*WITTGENSTEIN ON PRIVATE LANGUAGE AND PRIVAT MENTAL OBJECTS* Dale *Jacquette* The Pennsylvania State University, USA ABSTRACT Wittgenstein's private language argument in his PHILOSOPHICAL INVESTIGATIONS is explained and critically evaluated. The implications of Wittgenstein's conclusion that there can be no private sensation language are examined, in light of claims that Wittgenstein by the private language argument also proves that there can also be no private mental objects. The concept of a criterion of correctness is discussed as the key to Wittgenstein's reflections, and counterexamples are considered that raise doubts about the soundness of the private language argument. Difficulties identified in standard interpretations of Wittgenstein's argument indicate that the rejection of private sensation languages does not automatically imply a third-person hard psychological theory, such as logical behaviorism, nor does the argument effectively support reductivist or anti- intentionalist philosophy of mind. 1. The Private Language Argument Wittgenstein's private language argument in the PHILOSOPHICAL INVESTIGATIONS is scattered IN a number of related remarks. There is disagreement about whether Wittgenstein intended to support a thesis concerning the impossibility of private language, and about the significance of the private language argument in his later philosophy. It has been suggested that the private language argument has important negative consequences for phenomenology, phenomenalism, sense data theory, foundationalist epistemology, solipsism or skepticism about other minds, the existence of so- called private mental objects, and intentionality.If the private language argument is sound, then, according to some interpretations, there cannot be any private mental objects, nor a language in which private mental objects are intelligibly designated. Even if private mental objects do not exist, the fact that they can be intelligibly designated at least as nonexistent objects in ordinary language, and that inferences can be drawn from propositions in which they are designated, contradicts the conclusions of the private language argument as it is usually understood. Either the logic and semantics of everyday discourse must be amended in such a way that it is prevented from designating existent or nonexistent private mental objects, or else the private language argument must be refuted. The attempt is made here to interpret the private language argument, explain its consequences, and finally argue that although the argument is valid, it is unsound.Wittgenstein denied that he was offering philosophical theses,*1* and while some commentators have taken him at his word for this,*2* others have found it difficult to believe that he does not in fact, despite verbal protests to the contrary, argue in favor of the thesis that private language is impossible.*3* What is not open to doubt is that Wittgenstein's remarks in the Blue and Brown Books and Philosophical Investigations have led others to adopt the view that there cannot be a private language. The argument suggested by these remarks therefore needs to be evaluated on its own merits, without concern for the historical problem of Wittgenstein's actual intentions.*4* In the Philosophical Investigations, Wittgenstein writes: But could we also imagine a language in which a person could write down or give vocal expression to his inner experiences - his feelings, moods, and the rest- for his private use? - Well, can't we do so in our ordinary language? - But that is not what I mean. The individual words of this language are to refer to what can only be known to the person speaking; to his immediate private sensations. So another person cannot understand the language.*5* A private language is a language that can only be understood by the person who uses it to refer to his own private or internal psychological experiences. It must be such that it is impossible for anyone but the user to understand. That private languages are necessarily unintelligible to others at once rules out extraneous interpretations of the argument. The idea that a private language in Wittgenstein's sense of the word might be a secret code referring to sensations that could in principle be taught to others is eliminated, since it would not be logically impossible for another person to decipher. Also excluded are interpretations of the argument based on mere difficulties in learning to speak a language in circumstances of total social isolation. These problems are not central to Wittgenstein's concerns.*6* Another preliminary matter of interpretation has to do with the purity or mixture of putative private languages. It appears that for Wittgenstein, a private language may be embedded in or a proper part of a larger public language.Let us imagine the following case. I want to keep a diary about the recurrence of a certain sensation. To this end I associate it with the sign "S" and write this sign in a calendar for every day on which I have the sensation.*7* If a publicly understood calendar, with its square blocks or separate pages conventionally representing days of the year, is used as a medium or framework for recording purported private sensations, then the private language in question must be a mixed or impure private language, in which signs supposedly designating private incommunicable sensations are presented within the conventions of a public grammar. Castaûeda writes:It is not clear that Wittgenstein's is the issue between a public and an absolutely private language. Very naturally, one would expect to find many cases of private language all linked up by a series of family resemblances, ranging off from a language all of whose individual words refer only to private objects.*8* Wittgenstein insists that language does not exist in a vacuum, but in a context of social institutions that can be described as a form of life. He frequently emphasizes the background and stage setting without which the definition and use of a sign in a language are meaningless. Thus he asks:But what does it mean to say that he has 'named his pain'? - How has he done this naming of pain?! And whatever he did, what was its purpose? - When one says "He gave a name to his sensation" one forgets that a great deal of stage -setting in the language is presupposed if the mere act of naming is to make sense. And when we speak of someone's having given a name to pain, what is presupposed is the existence of the grammar of the word "pain"; it shews the post where the new word is stationed.*9* Since Wittgenstein's discussion of the diary of private sensations follows immediately after these observations, it may be concluded that the objections to the possibility of a private language suggested by his remarks apply to the concept of a mixed or impure private language.The reason for disputing the possibility of mixed private languages in Wittgenstein's argument depends on what might be called the criterion requirement for the meaningful application of terms in a language. Wittgenstein maintains that the person who attempts to keep a diary of private sensations does not have a criterion of correctness for the use of sign 'S'.I will remark first of all that a definition of the sign cannot be formulated. - But still I can give myself a kind of ostensive definition. - How? Can I point to the sensation? Not in the ordinary sense. But I speak, or write the sign down, and at the same time I concentrate my attention on the sensation - and so, as it were, point to it inwardly.- But what is this ceremony for? for that is all it seems to be! A definition surely serves to establish the meaning of a sign.- Well, that is done precisely by the concentrating of my attention; for in this way I impress on myself the connexion between the sign and the sensation. - But "I impress it on myself" can only mean: this process brings it about that I remember the connexion right in the future. But in the present case I have no criterion of correctness. One 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'.*10* Language is a rule-governed activity. But according to Wittgenstein, there can be no rules for the use of a term in a language if there are no criteria of correct or incorrect use of the term. To explain this, it is necessary to understand what Wittgenstein means by a criterion of correctness, and what difference the criterion or lack of criterion is supposed to make in the attempt to define or understand terms.Wittgenstein's concept of criterion is sometimes mistakenly regarded as having only epistemological significance. This may be the result of his frequent references to memory and the limitations of memory in attempts to reidentify a recurring sensation. Wittgenstein implies that an individual's memories, however internally coherent, are inadequate as a criterion for the correct or incorrect application of ostensible private language signs. He claims that memory is ordinarily relied upon with good justification only because of the logical possibility that memory may be verified or corroborated by checking it against information in nonprivate, nonmnemic sources. But this is impossible in the attempted use of private language terms, because the sensation in question by hypothesis has no external or publicly distinguishing behavioral manifestations by which its occurrence could be known even in principle by anyone but the individual who privately experiences it.It may then be inferred that since there is no way for the person to determine outside of his memories whether or not another sensation is the same or of the same kind as that originally supposed to have been designated by sign 'S', the use of the sign is not actually governed by a linguistic rule, and therefore cannot be part of a genuine language.Let us imagine a table (something like a dictionary) that exists only in our imagination. A dictionary can be used to justify the translation of a word X by a word Y. But are we also to call it a justification if such a table is to be looked up only in the imagination? - "Well, yes; then it is subjective justification." - But justification consists in appealing to something independent. - "But surely I can appeal from one memory to another. For example, I don't know if I have remembered the time of departure of a train right and to check it I call to mind how a page of the timetable looked. Isn't it the same here?" - No; for this process has go to produce a memory which is actually correct. If the mental image of the timetable could not itself be tested for correctness, how could it confirm the correctness of the first memory? (As if someone were to buy several copies of the morning paper to assure himself that what it said was true.)Looking up a table in the imagination is no more looking up a table than the image of the result of an imagined experiment is the result of an experiment.