into statements about the broomstick and the brush, but this should not lead us to think that the two sentences are contained im-- plicitly in the one sentence. As Wittgenstein says one would probably reply to a request for 11the broomstick and the brush which is fitted into it" with 11Db you want the broom? Why do you put it so oddly?" One might want to say that in the unanalysed sentence one missed the analysis, but the reverse might be equally true. 16. Marcuse, 110ne Dimensional Man", p.l75. t I I t - 16 ·-A description of a configuration of red, white and blue stripes just would not have the same point or meaning as the words liThe Union J~ckn. Behind the idea of the complete analysis lurks another: that the essence of language can be revealed by the dissection of sentences and can be expressed in terms of a theory of the general forms of all,propositions. This general form is supposed to be revealed in the component argu-ments of a truth functional compound. But as we have just seen, the sentences of an analysed sentence and the "sarne 11 un-analysed sentence, often belong to different language games. How then are we to find the essence of language? Wittgenstein argues that the essentialist endeavour is misguided. Instead of searching (in vain) for the essence of language we should recognise that languages are related, as games are, by family resemblances. The hopelessness o.f a search for the essence of language might lead us into saying such nonsense as, that the disjunction of all the properties of language games is what is common to them. "Something runs through the whole thread - namely the continuous overlapping of those fibres. 11 17• Wittgenstein caustically remarks. Of course I can give the term •game' a rigid defini-tion if I decide to draw a boundary~ The boundary cannot however be found, it can only be drawn. Stace deals with this point very well in his defence of empiricism. He writes: "For all classification, and therefore all oefinition, is relative to some purpose, which may be justifiable but may also be founded on mere prejudice. The decision to draw the line at any particular point is not dictated by the set of facts being classified, though of course it cannot ignore facts." 18. We might teach someone how to use the word "game" by telling him about. the game~ we know - we could not perhaps do better 17. P.I. 67. 18. Stace, nSome Misinterpretations of Empiricism". Mind VOL. LXVII, Oct. ~958. 1-- 17 -than this, "But this is not ignorance. We do not know the boundaries because none have been drawn." 19• When I "see what is common" to a number of instances I may have a picture of something which I use as a paradigm of essence. But the picture needs to be used in a certain way in order to be a paradigm of essence. i.e. it is by con-vention, or by my decision that THIS is considered to be the essence. "Essence is expressed by grammar.u 20 • When I say I know what· a game is, but can• t say exactly·, this does not mean that I have a vague picture which will eventually become trimmed to the exactness of a definition. My knowledge is in fact expressed in all the possible examples I could give. "Vagueness" writes Waismann, "should be distinguished from open texture ••• ·open texture •• is something like a possibility of vagueness. Vagueness can be remedied by giving more accurate rules, open texture cannot. .An alternative way of stating this would be to say that definitions of open terms are always corrigible or amendablen 21 • He goes on to explain what he means by the open texture of a definition, and his account summarises Wittgenstein1 s point so exactly that I should like to quote from it at some lenth: "Supp.ose for a moment that we were able to describe situations· completely without omitting anything (as in chess), then we could produce an exhaustive list of all the circumstances in which the term is to be used so that nothing is left to doubt; in other words we could construct a complete definition i.e. a thought model which anticipates and settles once for all every possible question o£ usage. As, in fact, we can never eliminate the possibility of some unforeseen factor em~ging, we can never be quite sure that we have included in our defini-tion everything that should be included, and thus the process of defining and redefining an idea will go on without ever reaching a final stage. In other words, every definition stretches into an open horizon ••• Thus the result is the in-19. P.I. 68. 20. P.I. 371. 21. Waismann, •rveri£iability11 , Logic and Language First Series. - 18 -completeness of the definition of the terms involved, and the incompleteness of the definition is rooted in the incompleteness of empirical description; that is one of the grounds why a material object statementPcannot be verified conclusively nor be resolved into statements s,s2 •••• s which describe evidences 22. n for it. Ttiese quotations from Stace and Waismann express the two ideas which Wittgenstein has used all through his attack on the atomist theory of language. Firstly, the definitional lines which mark off what is essential from what is non-essential are drawn by us, not found in the object. Secondly, the nature of all descriptions and definitions which apply to empirical objects is such that they are never complete. The philosophers' failure to remember these points results in a problem or "disquietude", for 11in philosophy we often compare the use of words with games and calculi which have fixed rules," 23 • and then forget that the rules of linguistic usage are not nearly as strictly predetermined as are the rules of the propositional calculus, or the rules of poker. It is, therefore, with these philosophic problems that Wittgenstein goes on to deal, in order to throw more light on what causes them and how they are to be treated. 22. Ibid. 23 P.I. 81 • • - 19 -C H A P T E R II PHILOSOPHICAL PROBLEMS AND GRAMMATICAL ILLUSIONS "Indeed, can't we define a philosophical discussion as a discussion as to whether any of us know what all of us know we all know.n 110ther Minds 11 - 111 Pages 60 & 61. · Wittgenstein's approach to the discussion of the character and causes of philosophical problems seem at first reading a little odd. He does not say explicitly that he is beginning a new section and so my division is perhaps a bit artificial. What he does say is that before we can achieve any clarity on the question of the aim and method of a logico-philosophic investigation we must examine what is involved in meaning, understanding and thinking something. The reasons for this will become clearer later but, anticipating a little, I should say that these topics are used as "reminders" of the attractiveness and compulsion of philosophical misconceptions -particularly the so-called essentialist misconception which Wittgenstein constantly attacks. The idea that there is only one way - one process - one interpretation, and that this one way is the way it ~be, exercises a power over the intelligence which can only be removed by reminding ourselves of alternative ways, processes and interpretations, etc. which ~also be. So he starts by discussing what it means to understand or follow a rule. The nrule by which someone proceeds" could mean: (a) the description we are able to give of this method after observing what he does (b) the table of the rules he consults (c) the answer he gives when we ask what rule he uses. * But (a) we might not be able to observe any clear rule by watch-* * ing what he does (b) he might consult no table (c) he might have no answer to our question or be prepared to alter the one he has. Rules then, seem "queer" or "intangible": Both I i ' - 20 -expressions would indicate to Wittgenstein that one was in the grip of a philosophical problem and that one was about to talk in an extraordinary (nonsensical?) way. A situation in which * * * (a), (b) & (c) apply might seem to be one in which there was no certain way of knowing what rules are being employed, if any, and whether they are being employed correctly. One might want to say that this particular game was a very loose one - hardly a game at all·. In fact this situation applies to a lesser or greater degree in every game. In some games we make up· the rules or alter them as we go along. And even in a game tightly bounded by rules such as chess or poker we can imagine a doubt as to how to apply the rules, i.e. we can imagine the need for rules for the use of rules. Whether this need arises is an empirical matter: some rules leave room for doubt, others do not. Every definition, for example, might itself stand in need of definition, but not everyone does. An explanation only requires to be explained if a misunderstanding occurs. This applies to orders as well e.g. 11Stand roughly here 11 might achieve its purpose very satisfactorily. If one is tempted to say "But it is inexact -it leaves·too much leeway for mis-understanding" then we must see what we mean by "inexact" in this context,- and what the point of a greater exactitude· would be. A thin line might be drawn on the ground but it still has breadth. Delicate instruments might be used to measure when someone had crossed the edge of the line. "But has exactness still got a function here: isn•t the engine· idling?n 1 • 11Exactnessn is used as a term of praise when a particular goal is in mind·, so it is the goal which determines what exactness will mean. One might say that the logic of 11exactness 11 varied according to the circumstances in which the term was used, and that the logic of the word revealed the meaning of the word. This last statement about the logic of the word (vague as it is) has a strong attraction, in Wittgenstein's opinion, for the philosopher. Logical investigations seem to 1. P.I. 88. - 21 -penetrate deep into things and to lay bare a necessary founda• tion which explains why things must be as they are. This sort of investigation does not involve the discovery of causal connections but rather the illumination of a priori patterns within things themselves, somewhat on the lines of Leibniz's 11Mondology"• Logic seems to lie at the heart of phenomena which need only to be penetrated for their natures to be re-vealed and their necessity made plain. And yet logical in-vestigations are not like those of natural science. The character of their problems, what Althusser calls their 11problematicn, is toto coelo different. As an example Wittgenstein· asks us to compare Augustin's problem with 11 the nature of time": (si nemo ex me querat scio, si quaerenti explicare velim, nescio") with any empirical problem. Augustin's problem seems, as Wittgenstein often says, to have . 2 the character of depth. • Its solution dangles before our noses and yet we do not come any closer to solving it. Why is this? Wittgenstein suggests that this state of affairs arises when the nature of the investigation is mistaken. 11We feel as if we had to penetrate phenomena: our investi-gation, however, is not directed towards phenomena, but, as one might say, towards the 'possibilities• of phenomena. We remind ourselves, that is to say, of the kind of sta~ement we make about phenomena". 3 • From this one can see that philo-sophical investigations are not concerned with natural science, but the grammar of· their probiems makes it look as if they were problems of natural science. It is this very similarity of grammatical form which Wittgenstein holds to be responsible for many of the misconceptions in philosophy and which should itself be the object of an investigation. So telling is this insight that it is illustrated by the very practice of philosophy itself which misconstrues its own role because the grammar of its problems is, on the surface, similar to tha~ of natural science. 2~ P.I. 111 3. P.I. 90 22 -"Our investigation is therefore a grammatical one. Such an investigation sheds light on our problem by clearing misunderstandings away. Misunderstandings concerning the use of words, caused among other things, by certain analogies be-tween the forms of expression in different regions of language. -Some of them can be removed by substituting one form of ex-pression for another:· This may be called an •analysis• of our forms of expression for the process is sometimes like one of taking things apart. 11 4 • The mistake of the Tractatus is to think that there is something like a complete analysis which will give the meaning of an expression without any possi-bility of ambiguity arising. To think this is to make the same mistake as we saw could be made with, 11Stand roughly here 11 • It is to seek what Aristotle warns against in the Ethics: an exactitude and precision which is out of place in the subject matter under consideration: nFor example it is absurd to demand logical demonstrations from a professional speaker; we might as well accept mere probabilities from a mathematician." 5 • ideal. Logical precision seems to stand before us as an Our philosophical investigations must leave no room for ninexactnessn or undefined terms. But when language is approached philosophically from this point of view it appears in a strange half light: on the one hand all the words and sentences are familiar, and ·On the other, they seem to have a ·hidden core which is what they are "in essence". We begin to look for the general form of a proposition, or·we think that a proposition must be, na pure intermediary between the pro-positional signs and the facts. Or we even try to sublime . 6 the signs themselves." • Thought too is treated in this way. Thoughts seem to be pictorial correlates of the world whose logical form they share. If this logical order did not obtain how could words have fixed meanings? nthis order is a super-order between Wittgenstein replies, so to speak - super-concepts. Whereas, of course, if the words 'language', •experience•, •world•, have a use it must be as humble-a one as the words, •·table•·, •lamp', 'door•." 7 • 4~. P.I. 90 5. Aristotle 1'Nicomachean Ethics 11 , Bk. I, trans. J.A.K. Thomson. 6. P. I. 94 7. P.I. 97 - 23 -This is the bewildering ambiguity of a philosophical perplexity. On the one hand we see that our ordinary sentences, sometimes vague, sometimes open-textured, have just the order they have, and operate pretty successfully. But on the other hand we want to say that if they have sense then the sense must be complete and perfect. "An enclosure with a hole in it is as good as none. • But is that true?" 8 • This is a brilliant analogy and if it is pursued it makes the point all the better. Fish nets for example are made with holes of varying sizes, depending on what they are used to catch. A fish net of tightly woven canvas or better still, of plastic sheeting (the plastic•s texture conforming to the ideal of "perfect sense"} would not be of very much use and might be so· cumbersome as· to be useless. The idea of nthe perfect sense" is the same as the ideal of the perfect game, one completely bounded by rules. But this misconstrues the part that the ideal plays in language. It leads us to say that, rrstand roughly here" bas no sense. We think that the ideal must be found in the language (in the signs} but since we do not find it there we come to think that the pure sense exists in the medium of the understanding, which is then conceived of as the intermediary between the propositional signs and the world. 9 • Thoughts, as it were, float between the inner and the outer worlds, being grasped by the former (the understanding} and reflecting the latter. Thus Frege writes: ·~o the result seems to be thoughts are neither things of the outer world nor ideas. A third realm must be recognised. What belongs to this corresponds with ideas, in that it cannot be perceived by the senses, but with things, in that it needs no bearer to the contents of whose consciousness to belong •••• When one apprehends or thinks a thought one does not create it but only comes to stand in a certain relation, which is different from seeing a thing or having an idea, to what already existed beforehand." 10. Here we have one form, a very sophisticated one, of trying to reduce thought and meaning to a bed-rock, to a level 8~ P.I. 99 9. B.B. p.3 10. Frege, "The Thought; A Logical Enquiry. rr - 24 where we can say, 11Well it must be like that otherwise language and thought are impossible.n nThe ideal as we think of it is unshakeable •••••• Where does this idea come from? It is like a pair of glasses on our nose through which we see whatever we look at. It never occurs to us to take them off." 11 • Now of course glasses can be very useful, but not if we don•t re-quire them, and not if we are given reading glasses when we are rather short sighted. Wittgenstein treats the philosophic problem as a conflict between the requirements of the investi-gation and the object of the investigation. The general form of the conflict emerges in the specific case of the philosophy of language out of the attempt to distil a crystalline logical purity out of the set of grammatical analogies and v~gue similari-ties which are found in ordinary language. The result is that language is tailored to the requirements of the investigation. 11We have got. on to slippery ice where there is no friction and so in a certain sense the conditions are ideal, but also, just because of that, we are unable to walk.. Back to the rough ground.n 12 • i.e. back from the ideal of "perfect language" to ordinary language, language at work, in use. He goes on to make a rather enigmatic and highly compressed statement, "The preconceived idea of crystalline purity can only be removed by turning our examination round. ·(One might say: the axis of reference of our examination must be rotated, but about the fixed point of our real need)" 13 • This implies, I think, an almost literal Copernican revolution in philosophy. The axis of reference of our investigation is indeed that clarity of which he speaks later. But our real need is to remove the puzzles which occur in philosophy because of misleading surface similarities between the grammars of sentences in ordinary language. The mistake hitherto has been to "perfect" language in various ways or to invent "ideal" languages so ignoring our real need and leaving the philosophical problems untreated. Philosophical problems he argues, are to be solved, nby looking into the workings of our language." 14 • They seem to be deep problems, incapable of ordinary solution, and our 11~ P.i. 103 · 12~ P.I. 107 13. P.I. 108 14. P.I. 109 I I I I I I i I I ! - 25 -inability to come up with extraordinary solutions leads to a frustration which Wittgenstein describes when he writes, n •But THIS isn•t how it is,• we say. •Yet THIS is how it has to be.• u15• When we read the phrases, 11the branch of the tree" and "the top of the tree" in the ordinary way, we know perfectly well what they mean (we know nhow things are" with each phrase). But when we are doing philosophy we ar~ 'disquieted' by the fact that there is nothing in the tree which we could cut .. off" and take home as 11the top 11 • We don't seem to know our way about· any more. 16·• If we look into ·the workings of the phrases when they are used in ordinary language we see "That though we can use quasi-descriptive phrases to enable us· to state where some-thing is, that the thin~~s there is a relational. character of the thing and not itself' a subject of characters" 17 • So that what appeared as a problem is shown up as a mistake about grammar. The most pervasively misleading mistake in the philo-sophy of language is to construe all sentences as being funda-mentally propositional in character. The reduction of language to one general form of propositions becomes a requirement of the' investigation which blinkers the philosopher to the extent that "One thinks that one is tracing the outline of the thing's nature over and over again and one is merely tracing round the frame. through which we look at it.n 18 • An alternative way of stating this which Wittgenstein us~s in the Blue Book and again here is to say that we derive a picture from a linguistic ex-pression which we try to transpose into another area of language, when in fact no application has been found for the picture there. Pole interprets this form of argument as being an attack on all pictorial ways of thinking. "Wittgenstein," he writes, "seems to recognise no positive r8le at_all as belonging to them; they· are nonfunctional themselves, but block our vision of the functioning of words. He constantly appeals to us to ignore the pictures which we associate with our words and look instead h . 19. at t e~r use." He goes on to say that he finds this approach 15~ P.r. 112 16~ P.r. 123 17. Ryle, "Systematically Misleading Expressions". Logic of Language, First Series. 18. P.r. 114· 19. Pole, "The Later Philosophy of Wittgenstein", p.91 I I· l - 26 -implausible but that it is in line with Wittgenstein 1 s "trade" which is demolition. Pole•s argument is persuasive at first but it seems to me that it is neither accurate nor consistent. It is in-accurate for the reason that Wittgenstein distinguishes between 11idle pictures" and pictures which have particular uses. 20• It is therefore incorrect to say that he considers all pictures to be non-£unctionale Furthermore, he is not concerned to disguise the fact that we constantly operate with pictures in the course o£ our ncalculatingn with words. The argument is inconsistent because Pole has earlier correctly stated that it is not with pictures as such but with their application or lack o£ application which Wittgenstein is concerned. He ex-plicitly says, 11To achieve his end, to wean us from our pre-conceptions and· break the power o£ those pictures which have come to dominate our thinking, Wittgenstein also employs various imaginative devices.u 21 • These devices are used to show us cases o£ linguistic usage which might not otherwise have occurred to us. They aid us in our attempts adequately to describe the workings o£ language, for it is Wittgenstein•s contention that philoso-phical problems cease to puzzle us when we get a clear view of these workings. of our language So we must give "perspicuous representations" by n£inding and inventing intermediate cases.n· 22 • In doing this one does not so much solve problems as dissolve· them, by seeing them for what they are. What is required is that the problem be adequately described. When for example we lay down rules for a game and then rind that things do not turn out in the game quite as we had expected (when we make logical purity a criterion of what makes sense in language) then it is our entanglement in the rules that we must get a clear look at. The WHOLE entanglement must be clearly seen. 20~ P.I. 291 21. Pole, op. cit. p.28. 22. P.I. 122 - 27 -As Wisdom notes, this point was made by Freud and I think that the psycho-analytic method throws quite a bit of light on what Wittgenstein means when he says that " .... the clarity we are aiming at is. indeed complete clarity. But this means that the philosophical problems should completely disappear.n 23 • This clarity is to be achieved by "assembling . d f . 1 . 24 • . b . reml.n ers or a part1.cu ar purpose." That l.S, y pol.nt-ing to examples which simply do run· against what we are in-clined in philosophy.to say, must be the case. Wisdom remarks, in a note to the Black-White debate, on how we come to know the invisible and what meaning is attached to, nknows what•s going on in the mind of another," "The whole difficulty arises like a difficulty in a neurotic;· The forces are conflicting but nearly equal. The philosopher remains in a state of confused tension unless he makes the effort necessary to bring them all out by speaking of them together. It isn't that people can•t resolve philosophical difficulties but that they won•t. In philosophy it is not a matter of making sure that one·has got hold of the right theory but of making sure that one has got hold of them all. Like psychoanalysis it is not a matter of selecting from all our inclinations some which are right, but of bringing them all to light by mentioning them and in this process creating some which are right for this in-dividual in these circumstances." 25 • ·I I To bring the various t~eories to light may of course require various techniques - some~eories lending themselves to diagrammatic representation and others not, for example. This is I think what Wittgenstein means when he says, 11There is not a philosophical method, though there are indeed· methods, like different therapies.u 26 • These therapies cannot them-selves be characterised as forms of theorising or of explanation -they merely put before us what we have always known. 23. P.I. 133 24r; P.I. 127 25. John Wisdom, uother Minds", p.l24 26. P.I. 133 - 28 -C H A P T E R III FOLLOWING RULES "Why are people so strongly drawn to believe, in the face of their own daily experience, that the intelligent execution of an operation must em-body two processes, one of doing and another of theorising? Part of the answer is that they are wedded to the dogma of the ghost in the machine. Sin6e doing is often an overt muscular affair, it. is written off as a merely physical process. On the assumption of the antithesis between •physical• and •mental•, it follows that muscular doing can-not itself be a m~ntal operation. To earn the title 'skilful,• •cunning,• or 'humorous,• it " . " must therefore get it by transfer from another counterpart act occurring not •in the machine' but • in the ghost; • for • skilful, • -• cunning, • and--- - . . -I humOrOUS t are ··certainly mental ··prediCateS e II Gilbert Ryle. 11The Concept of Mindtt P-.32 The discussion of what is involved in following a rule, serves a variety of purposes. (a) It throws light on what sorts of things rules are and so clarifies how a_philo-sophical entanglement with rules may come about. (b) It shows the inadequacy of the hypothesis of a mental mechanism to explain what Ryle calls "intelligent execution and operation". (c) It strengthens Wittgenstein•s argument that the meanings of signs are to be found in the way they are used. (d) It introduces the discussion of privacy by attack-ing the notion that it makes sense to talk of obeying a rule privately. Puzzled by "the nature of language" the philosopher tries to ease his discomfort by introducing a rule in the guise of a definition. He says that the general form of propositions is, "This is how things are 11 • He argues that - 29 although this is itself an English sentence it is used as the propositional variable "P" is, in symbolic logic. But we .can see immediately that we do not say that "P" is the general form of propositions because part of our concept of a proposi-tion is that it should sound like an English sentence. The correct thing to say is that there are many kind of propositions and we can give examples of these different kinds. But the question still nags at us, "How do you know whether they are all propositions? Isn•t one of the tests of a proposition that it must fit the concepts of true or false?" . This says Wittgenstein is a misleading picture. "True" and "Falsen might belong to our concept of a proposition just as "sounding like a proposition" belongs to it, but they do not "fit" it. They are not the forms into which the wordy matter is fitted so as to make that matter propositional in essence. 11Th is is how things are," fits in the way that L fits after K wh·en we recite the alphabet. And we might say to a child when teaching it to '"~ pick out a proposition," 1 Ask yourself ~:t' you can say "is true," after it. If these words fi"!: then it is a proposition.'" 1• Another tack would be to say that the meaning of a word fits the sense of a sentence which I understand. If the meaning of a word is the way it is used then this statement obviously makes no sense: the picture of something fitting something else is inappropriate. A tendency to think of meaning as some sort of shadowy picture grasped by the under-standing is revealed here. This way of thinking ignores that every picture can be applied and interpreted in various ways. For example, I may seem to misuse the picture of a cube if I point from it to a triangular prism, but with a suitable method of projection the picture does 11fit 11 the prism. , Even when a schema is supplied to represent how a picture is to be interpreted the possibility of the schema itself being interpreted remains wide open. When I teach someone an interpretation I judge whether he has understood me by the use he makes of the diagrammatic schema_ I give him l. P.I. 137 - 30 -or of the image he carries in his head. tions are fulfilled or they are not. Either my expecta-Only in the normal case, where he has ngot my meaning" is there no doubt as to the interpretation and just because of this the abnormal case illuminates the normal case. · Wittgenstein now moves to attack the idea of a mental process which is supposed to explain understanding. He suggests the following language-game: A teaches B the series of natural numbers. What does it mean to say that B understands the teaching of A? This, that he consistently writes down the series in the correct order. - Suppose I teach someone a series and he eventually continues it. How far must he go before I say that he•s got it -the thousandth place? One is tempted to say that continuing is only applyin_g the understanding and that "understanding is a state which is a source of the correct use,u· 2 • just as an algebraic formula may be ,thought of as the source of the series. If we have the idea that we know the series or the application of derivation rules apart from actual derivations, then we can be asked when we know the application - always? Sometimes? only when we think of the rule? We can only answer by saying that knowledge is a state of mind, a con-dition of a mental apparatus which explains our knowing. But as Ryle points out, nto possess a propositional property is not to be in a particula:r state, or to undergo a particular change; it is to be _pound or liable to be in a particular state, or to undergo a particular change, when a particular condition is realised.n 3 • And besides as Wittgenstein says,u •••• there are objections to speaking of a state of mind here,· in as much as there ought to be two different criteria fqr such a state: a knowledge of the construction of the apparatus, quite apart from what it does,n 4 • and a kn~wledge of the 2. P.I. 146 3. Ryle, "The Concept of Mind 11 , p.43 4. P.I. 149 I . - 31 -former is lacking to ·most of us. A more fruitful approach is yielded by his insight that the word 11knows 11 is often used like 11can" or 11is able to," and that its grammar· is also closely related to that of "understands". And here he uses a very good example. When I understand· how to continue this series 1, 5, 11, 19, 29, the following things might happen: (a) I try various algebraic formula and light on an = n 2 + n - 1 which is confirmed to be correct. {b) I find the series of differences 4,6,8,10, and say "Now I can go on" and do so. (c) I say, "Yes, I know that series as well as I know the series 1, 2, 3, ~' 5 11 and I continue. (d) I say no~hing but simply continue the series with the feeling, 11that•s· easy." But are these possibilities really understanding? We feel that there is something hidden behind (a), {b), (c) & (d) which is "understanding THE PRINCIPLE of the series,n and < yet we cannot· find this proce~s. We don 1 t know how o:r where to look for it. Wittgenstein suggests, ·~ry not to tnink of understanding as a •mental process• at all. - For THAT is the expression which confuses you. But ask yourself: in what sort of case, in what kind of circumstances, do we say, 'Now I know how to go on,• when, that is, the formula HAS occurred to me?n 5• It is the circumstances under which we have the ex-perience which justify us in saying we understando Strawson e.::1ters a caution here agairtst exactly the sort of misinter-pretation which Wittgenstein would want us to avoid. He writes, "This obviously does not mean that the words 'Now I can go onl' are short for a description of all these circum-stances, or that they mean 'I have had an experience which I know empirically to lead to--the continuation of the series;• 5. P.I. 154 - 32 -my certainty that I can go on is not a matter of induction. What we need (here I interpret a little) is to look at such first person utterances in a radically different way from the way in which we look at the corresponding third person utterances: to see them not as reports about myself for giv-ing which I have to apply criteria, but rather as 'exclama-tions• or •signals•, naturally and appropriately made or given-in certain circumstances ... 6 • The attractiveness of the hypothesis of a mental mechanism as it is used in philosophy is very clearly illus-trated in the case of explanations as to what reading con-sists in. We want to say that the difference between a fluent reader and a child who guesses correctly at words or knows the passage by heart is that there must be a different mechanism at work.in each of them. Now in the case of a reading machine we could say that it was reading when its mechanism was correctly adjusted, but in the case where we teach someone to read our cri~erion will be whether he con-sistently reads correctly and the concept of reading here will be independent of any hypothesis of a mechanism. "The change when the pupil began to read was a change in his behaviour.n 7 • If we fall prey to saying that the criterion of a·person•s reading is the presence of a certain conscious process with its various accompaniments such as hesitating, looking more closely, misreading, etc. then Wittgenstein asks us to remember that it is conceivable that someone should read an unfamiliar passage with a feeling that he knew it by heart. A slightly different but equally inadequate approach would be to say that in reading we derived sounds from letters according to a rule and that it is this derivation .which is characteristic of reading. But this only reproduces our difficulty for there is no one feature which is essential to 6. Strawson, "Review of Wittgenstein's Philosophical lnvestigations,n Mind_Vol. LXIII (1954) reprinted in Pi~~her, p.36. 7. P. I. 157 • I I. - 33 -deriving. A derivation schema can take all sorts of forms. In the same way there are a number of cases for which we use the word "reading". The psychological hypothesis reasserts itself with the suggestion that reading is characterised by the feeling of being guided by the letters. When we read English we nfeel the influence of the letters" in a way in which we do not when we scan Linear B: "••• that is really meant to imply that I, as it were feel the movement of the lever which connects seeing the letters with speaking." 8 • This feeling is of course not always or even often present in the case of fluent readers, so that it is quite impossible that it should be a criterion of reading. Saying that it is there unconsciously, or that closer inspection will reveal it is irrelevant as Wittgenstein shows when he says, "If I am supposed to describe . how an object looks from far off,- I don't make the description more accurate by saying what can be noticed about the object on closer inspection." 9 • What in fact happens in this case is that we latch on to one out of a variety of cases of being guided and refer to this as "the essence of being guided 11 • So we think perhaps of rails· guiding a train, or of the sensa-tion we get when we are led blind-folded and we try to apply these pictures in the case of reading or writing. Since these pictures do not really apply at all in this case we get the idea that the guiding takes place, - but not in a visible or tangible way. "'For surely,• I tell myself, 1 I was being. guided.• -Only then does the·idea of that ethereal, intangible instance arise." 10• This discussion of nbeing guided" is extremely im-portant to his treatment of following rule· and o£ understand-ing and meaning something, to which we must now return. The idea of a guiding influence is closely related to the idea. that when we mean someone to follow a rule, the steps he must 8~ 9. P.I. 170 P.I. 171 10. P.I. 175 (An argument with similar force is used by Berkeley in "A New Theory of Vision")• l ' .i l - 34 -carry out are predetermined by our order to him, in the sense that all possible steps in accordance with the order are some-how present in our minds, "As if they were in some unique way predetermined, anticipated·- as only the act of meaning can t . . 1 . 11. . . h . 1" k an ~c~pate rea ~ty.n The mean~ng ~s somet ~ng ~ e a final cause which determines the "end product in the beginning". This Aristotelian - Hegelian idea·, can be illustrated by the example of a machine, the movements of which may seem to some-one with a metaphysical squint to be in it before ever it moves. One is apt in this case to forget the possibilities of the machine•s melting, breaking, etc.; or at least the philosopher forgets·· them, - the mechanic does not. It begins to seem as if the "possibilities or movement" are something very close to real movement. The possibility of a movement is 11supposed to be like a shadow of the movement itself"• 12 • It· is not sufficient that the possibilities of movement be represented by a diagrammatic drawing ror this can be interpreted in various ( ways. No, the movements must be represented by what Wittgenstein calls in the Blue Book, a ·~icture by similarityn. 13• Such a picture permits of no interpretation and so is thought· to determine movement with the rigidity of the "logical must". This is an idea which is beautifully developed in the Founda• tions of Mathematics: "Suppose we represented the movement of the 'perfectly rigid 1 m~chanism by a cinematographic picture, a cartoon film. Suppose the picture was said to be PERFECTLY HARD, and this meant that we.had taken this picture as our method of description - whatever the facts may be, however the 14 parts of the real mechanism may bend or expand." • And he goes on to say "The connection which is not supposed to be a causal, experiential one, but much stricter and harder, so rigid even, that the one somehow already is the other, is al-. ~ 15. ways a connect~on o .... grammar." 11~ P.I. 188 12~ P.I. 194 13. B.B. p.36 14~ F.M. 121 15. F.M. 128 - 35 -In just this way one wants to say that when someone grasps what we mean, all the steps which we mean him to take are somehow determined. Wittgenstein objects to the "somehow" because this ma~es it look as though there is a "queer·u mysterious connection between what we say when we mean· some-one to do something (continue a series, play a game of chess) and what he does; when in fact there is no strange, super-rigid, mental shadow which determines what is to be done. intention and the thing intended, game of chess," exists, "••• in the teaching of it·, in the day to day The connection between my when I say, 11Let 1 s play a rules of the· game, in the practice of playing." 16• This bri~gs us to a clear view of the puzzle. It had·begun to look as though any rule could be interpreted in an unlimited number of ways, each perfectly justifiable, with the result that the idea of obeying a rule or following a rule became evacuated of meaning: "••• no course of action could be determined by a rule because every course of action can be made out to accord with the rule." 17 • What Wittgenstein has led us to see is that a rule and an interpretation of it do not constitute what we call obeying the rule, or ,following it correctly. A rule can only be obeyed in so far as the practice of doing so is part of an established system of such practices. This indicates two things, (a) "obeying a rule is a practice" 18• different from interpreting a rule (Wittgenstein prefers to reserve the latter for the expression of the rule); (b) a rule cannot be obeyed privately so that, "to think one is obeying a rule is not to obey i tn. 19• He asks whether, if mankind never played games, I could· invent a game which nobody ever played. The answer is •no•, because part of what we mean by a game is that . . it should have publicly ascertainable rules which are obeyed. So it is by reference to the common behaviou~ of mankind that languages, games and systems of rules generally, 16~ P .I. 197 17 ~ P.I. 201 18~ P.I. 202 19. P.I. 202 - 36 -are interpreted. And when we teach someone what a rule is we employ such words as "regular", "same", nuniformn, the meanings of which we teach by reference to that system of behaviour. We give in other words, "examples and practice" 20 • as the groundwork of our t~aching. We cannot do more tharr this, for we know no more of the matter than this. A point is always reached in teaching a rule where either the lesson is understood or we say, '~ell I can't give you any more teaching than .I've given,- this is simply what I do when· I obey the rules. 11 All attempts at inventing an unambiguous way for teaching rules are futile. The voice of intuition for example also has to be understood before it can be obeyed, just as an ordinary teacher must be understood. Intuition does not provide us with a reason for obeying a rule which is somehow more fundamental than any other sort of reason. Similarly to say that when a pupil merely has to do the SAME .. every time there can be no failure to follow the rule correctly, is really to say nothing at all. For example, we might say that in the series in 2, 2, 2, 2 •••••••• every step is the same and so there 'is no room for an interpretation: when you · see a thing you see an identity. Wittgenstein•s answer is quite devastating: ·~hen are two things the same when they are what one thing is? And how am I to apply what the one thing shows me to the case of two things?" 21 • This discussion of identity is extremely important for the treatment of whether two people can have the same ·pain, and we will return to it in a later chapter. For the moment what he is concerned to show is that the law of identity is a blind alley as far as the question of teaching rules is concerned. Ultimately the teaching of a rule is a practical matter. Rules are grounded in a customary system of practices. The reasons I give for how I know how to obey a rule will come to an end some time and my practice alone will remain. In 20~ ct. P.I. 337 on "intending". 21. P.I. 215 ' I I j I I I l I t - 37 -this sense my practice is 11groundless 11 • I take this way of putting it from On Certainty in which· Wittgenstein says a great deal which is significant for the concepts of under-standing, doubting and making a mistakeo Particularly im-portant for·the present discussion is the emphasis he lays there on established structures on beliefs and practices, 22 • and on the way that these structures are grounded on what is itself groundless. The close connection between what he is saying in the Investigations and what he is saying in On· Certainty can be seen £rom these two quotations. 11How can he know that he is to continue a pattern by himself - whatever instructions you give him? - Well, how do I know? - If that means •Have I reasons?• The answer is: my reasons will soon give out. Then I shall act, without reasons. 23 • 110ur rules leave loop-holes open, and the practice has to speak £or itself." 24 • 22~ o.c. 102 to 111 23~ P.I. 211 24. o.c. 139 \ \ - 38 -PART II PRIVACY The next five chapters are devoted specifically to discussion of certain aspects of the other .. minds problem -aspects which Wittgenstein comes to grips with in his later writings. My method of handling this part of the thesis will be rather different from Part I. I shall not follow the text of any particular work of Wittgenstein although the ·source material for these chapters, whether remoulded by other philo-sophers or not, comes in the main from the Blue and Brown Books, the Philosophical Investigations, Zettel and On Certainty. For instance, though the chapter on communication and in-tention is largely a summary of Professor Strawson•s inaugural lecture, my justification for using it in a· thesis ··on Wittgenstein is that, as I am sure Strawson would admit, the ideas and arguments which·he marshals so efficiently are ones which spring from Wittgenstein•s philosophising. I think a fairly close textual commentary on the middle sections of the Investigations, in which the private-language argument and the argument for private mental objects are discussed, would be possible·. . Wi ttgensteiri' s remarks only appear to have a certain randomness about them. In fact they are more often than not extremely closely integrated and it is the compactness rather than the looseness of his writing which makes the drift of his thought hard to follow. Instances of this seeming looseness and incongruity are in part attributable to his remarkable use of examples. Norman Malcolm says in his memoir that Wittgenstein was often amused by the bizarre examples with which he illustrated the points he was making, but that he would become concerned if his students were similarly amused. He wanted them to see the relevance of these examples to the topic under discussion. It is frequently very difficult not to mistake a fantastic example of his illustrative of one topic, for the beginning of a completely new topic. So I have decided not to use the commentary method of Part I, not because - 39 -such a commentary is not possible but because for one thing I want to discuss a number of arguments and points of view which have appeared in the literature subsequent to the pub-lication of the Investigations e.g. Strawson on meaning and truth, Ayer and Rhees on private language, Wisdom on doubt {although his essays pre-date the Investigations) etc. For another thing I have found reference to Wittgenstein•s latest writings in Zettel and On Certainty very valuable in trying to give an account of his dealing with problems arising out of the philosophical debate on privacy. They provide fresh examples and even an invigorated vocabulary (e.g. the use ofuground/groundle~s"; and "system of convictions" in On Certainty)··from which I have· drawn.- Also they contain further developments of his views on the character of conceptual, as opposed to factual, investigations. In what follows Wittgenstein•s views on language-meaning and the nature of philosophy, which were discussed in Part I will be constantly adverted to. In this way I hope Part I will serve as nstage-settingn for Part II. \ ' Hastings: Stanley: Hastings: - 40 -C H A P T E R IV THE CONCEPT OF A CRITERION His grace looks cheerfully and smooth this morning; There's some conceit or other likes him well When that he bids good-morrow with such spirit I think there's neer a man in Christendom Can lesser hide his love or hate than he; For by his face straight shall you know his heart. What of his heart perceived you in his face By any livlihood he showed to-day? Marry, that with no man he is offended; for were he he•.d have shown it in his looks. Shakespeare. Richard 111. Act 111 Sc.IV One of the most important and at the same time most elusive concepts in Wittgenstein•s post-Tractatus writings is that of a criterion. Its importance lies, for one thing, in the fact that it helps to clarify how we can know the sensa-tions of another person. I shall, however, argue in Chapter 10 that the role which the concept of a criterion plays in the problem of other minds has been badly misinterpreted. It is an elusive concept.because Wittgenstein typically does not de-vote himself to a sustained theoretical discussion o~ criteria and what he means by them. The most direct statement from him on this topic occurs on pages twenty-four and five of the 11Blue Book", and there is good reason to think that his notion of a criterion changed significantly in the "Investigations" from what it had been in the 11Blue and Brown· Books". Albritton writing on the notion of criterion in the "Blue and Brown Books" says that Wittgenstein•s ways of speaking there "imply that to be a criterion of X is-just to BE (what is called) X, in case - 41 -there is only one criterion of X, or to be (what is called) X, under certain circumstances, in case there is more than one criterion of X11 • 1 • He quotes a large number of in-stances where Wittgenstein speaks that way of criteria. Later he suggests that, "The dominant conception of a criterion in the •Remarks on the Foundation of Mathematics• and the •Investigations• is •••• : A criterion for a given thing's being so is something that can show the thing to 'be so and show by its absence that the thing is not so; it is something by which one may be justified in saying that the thing is so and by whose absence one may be justified in saying that the thing is not son. 2 • This seems to me to be an accurate description of what happens to the notion of a criterion and it shows that Wittgenstein was not from the first entirely clear as to how the term should be used, or at any rate it indicates that he chose to use it more than one way. In order to get some bearings on the question of what is meant by a criterion I should like to take as a starting point the following characterisation of Wittgenstein•s notion of criterion by Chihara and Fodor, although in the nature of the case it can be no more than a starting point. 11X is a criterion of Y in situations of type S if the very meaning of definition of 'Y' (or as Wittgenstein might have put it, if the grammatical rules for the use or 1 Y') justify the claim that one can recognise, see, detect or determine the applica-bility of 'Y' on the basis of X in NORMAL situations of type So Hence, if the above relation obtains between X andY, and if someone admits that X but denies Y, the burdon or proof is upon him to show that something is abnormal in the situation. In a normal situation, the problem of gathering evidence which justifies concluding Y from X, simply does not arise." 3 • Now this characterisation, which they say is only nro.ugh and schematicn, is not sufficiently detailed to provide 1. Albritton, 110n Wittgenstein•s Use of the Term •Criterion"' l'he Journal or Philosophy, VOL. LVI· .. (1959) __ . 2. (Albritton) Ibid~ 3. Chihara and Fodor, "Operationalism and Ordinary Language American Philosophical Quarterly VOL. II (1965). - 42 -the answer to a number of difficulties which arise out of the concept of a criterion, although coupled with Albritton's statement it gives a very neat account of the concept. · The difficulties in discovering how Wittgenstein uses the words 11cri ter ionn and "criteria" can I think be listed as follows: (a) If X is the criterion of Y, then is X, Y? i.e. is the criterion identical with that of which it is the criterion? (b) If a criterion is not of type (a) then what is its relation to Y? Does asserting that X entail asserting that Y? (c) If a criterion X is not of type (a) and its relation to Y is that of entailment, how does it come to be related in that way? (d) Is there always only one criterion X or one set of criteria for Y or is there nQ one criterion or set of criteria which is both a necessary and sufficient condition for Y? (e) What is the difference between a criterion and a symptom? Before I attempt to come to grips with these problems I should note that it is completely foreign to the spirit and intentions of the 'Investigations' to propound a completed theory. I partly ··agree with Anthony Manser that the notion of criterion presents a difficulty, "•••• only if it is assumed that Wittgenstein had a fully worked· out doctrine of criteria which has to be discovered from the scattered remarks in the 'Investigations•. If instead it is treated merely as a use-ful way of expressing certain insights there is less danger of puzzlement." 4 • On the other hand his statement (directed against Professor Norman Malcolm) is unhelpful from two view 4. Manser, "Pain and Private Language" Philosophy of Wittgenstein. Studies in the ed. Peter Winch. - 43 -points. Firstly, the problem with the concept is to.discover ~ it is employed to 11express certain insights" - what the function of a criterion is in Wittgenstein 1 s thoughts. Secondly, it is misleading to talk of n~ useful way" when there seem to be a variety of ways in which 11criterionn is used. And despite the fact that the remarks on criteria are scattered in the •Investigations•, they are sufficiently liberally scattered to warrant serious attention. 5 • At a number of points in the •Investigations• Wittgenstein talks of a criterion of identity and of identi-fying something by criteria. What precisely does he mean? Is the criterion of anything that, by virtue of which, it is what it is. To take an example adapted from Wisdom. Suppose a housewife goes in to the larder, sees two objects there, feels them, smells them and tastes them and then comes out and says, with no intention of deceiving, "Yes, there ar.e two loaves of bread there alright". Now as Professor Austin says, she has not discovered the signs or symptoms of bread, but the bread itself. 6 • ' And in the 110ther Mind" debate White quite correctly concludes; "•••• there is no way of finding out whether there are two· loaves in the larder superior to that of going and having a look, there is no other process suitable for this purpose which it is proper to dignify with the name of •coming to know•, in opposition to mere •concluding from the testimony of• our ~enses which happens in s~eing 11 • 7 • The question we must ask is whether that set of features of the object in the larder which (we will assume for the sake of the example) make up bread, and which the careful housewife checked on before she made her statement, could be called 11criteria of breadn in any Wittgensteinian use of the term "criteria". In other·words is there a use of the words "criterion of Y11 which makes them synonomous with "Y", so that where I say 11He· knows what the criteria of Y aren· r· could just as well have said; "He knows what Y isn? So At a rough count the word is used about fifty times. 6. Austin, "Other Mindsn, Philosophical Papers 2nd. ed. 7. Wisdom. ~·Other Minds.". p.66 - 44 -I suggest that this is how Wittgenstein uses the words when he talks about a criterion of identity. But if this is the only use he makes of the concept then its im-portance is considerably diminished since we can only say that X is a criterion of Y when X is Y, which is comparatively trivial. It is not trivial to decide what we are going to call nyu but we have not said very much when we state that what we· call uyn is also the criterion of Y. In fact this use of 11cri terionn makes giving the criterion for Y the same as giving a definition of Y. Wittgenstein himself relates criteria and definitions when he talks about "meaning": ·"For a large clas·s of cases - though not for all - in· which we employ the word •meaning' it can be defined thus: "The meaning of a word is its us"e in a language." 