*11* This appears to support the epistemological or verificationist interpretation of the criterion of correctness objection to private language. Yet it is not simply the difficulty of relying on memories to verify that a sensation is of the same kind as that supposedly designated by the first use of 'S' that prevents there from being a criterion of correctness or linguistic rule governing the private language term in the diary, but rather the lack of any criterion or linguistic rule is the result of deeper semantic trouble, of which the epistemological, memory, and verification problems are mere symptoms. The reason why memory fails to provide verification of the occurrence of sensations like that supposedly designated by the diarist's use of sign 'S' is that the sign fails to designate even in its very first attempted use.The epistemological interpretation of the criterion objection says in effect that the application of sign 'S' by the diarist stands in need of justification, which an individual's memory, the only possible source of justification under the circumstances, is unable to provide. This, in some unspecified way, is supposed to preclude the possibility of a linguistic rule governing the diarist's use of the sign in his putative private language. But this is not an accurate interpretation of Wittgenstein's argument. Rush Rhees writes:Wittgenstein did not say that the ascription of meaning to a sign is something that needs justification. That would generally be as meaningless as it would if you said that language needs justification. What Wittgenstein did hold was that if a sign has meaning it can be used wrongly.*12* The difficulty in the diarist's attempted use of sign 'S' is not just that memory alone is an inadequate epistemological basis for justifying subsequent applications of the sign, but rather that the sign has no meaning in any of its attempted applications, including the first, because there is no sense in which the sign can be used incorrectly.Another way of expressing the criterion objection is to say that there is no way satisfactorily to distinguish between instances in which an individual has correctly followed a linguistic rule involving the putative private language sign and instances in which the individual merely believes that he is correctly following a linguistic rule for the application of the sign.*13* Are the rules of the private language Impressions of rules? - The balance on which impressions are weighed is not the impression of a balance. "Well, I believe that this is sensation S again." - Perhaps you believe that you believe it!*14* It is not enough for an individual such as the diarist to believe or be under the impression that he is following a rule when using sign 'S' in attempting to record the occurrences of an incommunicable sensation. It must be possible to distinguish between situations in which the individual merely believes or is under the mistaken impression that he is following a linguistic rule for the putative private language sign from situations in which he is actually following a rule. But this, Wittgenstein suggests, because of the nature of the case, is precisely what it is impossible to do. Without an independent or external check on the application of the private language sign, it is not possible even in principle for the diarist to use the sign correctly or incorrectly.*15* If the argument is sound, it entails that the very first application of sign 'S' is equally ungoverned by any linguistic rule, despite its apparent stipulative character. This is different from the epistemological interpretation of the criterion objection, since the latter provides no basis for the claim that the original or very first use of a private language term also lacks meaning. The diarist does not actually make use of or act in accord with a linguistic rule in any of his attempts, including the first, but at best has the mistaken impression that he is following a rule.Then did the man who made the entry in the calendar make a note of nothing whatever? - Don't consider it a matter of course that a person is making a note of something when he makes a mark - say in a calendar. For a note has a function, and this "S" so far has none.*16* It is clear that for Wittgenstein, in the absence of whatever he means by criteria of correctness, the terms of a putative private language are without sense. This is worse than the epistemological inability to verify that subsequent occurrences of sensations are or are not of the same kind as a single instance originally supposed to be stipulatively designated by a purported private language sign.If the epistemological rather than the deeper semantic interpretation of the criterion objection to private language were accepted, then it would still be possible to formulate a restricted version of a private language in which individual sensations are designated by signs, but in which no terms are offered as predicate or sensation kind terms, and no attempt is made to reidentify sensations as falling under the same sensation kind term or private language predicate. Each sensation or sensation instance could then receive its own private logically proper name, and the epistemological restrictions of verification by memory comparison in the attempt to classify sensations by kinds would be taken as limiting reidentification beyond the capabilities of language. Nevertheless, the individually designated private sensations could still be said to have been named in a limited private language consisting only of individually designating terms. The semantic interpretation of the private language argument suggested by Wittgenstein's remarks avoids even this attenuated sort of private language by reflecting back on the meaninglessness of every attempted application of the sign. The absence of a criterion of correctness for the terms invalidates even the first attempted or original ostensibly stipulative designation of any private sensation.*17* 2. Phenomenology, Intentionality, and Psychological Privacy The most interesting byproduct of the private language argument is the claim that if private language is impossible, then there cannot be private mental objects. The inference seems to be that if there cannot be a language in which so-called private mental objects are designated, then necessarily there are no objects of the sort to be designated. O.R. Jones writes:The implication for the Cartesian viewpoint should be clear. Sensations, feelings, and so forth are not private in the way supposed in the Cartesian view. If they were, then only a private language involving private rules would be possible to talk about them. But such a language is impossible. Conversely, since we are able to talk about our sensations and feelings, they are not Cartesian private objects.*18* This would also be a serious consequence for object theory, since the domain of objects in the logic is determined by the terms and well-formed expressions it contains. If there are no terms designating existent or nonexistent private mental objects in the syntax of the logic, then there can be no corresponding existent or nonexistent private mental objects in the object theory domain. Yet there are terms in the object theory which do ostensibly designate private mental objects, and which appear to obey the linguistic rules of the formal system. If the private language argument is sound, drastic revision is required.The basic presuppositional foundations of phenomenology, sense data theory, phenomenalism, and intentional and subjective theories of many kinds, may also be undermined in the same way by the private language argument. The threat to phenomenology and the intentional outlook posed by the private language argument is acknowledged by Harry P. Reeder.Phenomenology often comes into criticism from linguistic analysts of the Wittgensteinian tradition who claim that the language used in phenomenological description must be an unacceptable form of language, due to the fact that the reflexive shift of the phenomenological reduction bars one from criteria of consistency of language use.*19* Alan Paskow observes:Despite the enormous body of commentary by analytic philosophers on Wittgenstein's theory of psychological privacy, I know of no phenomenologist who has attempted to deal with and respond to this theory, which is a threat to the very foundations of any philosophy that accords an honorific status to the data of subjectivity...What features of Wittgenstein's view of privacy constitute this challenge? Essentially those which point to the two following conclusions (which, if true, logically entail the falsity of critical suppositions of several varieties of phenomenological methodology): (a) that it is not possible to intuit apodictically the essences or patterns of sensory presentations; and further, (b) that it is not even possible to formulate with respect to such presentations true (or false) descriptions of fact. Indeed, Wittgenstein implies that subjective claims do not denote one's own private experiences and that the communicative function of such statements is different from what most ('pre-analytic') philosophers thought it to be.*20* Any philosophical theory based on data held to be of privileged epistemic access, or that attempts to reconstruct knowledge from phenomenological descriptions of the private contents of consciousness, is plainly contradicted if the private language argument is sound. The kinds of philosophical theories consistent with the argument by contrast then are extensional, behaviorist, physicalist, and materialist. Phenomenology and intentional philosophy are excluded.Philosophical behaviorism is supported if it can be shown that there are no private mental objects or private psychological experiences. It must then be possible at least in principle to find external or publicly observable distinguishing criteria for every so-called private mental event.*21* No thought, feeling, or occurrence in the mind could be concealed from the behavioral scientist equipped with sufficiently sophisticated monitoring equipment. It would always be possible for an outsider to know a person's innermost thoughts and sensations, no matter how carefully disguised by self-control or effort of will. The office of mind in the phenomenal world, a place of freedom and resort, would be as open to public empirical inspection as any external entity.It is also maintained that the private language argument refutes dualism, solipsism, and phenomenalist epistemology. Warren B. Smerud writes:Turning now to the claim that the anti-private-language thesis is philosophically important, the sort of significance which it is alleged to have is indicated by Norman Malcolm, who informs us that the possibility of a private language is presupposed in the formulation of a number of long-standing philosophical problems as well as in the sorts of attempts which have commonly been made to resolve them; for example, all traditional problems concerning inferring the existence of other minds, phenomenalism, and sense-data theory, presuppose the possibility of the sort of langauge which the anti- private-language thesis denies.*22* It is not worthwhile to exhibit all of the inferences needed to obtain these further results. In general, it must first be shown that a particular problem presupposes the possibility of private language. The 'problem' is then made unproblematic by the consideration that if the private language argument is sound, then private language is impossible.These consequences may be accepted by adherents of the private language argument, even if Wittgenstein would not have approved of them. But it should be observed that several philosophers have interpreted Wittgenstein's remarks throughout the later writings as strongly hinting that he would indeed have supported at least some of the secondary results of the private language argument. It may appear that as usual, Wittgenstein has deliberately left the most important things unsaid. He insists that readers of his unconventional philosophical investigations think for themselves and draw their own conclusions from the ideas he presents.*23* It has already been observed that if the private language argument is sound, then the domain of objects cannot contain any private mental objects. But there are well-formed expressions and well- defined terms in the logic of ordinary discourse which ostensibly designate objects that could only be the objects of thought of particular minds. The object necessarily thought about only by person A may be designated in the logic by the definite description, 'The x such that x is thought about necessarily only by person P'. If the private language argument is sound, then not only is there no such existent private mental object, but, since terms ostensibly designating the object are not governed by linguistic rules, the terms themselves cannot belong to a genuine language. This implies that the logic of ordinary discourse is not a genuine language. To amend the theory in conformity with the private language argument would require either rejecting the intuitive criterion for domain membership, or else revising the formation principles of the logic to exclude all possible terms ostensibly designating private mental objects. But any constant term potentially designates a private mental object if it is permitted to abbreviate a term such as the definite description mentioned above. Generalizations of the form 'For all x, ...x...' apply to all objects, and there is no suitable technical way of eliminating ostensibly designating private mental object terms by syntactical stipulation.These measures are unnecessary if the private language argument is invalid or unsound. Private mental objects may then be admitted into the domain of the logic, and the object theory can be used to formalize phenomenological and intentional theories so completely as to include their very foundations in private mental experience. Whether or not these theories are true, it must at least be possible to represent their distinctive logical structures.*24* 3. A Diary of Private Sensations and the Beetle in the Box If I am not in pain, I find that I do not know or rightly remember what pain is. If pain and sensation language generally were public rather than private, how could I not know what another person means when he says that he is in pain? I may be in great sympathy with the person, and I may have access to all the external behavior associated with pain, such as verbal reports, wincing, muscle tension, blood pressure. But it would be wrong to conclude that I know what the person is experiencing, except in the most general terms.The situation is not improved if I am told that the person has a sharp pain, excruciating pain, stabbing pain, flowing, burning, or traveling pain. I may know that the person is in pain, but I do not know what the pain is like, no matter what descriptive efforts are made in the public sensation language to help me understand. This is something that can only be known to the individual in the privacy of thought, and that can be expressed, if at all, only in a private language. The public sensation language used to communicate limited information about psychological states does not embody knowledge of what L.C. Holborow has appropriately called the infima species of sensation.*25* A man feels a twinge of pain and opens his diary to the page for that day. He writes down, 'burning twinge'. Then he scratches it out and rewrites, 'stabbing twinge'. He frowns, erases the words, and tries, 'wrenching twinge'. This too he rejects. He decides that no word in the language he has learned from others since his childhood adequately expresses the feeling he has experienced, and decides, arbitrarily, to write down the sign 'S' in the diary calendar space to indicate that on this day he experienced the sensation which was something but not quite like a burning, stabbing, wrenching twinge. He resolves for diagnostic purposes to write down the sign whenever he feels the pain.The next day he experiences another pain, and writes 'S' in the journal. Alongside it he makes the remark, 'I think it was the same as S, but cannot be sure. I will call it "S?".' He corrects the entry by writing 'S/S?' instead. Over a period of time he finally sees a doctor. He has not yet found a more apt way to designate the pain he experiences than by calling it 'S', 'S?', or indicating his uncertainty about whether or not the pain sensation is just like the first experience he chose to record. He tells the doctor that on a particular occasion he had a long S-seizure, and that he has no better way to describe it than to say it is something but not quite like a burning, stabbing, wrenching twinge. He says that for lack of exact description he calls it his 'S' pain.The doctor asks questions about the pain, and runs through his own vocabulary of sensation terms, attempting to explain to the man what each kind of pain is supposed to be like. But the man remains unwilling to accept any of these terms as more accurately naming the pain he feels. He does not give up 'S', and insists that the doctor also refer to it as 'S' when discussing the case. The doctor does not know exactly what is meant by the sign, but he complies with the request in order to humor his patient, and designates the pain, whatever it is, by 'S', even though he secretly believes that the man is stubbornly refusing to admit that the pain is in fact like one of the kinds listed in his professional lexicon.Several years later the doctor himself has a peculiar twinge of pain. In trying to identify it, he discovers that none of the terms he is accustomed to use adequately describes it, for the sensation is of a sort he had never experienced or heard of before. He feels the pain as something but not quite like a burning, stabbing, wrenching twinge, but admits that the description is not entirely correct. At last he recalls his former patient who had a pain he could not describe in ordinary terms, but was something not quite like a burning, stabbing, wrenching twinge. He remembers that the man had called it 'S', and he decides to call his own recent pain by the same name.The shared use of 'S' to designate their respective sensations does not mean that the doctor's pain and the patient's pain are of the same kind, nor that the doctor has finally learned the meaning of the patient's putative private language term. There is no way to tell, for as Wittgenstein and others have argued, there are no external criteria for private sensations. The doctor begins to use the same sign, and may even believe that in doing so he is identifying or designating the very same kind of pain as that experienced by his patient. But this does not entail that he has learned the meaning of the sign. It is a well-known linguistic phenomenon that the very same sign type can be used with widely divergent denotations and connotations. The fact that