8 • Then later he writes, "What is the criterion for the way the formula is meant? rt· is for example the kind of way we always use it, the way we are taught :to use it.n 9 • At this point it seems to me, it is possible to make a serious oversight in interpreting Wittgenstein. If one says th~t the relationship between a criterion and that of which it is the criterion is one of logical entailment, or if, in even more fasionable terms, one says that any statement giving the criterion of anything is a grammatical statement, and if one leaves it at that, then one glosses over and dis-guises another use of the term 11criterion11 o This second use is certainly very closely related. to the one I have just dis-cussed in the previous paragraph but at the same time it is subtly and importantly different from it. It is vital, I hope to show, to make this distinction in order to answer the question of whether Wittgenstein is a behaviourist and in order to deal with his attitude towards mental states and processes. A discussion of the second usage also serves as a good introduction to his "theory" of the relationship be-tween certainty and language-games·. This is important for its bearing on sceptical doubts about other minds. a~ P.r. 43 9. P.I. 190 - 45 -To make the distinction clearly I want to consider the difference between the following two cases. (a) Smith goes for a walk and sees $heets of water pouring from a cloudy sky. He goes back home and in answer to his wife's question as to why his clothes are wet he says, 11ft's been raining again 11 •· (b) Jones goes for a walk and sees that.the streets are wet, that all the trees are dripping, that the houses are wet, that there are pools of water in the fields, that the sky is an ominous grey, and he hears peals of thunder in the distance. He goes back home and in answer to his wife's question as to why his shoes are muddy he says, 11It 1 s been raining again." Now one could argue that Smith had the criterion for saying that it had been raining, viz. the fact that there were drops of water falling from the clouds, but that Jones had no criterion. Smith saw that it was raining, while Jones only assumed that it had been raining. And the argument might continue that while one circumstance justified Smith in saying that it was raining, however far Jones extends his description of dripping trees, p~ddles, etc., so long as he missed seeing the falling drops, he does not have a criterion for saying that it was raining. At best what Jones has is an experience of a very convincing set of symptoms from which it is possible to make the hypothesis that it had been raining. Here one could quote Wittgenstein in the Blue Book when he says, "I call a •symptom• a phenomenom of which experience has taught· us that it coincided in some way or other, with the phenomenom which is our defining criterion. Then to say, 'A man has angina if this bacillus is found in him,• is a tautology or it is a loose way of stating the definition of •angina•. But to say, 'A man has angina whenever he has an inflamed throat• is to make an hypothesis." The argument outlined in the previous paragraph is in line I think with the concept of a criterion which seems to have been most prominent in Wittgenstein•s thought at the time - 46 -of the lectures which form the 11Blue and Brown Books". But when we look at this argument in the light of his subsequent writings, particularly in the Investigations, then we can see that it is based on a concept of a criterion different from that which he later developed. He would not have said that his first usage was wrong (he seldom says that about a form of expression) but he clearly thought that it.was too limiting, and· found the need for another use. With the second use of •criteria~ it would be proper to say that Jones, having given a·suitably·detailed descrip-tion of his surroundings, had criteria for saying that it had rained. Given the prevailing climatic conditions (cloud, low pressure, etc.) it would be absurd to ask Jones, 11Are you sure that it had been raining?" Jones is not making an hypothesis which is more or less probably correct. He is fully justified in saying that it had rained even though he did not see what Smith saw. Jones can be certain that it had rained. In the Investigations, when the knowledge of other peoples sensations is under discussion, Wittgenstein makes the sceptic ask, 111 'But if you are certain, isn't it that you are shutting your· eyes in the face of doubt?' "·· His reply is apt 10 and curiously simple: "They are shut." · ·• We must interpret this to mean that they are shut by the· blinkers of the language-game. It is only in the context of a language-game that "certainty" has a meaning. (The break lever is such only given the rest of the mechanism.) This implies further that what is to count as a criterion will depend entirely on the language-game which we happen to be playing. Having said this we are in a position to raise the question of the nature of a criterion in a slightly different way, and a problem which I have avoided dealing with up till now must be raised. It is this: does a proposition describing criteria entail (in the technical sense) a proposition asserting the occurence of that of which they are the criteriao This is 10. P.I. Page 224 - 47 -the question which Professor Malcolm raises with respect to the behavioural criteria of pain. He writes, •IDo the pro-positions which describe the criterion of his b~ing in pain logically imply the proposition, •He is in pain•? Wittgenstein's answer is clearly negative. A criterion is only satisfied in certain circumstances. If we come upon a man exhibiting violent pain behaviour, couldn't something show that he is not in pain? Of course. For example, he is rehearsing for a play; or he has been hypnotised ••••••" 11 • This is a strong point but it is unfortunate that Malcolm defines a criterion in terms of the circumstances in which it is found: 11The ex-pressions of pain are a criterion of pain in certain •surround-ings• not in others.n 12 • He should not, therefore, ··say that a criterion is satisfied in certain circumstances, but rather that something IS a criterion o~ Y only in certain circumstances. Having made the co'ncept of a criterion dependent by definition on the situation S, he goes on to argue that we can never know the totality of circumstances in which a criterion (in this case pain behaviour) occurs. This argument is obviously circular since something IS a criterion only given the situation s. If, therefore, we can never know for certain when S is prevailing, we can never know for certain when a criterion X is present, and it makes no sense to talk as he does, of X occurring, as though X were independent of s. Finally he emerges with the conclusion that because the list of circumstances which make up the situation, is indefinite in length no entailment con-ditions for the criterion can be formulated. There are no such entailment conditions. In fact his argument, as it stands, has nothing to do with entailment conditions. It has to do with how we can know when a criterion occurs. Patently he asks what the criterion for a criterion is, so reproducing the original problem by pushing it back a step and making the concept X (what we normally call a criterion) dependent on the open textured concept of a situation, s. The conclusion which, as I have said, one 11. Malcolm, uwittgenstein•s Philosophical Investigations" 12. Ibid. ;i.n "Wi ttgenst~in", ed. George Pitcher 1968. . ·- 48 -would expect him to come to, is that we cannot know for certain what a criterion is, nor whether it is occurring. But this conclusion is strictly opposed to a statement he makes earlier about criteria viz. "What makes something into a symptom of Y is that experience teaches that it is always or usually associated with Y; that so-and-so is the criterion of Y is not a matter of experience but of definition. The satisfaction of the criterion of Y establishes the existence of Y beyond question." 13 • This certainly implies that there~ cases in which· we ·can say that the criterion is occurring, and say so with certainty. If the concept of a criterion is not an empirical but a gramma-tical concept, and if the definition of any criterion is to serve the purpose of linking it with that of which it is the criterion, then the definition must include a statement that the occurrence of the criterion, X, entails the occurrence of y. The confusion in Malcolm's argument arises because he does not seem to be sure whether the concept of a criterion is an empirical, or as I would rather say, open-textured concept, or one given by a stipulative definition. When he makes knowing the criteria dependent on knowing the whole situation (which is impossible) then he treats c~iteria as open-textured concepts. When he contrasts criteria with symptoms then he treats the concept as one delimited by a stipulative definition. Wittgenstein is well aware of the possibility of this confusion occurring. He says that, 11The fluctuation of grammar between criteria and symptoms·makes it look as if there were nothing at all but symptoms." 14 • This leads us to say for example that since we can only know a phenomenon through sense impressions and since our sense impressions are corrigible we can only form hypotheses as to whether Y will occur when X has occurred. In fact we can only say that it is 11highly likely" that X itself has oqcurred since our impression of 1 X1 could have been faulty. So that in the case o.:( say rain, we fail to see "that the fact that the false appearance is 13. Ibid. 14. P.I. 354 - 49 -precisely one o£ rain is rounded on a definition. The point here is not that our sense impressions can lie, but that we understand their language. (And this language like any other is rounded on convention.)" 15 • This seems to me strongly to argue for there being a· relationship o£ entailment between X andY. But the problem still remains as to how this entail-ment can be known, or makes itself apparent, in experience. It might seem as though there was a way out o£ this difficulty which happily combined the empirical and stipulative elements. It might be argued that the definition told one what the criterion o£ Y is, but whether the criterion was occurring was something which could only be empirically deter-mined. But this seeming way out is a cul-de-sac. It represents precisely the kind o£ argument which Wittgenstein attacks in the 'Investigations• and in •on Certainty'. In . -· . . effect it leads --to arguments such as the following: -·~e come across the victim o£ a motor accident who has a smashed kneecap. He is screaming in what sounds like agony. We know that his behaviour and his injury are in normal situations o£ this type, criteria o£ pain. What we do not and cannot know for certain, is whether this situation is normal. So we give him a pain-killer because there is a high probability that this is a normal situation, that therefore his behaviour is a criterion of pain, and we assume in con-sequence that he is in pain." There are two things wrong with this argument. Firstly and most importantly it is nonsense. Secondly it implies that only symptoms really occur in the world, -criteria never. Malcolm does not suggest that we employ the above argument as a way out. He shows quite correctly, I think, that it is inimical to Wittgenstein•s approach and quotes the following passage from the •Investigations• to prove the point: 15. P. I. 3 54-5 - 50 -"A doctor asks: iHow is he feeling?• The nurse says, 1 he is groaning.• A·report on his behaviour. But need there be any.question of whether the groaning is really genuine, is really the expression of anything. Might they not, for example, draw the conclusion 'If he groans, we must give him more analgesic• - without suppressing a middle term? Isn•t the point the service to which they put the description of his behaviour? '16• It is important to add to this the next little paragraph (Malcolm doesn•t do this) because it makes the link between the notions of criteria, certainty and language-games. Wittgenstein makes his interlocutor say: " •But then they make a tacit presupposition.• " To which he replies, "Then what we do in our language-game always rests on a tacit·pre-. . 17. supposJ.tJ.on. 11 A large portion of Wittgenstein 1 s writing in both the 'Investigations• and in 1 On Certainty•· centres very closely around this theme, and is characterised by his saying "The kind of certainty is the kind of language-game." 18 • This· idea is linked with another, perhaps t.he most important in Wittgenstein 1 s philosophy, namely that, ·~hat has to be accepted, the given, is - so one could say - forms of life." 19• I shall return to this topic later particularly in Chapters 10 and 11, but for the moment I should like to take up the Smith and Jones example again in order to characterise the second use of "criteria" more fully. If Jones did not see a shower but could give a full description of certain circumstances, then he did not see rain but was justified in saying that it had rained. He was justi-fied in so saying because the question: 1 Are you sure?' has no place when once he has given his description. 16. P.I. page 179 17~ P.I. page 179 18. P.I. page 224 Also o.c. 370 and 328-9 19. P.I. page 226 11This doubt - 51 -20 isn't one of the doubts in our game." We reach the bed-rock of conventions, and the conventions are such that a doubt is simply out of place in the circumstances we have described. In this case as in most others Jones makes the statement "It has been raining" on the basis of a number of criteriao ·This is another important point to noticeo Most interpreters of the notion of a criterion seem to agree that there is not always one criterion which must be satisfied, but rather that, "•••• we apply descriptive terms on the basis of several criteria which may be present or absent in varying degrees and no set of'which is both necessary and sufficient condition for the application of the termo" 21 • This state-ment, however, is true of both the uses of· "criterion". What distingu~shes the second notion (developed in the Investigations) from the first is the fact that the criteria are not necessarily parts of the phenomenon. The phenomenon need not be present to us for us to be able'to say that it is occurring or has occurred. criteria. The phenomenon is not necessarily reducible to its The importance of this-second usage for the dis-cussion of sensation and privacy is that it enables Wittgenstein to describe how· human behaviour'· and this includes a certain form of, what for want of a better expression I shall call verbal behaviour, is a criterion of sensation. When once this has been clearly described, he argues, then we can by-pass the theory that the only way of knowing whether a sensation is occurring is by introspection. This theory leads to scepticism since we cannot introspect another's pain, nor for the matter of that any of the other mental events and processes which the sceptic claims to identify in himsel·f o As we shall see, Wittgenstein discounts this claim by the sceptic to be able to know from his own case, but in dismissin_g it. he does not 20. P. I. 288 21. Wellman, 11Wittgenstein's Concept of a Criterion", l'he Philosophical Review, VOL. LXXI No .• 4. 1962. - 52 -dismiss the hypothesis that there are mental states and pro-cesses. He expressly says that he does not want to deny these states and processes, he merely considers them irrele-vant to the use (the grammar) of such terms as "consciousness", "understanding", "sensing", "feeling", etco When he says that "An inner·process stands in need of outward criteria," 22 • he is· pointing to the fact that without human behaviour which is publicly observable all talk of these inner phenomena is senseless. The grammar of inner states is such that it-in-volves reference to outward, behavioural criteria, which means that, it. o. o in order to understand the grammar of these states it is· necessary to ask: 'What counts as a criterion f b . . h . - ? 23. 1 or anyone's eJ.ng J.n sue a state.• " For examp ewe verify that someone understands the-principle of a series by getting him to continue it; or we verify that Jack has a sore knee by seeing how he winces when it is'pressed. "Asking when and how a proposition can be verified is only a particular way of asking 'How d 1 you mean?• The answer is a contribution to the grammar ·of the proposition. n ·24 o Again one is tempted by these and similar passages to ask whether he is denying the reality of inner states -trying to reduce them to the criteria by which they are veri-'· fied. Are there in fact two sets of phenomena, the inner and the outer, and if there are, is Wittgenstein guilty of saying that the inner are fictions? "If I do speak of a fiction," he says, "it is a grammatical fic;tion." 25 • It is a gramma-tical fiction in so far as the inner process is nothing with-out its criteriao Or_ to put it less dramatically and less misleadingly; we could say nothing.about it were it not for its criteriao It only makes sense to say that I am feeling this, thinking that (even thinking it to myself) because there are behavioural criteria for this thinking and feeling. Wittgenstein argues that it makes no sense to give oneself a 22~ P.I. 580 23. P.I. 572 24~ P.I. 353 25. P.I. 307 - 53 -direct, private exhibition of pain'itself through intro-spection. It is not so much that when we look in this way we find nothing but that we don't know how to look for it in the first place. (I shall deal with this more fully in later chapters) Of course this makes the case of inner states and their criteria different from our e~ample of the criteria which we would normally accept for the statement that it had· been raining. It is. possible to see rain falling whereas it is nonsense to say that we observe a sensation. (There is a use of sensation words in which we talk of seeing a sensation; 11 I saw the pain in his face" but this is not.the use I have in mind here) We say, 11 I saw a shower of rain," but never say 11I observed a pain." As Ryle puts it, 11To have a sensation is not to be in a cognitive relation with any object." 26• The point of my example was to show that Y is not always synonomous with its criteria. Wittgenstein does not want to say that pain behaviour is·synonomous with pain, and he does not want to deny that there are mental states of that mental proc?SSes take place. He merely argues that to talk about the mental state as though it was the sort of thing we could become acquainted with through closer observa-tion of our sensations, is not to talk about what is physically impossible, but to misconstrue·the grammar of "mental state". A mental state is not something like a frozen state. The · same goes for mental processes. We construe the mental pro-cess on the model of such processes as take place when an acid reacts with a base. We can come to learn how a salt is pro-duced by observing what happens when the acid and the base re-act together, and so we think that we can observe the mental process in a similar way, when in fact the nature of that pro-cess is totally different. Wittgenstein gives a b~illiantly lucid account of how the resultant puzzlement occurs: ••How do~s the philosophical problem about mental pro-cesses and· about behaviourism arise? - The first step is the 26. Ryle, 11The Concept of Mind" - 54 -one which altogether escapes notice. We talk of 2rocesses and states and leave their nature undecided. Sometime we shall know more about them - we think. But that is just what commits us to a particular way of looking at the matter. For we have a definite concept of what it means to know a process better. (The decisive move in the conjuring trick has been made, and it was the very one that we thought was quite innocent) - And now the analogy which was to make us understand our thoughts falls to pieceso So we have to deny the yet uncomprehended process in the yet unexplored medium. And naturally we don•t want to deny them." 27 • Just how important this Pcassage is, particularly the underlined sentence, I hope to show when dealing with Wittgenstein•s investigation o£ the concept o£ a human being. (Ch. 10) r-point out there that he never intended to give out that the other-minds debate could be resolved by a behavioural criteriology, as Chihara and Fodor seem to think. The im-portance o£ the concept o£ a criterion is not that it serves to bridge the Cartesian gap between body and mind, but rather that it show that the gap is .a metaphysical fiction. \ \ 27. P.I. 308 - 55 -C H A P T E R V PRIVATE LANGUAGE Professor Norman Malcolm writing on Wittgenstein's treatment of the private language argument says, 11 In order · to appreciate the depth and power of Wittgenstein1 s assault upon this idea you must partly be its captive. You must feel the strong grip of it. The passionate intensity of Wittgenstein•s treatment of it is due to the fact that he lets this idea take possession of him, drawing out of him-self the thoughts and imagery by which it is expressed and defended - and then subjecting those thoughts and pictures to fiercest scrutiny." 1 • Not only is this a very accurate description of the character of Wittgenstein's attack on pri-vate language, it also gives a good insight into his method and recalls strongly what Wisdom says about the treatment of a philosophical problem. The difficulty is not to hit on the right theory but to gather together all the theories and that in such a way as to be able to get a clear view of them. Wittgenstein puts it very graphically in a down to earth metaphor: "A main cause of philosophical disease - a one sided diet: one nourishes one's thinking with only one kind of example. 11 2 • In those parts of the Investigations in which pri-vate languages are discussed it is difficult to decide at first just where Wittgenstein stands - which of the voices which ask questions, make suggestions, argue for and argue against, is his. This initial difficulty in locating Wittgenstein•s position is an index o~ the mastery with which he practiced his philosophy. He assembles the whole problem by presenting all the arguments which have, each in their own way, constituted the problem, and the very moment the problem is completely assembled it falls down. I don 1.t think it 1. Malcolm, •rwittgenstein' s Philosophical Investigationsn .. p.67 2. P.I. 593 - 56 -would be unduly stretching his analogy (Investigations 118) if we said that the houses of cards which he destroys fall . because they are built too completely: the moment the card house is extended to the desired proportions it collapses. The same holds for the· philosophical problem - no sooner are all the arguments which go to constitute it as a problem assembled, than the problem collapses into nonsenseo "My aim," he says, "is to teach you to pass from a piece of disguised nonsense to something which is patent nonsenseo" 3 • Whether the nonsense is disguised or not depends on the· completeness with which the philosophical problem is presented. In the case of private language the point of Wittgenstein•s method is often lost and to talk, as Wellman does of his "refutation" of the private language argument 4• does not make it any easier to find. To say that he ·~efutes the argument" rather makes it sound as though he shows· that a set of perfectly sensible statements are simply false, as though private languages do not happen to exist because they are clumsy, or hard to learn, or confined in usefulness, etc. But the effectiveness of Wittgenstein 1 s treatment resides just in this, that it is a grammatical and ··not an empirical treat-ment. He is trying to get us and himself to reject a way of talking. He is not saying that that way of talking as such is false, "What could 'false• mean here?" we can imagine him asking. Rather he wants to --show that it is a nonsensical way. This needs to be shown since in philosophy n•)nsense parades as sense, and words which have every semblance of hav-ing a meaning have noneo As we saw in Chapter 2, what fre-quently happens is that we try to apply sentences and the pictures they suggest in areas where they have no application. The sentence appears to be in use when in fact it has no function at all. A somewhat analogous situation would arise if we put an animal in an evacuated box. . It would be odd to say that there is something wrong with its breathing, when we have put it in a situation where breathing is impossible. In a 3~ P.I. 464 4. Wellman, opo cit. - 57 -similar way it would 'be incorrect to say that propositions asserting the feasibility of private language are false since the propositional signs are not really in use ('The engine is idling• ). "Truth-or-falsity, are characteristics of a use of a sentence·." 5 • The sentences are being placed in an environment in which they cannot function. Sometimes the uselessness of the expressions we are using strikes us without our being.able to 'get a clear view' of what the matter is. nso in the end when one is doing philosophy. one gets to the point where one would just like to emit an inarti-culate sound" 6 • But of course this is not enough - a cure for our puzzlement needs to be effected, and the cure is grammatical and not empirical. For the reasons sketched above it is mistaken to accuse Wittgenstein of ambiguity, obscurity or of creating confusions. He does not create the confusions he merely describes them. Strawson, for instance, says that when Wittgenstein deals with private language and sensation, "one \ . 7 may well feel one•s capacity to learn coming to an end.n· • This feeling seems to me to result from a misunderstanding of Wittgenstein•s intentions •. It makes it seem as though the unclarity were his fault, whereas it is his achievement. Professor Cook argues quite correctly, I think, that Wittgenstein does not try to make the notion of a private language clear, "because the idea under investigation turns out to be irremediably confused and hence can only be suggested, 8 not clearly explained." 0 The idea of a private language is very closely linked by Wittgenstein with the idea of a private object, and he links the latter idea with the notion that sensations are private:· he describes a theory in which sensations are tre~ted as 5. Strawson, "On Referring", Logico Linquistic Papers, p.9-10 6~ P.I. 261 7. Strawson, "Review of Wittgenstein's Philosophical lnvestigations,u Pitch~r p.41 8o Cook, "Wittgenstein on·Privacy 11 , Pitcher p.286 ... 58 .. private objects. Professor Cook gives a brilliant and clear account of Wittgenstein•s position on private language from the point of view of his treatment of the privacy of sensations. This is what one might call an organic account, explaining and illuminating Wittgenstein•s thoughts on private language as they actually develop in the Investigations. ~nother method of dealing with private language is to show that it rests on an untenable theory of linguistic meaning. This method can only give an incomple·te account of his treatment and needs to be connected,·as Coo~ argues, with a treatment of the notion that sensations are private. In this chapter, however, I shall stick to the first point, viz. that the notion of private langu-age is in part the result of a mistake about language. In the following chapter I shall connect this with the notion of the privacy of sensation. Anyone who tries to present the Wittgensteinian position with respect to private language, and who tries to present it as the correct position, is immediately faced with the insoluble problem of trying to give an intelligible account of a notion which he holds is nonsensical. As far as .Wittgenstein is concerned the words "private language" have as little meaning as the words "square circle". One cannot then try to make sense of the d~scriptions of private languages offered by the protagonists of the notion, nor out of their arguments, because they are nonsensical. Wittgenstein warns us against the absurdity of trying to force sense out of non-sense when he writes, "When a sentence is called senseless, it is not as it were the sense that is senseless. But a com-bination of words is being excluded from the language, with-drawn from circulation." 9 • So in this instance the only recourse is simply to describe and exemplify the arguments of the private language protagonists, and in so doing try to show '~he bumps which the· understanding has got by running its head 10. . up against the limits o£ language." 9. P.I • 500 10. P.I. 119 - 59 -How then might a ~rivate language theoretician de-fine a private language? Possibly like this: a languag.