the doctor takes over the patient's sign to designate what he believes to be the same kind of pain does not establish that what was once private has now become public. Nor is it implied that the sign became public when the doctor first patronized the patient by referring to his pain by sign 'S'.It appears that 'S' may constitute a term in a mixed or impure private language, the precise meaning of which must remain obscure to everyone but the individual for whom it privately designates a particular kind of sensation. From this it follows that any sensation word, even in the so-called public sensation language, may be subject to the same indeterminacy of reference. In what is generally regarded as the public sensation language, we share signs that are believed to designate at least relevantly similar kinds of sensations. We teach portmanteau sensation words such as 'pain', 'excruciating pain', 'stabbing pain', and the like, color words, words for auditory sensations, and others, by teaching the use of signs that we have also learned under appropriate publicly observable circumstances that are thought to be causally connected to the occurrence of the sensations. But it may be that even these uses of language involve nothing more than a shared sign and an unjustified belief in the codesignation of public sensation language.The ordinary sensation language learning situation is perhaps no different in important ways from the story of the doctor and patient. At first, the doctor uses the same sign as the patient more or less to satisfy the patient's whim, though he is not really sure what the patient means to refer to by 'S'. Later, he adopts the same sign to classify his own sensations on the belief that they are the same. By analogy, the child says 'blue' when sufficiently coaxed to respond verbally in this way in the presence of what others have similarly been trained to recognize as blue. But who can speak for the color sensation which the child has when a blue object enters his visual field? Even if the experience is experimentally determined to be something we would otherwise like to call a shade of blue, who can speak for the child's experience in its infima species? This is concealed from all but the individual, who learns a public language to express inadequately what is in fact hidden away from public inspection.*26* A diary of private sensations like that described in the discussion above is thought by some to be outlawed by Wittgenstein's private language argument. But the objection must be critically examined. For this, an analysis is required of the concepts of following a rule, the recognition and reidentification of recurring sensation kinds, and the Wittgensteinian criterion of correctness. The criterion objection to private language states that putative private language terms do not belong to a genuine language because there are no criteria for the correct or incorrect application of private language terms. The validity of the argument is not in doubt, but its soundness is another matter. The assumption that if a putatively designating term belongs to a genuine language, then there must be criteria to determine correct and incorrect applications of the term, is sometimes regarded as an expression of the view that language or language use is a rule- governed activity. Wittgenstein's obsession with the practice of following a rule in the Blue and Brown Books, Philosophical Investigations, Philosophical Remarks, Remarks on the Foundations of Mathematics, Zettel, and Nachlass, indicates that the concept of a criterion, the concept of following a rule, and language games construed as forms of life, are intimately connected in his later thought. It is commonly held that Wittgenstein established the dependence of rule-governed forms of life on criteria of correctness, and that this logically excludes private languages.In order to challenge the soundness of the private language argument, it is best to begin with an understanding of what Wittgenstein means by the concept of a criterion. There is general agreement that a criterion is a kind of decision procedure, which in principle makes it possible to determine beyond reasonable doubt whether or not a particular term is correctly applied to an object. In a critical exposition of Wittgenstein's concept of criterion, Carl Wellman writes:To describe something is to specify what it is like and what is unlike. In this sense descriptive terms are always used to classify or divide things into kinds. This seems to imply that to use descriptive language a person must be able to distinguish between the different kinds of things. Unless a person were able to recognize members of a given class, he could hardly use the class name very effectively. To be able to use or understand descriptions one must be able to tell which objects fit a given description and which descriptive expressions fit a given object. But how does one know whether or not a specified description fits a given object? This is the question which Wittgenstein's conception of a criterion is intended to answer.*27* In what he takes to be the spirit of Wittgenstein's later philosophy, Norman Malcolm asserts:...one may be inclined to think that there cannot be a criterion (something that settles a question with certainty) of someone's having a sore foot or having dreamt, but merely various 'outer' phenomena that are empirically correlated with sore feet and dreams. This view, however, is self- contradictory: without criteria for the occurrence of these things the correlations could not be established. Without criteria the sentences 'His foot is sore', 'He had a dream', would have no use, either correct or incorrect.*28* But it is uncertain whether or not Wittgenstein meant to give 'criterion' a univocal technical sense. Paul Ziff observes:[Wittgenstein] meant by 'criterion' something like test, or standard or way of telling. That is, he meant what any speaker of his dialect would have meant if he were using the word in familiar ways. I am inclined to suppose that most likely his use of the word 'criterion' would fit his use of the word 'game', that is, one might be able to discern a family of cases.*29* Yet the private language argument requires a technical or quasi- technical definition of the concept. It may be regarded as suggested if not actually explicit in Wittgenstein's remarks about criteria of correctness.For purposes of evaluation, the following definition of 'criterion' is proposed. The concept of nondefective determination of the truth value of a predication must first be defined. This is necessary in order to avoid trivializing counterexamples involving the paradoxes of material and strict implication.*30* For any x and any predicate proposition p, x nondefectively determines the truth-value of p =df x is a nonempty set of propositions describing the steps of a finite procedure, together with the results of each step of the procedure in a particular application, that entails p or not-p, exclusively, and that does not imply any false or undetermined proposition.A special concept of criterion for this interpretation of the private language argument can now be defined.For any x and any predicate term 'P', x is a W-criterion of or for P =df for any object y, x in principle nondefectively determines that Py or not-Py, exclusively (determining nondefectively whether the term is truly predicated of the object, or not truly predicated of the object).The definition can be provisionally used in assessing the soundness of the private language argument on the criterion of correctness interpretation.Here is a reconstruction of the private language argument that accords with Wittgenstein's remarks, and with the most authoritative commentaries of critics and defenders of the argument. It is presented by means of the definition of a W- criterion.1. If a putatively designating term belongs to a genuine language, then there must be W-criteria that determine in principle whether or not the term is correctly applied to a particular object under particular circumstances.2. But there are no W-criteria that determine even in principle whether or not a putatively designating private language term is ever correctly applied to any purported private mental object._______________3. No putatively designating private language term belongs to any genuine language. There cannot be a private language.This version of the argument is valid, but not sound.*31* There is a number of ways to show that on this definition of 'criterion' or 'W-criterion' the premises need not, and perhaps cannot, be accepted as true.Defenders of the private language argument usually devote most of their energies to establishing something like premise (2). Reference is made to Wittgenstein's remarks about the inadequacy of memory in the reidentification of purported kinds of private sensations. Wittgenstein seems to believe that the internal comparison of sensation memories is the only thinkable criterion candidate, but that as a criterion it is as useless as consulting multiple copies of the same morning newspaper in an effort to determine the truth of reports published in any particular copy. This may be true. We can simply grant premise (2), and challenge the soundness of the argument by disputing the truth of premise (1), an assumption most accounts take for granted.A consequence of modern quantum physics and the indeterminacy principle is that, in the proposed terminology, there are no W-criteria for the reidentification of subatomic particles. The reidentification of subatomic particles requires the determination of their precise location in space and time. This in turn presupposes the ability to fix both the position and velocity of a particular subatomic particle at a given time. But such an exact determination cannot be made, because position can only be experimentally verified by interacting with the particle in a way that disturbs its velocity, and velocity can only be verified by interacting with the particle in a way that disturbs its position. Position and velocity can therefore never be jointly determined even in principle for a particular subatomic particle. The best that can be achieved is a statistical approximation of its location, insufficient for exact reidentification.The problem of reidentifying subatomic particles is much like the problem of reidentifying recurring sensations. Suppose that an arbitrary individual subatomic particle could be stipulatively named. It persists at least for a time in a swarm of subatomic particles, undergoing random alterations of velocity and position. According to the indeterminacy principle, there is no equivalent of a W- criterion for determining later whether or not any chosen particle is the same as the original, because it cannot be reidentified. But it would be strange to conclude that subatomic particles are anything but public objects. They are obviously not private mental objects, but are rightly thought to constitute the very stuff of the physical, objective public world.*32* There are other terms designating other kinds of public nonprivate and nonmental objects for which there are no W-criteria to determine even in principle whether or not the terms are correctly applied to objects.*33* Wittgenstein discusses family resemblance predicates as those for which there is nothing essential in common between objects falling under a given predicate.*34* But there are some indisputably public objects that fall in the shadowy areas between family resemblance predicates, where there is clearly no W-criterion definitely establishing them as belonging under one family resemblance predicate rather than another.As an illustration of the kind of public language terms in genuine languages for which there are no W-criteria, consider Wittgenstein's discussion of the public language family resemblance term 'game'. Wittgenstein allows great latitude in the kinds of objects that may reasonably fall under this predicate. But there are controversial cases of public objects for which there is no W-criterion that determines even in principle whether or not the predicate 'is a game' is correctly or incorrectly applied to the objects. Law offenders may describe their wrongdoing as a kind of game, in which the purpose is to violate civil or criminal statutes, or as many statutes as possible, without being detected, captured, or punished by the authorities. A number of different kinds of 'games' of the sort are possible. If the criminals are caught and convicted, they might regard the matter as of no further moral consequence than the loss of a game of chess or cards. Those who take a more serious moral attitude toward the law on the contrary may staunchly deny that crime is ever a game of any kind or in any sense at all, despite its family resemblance to certain kinds of games. (Is Russian roulette a game?)If someone learning English as a foreign language encounters the term 'game', and wants to know whether or not the term is correctly applied to crimes, there will evidently be no W- criterion to settle the question. If he asks criminals, they will tell him one thing; civil authorities may say just the opposite. There is no test or finite objective decision procedure which he can then apply to determine which of the two groups of language users is using the word 'game' (or 'crime') correctly, and which incorrectly. But the terms 'crime' and 'game' are undoubtedly public language terms in a genuine language.Similar cases arise in areas of dispute about moral and aesthetic value in public objects. There seems to be no W-criterion that determines even in principle whether or not the public language term 'person' is correctly applied to fetuses, and there is no W-criterion that determines even in principle whether or not the public language term 'beautiful' applies to particular artworks.*34* Some language users will say one thing, and others another. It would be extravagant in the least to insist that these persons are therefore speaking nonequivalent idiolects of a language rather than expressing substantive disagreements of opinion in the same language. Is there no single language in which this sort of disagreement can take place? If not, then either all language is private, or there is no genuine language. If there is such a language, then it is sufficient to refute premise (1) and the private language argument based on it. If public languages can be understood despite the lack of W-criteria for certain predicates, then purported private languages should not be expected to satisfy a more demanding requirement. Adherence to premise (1) of the reconstructed version of Wittgenstein's private language argument would have the unreasonable consequence that subatomic particles and many other more familiar kinds of things are not public and cannot be designated in any genuine language.*35* But if these consequences are unacceptable, then the private language argument is unsound, and provides no compelling reason for restricting private mental objects from the object theory domain. The possibility of a mixed or impure private language in the object theory then provides the necessary (but not sufficient) presuppositional basis for phenomenology, phenomenalism, sense data theory, and other kinds of intentional philosophical theories.Wittgenstein writes:Let us now imagine a use for the entry of the sign "S" in my diary. I discover that whenever I have a particular sensation a manometer shews that my blood-pressure rises. So I shall be able to say that my blood-pressure is rising without using any apparatus. This is a useful result. And now it seems quite indifferent whether I have recognized the sensation right or not. Let us suppose I regularly identify it wrong, it does not matter in the least. And that alone shews that the hypothesis that I make a mistake is mere show. (We as it were turned a knob which looked as if it could be used to turn on some part of the machine; but it was a mere ornament, not connected with the mechanism at all.)*36* The passage is difficult to interpret. But it suggests that a monitoring device such as a manometer might supersede an individual's subjective report about his own internal mental states.The transition is subtle. At first Wittgenstein seems to imply that the diary of private sensations might replace the external monitoring equipment because of its correlation with the person's subjective reports. The diary can be used as an indication of rising blood pressure, so the manometer becomes obsolete. In the sentences immediately following this, however, Wittgenstein maintains that recognizing the sensation incorrectly does not matter. All that appears to be crucial in this sudden change of emphasis is the correlation of sensation and manometer reading. The monitor somehow gains authority over the individual's interpretation of his own sensations, since, as Wittgenstein says, it eventually becomes indifferent whether or not the sensation is recognized 'correctly'. This is supposed to show that the concept of mistaken recognition is without meaning or application in an individual's assessment of his own psychological experiences. Wittgenstein's thought takes an unexpected turn here, since presumably it is the prior regular coincidence of sensation and manometer reading, based on a reliable recognition of sensations, that first justifies elimination of the monitoring device in favor of sensation reports recorded in the diary.A more sophisticated monitor is conceivable, which a behavioral scientist might use to read slight changes in the central nervous system of a subject. It could be like Wittgenstein's manometer, but atuned to factors more informative than blood pressure alone in evaluating a subject's so-called internal states. A machine of the sort might be connected to a person, so that a behaviorist could study the dials and readouts, interpreting them on the basis of his training in some experimental method, by translation manual, or in accord with external indicators, like a needle that rotates to positions on a panel marked with the names of different kinds of sensations.It is always at least logically possible that the machine and the person's introspective reports disagree, even on the assumption that the machine never malfunctions. The behaviorist's central nervous system monitoring equipment may indicate that a subject is experiencing sensations which the subject would positively deny. But there seems to be no better reason for saying that the machine is right about the infima species of sensation experienced, than that the person is right. Of course the monitoring equipment is sure to be right about something. It will have measured a parameter of the central nervous system of the subject to be interpreted by the operator of the machine. But if the machine indicates that the subject is in pain, and a needle or readout specifies a kind of pain called 'burning', though the person insists that it is not quite like a burning sensation but something different, then, leaving mechanical malfunction and subject dishonesty aside, it is always possible that the machine is wrong, and has not accurately evaluated the precise quality or kind of sensation which the person is experiencing.The machine is capable of measuring only parameters of the subject's experiences that are already part of the public language, since private experiences by definition have no distinguishing external behavioral manifestations. Whatever the physical mechanism used, it must always be interpreted by someone in a public sensation language. This means that it will necessarily fail to distinguish any private sensation from some other kind described in the public sensation language. The machine cannot settle the question of whether or not a person is having a private sensation of a particular sort, because its inherent limitations of interpretation and design logically exclude it from the mechanical determination of any but publicly defined sensations. The correlation of subjective sensation recognition and monitoring equipment results in the case of the manometer or more sophisticated imaginary devices does not make subjective sensation reports obsolete, epistemically inferior, or less trustworthy than the mechanical testimony of a behavioral monitoring device. The external monitoring equipment cannot provide evidence against the existence of private sensations or private mental objects. It is not to be relied on over and above introspective reports.In the Philosophical Investigations, Wittgenstein outlines the problem of the beetle in the box.Suppose everyone had a box with something in it: we call it a "beetle". No one can look into anyone else's box, and everyone says he knows what a beetle is only by looking at his beetle.