~ L is private if, when it is used by an individual I, the mean-inq of what ~s said in L by I is necessarily unintelligible to anybody but I. The point of adding "necessarily" is to distinguish this use of "private language" from the one where it is used to refer to some secret code or esoteric set of signs. When "private languagen is used philosophically it is used in such a way as to make it nonsensical to say that what is said in it is understood by more than one person yiz. the speaker. It is atlogicallytt private and not merely contingently so. This obviously means that it is impossible ,, to describe or exemplify such a language. A description of a language must include a description of the rules for the use of the language, and this it is impossible to do in the case of a private language. Ayer in his well known essay ucan there be a private language" purports to describe such a language as used by a solitary individual, but when we examine this description and the linguistic presuppositions on which it is based we come to see that the attempt to make clear what a private language is, is misguided. Ayer•s Robinson Crusoe uses this language to describe fauna and flora as well as using some of the words in it to nstand for his sensations". In the latter case Ayer says 1t •••• where these sen'sations are entirely private, in the sense that they have no 'natural expressions 1 which Man Friday can identify, it may well be that Crusoe fails to find any way of teaching him the use of the words which he employs to stand for themo But from the fact that he cannot teach this part of his language to Man Friday it by no means follows that he has no use for it himself. In a context of this sort, one can teach only what one already understands. The ability to teach, or rather the ability of someone else to learn, cannot therefore be a prerequisite for understan~ing." 11 • For·the moment I 11. Ayer, ncan There Be A Private Language? 11 , reprinted from Proceedings of the Aristotelian Soc. VOL. XXVIII (1954) in Pitcher p.261 - 60 -want to ignore what he says about the objects referred to in the language, whether palm trees or pains, and concentrate on just those features of our definition which he reproduces in his example. · Firstly, we are told that Crusoe uses the language to describe things and therefore he must use it in a certain way i.e. according to some rule of description. Secondly, he says things to himself which he and only he can understand i.e. the meanings of the signs in the language are private. Accordingly the first question to ask is whether it makes sense to say that one can follow a private rule? How does one know one is following it correctly? Wittgenstein asks how we undertake to use the word in a rule-governed way. 11 Is it to be assumed that you invent the technique of using the word; or that you found it ready made? 11 12 • It might seem as though one could answer quite straightforwardly 11Yes of course, I invent the technique,u but such an answer only sets the problem back to the question of what could be meant by a technique here. If it is used in the ordinary way then, nusing a technique" means, among other things, working to a rule, and the meaning of "rule" is related, as we saw in ' Chapter 3 to words like "same"·, "regular" and 11uniform 11 • The meanings of these words is, in turn, taught by·making reference to the "common behaviour of mank.ind;, 13 • But in the case of a private rule there can be no question of any reference to any pattern of human behaviour. What this means is that the combination of the words ''Private" and "rule" into a phrase is not a move in the language-game - it is like trying to promote a pawn to a king in chess. It is of course possible to invent a new word, but it is absurd to say that somebody has invented a customary usage of that word. The word comes to be used in this or that way, but its so being used is not something that an 12. Po I. 262 13. P.r. 206 - 61 -individ~al can invent. The point may be put in another way by saying that dictionaries primarily describe rather than prescribe uses of.words. The contents of a dictionary are derived from the practices of those who speak and w,rite the language. The lexicographer can prescribe how a word ought to be used only on the basis of a description of how it has been used. If someone invents a new word, and not just a new ' sound, then he must invent a use for this word. Or if he wishes to retain a familiar word, but give it a new meaning then he must prescribe a new use for this word. This might seem contrary to what I have just said about the lexicographer but it is noto Everyone (lexicographers included) can in-vent new words and new uses of words, but no one can invent a custom of linguistic usage. To make what I have just said a bit clearer I shall take an example fro~ Lewis Carrol: Humpty Dumpty and Alice are discussing the former 1 s peculiar use of the word "glory'': • "When I use a word," Humpty Dumpty said in rather a scornful tone, "it means just what I choose it to mean - neither more nor less." "The question is," said Alice, "whether you CAN make words mean different things·.u "The question is,u sa:Ld Humpty Dumpty, "which is to be master - that•s all." 1 He goes on to discuss the "temperaments" of different parts of speech and then exclaims: ' "••• Impenetrability~ That's what I sayt" "Would you tell me please," said Alice, "what that means?" "Now you talk like a sensible child," said Humpty Dumpty, looking very much pleased. · 11 I meant by •impenetrability' that we've had enough of that - 62 -subject, and it would be just as well if you'd mention what you mean to do next, as I suppose you don.' t intend to stop here all the rest of your life. n ' 14 • So not only does Humpty Dumpty know the meanings of new words, such as those in Jabberwocky, but he invents uses for old words, and what is more, he teaches them to Alice, who understands them. What he does not and cannot do, is invent the technique of usage. He can invent languages but he cannot invent language. And he cannot do that because it makes no sense to say that someone invents a custom of using words, or that someone invents an established practice of usage. What enables him to use a word in a new way, according to a rule different from the normal rule, is the fact that there exists a practice which is the practice of following rules. Within that practice enorm~us varieties of uses of words are possibleo The problem is that we are too apt to forget what is involved in the phrase "within that practice." We forget that all particular uses of language, such as· naming, describing, ~ying, asserting, asking questions, etc. presuppose "a great deal of stage setting." 15 • This stage setting is one or other form of social life~ Social life is the terminus a quo for language. Language is part of a social practice. This is what Wittgenstein means by saying that when human beings agree in the language they use, 11this is not agreement in opinions but in forms of life. n 16 •· It is one thing to invent words and uses for them, it is· another to invent forms of life. As Rush Rhees argues, one can invent a vocabulary, 11in circumstances of a social life which has in fact grown up with language and could no b . t d th J.~ CO''lld.11 17 • more e 1nven e an anguage ~ The private language argument must exclude the possibility of the language being part of a social practice. The consequence of this exclusion is that it can make no ·sense to talk of following the rules of the language for the simple 14. Lewis Carrol, ."Through the Looking Glass." 15~ P.I. 257 16. P.I. 241 17. Rhees, 11Can There Be a Private Language?" Pitcher p.276 - 63 -reason that there is nothing which could count as a correct or as a mistaken application of the rule. Being private, in the sense of the definition, means that there can be no check-ing by another on whether a word is being used correctly or not. In fact the notion of correct and incorrect is out of place here. With private languages there is no criterion of what is correct. 110ne would like to say: whatever is going to seem right to me is right. And that only means that here we can't talk about •right'•" 18 • From this we must conclude that the statement that a private language is used in a par-ticular, rule governed way, is nonsense. The basis for this conclusion has been laid by Wittgenstein in his treatment of rules: where any action can be made to accord with a rule, no course of action can be said to be prescribed by the rule and the concept of a rule becomes inappropriate in these circumstances. The stock objection to Wittgenstein•s argument is that it IS quite possible for an individual to check his own use of a private word. The protagonists of this argument, like Ayer for example, usually concentrate on private languages which they treat as descriptive of private mental states and processes, namely sensations and/or sense data. They argue that the private language user can check whether he is using his private word correct1y by remembering the cir-cumstances in which he "undertook" to use the word in the first place. Wittgenstein•s rep1y, so often quoted, is that such a man is like someone who buys several copies of the same newspaper in order to check that what the first copy said was true. This is meant to show that as long as there is no standard against which to check the use of a word on any par-ticular occasion, it makes no sense to talk of "checking" at all. Ayer•s reply does not really seem to me to be a counter to those he·is attempting to refute. First, he admits that the important difference between private usages and newspaper reports is that, "••• the facts which the newspaper reports are independently· verifiable, in theory if not always in 18. P.I. 258 - 64 -practice. But," he continues, "verification must stop some-where. As I have already argued, unless something is re-cognised, without being referred to a further test, nothing can be tested.. In the case of Crusoe•s.sensations, we are supposing that beyond his memory there·is no further test. It does not follow that he has no means of identifying it, or that it does not make sense to say that he identifies it ·19. right or wrong.n But a memory must surely be a memory of some object or event, the existence or occurrence of which must be capable of a verification other than the testimony of that memory. And I do not mean that the object or event must IN FACT be independently verified but that is possible in principle to verify it. If this is not so then whatever one remembers to have been the case must have been the case -memory would be incorrigible. however. We do not treat memory this way, Wellman formulates the same argument as Ayer's slight-ly differently. Like Ayer he argues that one memory can be used to check another, "provided that each has some initial probability." 20• Now· one would expect this to mean: provided each is a memory of some objective, publicly observable event. But apparently this is not what Wellman has in mind. The criterion for the correctness of the memory need not, in his view, be independent of the memo~y - the memory is its own sufficient warrant: "Wittgenstein seems to assume that any genuine criterion must ·be something independent of that of which it is the criterion, external to the person using the criterion, and accessible to all persons equally. Until some reasons are produced for this view it remains a dogmatic essumptio•l which must he questioned.a 21 • He then goes on to restate Ayer's argument that Wittgenstein's view results in an . infinite regress since each act of checking the use of a sign, itself stands in need of checking. "Are there standards for the use of standards? Unless at some point there is no longer 19. Ayer, op.cit. pp.260-61 20. Wellman, op.cit. P• 21. Wellman, op.cit·. ' \ - 65 -any need for an external standard all criteria become point-less." 22 • Both Ayer and Wellman seem to imply (a) that Wittgenstein has given no justification for insisting upon the publicity of criterial standards; {b) that he is unaware of the problem of regressive verifications; (c) that their view, which I shall call sense datum empiricism is not beset by any of the difficulties with which Wittgenstein argues it is beset. Leaving aside the last point for the while we should notice that both (a) and {b) are major topics of in-terest in the Investigations and in On Certainty. The fact that the latter was only published in 1969 perhaps accounts for their overlooking Wittgenstein•s treatment of these topics. In On Certainty Wittgenstein constantly stresses that verifications alway·s take place within a structure of nconvictionsu.· 23 • Unless this system is given, no judg-ments are possible, because no confirmation of hypotheses would be possible. 24 • This system of convictions arises out of specific conditions of existence - it is a function of a specific societal practice. Language is a part of this practice. In a language-game some judgments are not put in-to question, ·they are the groundless grounds which make the practice of language possible.. Our assent to these grounds is not primarily an intellectual assent - the ultimate criterion for what we accept as certain is what we do: "Sure evidence is what we accept as sure, it is evidence that we go by in acting surely, acting without any doubt." 25 • Thus a language-game can only be played on the basis of a form of life in which certain things' are not put into question: 11 I really want to say that the language-game is possible if one trusts something (I did not say •can trust something')" 26 • 22. Wellman, op. cit. 23. o.c. 102 24~ o.c. 105 (This theme recurs throughout the book). 25~ o.c. 26. o.c. and 344 and 524. - 66 -The parenthetical remark is very important. It shows, I think, that Wittgenstein is not committing himself to a static and idealistic theory of knowledge. He is not saying that human knowledge constitutes reality i.e. that what we take to be the case (what we "trust~')becomes the case. All he is asserting is that that knowledge can only be knowledge within a structure. The structure itself may change when what served as foundations are put into question and a new set of evidences take their place. The structure is what he calls a world picture. Propositions describing the world picture are from one point of view prescriptive: "••• their role is like that of rules of a game; and the game can be learned purely practically, without learning any explicit rules." 27 • The game is the abstract expression of a practice. When the game changes, the practice changes. That at any rate is the way the change is thought, in theory. In practice the division does not really exist. These themes need a development which is beyond the scope of this chapter. What I wish to establish as far as the present discussion is concerned .is that the regression of doubts must come to a halt if a language-game is to be played. But further, that it cannot come to a halt in private experience. If the word ntruen could only be used .on the basis of private experience then there could be no sense in talking about agree-ment in judgments. If the verification of whether a word has been used correctly is_ultimately a private affair then it is very hard .to see how language (public as opposed to private) functions. Wittgenstein•s remarks in the Investigations are intimately connected with ··those in On Certainty. · His reply to critics such as Ayer and Wellman comes in the form of a question which he puts to himself: 11 •so are you saying that human agreement decides what is true· and what is false?• " and an answer: "It is what human beings SAY that is true and false; and they agree in the language they use. That is not agree-ment in opinions but in form of life. If language is to be a means of communication there must be agreement not only in definitions but also (queer as it may seem) in judgments." 28 • 27 ~ o.c. 28. P.r. 241 - 67 -Another line of objection to Wittgenstein•s treat-ment of rule-following and checkability, is that taken by Judith Jarvis Thomson in her essay, "Private Languages". The essay is interesting in its own right, but her criticisms also make necessary a close examination of Wittgenstein on privacy and rules, and this re-examination requires, I think, that the interpretation of his thought given.by Malcolm be, if not revised, then made somewhat more rigorous. The view which she is attacking and which she attributes to the Wittgensteinians, particularly to Malcolm, is the following: 11A man•s use of a sign is not governed by a rule unless it is not merely possible that he should violate the rule but more that he should violate it unwittingly. That is it must be possible that he should think he is following the rule and not in fact be following it; from the fact that he thinks he is following the rule; it musn•t follow that he really is follow-ing it. 11 29 • Now it certainly does seem to be part of Wittgenstein•s conception of a rule that it must be logically possible for ··the rule to be broken, and the possibility must therefore remain open, in some cases, for someone to think he is following a certain rule and to be mistaken. I say "in .. come casesn because Thomson tries to show that there 'are· rules· which it would be impossible to break without knowing one was doing so. This sort of rule is meant to provide the exception to the maxim that i't must always be possible to think one is obeying a rule and not be doing so. As an example she gives: "Always decide to do what you think at the time it would be most fun to do.n 30 • It is obviously impossible to think one is deciding when one isn•t, and one can accordingly classify under this heading all rules of the type: ·~hink X whenever ••• o ", "Decide Y whenever. o •.. o ", "Imagine Z whenever o • o. 11 and •, so on. · Generalising these into· a formula one might say· that there was a class of rules which were of the type: 11Do X, which it is impossible to do without being aware that ·one is doing it whenever ••••" 29. Thomson 30: Thomson, op. cit. 68 Now Malcolm's statements on rules imply that cases such as the above are not rules. He says that, "••• the concept of a rule implies that there be a difference between, •He is following a rule• and •He is under the impression that .. . 31 he is following a rule• ··••• ...... • I suspect that Malcolm overlooked cases such as tho-se put forward by Thomson, and that he had only those rules in mind, obedience to which can be checked independently of a report by the person who is supposed to be following the rule. This is quite a large oversight but it is not mortally damaging to Malcolm, Cook and Rhees•s arguments against private rules and private languages.; The possibility of a rule of this type 't' being broken still remains open. For instance, if the rule is: "Remember me when I am gone away", it is surely possible that it should be broken? Whether or not it has been, can be discovered by asking the person who is supposed to have followed the rule. But Thomson disagrees. She formulates her counter argument as a dilemma. Either, she argues, it is not possible to break such a rule knowingly, or, (should this not be thought a serious objection) that such a rule cannot be broken at all, it can only be forgotten. She writes, nfor notice that one can't even violate this rule unwittingly unless one has, in a sense forgotten the rule. And now what rules - however private - can't be violated in this way? 11 32 • I do not see that the first horn of her proposed dilemma· really constitutes a problem. No one (except Malcolm through an oversight) has claimed that one must be able to know one is breaking a rule at the moment of breaking it. Wittgenstein certainly never argued that it must be nossible in all cases to think one is.obeying a rule and not be. The second horn seem a bit trivial: it' the ·rule is, "Remember X wh~never • o •• " then to say that the only way of failing to remember X is to fail to remember the rule is nonsense. In such a case the person involved has simply forgotten X i.e. failed to follow what the rule lays down he should do. It is true that it is 31. Malcolm, op. cit. p.68_ 32. Thomson, op. cit. P• - 69 ... not possible to remember the rule and still break it un-wittingly but it is none the less perfectly sensible to make a distinction between the fulfilment of what the rule lays down and ·remembering the rule. In Investigations 201 and 202 Wittgenstein dis-tinguishes between the expression of a rule and obedience to the rule. And Thomson herself says that it is possible for any rule to be forgotten, and rules of type •t• are not peculiar in this. If, therefore, she means-to recommend that we never talk of breaking a rule of type 't' but only of forgetting it, then she i& prescribing a new usage, contrary to the ordinary one, but she is not really attacking the Wittgensteinian position that it must be possible for a rule to be broken. It is possible to verify whether a rule of type 't' has been kept or broken by asking the individual involved, "Did you think, decide, remember, ima,gine, etc. ••••• as it is laid down that you should?" And it is possible to find out whether the meaning of the· rule has been grasped by the in-dividual by his answer to this and to further questions. What then is the distinction between these cases and those of the so-called private rule following? After all Wittgenstein does say, nto think one is obeying a rule is not to obey a rule" and this could be reformulated, 11Saying one has obeyed a rule is not to have obeyed a rule."· The difference is this: we know what is meant by, "Imagine the picture when'"' ever ••••"• 11Think of cheese whenever ••••", "Remember Rosy whenever · •••• 1•, and we can verify whether these· things have or have not been· done. We know what must be done and we know the conditions under which it must be done. But the problem with private rules, whether private language rules or private rules for any other practice, is not so much that obedience to them can•t be verified but that the rules can•t be intelligibly formulated in the first place. Private·Rulei "Do X whenever Y occurs.u Qua private"obedience to this rule cannot be verified because conditions for following the rule cannot be - 70 -formulated. We cannot say what is going to constitute an occurence of Y because we have no language in which to say this. So the private rule cannot be formulated. Talking about the private rules for a private language Rush Rhees says, "I say I cannot know a language privately, for what would there be to KNOW? In language it makes a difference what you say. But how can it make any difference what you say privately? {I do not mean talking to yourself) It seems that in a private language everything would have to be at once a statement and a definition. I suppose I may define a mark in any way I wish. And if every use of the mark is also a definition - if there is no way of discovering that I am wrong, in fact no sense in suggesting that I might be wrong - then it does not matter what mark I use or when I use it." 33 • .Now in the case of rules of type •t• {the ones which were beginning to seem indistinguish-able from private rule·s) it does make a difference what is done by the person to whom the rule has been given. What he does whenever •••• is either in accordance with the rule or it isn•t. This is not so with private rules. The point which Wittgenstein establishes is that a "private rule 11 only seems to be a rule. It is not so much that a private rule is a rule which one can't break. To put it that way is misleading because it makes it look as though there WAS such a. thing as a private rule which had this peculiar quality .about it: it can't be brokeno If we say that a private rule can•t·be. broken then we should add as qualification, that neither can it be obeyed, because it cannot be formulated i.e. conditions for rule governed behaviour cannot be formulated. As I said at the beginning of the chapter, the words "private rule" can have no intelligible , meaning if we are attempting to use them in the ordinary wayo This particular combination is being excluded from the language. 33. Rhees, op. cit. p. 274 - 71 -Much of the confusion about, and resulting dis-agreement on, Wittgenstein•s treatment of privacy has re-sulted from failing to distinguish between two statements in the Investigations. The statements are 202 (about thinking one is obeying a rule) and 259: "Are the rules of the private language impressions of rules?· - The balance on which impressions are made is not the impression of a balance." I formulate the distinction between them in this way: the reasons why thinking one is obeying a public rule not of type •t•; and thinking one is obeying a private rule, do not constitute obedience to a rule, are different. Thinking one is obeying a public rule is not the same as obeying it because there are objective criteria for de-ciding whether the rule has been followed, and thus the possibility remains open that one may think one has obeyed the rule and yet be correctly judged not to have obeyed it. Thinking one is obeying a private rule involves a mistake of a different sort. In this case one mistakenly thinks that there IS a rule to be obeyed. The "rules" of the private language are not rules although they·may appear to be. So when we try to use a private languag~ we think we are obey-ing rules, only there are no rules to be obeyed. When we use ordinary language the possibility exists that we may think we are obeying its rules, and not be. language nothing can be sa~d. \ In a private ... 72 -C H A P T E R VI LINGUISTIC FORMALISM AND THE THEORY OF COMMUNICATION-INTENTION In the chapter on private languages I have tried to show that the idea that it is possible to follow a private rule has unacceptable consequences. It is not possible, I have argued, to enunciate clearly what one means by "follow-ing a rule privately," any more than it is possible to say what measuring is if the measuring is being done with a piece of elastic. Or perhaps I should put it another way by say-ing that it is possible to say what "following a rule private-ly" means, or what "measuring with an elastic ruler" means, provided one accepts that the words are not being u·sed in the .. ordinary way. The game of measuring with a meter rule is not like the game of measuring with a piece of elastic. The conclusion I carne to in that chapter is pretty disasterous for any theory of language which claims that the rules of a language need not be public, and my concern there was chiefly to examine the arguments of philosophers who think that private rules for private languages do not entail such consequences. Now I want to take a look at a semantic theory which seems to me to result in the positing of private .rules. My purpose in doing this is not simply to try to· r e·· ject this theory because of its untenable consequences but (a) to show how these consequences follow from the nature of the theory (P) to suggest an alternative approach to languag0 and linguistic meaning which is opposed to the first and which does not entail the privacy consequence (c) ·to show that the alternative is _{i) Wittgensteinian, (ii) that it needs to be supplemented as Wittgenstein saw by an account of intention and (iii) that it involves the notion of language as social practice -a notion implicit in Wittgenstein's later work. - 73 .. A further, more general, intention behind this chapter is that it should serve to link up the main points rna~ •. in the thesis so f~r. The discussion will involve the ubiq~-tous notion of rules~ their function and what it means to follow them, the concept of a criterion will be invoked in dealing with the question of how we know when something , ~ been said, and perhaps most important of all, semantic theories of the formal type found in the Tractatus, will be criticised from the stand point of another theory. But before I go on with the discussion a word of explanation is necessary. A few lines earlier I spoke of an •alternative approach' to language, thus assiduously avoiding the imputation of advancing a theory, but now it may look as though I have come out in the open, so to speak and talked of criticising one theory from the stand-point of another theory. How is this second, seemingly more ingenuous way of putting it,to be made to square with Wittgenstein 1 s intention not to "advance any kind of theory.n 1 • The answer lies I think in the descriptive character of the · Wittgensteinian treatment ("theoryn) of language. He is not· advancing an hypothesis as to how ·language is produced by us. He is saying that if by •language' we mean what we ordinarily mean (if we use the word in the ordinary way) then his is an accurate description of what using a language, or what saying something consists in. He is trying to give us the concept of language. He is not trying to EXPLAIN how language is acquired, or to explain the workings of the mechanism which enables us to use it. Such explanations are the business of the scientist, particularly the psychologist and the genetic epistemologist. In so far as they employ a hypothetico-deductive method, and in so far as their explanatory theories are subject to verification at the descriptive an'd predictive level, Wittgenstein•s treatment cannot be applied to them. As he so often stresses, his treatment is a grammatical treat-ment, his aim is nto bring words back from their metaphysical to their every day use." 2 • Pole can see nothing in this but destruction and the·.ha1t of all speculation, and in a way 1 ~ P. I. 109 2. P.I. 116 - 74 -he is right. Metaphysical systems are being destroyed and metaphysical speculation is being halted. But it is quite incorrect, it seems to me, to contrast ordinary language with the technical languages of the sciences, in an attempt to show that Wittgenstein wants to stop scientific investi-gation. He has no wish to make a contrast between ordinary and technical language, in fact he points out that the, ·"Symbolism of chemistry and the notation of the infintesimal calculus 11 3 • have become part of our language. The con-trast he·makes is between ordinary language and metaphysical language. The latter is extraordinary not because its terms are technical or outre but because the language bears the ( ~ mark of conceptual confusion and itself perpetuates 'and elaborates this confusion. Wittgenstein's technique is geared to the purpose of enabling "the reader to shift for himself when he encounters conceptual confusions." 4 • And in line with this, his aim in dealing with language is not to dispute any scientific hypothesis about how in fact • language is acquired, but rather to clear away certain con-ceptual problems which arise in the discussion of language. For these reasons 11Wittgenstein's theory of language, 11 is something of a misnomer. His general-intention is to describe the actual use of language, which description will include the use of tl:le word "language" in the language of every day. 5 • One of the things involved in this is a des-cription of "meaning": the meaning of a word comes out in its use and so a description of the use of words involves talking about meaning. A semantic theory which separates word, use and meaning will result in the conceptual confusion which he warns against when he writes: 11You say: the point isn•t the word but its meaning, and you think of the meaning as a thing of the same .kind as the word though also different from the word. Here the word, there the meaning. The. money, and the cow that you can buy with it. (But contrast: money, 3~ P.r. 1a 4~ P.I. page 206 5. P.I. 120 and 124 - 75 -and its use.)" 6 • When we think about semantics in one v,·~. it seems as though the signs we use in language were lifeh· -. in themselves and that we put life, i.e. meaning, into them, through using them according to the rules dictated by the nature of a linguist'ic mechanism. The gap between word and meaning, is, as it were, closed by a mental mechanism which connects sign and syntax with meaning. Wittgenstein is con-cerned to show that the hypothesis of a mental mechanism (which might or might not exist) will not help us in getting a clear view of the c?ncept of meaning in langua~. In so far as he believes that his treatment of confusions about language and meaning enable us to get a clear view of this concept, it may, I suppose, be regarded as a theory of semantics. The two theories which I have in mind are those discussed by Strawson in his inaugural lecture on "Meaning ' 7 and Truth" • and I shall follow his argument through here in some detail. The meaning of a sentence depends, in a systematic way, on the meanings of the words which compose it:.- Con-versely a word has meaning if it contributes systematically to the meaning of any sentence in which it occurs. So far there is pretty common agreement among all schools of thought on the topic. But at this point disagreement breaks out between what Strawson calls the theorists of communication-intention and the theorists of formal semantics. The former argue that there can be no account given of meaning without reference to an audience directed intention to communicate something, on the part of the speaker. They are quite will-ing to grant that words and sentences are bounded by, or used according to, rules, but they insist that these rules can only be understood as rules for communicating and must therefore be understood in terms of an intention to do so. 6~ P.r. 120 7. Strawson, "Meaning and Truth" Logico Linguistic Papers. - 76 -The formal semanticists on the other hand, say that the intention to communicate, far from being essential to the concept of linguistic rules, is quite incidental to them. For them, lin.guistic competence can be dealt with, without any mention of communication. Taking these points of view in turn and starting with the communication-intention theorist, we must ask what sort of account is required of him. Firstly he must elu-cidate the concept of communication-intention without making any reference to linguistic meaning. Then he must show that linguistic meaning must be explained in terms of an in-tentional communication act, and this of course further im-plies that the theory must contain as primary and essential, . the concept of a speaker's meaning something by an audience directed utterance. What the speaker means is related essentially (conceptually) to what he intends. Finally, since we speak according to rules, and since utterer's mean• ing, as Strawson calls it, is in this sense governed by rules, it must be analysed in terms of these rules of grammar. But the rules are rules for communicating and their nature can only be understood if they are seen as rules which enable those who follow them to get across some or other belief or wish or puzzlement or annoyance, etc. to an audience. Here the formal semanticist might object that since we communicate very complex things we must have very complex intentions to communicate, which, if we are to avoid circu-larity must be posited independent of any means of communi-cating them. We are called upon to posit a Shakespeare with the intention to communicate the ideas found in Hamlet, or an Einstein with the intention to tell the world about relativity, but with no means of actua~ly communicating. This, it might be argued is absurd. But the answer to this is that no such positing is required. All that is required is the concept of a primitive, pre-conventional (pre-rule governed) communication. - 77 -This concept of a pre-conventional utterance fi-'-s into what Strawson calls an account of the analytic-genetic variety. Such an account might run as follows: An utter~who has an audience directed intention successfully communi-cates pre-conventionally by an utterance x. Let 1 s say he meant P by x. On the next occasion upon which a communica-tion intention arises the utterer utters 'x 1 in order to communicate P. So •x• is uttered and works, then it is uttered because it works, and finally it works because 'it has become an established utterance. But what of syntactic structure, since the meaning of a sentence depends on the arrangement of its parts? Well a pre-conventional utterance could have a structure such that the utterer might go on to have another communi-cation success by retaining one part and varying another, thus saying somethin'g which is partly the same and partly different from what was meant by the .first communication. If we step back at this stage of the discussion and make a general reconnoitre of the ground covered we can see the areas disputed by the two parties. They agree (i) that the meaning of a sentence is determined by semantic and syntactic rules of language; (ii) that all members of a community with linguistic competence can communicate with and influence one another; (iii) that communication is determined by convention; in that, what is intended one utterance should communicate is related to the conventional meaning of that utterance. They differ on the relation-ship between meaning-determining rules of language and the function of communication: · Communication-intention theorists insisting that the nature·of these rules can only be determined by making reference to communication-intention and formal semanticists denying this. Formal semanticists define meaning in terms of the conditions under which any sentence is true, and even those - 78 -forms of expression such as imperatives, optatives, etc. which are not propositional in character can be dealt with in terms of some related set of conditions as Hare shows in "Meaning and Speech Actsn 8 • So when an account has been given of all the possible transformations (from truth conditional statements to say fulfilment conditional state-ments, in the case of optatives), then the framework of semantic theory will be complete. The difficulty, that the sentences that we are talking about are type sentences, sentences not actually in use, and as such cannot be true or false, can be overcome by relativising their truth con-ditions to the particular conditions pf utterance i.e. a general statemept can be given of the type of conditions in which a $entence may be utte~ed truly. The obvious move now for the formal semanticists .. is to say what truth is, in order to pe able to understand what meaning is.. He cannot evade this by saying truth is what is laid down in the rules governing truth conditions in a language- this'is j\lst ciroular. As account of truth must therefore be given in terms of existing states of affairs. Strawson has in mind here, I think, the sort of account which Wittgenstein gives in the Tractatus: 4.062 "Can we not make ourselves understood with false proposi-tions just as we have done up till now with true ones? -So long as it is known that they are meant to be false. No1 For a proposition is true if we use it to say that things stand in a certain way, and they do; and if by 'P' we mean -p and things stand as we mean that they do, then construed in the new way, 'P' is true and not false." So we must say that tttrue" in the expression "true statement" refers to the circumstance that things are as the utterer of the statement states them to beo And the meaning of a sentence is determined by the rules for how things are stated to· be by one who makes a statement, (or by rules by how things are expressly supposed to be by one 8. Hare, "Meaning and Speech Acts" In "Practical Infe::::-ence~' - 79 -who expresses a supposition) etc. The concept of a sta te.!r.:·.,·: only makes sense in terms of conditions of truth and falsity -a sentence which is neither true nor false is not a statement. These truth conditions coupled with a rule or set of rules for the formation of sentence-statement, determine the meaning of the sentence. But as we have noted, sentence types are by their nature not subject to truth conditions and so we must relativise the formation rules for statements, to contexts, so that ntruthn and nfalsityn may become applicable. The ·mean-ing of a sentence·must be·determined by those rules which state WHAT statement is made by one who in uttering a sentence in certain conditions makes a statemeqt. (And a si,milar form-ulation of meaning can be given in th~ case of wish~sv orders, questions~ etc.) In case 'this is too abstract and unclear lets take this example. .In overcast conditions if I want to make a statement about the colour of the heavens then I must utter the sentence, 11The sky is grey", which sentence is composed of words put together in a way· determined by the rules of syntax. Unless we are given a set ·of circumstances we cannot determine what sentence must be uttered if a true statement is to be made, and for this reason truth conditions cannot be formu·la ted for type sentences. In this way, the formal semanticist argues, the meaning of the statement is determined by rules which specify what sentence is to be uttered in given circum-stances and rules which specify how such a sentence is to be put together. Were there not such rules then the utterance "The sky is grey" would be a random s:tring of phones arbitrarily uttered, and as such would have no linguistic meaning. And of course they are right in what they say. The weakness in their position is more in the nature of an ommissior!" Their theory is partial in a way which makes a conceptual account of language in terms of it, impossible. The prota-gonists of communication intention would point out that when - 80 -once reference has been made to the content of a speech act (what is said) we must refer also to some audience directed intention. For example, in the case of assertions: we have said that their meaning is determined by rules which say what statement is made by one who, in uttering a sentence in.given circumstances, makes a statement. Thus the rule would go something like this: "If you want to state that you are sweating and state it truly· then in circumstances in which perspiration forms on your body you must say, "I am swe·ating", which string of words is arranged according to· the following· rules of syntax •••••" In this case, as in the case of the cloudy sky, included in the rule must be a reference to some intention to state something on the part of the utterer. If the intention to communicate is excluded from the antecedent or if the rule is not stated in the form of a hypothetical such as we h~ve invented, then it does not fulfil the re-quirement of a meaning-giving rule. will come out shortly. 9 • Further reasons for this Two ways of meeting this objection are open to the formal semanticists (a) he·can make meaning dependant on truth conditions without giving any further account of the latter. This we have seen is unsatisfactory. (b) Truth conditions can be subsumed under the notion of correlation, i.e. the elements of any given sentence can be correlated with any actual state of affairs. This notion of correlation, however, is too unspecific to be useful. There are many kinds of behaviour·which are .correlated to states of affairs, without this correlation conferring on the behaviour the character of truth or falsity. The notion of·correlation alone is in-sufficient for the formulation of truth conditions. Secondly the hoary problem of the false sentence arises. Strawson gives the example of the sentences, 11 I am tired" and "I am not tired" both of which can be correlated to the same possible state of affairs viz. that of my being tired. Granted the 9. Waismann, . "The Principles of Linguistic Philosophy" The Chapter on "The. Causal Interpreation. of Language contains an excellent treatment of language, meaning and intention. - 81 -correlations would be different the one yielding a true and the other a false assertion. But this only shows that the notion of ciorrelation alone is not sufficiently specific to give an account of what makes a sentence have the meaning of a true statement. And in the case of a sentence which can be correlated with many different states of affairs in which case it will have many different meanings, the notion of .. correlation is again insufficient. We need to be able to specify how, in each particular case, a true statement ·can be uttered. That is to say we need a method fo~ correlating sentences with s\ates of affairs in such a way as to yield a true statement every time. the notion of correlation alone does not equip us to qo this. Strawson formulates the formal semanticists dilemma at this point like this, nhe sees that he cannot stop now with the idea qf truth. ·That idea leads straight to the idea of what is said. The ~ontent of what is said, when utterances are made; and that in turn to the question of what is being done when utterances are maoen, viz. to the idea of the communication o~ some belief (in the case of assertions which we are taking as our model case). The only way out for the formal semanticists now would seem to be to give an account of truth conditions which involve the notion of an assertion, since correlation alone will not do. In other words he must use the idea that a true assertion is made when things· are as one states them to be. He must go along with the theorists of communication intention in accounting for assertion in terms of belief, saying that an assertion is made when a belief is expressed. But he must stop short of saying that a belief-expression need in any way be linked with. the intention ::to communicate the belief o And he might justify his stopping short here on the grounds that an audience directed intention to communicate a belief is merely an appendage to the expression of belief; what we do WITH the intention to communi~ate, we can do without such an - 82 intention. But these grounds are unsatisfactoryo If there is a concept of belief expression independent of communication intention then we must be able to give an account of it, and how are we to give such an account when the expression may not be considered as having any communication goal? The striking lacuna in the formal semanticists theory becomes evident when we cons.ider that asking, "What is Jack trying to say?" is another way of asking "What does he rneari? 11which_ is another way of asking "What is he trying to tell us? 11 Saying something is different from uttering a series of· signs, which series might or might not be subsumable under a set of I rules. Which brings us to a second, extremely closely re-lated point, namely that the behaviour we are dealing with should be capable of being formalized in terms of rules, and these must be rules for the expression of something. What · the utterer does must be recognisable as an expression of belief if we are going to be able to ascribe to him the practice of following the rules for the expression of belief i.e. for making statements. But the formal semanticists theory makes it impossible for us to have the notion.of belief expression since there is nothing in terms of which we can say that the utterer intends to express belief. To take Strawson•s example: Say a man hoots whenever the sun rises. We can say that his practice might be systematically be re-lated to his belief that it is rising (compare the case of a dog salivating when food is presented to it) but we could not say that he was expressing his belief. He might simply salute the rising sun in this way - why he does what he does we don't and can't know. Of course we might answer why having interpreted the question as one requiring an explana-nation similar to the kind of explanation a zoologist would give to a dog salivating, or that a botanist would give for the behaviour of plants with respect to sunlight. Such an explanation would have nothing to do with a concept of lingu-istic practice - language is not a pattern of organismic .. - 83 -responses to stimuli. This is what I think Rhees is driving at in a rather mysterious sounding way, when he writes of the people using Wittgenstein•s primitive language-games, "Those people are not just going·through a complicated trick;· what they say depends on what they find. ·They are not just carry-ing out orders. They use the expressions they do because they have something to say and because that useis understood by all parties. Whereas you may train animals to make the •correct• responses to different words or signs, the animals themselves do not use different words." 10 • When a dog salivates or wags his tail, when he barks or when his hackles rise, he does not know what he means nor does he mean what he says because he does not say anything. If we say that he is telling us something then we say this by construing the be-haviour of the dog on the model of intentional human behaviour and his meaning something by salivating or barking, is only intelligible if we think that he is communicating with us. The question of whether it makes sense to attribute intentions, and by implication communication intentions to a dog, we will deal with later. Now if despite all this we stick to the formal semanticists theory then it must follow that we believe that the rules that are employed in language are private. There must be nothing in the concept of a rule of language which makes it impossible conceptually for a speaker to speak a language which he and only he understands or can understand. Since this is so we must ask as we have asked before in what sense an utterer follows these rules, and in fact we must deny that these rules are linguistic rules at all - language is something more than mere·patterns of regularity which we describe. 10. Rhees, op. cit.\ p.2761 - 84 -C H A P T E R VII SCEPTICAL DOUBTS: THE INNER-OUTER DOUBT In the last chapter I mentioned that the discussion of private language takes place in the Investigations, in the broader context of a discussion of the notion of privacy itself • in particular of the notion that sensations are pri-vate. In order to see the full significance of the treat-ment of private language and in order to get a fuller under-standing of what that treatment is, we must examine what the notion of privacy involve~. More particularly we will con-centrate attention on the question, 11 In what sense are my sensations private?" 1 • From a scrutiny of this question the outlines of its· various senses will emerge as well I hope, as its various non senses. The statement that my sensations are private, when made by someone not doing philosophy (and it is seldom if ever made in that case) is quite prosaic, un-exciting, platitudinous and easily translatable into a sent-ence which makes no use of the word ''Private". Furthermore the sentence makes sense. When it is made philosophically it seems revelatory, profound and slightly puzzling. Un-fortunately it makes no sense or at any rate very odd sense. To pass from non sense to sense in the case of the notion of privacy means to pass from profundity to platitude. This is the price which we must pay for getting rid of philosophical perplexity. From one point of view the investigation "seems only to destroy everything interesting, that is, all that is great and important." 2 • The reward of the investigation is, that it makes it possible for us to stop bumping our heads against the limits of language. In order to get a clear look at the philosophical idea of privacy we must follow the notion to its source in a 1 ~ P .I. 246 2. P.I. 118 - 85 -disquietude. This disquietude is perhaps THE disquietude in philosophy. It is the disquietude produced by doubt. The argument that all sensations are private is used to bolster the argument that there is no way of knowing ·what sensations another has, if he has ani at all. Scepticism with respect to other minds goes hand in hand with the notion of private sensations. things: In this chapter I shall, therefore, try to do· three (a) discuss the character of the philosophical doubt; (b) deal with a specific form of this doubt, ioe. with the doubt about sensations of others; (c) show how this leads to the idea that sensations must be private and show precisely how this notion of privacy differs from, and how it is similar to, the ordinary notion of privacyo A fourth point will I hope emerge out of the other three, viz., the integral connect-ion between the private language argument and the p~ivate sensation argument. This will ··mitigate somewhat the arti-ficiality of my having dealt with private language in a separate chapter. In chapter 4 we discussed the problem of how criteria for Y are related toY. What, we tried to find out, would constitute such an evidence for Y anq how could it come about that any~hing should become an evidence. In chapter 3 the question a1~ose as to how we could ever know that a rule •,•muld be followed currectlyo Would one perhaps need rules for .the use of rules for the use of rules etc.? And again in chapter ~ ~ question came up as to what would constitute a sufficient verification of an empirical assertion, a problem which we had mentioned in chapter 1. The answer to _these questions tended to come in the form of a discussion of lang-uage-games and the way they are embedded in social forms of life. This way of finding a solution might not seem very satisfactory, it has been too briefly and too vaguely stated at this stage even to merit the name of a solution. I shall try to remedy this in a later chapter. For the present I - 8,6 -want to ignore the solution aspect of these problems and concentrate on the problems themselves. They form part of a family of philosophic problems characterised by doubt, we may call them accordingly ''Philosophic doubts." This term indicates that philosophic doubts must be distinguished from the doubts that we ordinarily have. For instance, philosophic doubts seem, when we are not doing philosophy, to be pointless. I may doubt, whether the· creature I have glimpsed in the fading bushveld light was a lion or a cheetah - this is a quite natural doubt. But if I went into a zoo at noon and doubted whether what I was seeing was a lion then there would be grounds for thinking that I might be in the grip of a philosophical doubt. And here it is impossible not to invoke the name of Descartes and with the name the spectre of those amazing doubts which he sets out so methodically in his writings. Take this passage from the Discourse: 11 (A)s I then desired to give my attention solely to the search· after truth, I thought ••••• that I ought tore-ject as absolutely false all opinions in regard to which I could suppose the least ground for doubt, in order to ascer-tain whether after that there remained aught in my belief that was wholly indubitable. Accordingly, seeing that our senses sometimes deceive us, I was willing to suppose that there existed nothing really such as they presented to us; and because some men err in reasoning, and fall into paralo-gisms even on the simplest matters of geometry, I, convinced that I was as open to error as any other, rejected as false all the reasonings I had hitherto taken as demonstrations; and finally when I consider that the very same thoughts which we experience when we are awake may also be experienced when I we are asleep, while there is at that time not one of them true I supposed that all the objects" that had ever entered into my mind when awake, had in them no more truth than the illusions of my dreams." 3 • 3. Descartes, "Discourse on Method." - 87 -For our purposes this quotation is interesting in two respectso For one thing it is the preface to Descartes' theory of the Cogitatio, a sense datum theory which is basic, 4. . h f . as Kenny has shown to the Cartes1an t eory o pr1vacy. Secondly, and this is the asperit I·want to concentrate on now, the doubts which Descartes expresses in the passage quoted, are such as no one in their right mind would, in the ordinary course of events, entertain. And yet in philosophy they came to be accepted as perfectly understandable doubts - doubts all of which not every philosopher actually entertained or thought he should entertain~ but doubts which, nevertheless did not seem out of placeo This is so much the case that people who are not doing philosophy often tend to think (quite rightly) that philosophers spend most of their time wondering whether 11the table is really there". And this has become something of a standing joke about philosophyo So one of the characteristics of the philosophic doubt is that it appears to question things which it ordinarily makes no sense to question. It, as it were, holds up the process of learning and of knowledge, by putting artificial barriers in the way. It is almost as if the philosopher does not know what it means to know, and therefo·re by implication, what it means to learn anything. Wittgenstein gives the example of a school boy who holds up the history class with doubts as to whether the earth has existed for more than a hundred years. 