- Here it would be quite possible for everyone to have something quite different in his box. One might even imagine such a thing constantly changing. - But suppose the word "beetle" had a use in these people's languages? - If so it would not be used as the name of a thing. The thing in the box has no place in the language-game at all; not even as a something: for the box might even be empty.- No, one can 'divide through' by the thing in the box; it cancels out, whatever it is. That is to say: if we construe the grammar of the expression of sensation on the model of 'object and designation' the object drops out of consideration as irrelevant.*37* Here again Wittgenstein makes an unexpected and apparently unjustified leap. As he describes the case, there is nothing objectionable about everyone having an inscrutable, different kind of beetle, even one that constantly changes. The problem arises only on the further supposition that the word 'beetle' may have a use in public language. The conclusion drawn is that if this were true, then the word could not be the name of a thing. This is left unexplained until the following sentence, in which it is said that the word could not have a place in the language game because an individual's box might be empty.*38* But is unclear even in this situation why the word 'beetle' would not still function at least to designate the contents of an individual's box, whether empty or not, existent or nonexistent. The Internal Revenue Service intelligibly uses the word 'income' to refer to a person's yearly earnings regardless of amount, including as a limiting case the possibility in which a person has no earnings at all. (The analogy with the beetle in the box can be reinforced by supposing that a person's income is entirely inscrutable by outsiders, that it may change constantly, and no one else, including the IRS, is permitted to know exactly what another person's real income is.) The analogy with the beetle in the box remains intact, because although there is nothing which in principle prevents a person's income from being publicly known - a disanalogy with the case of private mental objects - this is also true of physical objects like beetles kept in boxes. In this respect, both income and the beetle-box have crucial disanalogies with private mental experience. But these features are not relevant to Wittgenstein's claim that the terms in question could not be used as names of things, and that they could have no legitimate function in any kind of language game. Nonexistence of the thing is not decisive, since words like 'income', 'Pegasus', 'phlogiston', and others, function perfectly well in public language and public language games, despite the nonexistence of designated objects.The beetle in the box presents no theoretical difficulties for intensional logic. The beetle is an appropriate metaphor for the privacy of sensations and the inclusion of private mental objects in the domain of the logic. The beetle in the box is the private and incommunicable psychological experience of a person that makes possible the learning of public language, and the development of phenomenological and related intentional theories.The epistemological problems connected with the private language argument raise further questions about the limitations of what the mind can know by introspection. Is it possible for the mind to be mistaken about the identification or reidentification of a private sensation or private mental object? The anti-private- language thesis seems to entail that the very concept of mistaken identification of private sensations or private mental objects is meaningless, and that because of this correct identification is also meaningless and impossible, placing ostensibly designating private language terms beyond the pale of linguistic rules of application. This is expressed by premise (2) of the reconstruction of the private language argument. In previous discussion, the premise was simply granted in order to concentrate on premise (1). Now premise (2) must also be examined.It may be objected against the premise that private experience and the recognition of private sensation kinds is a prerequisite for public language and for public or external criteria of mental phenomena, and that there is a sense in which the mind may be mistaken in hypothetical applications of private language terms to private mental objects or so-called internal psychological experiences. Wellman writes:One cannot claim that the credibility of the identification of recurrent kinds of sensations depends entirely upon the possibility of public corroboration, for of what value is checking one identification against another unless each has some independent credibility? Actually, corroboration is a test of correctness only because the identifications which support one another each have some antecedent claim to correctness.*39* If this is true, then it must also be possible for a person to be mistaken in the application of a private mental term to a private mental object. There are conceivable circumstances in which the mind may wrongly describe a sensation in the vocabulary of its private language.An individual uses private language term 'S' to denote an incommunicable pleasure, and another private language term 'E' to designate an incommunicable pain. It is possible for the person to mistake one for the other, and later correct himself. This is not to say that the individual could have a sensation and not know it, but that he could have a sensation of a particular kind and mistakenly identify it at least temporarily as a sensation of a different kind. Deception can also occur if a subject is preconditioned in appropriate ways. Suppose that stroking the surface of the skin with a feather-edge produces an incommunicable pleasure designated 'S' in a person's private sensation language. A paper-cut on the other hand produces an incommunicable pain, which the person designates 'E' in his private sensation language. If the person is preconditioned to expect the pain of the paper-cut as part of an experiment or initiation ceremony, but in fact receives a light feather stroke, then he may utter 'E' at the moment the skin is touched, and wince as if in pain. He may at once realize the error and correct his prematurely mistaken identification of the sensation. He may notice the faintly pleasant glow that normally follows an S-type sensation, instead of the throbbing ache of the E-type, and admit that it was not E after all, but S. A subject might also prepare a questionnaire, in which he asks for information to be recorded in his diary of private sensations as a kind of routine introspective procedure. One of the questions asks, 'What sensation did I experience at 12:00 noon?' The person waits until what he takes to be noon, unaware that his watch is broken or that daylight savings time has been instituted that day, and so writes 'S' when in fact at noon by the correct time he does not experience an incommunicable pleasure, but an incommunicable pain. Later he may realize that his judgment of time had been inaccurate, check the diary for a record of what his sensation report would have been at that time, and correct his answer on the questionnaire, erasing 'S' and writing down 'E' instead. This provides yet another sense in which the significant correction and possibility of error in private language ascription and private mental object identification can occur. The possibility of error in the application of private language terms establishes the significance of private sensation language. This in no way contradicts the presuppositions of phenomenology, phenomenalism, or foundationalist epistemology, since it remains incorrigibly true and directly evident that it seemed to the individual that he was in E pain, and later that it seemed he was not. And incorrigible directly evident seeming is all that philosophical theories of the sort usually require.The philosophical objection to the existence of private mental objects in the private language argument is that there cannot be private mental objects because such objects could not be designated in a language or fall under any linguistic rule of application. But in a logic of existent and nonexistent objects, there are many linguistic rules regarding the application of ostensibly designating private language terms. Whether or not private mental objects exist, they can intelligibly be talked about in such a logic. The linguistic argument against the possibility of private mental objects, that there are no linguistic rules governing their application, is ineffectual. In order to show that private mental objects do not exist, the defender of the private language argument must establish that the objects could not exist because they have some metaphysically incompatible combination of constitutive nuclear properties. Otherwise, and this would make the anti-private-language position philosophically less interesting, it may be necessary for the defender of the private language argument to prove that there is some contingent reason