11The teacher", he says "would feel that this was only holding· them up, that this way· the pupil would only get stuck and make no progress. - And he would be right. It would be as if someone were looking for some object in a room; he opens the drawer and does not see it there; then he closes it again, waits, and opens it once more to see if perhaps it is not there now, and keeps on like that. He has not learned to look for thingso And in the same way this pupil has not learned how to ask questions. He has not learned the game we are trying to teach himo" 5 • When once the teacher gets 4. Kenny, 11Cartesian Privacy" Pitcher. 5. o.c. 315 .. - 88 -the pupil to stop asking these questions, to stop having these doubts, and makes him see that they are pointless and obstructive, then he has also put him in a position to understand what learn-ing and knowing is •. The pupil can then continue or begin his course of education. The description which Wittgenstein gives is strongly reminiscent of Freud's description of the situatio~ which often occurs during psychoanalytic therapy in which the p~tient ob-structs the course of an apparently smooth-running analysis by refusing to believe what he is telling the analyst: 11Then we discover that the resistance has fallen back on the doubts characteristic of the obsessional neurosis and is holding it successfully at bay from this vantage point." 6 • In a similar way the philosopher obstructs the course of knowledge by questioning and in fact doubting whether we know what we all know we know. It is interesting to notice how Descartes al-most exactly reproduces Freud and Wittgenstein's portraits in his own actions. In part three of the Discourse he describes how he follows certain opinions and maxims in his daily life while suspending his judgment as to their truth. As soon as he begins to meditate, however, his normal practices including the acquisition of scientific knowledge comes to a halt, ob-structed by the doubt. What I have been saying here clearly takes its in-spiration from Wisdom, and ultimately from Wittgenstein. Wisdom points out both the similarities and differences between the purely psychogenic case of neurosis and the case of philo-sophical perplexity. The philosopher like the neurotic clings to the problem or disquietude which is disturbing him so that, 11in spite of his evident unhappiness and desire to come from hesitation to decision he also desires the discussion never to end and dreads its ending." 7 • But unlike the neurotic what the philosopher says about· the cause of his doubts and problems 6. Freud, "Twenty Eight Lectures on Psycho-Analysis". page 244., ·7. Wisdom,. uphilosophy and Psycho-Analysisn, page 172. - 89 -does not so much have the savqur of an empty rationalisation. Instead he gives something very much more like reasons, and impressive reasons at that. Only, the reasons he gives for holding a position can be countered by equally convincing reasons against his holding it. His position is thus para-doxical - now one horn of the dilemma appeals to him, and now the other. Like Buridan's ass he stands between two equally attractive alternatives and just because of this he cannot move at all. He becomes a victim of what Wittgenstein·so aptly called "mental cramp." The paradoxes of philosophy· are, Wisdom argues, 8• rather more like riddle problems than like mathematical problems. Their solution lies in "seeing and then forgetting about the whole· story. good model of a philosophical perplexity is following problem: the point", grinning I think· a fairly provided by the Before setting out on a flight from Jan Smuts all the passengers take out life insurance policies with an inter-national insurance company. The aircraft trag·ically crashes exactly on the border between South Africa and Botswana, so that the tail part falls into South African territory and the front section falls into Botswanan territory. Must the South African or the Botswanan branch of the company pay out the survivors? A decision seems to be called for and yet any reasons one could advance for the answer that the South African branch o must pay, could equally be advanced for the Botswanan branch having to pay. But of course the dilemma only remains for as long as one does not notice that the problem is a joke; for no one has to pay the survivors anything since they took out life insurance policies and since they have survived. This example is slightly off centre, although it is so in quite an illuminating way. There is a· fact in the story 8. Wisdom, op. cit. pp.l76-7 , - 90 -itself which, when noted, enables us to see that no decision is called for and that the problem is a trick. In a philo-sophical problem, the doubt as to which answer should be given cannot be resolved by noticing any stated fact internal to the problem itself. So it is not so much as though we overlook one vital sentence or phrase in the statement of the problem, as that we overlook the jokelike character of the problem as such. All analogous examples of the philosophical problem limp in one or another important respect it seems to me and the statements of the problems themselves; in the long run, best illustrate their own peculiar character. The doubt about the existence of other minds, like most philosophic do~bts is criteriological in character. It is a doubt as to how from the knowledge that X I can know that Y: 11Do I have sufficient -can I ever have sufficient, criteria .. for Y, short of being given Y itself? 11 11What constitutes my having sufficient criteria?" "How can I· make the leap from the criteria of Y to Y? 119 •In the case of sensations: Given thar. Jack is groaning,' writhing, etc. - are these not after all t r>e outward manifestations of what I HYPOTHESISE to be some inward state of Jack's?" As far .as the other-minds empiricist is con-cerned the evidence for Jack's being in pain ioe. for his in-ward state, is always some behaviour which is in principle observable by someone else. Now as Paul Ziff points out in his short article "About Behaviourism" 10• a phrase, 11you can in principle find· oritn does not have· a very clear meaningo . I will attempt a few· translations of it which more or less bring out what I think those who use the phrase mean by it. "You can in principle find out" - 11it is not inconceivable that you should find out" or "it is no't logically impossible that you should find out1• or "it is not self contradictory to say that you can find out" or· "it makes sense (it is not absurd) to say that you can find out·." 9. Wisdom, "Other Minds," pp.l-2 (notes) 10. Ziff, "Apout Behaviou;rism", Analysis XVIII (1957-58) ;reprinted in The Ph;ilosophy of Mind" ed. by V.Co Chappell, page 148. - 91 -The implication of this sort of talk is that in principle you CANNOT directly find out what Jack's INWARD state is. Our knowledge of one another's sensations is mediated by our knowledge of one another's behaviour and the situation is conceived to be something like that in which an uncle jingles a few coins in his fist and .then asks his nieces and nephews to guess what coins are there. But this fist cannot only not be prised open by niepes and nephews, it cannot be opened by all the energy in the uni-verse, it is kept clenched by the muscles and sinews of · 1 . 1 . . 11 • s h h .ld 1 og1ca necess1ty. o t e c 1 ren can on y guess at I the number and the value of the coins but they can never KNOW unless their uncle tells themo Their guesses might become more and more informed as they play the game more and more often, but these guesses will only be based on in-ductions from past jingling of the coins and subsequent confirmations or corrections of guesses. For some sceptics this analogy is more strictly· correct than for otherso But all protagonists of scepticism with respect to a ~owledge of another's mind would agree on the logical impossibility of the contents of another's mind ever becoming patent to anyone else. This of course raises the problem of how the mental objects ever got into the mind in the first place (the uncle must at some stage take the coins from his pocket)o This is an objection which I will not pursue. It has traditionally been met by say that the sensations ioe. the mental contents are the effects of the ~xterior causeso The welter of confusions and difficulties which this answer raises, and the various critiques and sophistications of it by empiricists them-selves, I also shall not deal witho The main point is that the contents of other minds are hidden from me and can only be inferred by me from behaviour. Where the empiricist's argument leads into scepti-cism is just the point at which the analogy with the uncle's 11. See Wi ttgenstein on 11the hardness of t_he logical must" ·· in F .M. 113-141 ... 92·-logically tight fist can be made to break down. We said that the children could never know the number of coins un-less he told them and just sofone may think that the really only satisfactory way to know the mind of another is through what she tells us. But of course the radical philosophic doubt cannot stop sh~t at a verbal report and there is no good reason why it should. It is quite natural for the philosopher, hell bent for certaint.ies, to ask, "How do I know that she means what I mean by pain, ache, tingle, warm, blue, black, green, etc.?" or "True he CALLS it red, but how do I know he does not see· it as green?" These and a host of similar doubts begin to crowd in until the gentle but per-sistent question can no longer be drowned out or circumvented: "How do I know that there A!S any other minds?" And with _that question as Joh~ Wisdom so poignantly puts it we, 11 1 look down that lonely road• which leads past abandoned illusions to the security of Solipsism". 12 • Clearly.there is no valid case to'be made for an ·induction from a set of behaviours, where "behaviouru is used in the very broadest sense,to mean observable movements, to the existence of a mind. What sort of probability could these observations yield if,in principle there is no possi• bility of an independent confirmation of the existence of another•~ mind? 13 • To the sceptic there can be no reason for thinldllg that what he calls 11another person" is anything more than a very co~plicated machine which responds in ways more or less predictable to various stimuli. For instance, when he observes a human organism responding to a stimulus such as a low temperature by shivering and saying, "it is freezing", this response is strictly on a par with the flash-ing of a· light on·a car dashboard indicating that the radiator is about to boil over. And just as it is possible ··~to open the car bonnet and measure the temperature of the cooling system with a thermometer so it is, in theory, possible to 12. Wisdom, "Other· Minds,np.l41 13. See Malcplm "Knowledge of Other Minds", Pitcher p.371 - 93 -correlate the verbal and non verbal behaviour of the human organism with some state possibly a brain state, of tha.t human organism. Now while this correlation of the sensa• tion responses of a human organism with a brain process might de facto be possible 14 • the firing of synapses the passage of neurons, the messages from the afferent and efferent nerves are not what we ordinarily refer to when we say of someone that she is in pain. We do not use the word •tpainn like this. . Wittgenstein puta it this way: 11Could a machine ttlink? - <;ould it be in pain? -Well, is the hqrnan body to b~ called such a machine? It surely comes as close as pos~ible to being such a machine. But a machine surely cannot think\ • Is that an empirical statement? No. We only say of a·hu~an being and what is like one that it thinks. W~ also S&Y it of dolls and no doubt of spirits too. Look at the word 11to think" as a tool." 15 • And in the Philosophical Investigations Part 2 he elaborat~s on the attitude we take up when we use sensation words about others, such as "Paddy is in pain" and writes: "Suppose I say of a friend: • He is not an automaton• • -What information is conveyed by this; and to whom would it be information? To a HUMAN BEING who meets him in ordinary circumstances? What information COULD it give him? (at the very most that this man .always behaved like a human being and not occasionally like a thing.) •I believe that he is not an automaton,• just like that, so far makes no sense. My attitude towards him is an \attitude towards· a soul. I am not of the OPINION that he has a soul "• 16• I shall go on to deal with Wittgenstein•s position on this problem in the next chapter and I merely intersperse 14. See·u.T. Place 11 Is Consciousness a Brain Process" reprinted rrpm The British Journal of Psychology in Chappel p.lol. 15. P.I. 359-60 16. P.I. page 178 - 94 -these two remarks of his to show what the argument of the empiricist driven to scepticism comes to - how odd it can be made to sound. But if the mind and more particularly the sensations of others are hypothetical objects the existence of which it is logically impossible for me to verify, then by that very fact, I make no real hypothesis. What I assert about another's pain can be neither proved true nor false and so, to put it strongly, I assert nothing, or to put it less strongly, I say something which we would nor ordinarily des-cribe as an assertion. From where then do we derive the concepts of mind, of consciousness, of sensation? And now the sceptic is ready with an answer which is both a solution to that last problem and, I think, the cause of his scepticism. It is in a sense, the alpha and omega of sceptical doubt about other minds. He says that he does and can only know about 'these things FROM HIS OWN CASE. This is not only true, he argues, but necessarily true. The only sensations he can know are his own, and what is more, he is the only one who can know them. He can hypo-thesise about the behaviour of others that they have THE SAME as he has when he is in pain, but he can have no direct awareness of their pain.· He can perceive their behaviour but in his own case he can CONCENTRATE HIS ATTENTION INWARDS and directly and immediately know that he is in pain. His sensations are private in the sense that only he can have THEM. When these things have been said then any'resemblance between the ordinary (though infrequent) statement that sensa-tions are private and the philosophical statement disappears. Imagine a doctor walking into his waiting rooms and saying, ·uwell, whose aches and pains want attention first?" at which t . Robinson despe,rate for a pain killer. groans, 11Mine1 11 - That - 95 -is a perfectly ordinary use of the possessive adjective: his pain will not be alleviated by the doctor treating Smith or Jones, and because his pain is severe he wants it attended to before the doctor attends to the alleviation of Smith and Jones• pain. In this rather trivial sense Robinson's pain is private. He can identify himself as the one who·is in pain, and the relief of his pain is not the relief of anyone else•s. (Even this last point has to be very carefully watched} How would this example be handled by the sceptic. He would have to say that when Robinson said that his pain was most deserving of treatment, then-the doctor, who proceded to treat him was trying to eliminate a certain behaviour in Robinson, viz. his groaning, saying that he is in pain, call-ing for an analgesic, etc. If he was a behaviourist then he would say that third person descriptions of pain were descrip-tions of certain measurable behaviours, and in this he would be correct to a certain and rather inisleading extent. But he cannot be content to rest here, ·for the problem of the first person statement has to be.met with by him. What, he must ask, is the meaning of Robinson's description to Robinson, for clearly he does not decide on the basis of an observation of his own behaviour that he is in pain. It must be that the first person sensation statement is a description of Robinson's private inward state. So, when Robinson uses the word 'pain' of himself he uses a name which refers to a private object the existence and character of which only he can know. Of course here one wants to ask how someone else could know that Robinson was describing any inward; private state, and as usual the question drives the sceptic into solipsism. He then grants that he can1 t KNOW that Robinson describes his in-ward state, or for that-matter that he has any inward state to describe, strictly all he can say is that this is what HE HIMSELF does. When he turns his attention inwards he recog-nises something which he knows to be pain. Although the que~?tion is premature (I want to reserve - 96 -any criticism until the next chapter) one wants to ask the sceptical philosopher what this "something" is that he recognises. "If you say that he sees a private picture before him which he is describing, you have still made an assumption about what he has before him. And that means that you can describe it or do describe it more closely. If you admit that you haven'.t any notion what kind of thing it might be that you have·before him- then what leads you into saying in spite of that, that he has something before him? Isn't it as· if I were to say of someone: 'He HAS something. But I don't know if its money, or debts, ·or an empty till.'n 17 • The reply to this is usually, that we do know what our own private pain is like and can describe it to our own satisfaction although this description is in a language as private as that which we described in chapter 5. In fact, he argues, this knowledge is one of the best examples of knowledge which we have, for here, when I say I know I am in pain then I can't be wrong. I make an identification of my inner state to which I attach a name·- "pain," 11hot 11 , "itchy 11 , etc. These names when used about another are descriptions · of overt behaviouz:, used o£ myself they are names in a private language. Whether they mean for others what they m,ean for me I can never know. And in an attempt to be fully consistent the privacy argument has been extended by some philosophers, Hume for example, to cover descriptions o£ the external world including those very descriptions of the behaviour of others. For when I have an experience of the blue sky there is no way for me to know whether any other organism has the same ex-perience which I have or any experience at all. The language I use about the noutside world" is as private as that which I use to describe my sensations.· This, it goes without saying is an extremely "philosophical" idea. Wittgenstein both sketches its out"lines very vividly and captures its atmosphere in these paragraphs from the Investigations: 17. P.I. 294 - 97 -11Look at the blue of the sky and say to yourself 'How blue the sky isl 1 - When you do it spontaneously -without philosophical intentions - the idea never crosses your mind that this impression of colour belongs only to you. And you have no hesitation in exclaiming that to someone else. And if you point at anything as you say the words you point at the sky. I am saying: you have not the feeling of point-ing-into-yourself, whicq accompani~s 'namin9 the sensation•· when one is thinking abQ4t 1 •pr:j.va~~ :).angua.g'P·', Nor do you . . think that really you 01.Jght ·pot to point to the colour with your hand, b4t with .your attention. (Consider what it means •to point to something with the attention•.) 11But don 1 t we at least MEAN something quite definite when we look at a colour and name our colou·r-impression? It is as if we detached the colour-IMPRESSION from the object, like a membrane. {This ought to arouse our suspicions) •o••• it is easier to produce'this experience when one is looking at a bright colour, or a:t an impressive colour~scheme." 18• Of particular significance in this passage is his reference to "naming the sensation" which takes place in private language. This idea marks the reappearance of that atomistic theory of language which he deals with right at the outset of the Investigations. (see Cho 1) Accordi"ng to this theory the meanings of my words are the objects for which these words stand. Only propositions have sense, and there is only one way ~n which a proposition can be taken; the elements of a proposition. must be correlated with the world in such a way that there is a one~to-one correspondence be-tween those elements (names) and the objects ·which occur in the fact (the state of affairs) which the proposition depicts. There is only one way in which a proposition can be projected onto a state of affairs if it is to be meaningful, and that way is such that the names in the proposition must stand for objects: "A name means an object. The object is its . . ).9. mean~ng.n 1a. P.r. 275-6 19. Tractatus, 3.203 - 98 -No cognisance is taken here of the way in which the words are used. If the objects are private then the meanings of the words are private. The connection between my words and the world will be private if I construe the world qS somehow inside me i.e. as an internal·possession. But to use Wittgenstein's favourite way of putting I it there is no way in which this connection can be set up. And this statement, to use another typically Wittgensteinian expression does not refer to an empirical difficulty but to a grammatical impossibility viz. to the fact that there can be no intelligibly formulable rule for the correlation of names with private sensation and to the further fact that the concept of a private sensation is not intelligible. There may seem to the philosopher in the grip of a disquietude 'to be a gap bet~een my words and the mental objects to which the~ are supposed by the,private language argument to refer. But as Melden remarks in another context "•••• the appearance of a gap is symptomatic of confusion." 20 •· And in this case the conceptual confusion results from mixing up the categories of physical object.s and sensations. This confusion makes it seem as though there is an unbr.idgable gap between the names I .use for sensations and the sensations themselves (a quasi physical gap). Of course there does•nt appear to be a gap to the uncritical exponent of the private language argument, but there is a paradoxical position in the philosophical twilight between acceptance and rejection of that argument, in which the attempt to use language to describe private mental objects does seem unsatisfactory, - there IS a gap between the name of a pain and the painful sensation, but \ the nature of thegap and why it is there is not clear. 20. Melden, "Free Action", P• 57 .. 99 -CHAPTER VIII SCEPTICAL DOUBTS: WITTGENSTEIN 1 S "TREATMENT" I have attempted to outline the argument that sensations must be private without saying what is objection-able about it. Apart from one or two quotations from the Investigations which put the argument in a rather odd l.ight I have given no systematic critique of it as such. The critique of the private language argument, though closely connected with the problem of private sensations, will not entirely serve as therapy ·for the latter. In fact all the compelling force of the idea of a private language comes back into play· when the ground is switched from the rather abstract discussion of rule-following, speech acts and ling-uistic meaning to sensations and the language we use in connection with them. So long as the idea of private sensations has a hold on us we will never really rid ourselves of the hankering to propose that sensation language must in some way be private. When we come to think about sensations. then we are shocked back into a way of thinking about language, which, to use Wisdom's Freudian idiom,we unsuccessfully re-pressed when we countered the private language argument. There still lingers in our philosophical unconscious a desire to resurrect private language, and the notion that sensations must be private serves as an opportunity for bringing the old arguments into the light. ttAfter alltt, we want to say, npeter·knows his pain in a way which Susan does 1 nt simply because Peter is Peter and Susan isn• t Peter. ·And of course \ . the converse applies which only makes the gulf between his and her pain the more obvious.n If we have thought quite hard about the problem ·and have read the current literature on it we will even dismiss w.i th sophisticated scorn the naive idea that the barriers to Susan knowing Peter's pain are ;physical or even psychological. Ayer, for example, referring to Stace•s empiricist arguments in favour of privacy says that ' - 100 -his criticism of Stace•s propositions, 11 •••• is not that they are false, or even dubious, but that they are trivial. For what is it after all, that prevents one person from having the experiences of another? •••• This is not a case of physical incapacity, like my inability to see through a brick wall, or of a psychological incapacity, such as my inability to remember the events of my childhood. The barriers that prevent us from enjoying one anothers experiences are not natural but logical." 1 • There is something peculiarly subtle and at the same time particularly dangerous about this suggestion. It repre-sents what might be termed a false. advance on the problem. As though one had gone through all the motions of taking a great stride forward while being unaware that one was trying to ascend a down-going escalator.· I shall try to show that Ayer and those who argue like him have not really freed ·themselves from the power of the physical metaphor (which clearly has a strong hold on Russell); that had they done so they would not speak of a nlogical barrier u preventing us from· enjoying one anothers sensations. Instead they would have investigated the grammar of our sensation language~ as is actually used and dis-covered in that the sources of confusion which lead to the privacy argument, for it is an argument which Wittgenstein has shown, I think, to be irremediably confused. It is permeated by analogies and metaphors which often cease to be what they were intended to be and become instead the very basis of the argument. And the difficulties of dealing critically with these sorts of argument are increased by the fact that they DO give.important insight into the nature of our thinking, feeling, talking and the things we think, talk about. and feel. This means that to dismiss them for the wrong reasons is· as bad and possibly worse than accepting them. The argument with which Wittgenstein deals in the Investigations is well summari~ed by Cook in the following 1~ Ayer, "The Foundations of Empirical Knowledge", page 138 - 101 -three premi'ses: (Pi) · No one can feel (experience, be acquainted with) another personJs sensations. ; (Pii) The proper ·and necessary means of coming to know what sensations another person is having is to feel that person's sensations. (Piii) Anyone who has a sensation knows that. he has it because.he feels it, and whatever can be known to exist by being felt cannot be known (in the same sense of "known") Conclusion: to exist in any other way. No one can know what sensations another person . h . 2. l.S avl.ng. These premises are countered by Wittgenstein in the much quoted section 246 of the Investigations: 11 In what sense are my sensations private? - Well, I can know whether I am really in pain; another person can only surmise it. - In one way this is wrong, and in another nonsense. If we are using the word •to know• as it is normally used (and how else are we to use it?) then other people very often know when I am in pain. - Yes, but all the same not with the certainty with wh~ch I know it myselfl - It can't be said of me at all (except perhaps a:s a joke) that I KNOW I am in pain. ' What is it su~posed to mean - except perhaps that I AM in pain? 110ther people cannot be s8:id to ~earn of my sensa-tions ONLY from my behaviour, - for I cannot be said to learn of them.. I HAVE them. "The truth is: it makes sense to say about other people that· they doubt whether I am in pain; but not to say it about myself." 2. Cook, 11Wittgenstein on Privacy" in Pitcher pages 289-90 - 102 -This then is the crux of Wittgenstein•s position. Where 11 I known is used·as an expression of the fact that in-formation has·been acquired (and not as an expression of ex-asperation, 11Damn it I know I've got a headache but I can't go running off to the·doctor for that") then it makes sense to say, (i) that I came to know throu·gh the exercise of • certain perceptual and cognitive faculties; in a good position to experience the object (iii) that I have learned to recognize this (iv) that I could on this occasion correctly (ii) that I was. of knowledge; .. object as an ••• ; identify this as an •••• These instances recall the variety ?f meaningful ' answers which Austin suggests can be given to the question "How do you know?" In his example the question put is, 11How do you know there• s a bittern at the bottom of the garden?" and he lists.the possible answers as: (a) I was brought up in the fens. (b) I heard it. (c) The keeper reported it. (d) By its booming. (e) From the booming noise. (f) Because it is booming. 3. I would suggest a further retort (g) which emphasises (i) above, namely (g) I've got eyes in my head haven~t I? · The first three answers and (g) say how one is in a position to know and the last three say how one can tell at the time i.e. how one justifies the identification of this· particular instance as an instance of X. But the central point to notice, it seems to me, is that Austin•s philosophic and verbal acumen can only be exer-• cised here because there is a primitive sense of 11to know 11 which implies, •'Where I know I can be mistaken, 11 or better •'Where I thought (said) I knew, I could have been mistakeno 11 3. . Austin, "Other Minds" in Philosophical Papers, page 79. - 103 -Just because of this it makes sense'to justify that one knows in ways (a) to (g) above.. I would not go so far as to say that where "I know" means "I can 1 t be wrong" then it is an incorrect employment .of "known, but it is, at any rate, a use carefully to be distinguished· from the primitive one. Where "I known means 11 I can't be wrong" then there is clearly nothing like an· objective test of whether I DO know: I say I know and in this case rule out the possibility of a falsification and therefore of verification. This use of "I known occurs ~n Moore's "Proof of an External World", and·wittgenstein gives a most brilliant illumination to this· use in ~'On Certainty" where he says: "Moore's view really comes down to this: the concept 1 know 1 is analogous to the concept~ 1 believe 1 ~ •surmise,• •-doubt•-r •be convinced•, in that the-statement 1 I know ••• 1 canf't be a. mistake. And if this IS so, then there can be an inference from such an utterance to the truth of an assertion. And here the form 1 I thought I knew' is being overlooked. -But if this latter ··is inadmissible, --then a mistake in the asser.tion must be logically impossible too. And anyone who is acquainted with the language-game must realize this -an assurance from a reliable man that he knows cannot con-tribute anything.'-' 4 • In the game which Moore is playing with "I know" there is no move which is discovering that he is mistaken,· and so far his use of 11 I know" is something in the nature of an emphasis. The case is somewhat similar with 11 I know" used in first person sensation statements. Here there is no room for "I ·doubt". Doubt makes no sense here and neither therefore \ does knowledge in the ordinary use of that word. It is just this which makes (Piii) unsatisfactory, and which destroys the whole edi£ace of sceptical argument against knowledge of another's sensations. What am I adding when I preface 11 I am in pain" with "I know", .. nothing. Pain-is neither discovered 4. o.c. 21 - 104 -nor observed by the one who is in pain. He may well observe how long it lasts, remember that it was worse last time, notice the ravages it produces in his constitution, but none of these constitute "knowing painu and none of them could be-cause, as far as the' sufferer is concerned, that latter con-cept makes no sense. If I can identify my pain then I can also misidentify it, but neither concept makes sense in the first person language-game of pain. Of course this does not mean that self-diagnosis is infallible, but then diagnosis does not. consist in identifying pain but in tracing its causes. I might well diagnose a pain in the ear as ear-ache and mean by this .that there is something wrong with my ear when in fact the pain is alleviated by treatment of an infected molar.. However, it is quite absurd to imagine that I make any sort of identification error in my ·first person pain language. And this is not because I am in a very good position to judge, or because I have superbly sharp and re-liable introspective faculties, but because WHAT I SAY is the criterion of the intensity, nature, duration and location of my pain. Pain language in t~e first person is not descriptive in character. When I say, "I've a stabbing tooth ache 11 I am in no way identifying, naming and describing anything. · I am, Wittgenstein maintains, replacing a primitive pain behaviour, such as crying and clutching my cheek, with a new, verbal, be-haviour which expresses my pain. 5 • And this ~hould not be taken ~:to mean that ttpainn means crying, since it does not describe it but replaces· it. Nor should it be taken that the behaviour is the pain, as certain sorts of behaviourists, bent on reductionism, might argue. But why should it not be .taken in that way? Can there be any safe passage between the Scylla of private pain and the Charybdis of mechanistic reductionism? When a physicist says that terrestial bodies have weight, or fall because of gravity they use these terms to s. P.r. 244 - 105 -describe the behaviour of bodies near the earth e.g. that the acceleration of their fall is 9,8 m: sec2• When we say of someone that he is in pain why should we not mean that he is behaving in a certain way. What.I shall call the strong behaviourist might answer in this way: When we say of someone that he is in pain we mean no more than that he exhibits specified behaviours which are in principle publicly observable. The supposition that ''Pain" refers to a mental process or event, or to the presence in the person of a mental object, is logically im-possible to verify and is, therefore, unintelligible as an hypothesis~ Among the behaviours which form the class "pain behaviour" are the verbal behaviours in which the person says "I am in pain" etc. The latter is not a report on a mental state (and here they agree with Wittgenstein). Furthermore our understanding of :"pain" is neither dependent on our having had pain, in the sense of being in a certain mental condition, since this is unintelligible, nor is it dependent on our ever having exhibited any pain behaviour ourselves, any more than our understanding of the behaviour of a billiard ball is dependent on our having rolled around a table. Pain simply means a re-sponse of a certain type to a stimulus. The mistake in this theory is that it ignores the fact that one who has never been in pain does NOT understand the statement "I am in pain" in exactly the way that one who does feel pain· does. This· might seem ·to beg the question since the strong behaviourist denies that feeling pain means anything more than exhibiting pain behaviour, which can include among other things electro-~hemical and neurological behaviour in the nerves and brain. It is not question-begging however. This criticism of strong behaviourism merely points to the inability of that theory to adequately distinguish between first person and third person pain language. When I say "I am in painu I am not describing a mental event nor am I des-cribing my· behaviour. There are cases in which one could - 106 -imagine "I am in pain" being an intelligible description of his behaviour by one who utters the sentence. Say a drama teacher walks into a room in which a student is re-hearsing Claudius• death agonies in Hamleto He stops the scene and says to·the student, 11What in God's name are you doing?n arrl receives as an explanation the offended reply, "I am in pain", meaning, nr am portraying Claudius• pain ••• am behaving in a way appropriate to an imitation of someone's death agonies." Here the surface grammar is identical with the grammar of· the statement uttered by someone who is in pain, but that is where the similarities end. The function of the sentence in the latter case is expressive not des-.. criptive, and as far as the depth grammar goes the 'former use is more correctly assimilable to the third person statement, "He is in pain." But how far does this take us? Left at that, not very far. What I am presenting as the Wi ttgensteinian a::tter-native to behaviourism is still open to objections both fro@' behaviourists and mentalists. When Wittgenstein says that first person pain language is expressive the behaviourists counter with, "Well, if you say it is expressive then the verbal behaviour which consists in saying, 'I am in pain•· must be taken by you as the expression of something else viz. ·the presence of some mental object, process or event. And the private-mental-sens~tion faction will say the words which Wittgenstein puts into their mouths: '''But you will surely admit that there is a difference between ··pain-behaviour accompanied by pain and pain-behaviour without pain?• -Admit it? What greater difference could there be? - And yet you again and again reach the conclusion that the sensation is a nothing." 6 • ••••"Are you not really a behaviourist in disguise?" 7 • · Now is the time to take stock of the position. I said earlier that we were trying to steer between behaviourism 6~ P.r. 304 7. P.r. 307 - 107 -and the theory that sensations are private affairs ~n t~e mind. In fact putting it like that makes the difference between them appe,ar too great. Both positions suffer from too restricted a diet. Both have the idee fixe that either one denies sensations or one thinks of them as etherial things existing in the medium of a mind. But as Wittgenstein might have said, the first mistake is to ask the question as to what sort of things sensations are, and the second is to try to answer it. His actual remark about pain is, "It· i.s not a SOMETHING, but not a NOTHING eitherl The conclusion TNas only that a nothing would serve just as well as a something about which nothing could be said. We have only rejected the grammar which tries .to force itself upon us here". 8 • This grammar ·suggests to us that pain must be a something (another entity) accompanying pain-behaviour. If we have thisidea then the rejection of the behaviourists in the last paragraph but one, in wh ic_h it was said that someone A who has never himself had pain, ·will not understand another B saying 1 I am in pain', in the same way as someone C who has felt pain, will be misunderstood by us. We will take it that the difference lies in A's having something which B has not got and never had? but which-c has got or has had. To obviate this misunder-standing we must reiterate that pain is not something. When we say, "his pain" or "her pain" we are not using the possessive pronoun in the way it is used when we say "his hat 11 or 11her room"• It is just t:his which, if ignored·, leads philosophers to say either that pain IS behaviour, or that pain is a logically private object. When it is said, as Ayer does, that it is logically impossible literally to have another's pain, the grammatical confusion about the way the genitive··is used is clearly in evidence. Cook argues quite correctly that to say what Ayer says is equivalent to saying that the SENSE of 11 I have her painn is SENSELESS. The meaning of 11the literal sense" is usual.ly explained he says through, "presenting the parts of a. P.I. 304. See also P.I. 293 - "the beetle in the box" example. - 108 -the sentence (either words or expressions) in som8 fani.liar context in which they have the desired meaning and then specifying that it is when the sentence in question combjr.es the words or expressions as used i:J;'l these contexts that it ~,~,. its literal sense. But what could it mean to speak of trans-ferring a word or expression AND ITS MEANING from a context .in which it has a particular use to a sentence in which it ha.s no use at all (except as a part of speech) - and certainly ne>t the use it had in the context from which it was allegedly transferredn. 9 • In a most effective reductio. ad absurdL.a he shows that s_aying that I logically cannot, literally havE· ) . another's pain, for the reason that they have got somethinq which I haven't, is equivalent to saying that I cannot have my fa.ther' s build o His argument is that if builds are as simi-lated grammaticailyto say overcoats.then it will seem that since it is possible to misidentify the owner of an overcoat, and since.it is impossibl~ to misidentify the owner of a build, builds must be things like overcoats which we possess but unlike overcoats whey must be among the most clearly marked and inviolable forms of private propertyo (This is an extreme compression of his most lucid argument) The obvious absurdity resulting from this assimilation of the grammars o"' possessives (his coat/his build) reduces us to saying that no one can have anyone ~lse•s build. Now what should be strong-ly noted is, that if it is senseless to say that I have anyon2 else•s build then it is equally senseless to say that I have my own build, in which case it is senseless also to say thaT a build is some private possession of mineo This can all be applied mutatis mutandis to 11my· pain" where this is·. taken to mean a ttlogically private" possession of mine. In the +anguage-game played with pain it makes sense to say that I have a pa;in, but not to say that I possess a pain. It makes sense to talk of my pain, her pain and their pain. It also makes sense to talk of her having the same pain as I have, provided this is not taken to mean that 9. Cook, op. cit. page 301. See also Zettel 448. - 109 -she and I possess the same object. Pain, like joy, is not a thing in the sense of an object, which is why Wittgenstein writes in Zettel: " •But I do have a real feeling of joyt' Yes, when you are glad ·you really are glad. And of course·joy is not joyful behaviour, nor yet a feeling round the corners of the mouth and eyes. " ' But rtjoyn surely designates an inward thing (etwas Inneres) .• -· No. 1 Joy1 designates nothing at all. inward nor any outward thing." 10• Neither any The suggestions put .forward here dispose once for all, I think, of the idea that pain is an object and con-sequently of th~ argum~nt that it is a private object. It still remains to deal with the question of whether pain is pain behaviour, since it would seem that having shown pain not to be an object one is committed to the view that it must be a behaviour. And yet as the quotations from Wittgenstein have shown he does not say this. To deal with this problem I should like to turn to ,Wittgenstein•s writings in Zettel, which to date, possibly due to their-comparatively recent publication, have received far less attention than have the Investigations. In this work the character of his philosophical technique is more ex-plicitly stated_than anywhere else. He says for example: "Philosophical investigations: conceptual investigations. The essential thing about metaphysics: it obliterates the distinction between factual and conceptual investigations." 11 • ' As is evident from the way he treats the problems of the philosophy of mirld in this book, he means by "conceptual investigation" an examination of how words are used in ordinary language. It is vital to keep this in mind when dealing with the question of sensations and behaviour. Wittgenstein is not concerned to answer the question, "What is pain?", he 10. Zettel 487 11. z. 458 - 110 -would characteristically follow this with another, ·far more philosophically viable question, "How is the word 1 pain 1 used? 11 Now we have already said· that in' first person state-ments· it is used expressively, and have further argued against the picture bf pain as an inward process or object. If we say in addition that it is not behaviour then the temptation to ask, 11Well, WHAT is it?" becomes almost irresistible. The hardest thing as Wittgenstein says on a number of occasions is to stop asking a particular sort of question, and yet this is what we must do ~f we are to make any progress in clearing up the philosophical problem about the concept of pain. A remark in Zettel about fear (in the section on emotion - predicates which immediately preceeds the remarks on pain-language) is extremely useful in coming to grips with this problem. He says: "It might be asked whether this word would realy relate·simply to behaviour, simply to bodily changes. And this may be denied. There is no future in simplifying the use of this word in this way. It relates to the behaviour under certain circumstances. If we observe these circumstances and that behaviour we say that a man is h 12. or as •••••" This remark drives out the idea that the concept of pain is simply related to bodily movements. Only in appropriate circumstances is human behaviour expressive of pain: "The concept of pain is characterised by its pecu-liar function.in our life". 13 • "Pain has THIS position in our life; has THESE -connections; ·(that is to say: we only call 'paint what has THIS position, THESE connections)" 14 • "Only ··surrounded by certain normal manifestations of life, is there such a thing as an expression of pain. Only surrounded by an even more far reaching manifestation of life such a thing as the expression of sorrow or affection. And so on." 15 • These statements give an important insight into what Wittgenstein means by language-games being part of a form of life, and their I 12o 13. 14. 15. Zo 523 z. 532 Z.533 z.534 - 111 -theme is taken up a few pages further on: 11Being sure-that someone .is in pain, doubting whether he is·, and so on, are s~ many natural, instinctive, kinds of behaviour towards other human beings, and our language is merely an auxiliary to, and further extension of, this relation. game is an extension of primitive behaviouro language-game is behaviour) 1116 ·-Our language-(For our What disquiets us about the statement that "painn means pain-behaviour (and it is a well founded disquietude) is that it is possible to simulate pain behaviour. Some-one can exhibit pain behaviour without being in pain, and yet we do not want to say that in that case there is an in-gredient, pain, lacking, for to say that would be to re-introduce the idle picture of pain and the idea that you can give yourself a private exhibition of it. The quotations from Zettel above imply that the CONTEXT in which behaviour occurs is a vitally important factor in differentiating be-tween someone who is in pain (the victim of a motor smash) and someone simulating pain, but at first sight context may seem relatively unimportant -. everyone can be in pain any-where, anytime, or all the time. But is this so? - Is it conceptually possible? To answer this I. suggest the following case: Let us suppose that the world is a vast stage and that all the people in the world are superlatively good actors, with the exception of one or two who comprise the "audience"._ Suppose that you fall into the latter group. You watch the scenes of life as you watch the scenes of a play. You come upon a motor accident ''where a man is lying. writhing on the ground, moaning and sometimes screaming 110 my 'leg, my leg." Let us assume you are a doctor. oo·you give the man a morphine injection? Well you might, but it would be a mistaken thing to do - one might say that the script didn~. t call for it. No play script calls for the audience to act in the play& But say you did what the script didn'.t · call for and administered a pain-killer, and say the actor 16. z. 545 - 112· -stopped his writhings - then we would also make a mistakeo He would have forgotten that the play must go on. But would he be mistaken in all circumstances? What if the play were performed in order to get you, the audience to do something? ("The Murder of Gonzagon in 11Hamlet") What if the playwright and players intended you to think that the play was reality? In that case the actor would have missed his cue if he did not respond to your treatmento But note also that if he does respond then it begins to be uncertain who is directing the play - the play is ceasing to be a play. from his job as director. The mal genie has been fired If you strike someone and she goes on laughing and chatting as the script demands, then you will know that she is acting. Her behaviour is not normal. The point of this example is to show the primary importance of context for the grammar of pain language. {It has other, _broader implications for our notions of con-cept, grammar and forms of li£e, which will be discussed in Part III). In certain problematic cases only the context can enable us to decide whether someone is in pain. I£ an actor is playing the part of someone who has a headache and he has a very, very bad headache himself then only the further context of his off-stage behaviour will enable us to say of him, "He wasn'.t acting after all." So while it is true that someone can be in pain at any time, the question of how it is to be decided whether he is in pain cannot be answered without reference to context. And further, it is obvious that we couldn't all be in pain all the timeo - It is not conceptually possible. Similarly it is nonsense to say that we could all simulate pain all the time. -~ The language-game of lying, or acting has to be learned, and the preparation for learning it consists partly in acquiring the :concept of pain, ~hich means among·other things, understand-' ing the difference between suffering pain and not suffering it. - 113 -The sceptical argument, whether o£ the 11private sensation" type or o£ the "strong behaviouristu type, turns on the refusal to supply any criterian for the· correct ascrip-tion of pain predicates to a person. The rules o£ ordinary language are abrogated by the sceptic who then finds to his surprise that the game cannot be playedo He is rather like someone who, whenever shown something which is claimed to be a diamond always declares it to be a fake, and who says, even when shown the Cullinan that it is glass. \ . _, - 114 -PART THREE LANGUAGE AS INTENTIONAL ACTION Having tried to show the Wittgensteinia~ technique at work in the field of arguments for and against the privacy of sensation, it is necessary that certain other but related investigations of his should be briefly covered. In the following three chapters I propose to deal with the topics of (a) meaning and intentional action; {b) the concept of a person; and (c) language-games and forms of life. The choice of these topics was guided by a number of considerations. Firstly, they all feature prominently in Wittgenstein 1 s later writingso The investigation of intention forms a large part of the end sections of Investigations Part 1 and re-appears frequently in Part IIo Considerable.attention is also given to intention in Zettel {227ff and 570ft). The im-portance and fertility of these investigations can in part be gauged by the amount of literature on the topic of intention and human action which has appeared since the publication of the Philosophical Investigations. Very much the same applies to his treatment of the concept of a person - and I think here particularly of Strawson 1 s work in this area. Wittgenstein is hard to pin down textually here and the importance of the sub-ject in his thought is more often implicit than explicit, but it does seem prett~ clear that as far as his conceptual studies ' in the philosophy of mind are concerned he held the·investi-gation of the concept of a person to be of very considerable importance. As for the third topic, its importance is apparent even from the most' cursory reading of the later works. Not only is the technique of using language-games as illuminatino examples employed in the Investigations, Zettel and On Certainty, ·but the relationship between these games and what Wittgenstein calls forms of life, is constantly adverted to by him. - 115 -In the second place I have selected these topics because of the close connection which they have with one another. I might for instance have chosen the quite lengthy treatments of colour concepts or emotion-predicates in Zettel, but they do not have the same intimate and fundamentally im-portant relationship with each other or with any of the three chosen topics which the latter have with each other. That human action and the concept of a person are related is ob-vious from the very phrases themselves, but their connection with the third topic may seem at first a little tenuous. I hope to show, however, that the investigation of language-games and the conceptual structure which they imply greatly facilitates the investigation of the concepts dealt with in the other two chapters. Thirdly my choice was dictated.by certain important lacunas in Part 11. To be specific: in Chapter six I spoke constantly of iqtentions in connection with communication in order to suggest an alternative to what Waismann calls the causal interpretation of language, but it was beyond the scope of that chapter to deal with the concept of intention as sucho Consequently the question of the meaning of sentences such as, "I intended he should leave when I said 1 Get out' n, or, 11 I mean THAT table not that one", or "Her words mean· nothing she is just babbling," still remain largely unclarified, and I hope to go some way in remedying this. Another loose end whic~ needs to be tied up is the problem raised by Wellman and Ayer which I mentioned in Chapter five, of where a verification stops, why it must stop s~mewhere, etc. This is part of a more general problem - I shall call it the problem of conceptual limits - which greatly occupied Wittgenstein and which we have mentioned in discussing rules for the use of rules for the use of rules o ••••• , and explanations of explanations •••••• o, and similar problems. This area will be dealt with in discussing language-games and forms of lifeo The chapter on the concept of a person is meant to supplement the discussion of sensation particularly where it touched on the behavioural criteria of - 116 -sensation. It will also be used to give another slant on thP problem of conceptual limits in that it will deal with the question of why we do not predicate psychological terms of amoebae or even of worms, whereas we do use some of them of dogs and employ the full range,of human beings. One further point might be made before I begin: these chapters, like those in the other two parts of the thesis are meant, above everything, to illustrate the strictly-con-ceptual character of Wittgensteinian philosophisingo Toulmin, 1 • in stressing the unity of Wittgenstein's thought, makes this point strongly. He argues, for example, that the transition from the Tractatus to the position in the Investigations was a shift in the method of handling conceptual problems in their connection with language - the shift from the conception of language-as-Gleichnis to the concept of language.:.as-Handlungo In my opinion the radical nature of this split should not be underestimated or clouded by an unbalanced emphasis on the unity of Wittgenstein•s thought, but this much at any rate is clear: in the Tractatus he had already broken with the con-ception of philosophy as a ·supra-scientific discipline which dealt with factual matters of a subtle and rari.fied kind: He writes there: 4.111 Philosophy is not one of the natural sciences. 4.112 Philosophy aims at the logical clarification of thoughts. · Philosophy is not a body of doctrine but an activity. A philosophical work consists essentially of elucidations. Philosophy does not result in philosophical propositions but rather in the clarification of propositions. Van Peursen who presents an account of the runture in 1'A7i ttgenstein.' s work 2 • at the same time never suggests that Wittgenstein -·had at one time regarded his philosophy as a 1~ Toulmin, "Ludwig Wittgensteint•, Encounter VOL.XXXII No~le .Jar:..~As. 2. van Peursen, "Ludwig Wittgens:tein. An Introduction to his philosophy. - 117·-. metaphysic and then later decided to reject that view. So however different, and significantly so, his later work was from his earlier, he was all the time working within the framework of philosophy as a conceptual discipline. In this way Wittgenstein•s practice reflects Engels' conviction . . 