why private mental objects do not exist. But this again goes beyond anything that has so far been represented as a version or interpretation of the private language argument.The linguistic argument alone will not do, because even if private mental objects are nonexistent, they can be designated in a rule-governed way in the object theory. Since there are terms ostensibly designating private mental objects, the domain of the logic contains at least nonexistent private mental objects. It remains for the anti-private-language theorist to demonstrate that there are not also existent private mental objects. The defender of Wittgenstein's private language argument may point out that Wittgenstein did not mean to suggest that private mental object terms could not be part of any genuine language or fall under any linguistic rules, but the less sweeping claim that private mental object terms do not fall under any linguistic rules for the application of such terms to actual objects in experience. According to these semantics, a predicate is true of an object if and only if the state of affairs obtains in which the reference class or intension of the predicate includes the object. The epistemological problem of how we can tell whether or not the state of affairs in question obtains need not entail that there are no linguistic rules governing the application of private mental object terms in the logic. Wittgenstein's criterion objection to private language embodies a somewhat jaundiced view of the reliability of the private language user. Wittgenstein maintains that there can be no criterion of correctness in the case of private sensation terms because there is no independent check on whatever a person says is true about his internal states. An individual could conceivably insist that any two sensations, no matter how phenomenologically distinct, were of the same kind, and that two sensations phenomenologically indistinguishable were actually distinct. But this is not a difficulty for the hypothesis that private sensation language subconsciously orders experiences within the mind in a rule-governed way, nor for the view that the mind is equipped with a private mental language or acquires its own private mental language prior to socialization.*40* A less intransigent individual conscientiously recording sensations in a diary of private sensations, sensitive to fine- grained phenomenological distinctions in experience, and honestly attempting to apply private language terms as family resemblance predicates for similar kinds of sensation in an accurate description of his mental life, will not be likely to confuse private pain and private pleasure, or other distinct sensations, as incommunicable experiences of precisely the same kind. Nor is it probable that he will classify similar incommunicable experiences as radically different, though mistakes may occur. The objection that phenomenological misdescriptions can arise presupposes the possibility of error, which is just what the private language argument is supposed to preclude. By Wittgenstein's reasoning, the possibility of error also implies the possibility of correct judgment and accurate reidentification of sensations. Epistemic limits of memory aside (which are equally problematic for the users of any public language), there is no reason to think that at least some kinds of private sensations could not be identified wrongly in a private language vocabulary. This also makes meaningful the possibility of correct sensation description in a private mental language.*41* NOTES *1* Ludwig Wittgenstein, Philosophical Investigations, 2nd edition, translated by G.E.M. Anscombe (New York: The Macmillan Company, 1958), §109, p. 47e: "And we may not advance any kind of theory. There must not be anything hypothetical in our considerations. We must do away with all explanation, and description alone must take its place." Also §128, p. 50e: "If one tried to advance theses in philosophy, it would never be possible to debate them, because everyone would agree to them." *2* Judith Jarvis Thompson, "Private Languages", American Philosophical Quarterly, 1, 1964, p. 20: "On the contrary, the disservice was done by those who credited the [anti"private"language] thesis to him [Wittgenstein]. If nothing else, they failed utterly to take seriously his claim that he held no opinions and put forward no theses in philosophy." Timothy Binkley, Wittgenstein's Language, (The Hague: Martinus Nijhoff, 1973), p. 172: "The discussion of private language is...contrapuntal, and not the development of a theory which claims that private language is impossible." And, p. 173: "If [Wittgenstein] does not want to put forward theses, he avoids negative as well as positive claims: his task is ultimately neither to affirm nor to deny philosophic claims or theories. If p is a philosophic thesis, so is - p, and Wittgenstein will have nothing to do with the assertion of either one." F.A. Siegler, "Comments" (on Newton Garver's "Wittgenstein on Criteria"), in Knowledge and Experience: Proceedings of the 1962 Oberlin Colloquium in Philosophy, edited by C.D. Rollins (Pittsburgh: University of Pittsburgh Press, 1963), p. 77: "Has Wittgenstein a logical theory? He disavows having any sort of theory at all...the fact that Wittgenstein disavows any logical theory should lead one carefully to question assertions that he does have one." Colin McGinn, Wittgenstein on Meaning: An Interpretation and Evaluation (Oxford: Basil Blackwell, 1984), p. 2: "We must avoid the temptation to regard the text as a sort of cipher through which we must penetrate to reveal the linearly ordered argument beneath. It is not that Wittgenstein really has an argument of orthodox form which for some inscrutable reason he chose to present in a disguised fashion." Jaakko Hintikka, "Wittgenstein on Private Language: Some Sources of Misunderstanding", Mind, 78, 1969, pp. 423-25. Warren B. Smerud, Can There Be a Private Language? An Examination of Some Principal Arguments (The Hague: Janua Linguarum, Series Minor, 1970), pp. 14-5. *3* Norman Malcoln, "Wittgenstein's Philosophical Investigations", in The Philosophy of Mind, edited by V.C. Chappell (Englewood Cliffs: Prentice-Hall, Inc., 1962), pp. 74-100. James D. Carney, "Private Language: The Logic of Wittgenstein's Argument", Mind, 69, 1960, pp. 389-96. C.W.K. Mundle, "'Private Language' and Wittgenstein's Kind of Behaviourism", The Philosophical Quarterly, 16, 1966, p. 35. Hector-Neri Castaûeda, "The Private Language Argument", in Knowledge and Experience, pp. 88-132. John Turk Saunders and Donald F. Henze, The Private Language Argument: A Philosophical Dialogue (New York: Random House, 1967), p. 5. Anthony Kenny, Wittgenstein (Harmondsworth: Penguin Books, 1973), pp. 178-202. *4* Smerud, Can There Be a Private Language?, p. 15, adopts a similar provision. On some interpretations, Wittgenstein's remarks on private language are not even supposed to suggest the impossibility of private language, but show only that a private language, if there were such a thing, could not provide the basis for a phenomenalistic reduction of public language. See Moltke S. Gram, "Privacy and Language", in Essays on Wittgenstein, edited by E.D. Klemke (Urbana: University of Illinois Press, 1971), pp. 298- 327. Andrew Oldenquist, "Wittgenstein on Phenomenalism, Skepticism, and Criteria", in Essays on Wittgenstein, pp. 394-422. *5* Wittgenstein, Philosophical Investigations §243, pp. 88e-9e.. *6* A.J. Ayer may have inadvertently misled the private language controversy in his essay, "Can There Be a Private Language?", in The Private Language Argument, edited by O.R. Jones, (London: Macmillan & Co. Ltd., 1971), pp. 50-61. Ayer complicates Wittgenstein's example by asking whether a Robinson Crusoe, left alone on a desert island, could invent a language in total social isolation in which private or internal sensations are described. *7* Wittgenstein, Philosophical Investigations §258, p. 92e. *8* Castaûeda, "The Private Language Argument", p. 136. *9* Wittgenstein, Philosophical Investigations §257, p. 92e. *10* Ibid. §258, p. 92e. *11* Ibid. §265, pp. 93e-4e. *12* Rush Rhees, "Can There Be a Private Language?", Proceedings of the Aristotelian Society Supplement, 28, 1954, p. 68. *13* Carney, "Private Language: The Logic of Wittgenstein's Argument", p. 563. Smerud, Can There Be a Private Language?, pp. 28f. Ross Harrison, On What There Must Be (Oxford: The Clarendon Press, 1974), pp. 158-59. *14* Wittgenstein, Philosophical Investigations §§259-260, p. 92e. *15* Castaûeda, "The Private Language Argument", pp. 144-45. *16* Wittgenstein, Philosophical Investigations §260, pp. 92e-3e. *17* See Saul A. Kripke, Wittgenstein on Rules and Private Language (Cambridge: Harvard University Press, 1982). Kripke's interpretation is mistaken in several ways. Kripke does not accept, and notes that Wittgenstein would not endorse, outright skepticism about the meaningfulness of language. But unless Kripke's community of language users solution to Humean skepticism about language is satisfactory, his exposition leads inevitably to this conclusion. That the proposal is unworkable is sufficiently indicated by the fact that absolutely any reaction of the language community can be regarded by the Humean skeptic as expressing approval or disapproval of the use of a term (say, the continuation of the series 2, 4, 6, ...), just as any number placed after 6 in the series can be regarded by the Humean skeptic as correctly continuing the series. Thus, if Kripke's analysis is correct, it provides the basis for an immediate reductio ad absurdum of the private language argument. Further criticisms of Kripke's interpretation are given by McGinn, Wittgenstein on Meaning. *18* Smerud, Can There Be a Private Language?, p. 107f. *19* Harry P. Reeder, "Language and the Phenomenological Reduction: A Reply to a Wittgensteinian Objection", Man and World, 12, 1979, p. 35. *20* Alan