3 that philosophy has been "expelled from nature and history 11 o and that its proper task was conceptual and logical investi-gationso \ ' 3. Engels• "Ludwig Feuerbach and the End of Classical German Philosophy", Part 4. - 118 -C H A P T E R IX INTENTION We have said in Chapter 6 that the meaning of a word is given by the rules for its use. But this, as we saw, needs further qualificationso In general a word can only be said to mean something when it is combined with other words to form a sentence. Therefore the rules which de-termine the meaning of a word can be divided into two sorts: formation rules from which sentence schemata can be derived i.e. the formal rules of syntax; and rules for the actual em-ployment of the sentence in a given context. We saw further that the formulat.ion of this latter sort of rules is only possible if reference is made to what the utterer w,ishes to communicate. Together these two sorts of rules determine the meaning of what is said by an utterer. Thus a machine so constructed as to articulate sentences in accordance with the formal rules of syntax cannot yet be described as SAYING anything, since in this case no reference is possible to the intention with which the sentences are articulated. We may say of the man that he means what he sayso We do not say of a machine that it says anything at all or that it means any-thing by the sounds which it articulates. But is this true? Don 1 t we often say· things like, "The clock tells me it•s time to leave", or "According to the speedomete:~;: we' ~e hitting 90", or .,,The pressure gauge says that the locomotive has got· up a good head of steam"o How is the meaning of such sentences, which are grammatically similar to a sentence such as 11He tells me it is time to leaven, to be distinguished from a sentence of this last sort? Here it would only be a partially satisfacto~y answer to say that the difference lay in the different objects to which the words nthe cl ockn and "he" referred: we have not yet dealt with the difference between the concept of a person and the concept of a machine. Before we can make that distinction - 119 -we must clear the conceptual ground by.making certain others. We will approach this task by way of investigating the logic or. grammar of the two sentences, 11He tells me ••• o. 11 and "the clock tells me •••o" The first major difference between these sentences is that there is a certain sense of the question "Why?" 1 ~ in which it is proper to ask, "Why does he tell me o •• ·.? u and in which it is not proper to ask, "Why does the clock ' tell me •••• o ? 11 in order to clarify thi~ difference, and at the risk of cluttering the investigatioq with prior quali-• fications, we must first distinguish th6 use of the question 1'Why? 11 in which it .£illl properly be asked of both statements .. The sense in which it can be used in both cases is what is ~~i~called 11the causal sense". If I ask, "Why does the clock te-~2. me it i-s time to leave? 11• and mean by this, 11How do the clock• s workings come to be poised it+ this way", then someone mig.1 ·::. answer, "Because it has been running fast for ages", or "Because· the workings have j~st been adjusted". In other words he gives a causal explanation "in order to satisfy my question. Similarly in answer to "Why does he tell me it .! • .;; time to leave?" someone might say, "Well a certain set of synA::.,~2:s ' c have just fired in his brain", or even "Poor chap, he has jus_ had a brain operation you know, and he keeps coming out with these embarrassing orders "• · Now explanations of this sort are usually called causal in order_to distinguish them from explanations which, it is said, give a reason as opposed to a cause. But this is not quite good enough because there is a use of "reason" 1n which it is proper to say that 11He has just had brain surgery" IS giving a reason for his behaviour. At the same time it· is necessary to distinguish between what, to be pedantic, we might then term "causal re·asons 11 and reasons of another sort, and having done this we will find that the distinction corresponds to the distinction between our two uses of the lo Anscombe, "Intention". - 120 -question "Why?u How is this distinction to be made? Wittgen~tein invented a technique for solving this type of grammatical.problem (which occupied a great deal of his attention). 11What is the difference between cause and motive? 11 he asked,· and he.:then suggested a method for answer-ing by putting a further question: "How is the motive DISCOVERED, and how the cause?" 2 • · (Unfortunately however, the introduction of a concept of a motive raises a number of thorny problems which it is beyond the scope of this chapter to cope with; 3 • for instance, we might ask questions as to the relation of motive to intention. But we will not deal with this area of problem) Turning to Wittgenstein 1 s remark and following through its implications we can come up with the following points: One of the ways in which we discover the difference between an intentional behaviour and an unintentional one is by asking why it occurred. (a) If the explanation given in answer to the question is of a sort which says that the par-ticular behaviour is the effect of a prior cause then it is always logically possible (i.e. it makes sense) to ask why the cause occurred. So the question "Why?" interpreted causally can, in principle, generate an infinite series of questions of a similar causal nature. {b) But if the question is interpreted as an attempt to discover the intention with which the behaviour occurred (and I use the awkward locution 11intention and behaviour occurring" with a purpose) then it does not always make sense to continue asking 11Why? 11 of the explanations given. Here are two examples which illustrate (a) and (b) but which also show the inadequacy of (a) and (b) for establish-ing how we discover the difference between causal reasons and that other sort of reasons which are bound up with intention and motive. 2~ P.I. page 224 3. See Kenny, "Action, Emotion and Will", pages 76-99. - 121 -Example {a): A steam pressure gauge registers a high pressure Q. Why does the gauge register so high? A. Because the steam is exerting great pressure in the boiler. Qo Why is it doing that? A. Because the fire has been burning for many hours. Q. Why has it been? A. Because the stoker lit it early. Q. Why did he do that? Ao Because his brain behaves in a certain way. Q. Why? A. oooo• {this leads off into a series of questions as to the causes and effects of neurological phenomena.) Now this does not, of course, constitute a proof that the series can be infinite, but it does indicate how misplaced it would be to look for an explanation of this sort which could stop all further questions. Let us look at anotper question and answe·'r series -Example (b): Jack shakes his fist at Jill. Q. Why did Jacl< do that? ' A. I don• t know, ask him. Q. Jack, "Why did you shake your fist at· Jill? A. o, I don 1 t know, I just wanted to. Q. Thank you Jacko I simply was curious to know In example {b) the questioner might have pressed on with, ••Why did you want to?n If Jack answers quite soberly, nYou know I really can•t say, but I DID want to shake my fist at her,n then the questioner will realise that nothing more can be done by way. o,f question and· answer to find out from Jack why he wanted to. But is this true? Yes and no. And it is precisely in the ambiguity of this answer that the inadequacy of (a) and , ~Y· " - 122 -(b) for distinguishing between causal reasons and intention reasons comes out. For let us assume that the questioner is a psychiatrist, .then he might either go on to ask Jack 1 s friends and family why he wanted to shake his fist or ask Jack, not why he wanted to shake his fist, but why say he keeps staring at a paper knife lying on the tablet But now an important difference between example (a) and example (b) comes out. In order to continue asking 11why" in example (b) the focus of interest has to be changed and with it·the direction of the question asked. The questioner cannot sensibly persist with the question, '~hy did you want to shake your fist Jack?" When once Jack has said he doesn't know. But at this stage somebody might counter that it would not be sensible of someone to persist in questioning Jack as to why volcanos erupt when once he has said he doesn't know. Yet here one wants to say 11But this is different" ioe. the difference between sensibly persisting with the question about volcanos and not sensibly persisting with it is different from the difference between sensibly persisting in the question "Why did you do X?" and not sensibly persisting with it. How is it. different? When we said that there had to be a change in the direction of the question "Why?" if the series of question and answer was to coqti~ue· in example (b) we said something which was slightly but very importantly inaccurate. It is not so much that the change in direction enables the original series to continue but that it enables a new sort of series to begin. Having elicited from Jack that he shook his fist because he wanted ·to, we have established the connection be-tween two things, his wanting to and his shaking his fist, which is not of a causal sort. We have satisfied ourselves from the point of view of finding out whether Jack's action was intended that it was so intended. Having discovered this we can go on to discover the causes of this intention, but then the kind of explanation given will be of a different sort. - 123 -For instance, an important difference betwee:- :::. explanations and the giving of intention-type reasons :i.s ·::-it is quite possible to give a mistaken causa:·~ '--x:-;::.:._2,r:etio:r: whereas it is not possible for someone tell in'] with which his action was performed to maLE: e-. . -,·.., . ~· is why we want to say that the series in eY.;,·.c.·: · ,:- -,·--'; ;::..::,:: .. :::_ a halt, even when more information can stij_ ~- ., The information· acquired about why Jack wan tee' fist, if it is in the form of causal explan2_-;.;:i.or.s, ~_,a corrigible nature, whereas it makes no se::-"se to sc:.-is in error when he tells us the intention or ffiot~ve je his action.So if we want to find out whether someon~ re~. do something and what it was they meant to do, we ~~nt out something different from what we want to fir:d -::;u-c ·, ask what caused him to do it. ..... There is another difference between t~e ~wo explanation. · In example (a) it is always possible to stitute "How? 11 for "Why?n and further to put the interr,: into the· third person passive. Thus: Q. How is the needle on the gauge moved to ~: · A. By the pressure of. the steam in the cylJ.r;'-;"" ... Q. How is that pressure exerted? A. By the motion of molecules. Q. How •ooo••••o etc. But, and this point is made by Melden, 4 • w.;,;._: -~ vie: .... ~- :. ·: to discover whether an action was intentional and h'hat t: . ...: tent ions ~ere we cannot sensibly ask ouest ions li.:~c, ~~~.-:~')W , your fist clenched?" The answer to this quest:.:. c. -:_s a :•.a.J.- · _ for a physiologist.· Still less can we ask, s:r-lo;.;; ,---; ic jOt. clench your fist?n This question i-s only intell:i-:_;}.bl<::. if t~ is something the matter with the fist e.g. that t1~e pt. . .:son -~.~ painful arthritis in the joints of his fingers. Or, ~n another cas.~, it ~an be asked when there is an .: .. JStdCl-2: :·_:-way of the fist being clenched - powerful springs hold:.:..~·.; .. 4. Meld en, "Free Action 11 , p. 26 ff. - 124 -fingers out straight. In these cases "How did you clench your fist? 11 can always be substi t"uted by, "How did you ma_nage to clench your fist?" But when we want to know the in-tention with which the fist was clenched we neither have to consult a physiologist (nor even a psycho-physiologist) nor do we want to know how the fist clenching was achievedo So when we enquire a"Qout intentions we enquire after reasons which, as Waismann puts it, will JUSTIFY the action. 5 • This is very important for our discussion of what it means to say of someone that he says something, as opposed to saying this of a clock. It connects for one thing with the rules for the actual, contextual use of a sentence. To sum up so.far: (i) there are different senses of the question 11Why?" (ii) the ones we are interested in are the causal use and the use in which a reason is demanded; (iii) the series of causal questions and answers cannot be terminated in the way that the series demanding a reason ca~ be; (iv) answers to questions about causes are corrigible in a way in which first person giving of reasons are not; (v) tne causal use of "Why?n permits of a translation which is effected by substituting "Hot-v" for "Why" and putting the rest of the sentence into the passive v'oice. There is another feature which we must examine which differentiates the two uses of 11Why?" namely the difference between the relation of cause to effect, and the relation of intention to action. This, of all the differences we have dealt with, is the most important and the most obvious - for which reason, as so often happens, it is the most easily over-looked. Wittgens~ein treats the problem like this: "I should not say of the movement of my arm, for example; it comes when it comes, etc. And this is the region in which we say significantly, that a thing doesn't simply happen to us, but that we DO it. I don 1 t need to wait for 5. Waismann, "The Principles of Linguistic Philosophy, page 122~ - 125 -my arm to go up - I can raise it. 11 And here I am making a contrast between the movement of my arm and say, the fact that the violent thudding of my heart will subside. "In the sense in which I can ever bring anything about, (such as stomach-ache through overeating), I can also bring about an act of willing. In this sense I bring about the act of willing to swim by jumping into the water o Do.ubt-less I was trying to say: I can 1 t will willing; that is it makes no sense to speak of willing willing. "Willing" is not the name 0f an action; ahd so not the name of· any voluntary action either. And my use of wrong expression came from our wantina to think of willina as an immediate non-causal brina-ing about. A misleading analogy lies at the root of this idea; the causal nexus seems to be established by a mechanism connect-ing two parts of a machine. The connection may be broken if the mechanism is disturbed 11 o (P.r.) 6 • If we think of the connection between the intention to do something and the doing, as being of a quasi-causal sort .then the matter becomes extremely puzzlingo The difficulties in Pritchard's treatment of voluntary action 7 • all arise from his approaching the subject 'in this way. If the intention is construed as operating as a causal agency for action then there seems to be a miraculously invariant relationship between ln-, tention and action. And yet this harmony seems queer and un-canny. For instance, why. is it in the case of intentions that when we ask someone the intention with which he performed an action, his answer can never be wrong? Is it because his in-trospections have yielded him overwhelmingly strong empirical evidence that whenever he intends to do X, the intention results in the action x? No - this is incorrect for two reasons. Firstly we qriite often intend to do, or say som~thing and yet do not do or say it. Secondly, no amount of empirical· evidence of the relation of a cause,to its effect can be so overwhelming 6. P.r. 612-13 7. Pritchard,· 11Duty and Ignorance of Fact", and 11Acting, Willing, Desiring 11 ·in "Moral Obligat:;i.on 11 • · - 126 -that it is not possible to correct the statement that in a given case an effect e is the result of a cause c. So take the first point. When I intend to do something I may or may not do it. I may intend to go to beach and not manage to get there. But if I do go it is the not as though I inferred from my intention that it was likely that I would go. The intention to go doesn't produce the action of going, in the way that the action of·the cue produces a movement in the billiard ball. If we are tempted to think of 11willing 11 as a sort of producing - not however as a case of causation, but I should like to say, .as a direct, non-causal producing 8 • then we should try to give up this picture by remembering Wittgenstein's little example: 11 If someone meets me in the street and asks 'vJhere are you going? 1 and I reply 1 I don't know,' he assumes I have no defini~~ INTENTION, not that I do not know whether I shall be able to t · t· 8 • d ' carry ou my ~nten 10n". The case where I o not KnOw vJile"thPr I shall be able to carry out my intention is one where someon8 at the snooker table asks 11Are you going to get position on the black? 11 and I say "I don't know 11 • And here the uncertair.t:.· arises because causality is involved~ viz. my inexpert use of the cue to propel the white ball towards the red, etc. The second point is connected with the first. In tl--,c~ case of cause and effect I DISCOVER that THIS is the effect nf THAT, through observation of a more or less sophisticated kind. In the case of voluntary action I do not discover that this action of mine is the result of that intention of mine (I DIS-COVER this about other people.) That this must be so v:e c.<>. n see from the grammar of the language-s~ame played with i;:-tentions. Certain grammatical features mark it off from the language-game of causality. For example I cannot sensibly say 11 1 intend" but I can say "I push". Even though 11intend 11 is a finite verb when considered as· a part of speech", it r.mst al-ways be completed by an object - "I intend to push 11 • Thus :10 . . description can be given of an intention independent of wh2~ 8. z. 580 - 127 -is intended. Whereas I can describe my action of pushing independent of whether I am pushing a rock over a cliff or a mower over a lawn. This grammatical rule accounts for the· "queer 11 feeling that one gets if one thinks about in-tentions as being ACTIONS which CAUSE movements: there seems to be such a wonderful harmony between ~he cause and the effect - as though the effect was already 11 in some way" present in the cause. But the harmony between the two, like all metaphysical harmonies is found in the grammar of the langua0e. The connection between intention and action is logical not causal 9• So we can say "I intend to wri te 11 a.nd 11 I write intentionally" and "I intend to write a letter of thanks." We do not discover that we are doing something in-tentionally. Were we to do this there would be occasions on which we were surprised or astonished that THIS action was l· .... _ " tentional. This is what Wittgenstein means when he says: "Writing is certainly a voluntary movement and yet an automatic one. And of course there is no question of a feeling in each movement of writing. One feels something, but could not possibly analyse the feeling. One's hand writes; it does not write because one wills, but one wills what it writes. One does not watch it in astonishment or with in-terest while writing, does not think 'what will it write now?' But not because one had a wish it should write that. For that it writes what I want might well throw me into astonish-t 10. men • " The intention is neither a feeling which accompanies the action nor is it any other peculiar experience, although it would be psychologically interesting if certain feelings occurred in someone who acted intentionally or if certain immages occurred to him. 9. See Meldent s treatment of this topic in "Free Actionu~ 10. z. 586. See also P.I. page 185 on anaesthetic sensations. - 128 -We are now in a better ppsition to return to our ori0inal problem of what it means to say of someone, that he means what he says or, to put it in the form 'in which the question was first posed, of what the difference is between a man telling me something and a clock telling me something. Quite simply the difference is this: the reason why the clock shows eleven o'clock is that it is caused to function in a particular way by the construction and adjust-ment of its mechanism; the reason why a man tells.me this might be that he intends that I shouldn't miss my train. Because we say of a man that he has intentions to communicate we teach him ways to use language, we teach him strategy of verbal action. We do not teach these to a clock because a clock does not act, it moves. So when we ask a man why he says what he says, he answers by giving a justification in terms of what he intends by his utterance (what he wants the utterance to achieve) and by connecting this intention with the rules for the actual employment of the sentence which he is uttering (communication-intentions rules). This is a way of discovering the meaning of an utterance which for some or other reason we don•t understand. circle. This discussion has in a sense brought us full Throug;.1 an analysis of the difference between the two senses of the question 11Why? 11 we have seen why the notion of intention is essentially· connected with the language-game in which we say of people that they mean something by what they utter. But having come back to the original point we are forced into . a new topic o. Earlier I said that the locution "such - and - such and intentional behaviour occurred in someone" was odd. We do not normally say this. We say "He DID such-and-such 11 or uSuch-and-such an ACTION was performed. 11 Clearly there· is something in this which connects with the fact that we cannot translate the intention-seeking question into the "How?" .form. - 129 -And this in turn connects with the statement that machines cannot act, but only move. Finally this last point is ~eant to serve as a contrast to the fact that vre only attribute feelings, thoughts a.nd intentions to human beings and what resemble them. So it is to the question of what the concept o:f a human being involves. that we must now turn. \ - 130 - .-£..liAPTER X HUMAN BEINGS There frequently appear in Wittgenstein 1 s later writings remarks such as the following: 11We only say of a human being and what is like one h . h" k 1 • t at 1t t ln s.n 11Say to yourself, for example: 1 The children over there are mere automatons; all their liveliness is mere auto-matism.' And you will either find these words_becoming quite meaningless; or you will produce in yourself some kind of un-canny feeling, or something of the sort." 2 • "••••• only of a living human being and what resembles (behaves like) a living human being can one say: it has sensa-tions; it sees; is blind; hears; is deaf; is conscious or un-. 3. consc1ous.n HThe human body is the best picture of the human 1 4. sou ·" uour criterion of someone 1 s saying something to him-self is what he tells us and the rest of his behaviour; and we only say that someone speaks to himself if, in the ordinary sense of the word .he CAN SPEAK. And we do not say it of a f h 5. parrot; nor o a grammap one." How are these remarks and the many others like them to be interpreted? ' 6.' Some philosophers like Chihara and Fodor have seen in them evidence that Wittgenstein is arguing for a sort of behavioural criteriology with a strong operationalist bias, and in consequence they attack what they term his "logical behaviourism". These interpreters often make the mainstay of their argument Wittgenstein 1 s statement that an 1. P.r. 360 2. P.r. 420 3. P.r. 281 4. P.r. page 178 5. P.r. 344 6. Chihara and Fodor, op. cit. - 131 -inward process (a mental process) stands 1n need of outward criteria. In apparent oppositio~ to this argument, other philosophers have sought to defend Wittgenstein by arguing for the correctness and coherence of what they take to be lltheory n of criteria in the ·field of the problem of other minds. · A third school dismiss the whole debate on the grounds that (a) its protagonists pose the problem of other minds and the question of what a human being is in the wrong way (b) that they completely misinterpret Wittgenstein by placing his writ-ings in the Cartesian problematic, when this was just the prob-lematic from which he so radically parted company. Among the proponents of the third view are Strawson 7 • Cook 8 • and Reinhardt. 9 • I take their position to be the correct one. In 11Persons 11 Strawson discusses the fac.:t that philo-sophers have found it· puzzling that we ascribe both material (M-predicates) and personal psychological (P-predicates) predi-cates to a certain class of entities, namely human beings. Their ways out of this puzzlement have taken two superficially different forms. The Cartesian way out has been to suggest that since M-predicates are ascribed to both human and non-human entities, but since psychological words are only pre-dicated of human beings, the defining criterion of a human being must be that psychological words can be predicated of him. Therefore, Descartes argued, there is no way of deduc-ing from the fact that there is present to us, what, in the ordinary way we call a human body {the subject of M-predicates) the fact that a person is present. In this, as critiques of l\1ill• s attempt to make such an inference have shown: he was correct. Descartes was thus forced to say that only the think-ing subject himself could know that he was a thinking subject, i.e. a person. This knowledge is gained from introspection and is private and incommunicable • . The apparent alternative to'this line of approach has been to say that P-predicates have meaning if and only if the statements in which they are used,·are translatable into statements containing only M-predicates, i.e. predicates in a 7. 8. 9. Strawson, 11Persons" in nindividuals" Cook, "Human Beingsn in ·Studies in.the Phil. of Wittgenstein ed. Peter W1nch. ~ · · Reinhardt, 11Wi ttgenstein and Strawson on Other Minds u in ''J."< ~:·. - 132 -ourely physical description of what we call the human body. This description will, of coursef not contain any reference to feeling, intention or action. A behaviourist such as Hull :for example, wishes to describe what in the ordinary way we would call upurposive behaviouru as "colourless movements". F'or him there is a logical and cri teriological relationship · between P- and J'vl-predicates, which Cook describes ·when he says: 11 'It is Hull's stated aim to begin from colourless movements and mere receptor impulses as such' and build up (or 'deduce') such concepts as purposive action, ·intelligence, intention and other mental verbs and predicates." 10 • Both the behaviourist and the Cartesian approaches have a noteable common feature - they involve a radical de-parture from the ordinary use language. To be fully con-sistent Descartes should not talk about it, and the behaviour-ists should not really have to talk (except by way of abbrevi-ating a language of bodily movements constructed in terms of co-ordinate Geometry) of the human mind. Another feature which the behaviourist shares with the Cartesian is his preparedness to work in a problematic in which the concept of a person is dealt with in terms of a body -mind dichotomy. For the Cartesian the body hides the mind (the person) in the sense that it is an irrelevant and a decep-tive appearance. For the behaviourist the mind hides the body in the sense that it is a superfluous ·unverifiable and confused hypothesis which tends to bar us from seing clearly that all P-predicates are reducible to J'vl-predicates. Wittgenstein, as we have seen rejects this way of conceiving a human being. Of a human being he says: 'My attitude towards him is an attitude towards a saul. 11 (P .I. page 178) 11 • and 11 •I noticed that he was out of humour'. Is this a report about his behaviour or his state of mind? ('The sky looks threatening.' Is this about the present or 10. Cook, uHuman Beings 11 , page 131. 11. P.I. page 178. - 133''~ the future?) Both; not side-by-side however, but ahout the ( ) 12. one via the other." P.I. pa

Wittgenstein on private language.

Abstract

Extracted data

Wittgenstein on private language.

Abstract

Extracted data

Related items

Related items