Paskow, "A Phenomenological View of the Beetle in the Box", The New Scholasticism, 48, 1974, p. 277. *21* Mundle, "'Private Language' and Wittgenstein's Kind of Behaviourism", p. 35: "Though Wittgenstein professed to eschew philosophical theories, he seems to have accepted a theory which could not be confirmed simply by observing how people talk when not doing philosophy, namely a form of Behaviourism...If Wittgenstein had formulated this view explicitly, it would presumably have run something like this: that words which are ostensibly used to name or refer to private experiences can have meaning only by referring to overt behaviour (as in 'he is in pain') or by deputising for other, and 'natural' forms of behaviour (as in 'I am in pain'). Wittgenstein's main reason for putting forward this account seems to have been his acceptance of a questionable theory of meaning, namely that statements can be meaningful only if they are publicly verifiable." *22* Smerud, Can There Be a Private Language?, p. 16. Also, Malcolm, "Wittgenstein's Philosophical Investigations", p. 75. *23* Wittgenstein, Philosophical Investigations "Preface", p. xe: "I should not like to spare other people the trouble of thinking. But, if possible, to stimulate someone to thoughts of his own." *24* Nicholas F. Gier, Wittgenstein and Phenomenology: A Comparative Study of the Later Wittgenstein, Husserl, Heidegger, and Merleau- Ponty (Albany: State University of New York Press, 1981). *25* L.C. Holborow, "Wittgenstein's Kind of Behaviourism", The Philosophical Quarterly, 17, 1967), p. 128: "We must now return to Mundle's diarist and his stomach pains. It is now clear that his claim that this pain is just like the others that have troubled him all morning is a 'result of observation' in Wittgenstein's sense. But the claim rests on the contention that this is a specific type of pain which can be discriminated, but not publicly described. Others can be told that it is a sharp type of stomach pain, but this does not give its infima species." Wittgenstein, "Wittgenstein's Notes for Lectures on 'Private Experience' and 'Sense Data'", edited by Rhees, The Philosophical Review, 77, 1968), p. 233: "It is as though, although you can't tell me exactly what happens inside you, you can nevertheless tell me something general about it. By saying e.g. that you are having an impression which can't be described. As it were: There is something further about it, only you can't say it; you can only make the general statement. It is this idea which plays hell with us." *26* This is the problem of partial spectrum inversion. See Keith Campbell, Body and Mind (Notre Dame: University of Notre Dame Press, 1980), p. 74: "When I see that the traffic light has changed, more has happened than just the acquisition of a new set of dispositions to acts in which I discriminate one state of the traffic light from another. If I have a curious sort of color blindness, in which I see as many different shades of color as you do, but different ones, then when we both see the traffic light (or anything else) we will each acquire the very same discriminative dispositions. Yet there are great differences in our mental lives..." *27* Carl Wellman, "Wittgenstein's Conception of a Criterion", The Philosophical Review, 71, 1962, p. 435. *28* Malcolm, Dreaming (London: Routledge & Kegan Paul, 1959), p. 60 (emphasis added).*29*Paul Ziff, "Comments" (on Garver's "Wittgenstein on Criteria"), in Knowledge and Experience, p. 84. See also Rogers Albritton, "On Wittgenstein's Use of the Term 'Criterion'", The Journal of Philosophy, 56, 1959, pp. 845-57. Albritton emphasizes the apparent plurality of meanings of 'criterion' in Wittgenstein's writings, and collates passages from many places throughout the later works to support a multiplicity of interpretations of the concept, and in this sense agrees with Ziff. But compare Kenny, Freewill and Responsibility: Four Lectures (London: Routledge & Kegan Paul, 1978), p. 11: "To use Wittgenstein's technical term, the physical expression of a mental process is a criterion for that process: that is to say, it is part of the concept of a mental process of a particular kind (a sensation such as pain, for instance, or an emotion such as grief) that it should have a characteristic manifestation." *30* The term 'nondefective determination' is borrowed from and suggested by Roderick M. Chisholm's definition of 'non-defectively evident' in his solution to Gettier-type counterexamples to the traditional analysis of knowledge. Chisholm, Theory of Knowledge, 2nd edition (Englewood Cliffs: Prentice-Hall, Inc., 1977), p. 109. *31* This account has the advantage of presenting the structure of the private language argument as an indirect proof, as Castaûeda and others have recommended. See Jones, editor, The Private Language Argument, Part V, 'The Private Language Argument as a Reductio ad Absurdum', pp. 132-82. *32* The indeterminacy thesis in quantum physics is sometimes said to have idealistic implications, but this contradicts Wittgenstein's private language argument construed as a refutation of idealism and methodological solipsism. See Hacker, Insight and Illusion, pp. 245-75. *33* Robert J. Richman, "Concepts Without Criteria", Theoria, 31, 1965, pp. 65-85. *34* Wittgenstein, The Blue and Brown Books (Oxford: Basil Blackwell, 1958), pp. 17-20, 87, 124; Philosophical Investigations §§ 23, 65- 73, 77-8, 87; Remarks on the Foundations of Mathematics, edited by G.H. von Wright, Rhees, and Anscombe, translated by Anscombe (Cambridge: The MIT Press, 1956), I, §§ 64-7, IV, § 8, V, §§ 26, 36; Philosophical Grammar, edited by Anscombe and von Wright (Oxford: Basil Blackwell, 1967), §§ 326, 472, 474-76. *35* Wellman, "Wittgenstein's Conception of a Criterion", p. 438: "Since there is no sharp line between essential and nonessential characteristics, it is a mistake to look for some essence common to all instances of a term. Instead, a term is usually applied on the basis of many overlapping characteristics which form a family likeness. As a rule there is no such thing as the criterion for the use of a descriptive expression. This implies that in justifying the use of an expression by giving its criteria one will normally have to give more than one criterion...Upon occasion these various criteria may even conflict with one another. Which criteria are relevant to the use of a term on any particular occasion will depend primarily upon the circumstances under which it is used." See also Leon Pompa, "Family Resemblance", The Philosophical Quarterly, 17, 1967, pp. 66-8; Richman, "'Something Common'", The Journal of Philosophy, 59, 1962, p. 828. Nicholas Griffin, "Wittgenstein, Universals and Family Resemblances", Canadian Journal of Philosophy, 3, 1974, pp. 644-46; Hubert Schwyzer, "Essence Without Universals", Canadian Journal of Philosophy, 4, 1974, pp. 69-78; R.I. Aaron, "Wittgenstein's Theory of Universals", Mind, 74, 1965, p. 251. An example involving art objects is suggested by Maurice Mandelbaum, "Family Resemblances and Generalizations Concerning the Arts", American Philosophical Quarterly, 2, 1965, pp. 220-22. Wittgenstein, in Philosophical Investigations §66, p. 31, mentions Kampfspiele, which Anscombe translates as 'Olympic games'. *36* Wittgenstein, Philosophical Investigations §270, pp. 94e-5e. *37* Ibid. §293, p. 100e. *38* C.A. Van Peursen, Ludwig Wittgenstein: An Introduction to his Philosophy (New York: E.P. Dutton & Co., Inc., 1970), p. 91: "...Wittgenstein compares this whole problem of mental images, that are supposed to give meaning to words, with the following game. A number of people have a box with a beetle inside. Each person can look in his own box, but in no one else's. They tell each other what their beetles look like, what color they are, and so on. This language game can continue smoothly, even if all the boxes are empty. The thing in the box, which must not be public, accessible to others, is not essential for the game." (Emphasis added.) Note that Van Peursen has distorted Wittgenstein's description of the example. See Wittgenstein, Philosophical Investigations, II, xi, p. 207e: "Always get rid of the idea of the private object in this way: assume that it constantly changes, but that you do not notice the change because your memory constantly deceives you." *39* Wellman, "Wittgenstein's Conception of a Criterion", pp. 446-47. Wellman writes, p. 447: "It is clear...that Wittgenstein is faced with an awkward dilemma. Either there is some justification prior to corroboration for trusting one's memory or there is not. If there is, even a private identification has some claim to validity; if there is not, even a public identification has no claim to validity. Therefore, either Wittgenstein's objection to private sensations serving as criteria is mistaken, or his own theory that publicly observable characteristics serve as criteria is inadequate." *40* Jerry A. Fodor, The Language of Thought (Cambridge: Harvard University Press, 1975), pp. 55-205. *41* Wittgenstein's later diagnosis of philosophical malady identifies the grammar of a language game in its natural habitat of convention and practice, and exposes the conceptual confusions that result when such a grammar is inappropriately projected out of its depth onto another language game. Ironically, this is just what Wittgenstein seems to do in taking the criteria of correctness grammar involved in the naming and reidentification of some material objects and applying it to phenomenology. This produces havoc when private experience fails to satisfy alien grammars transferred from material-object-description language games. The affliction is marked by the private language argument itself and the philosophical confusion it fosters, the symptom of which is the wide variety of interpretations of the argument and the ingenious but usually strained efforts to reconcile it with ordinary ways of speaking about psychological experience.