WITTGENSTEIN'S VIEW OF METAPHYSICS WITH REFERENCE TO THE PROBLEM OF SOLIPSISM

 

 

 

          Wittgenstein's conception of metaphysics and the role that it plays in the philosophical enterprise has been a persistent source of confusion and misunderstanding among many of his followers.  His own contempt for the direction in which the Vienna Circle developed the doctrines of the Tractatus indicates that this is so for at least the early part of his work.  The Vienna positivists interpreted Wittgenstein's view that metaphysical questions are nonsensical in that they violate the logical syntax of language as implying that what metaphysicians attempt to say also lacks any significance.  But as Carnap, Feigl, Schlick, and others learned in their strained encounters with the author of the Tractatus in the 1920's, this is not at all Wittgenstein's intention in that treatise.  The fact that metaphysical issues cannot be given a syntactically correct formulation does not mean that they are nonexistent, trivial, or unworthy of consideration.  As Wittgenstein remarks in the "Preface" to his early work, very little is achieved by the logical clarification of language, with its banishment of the questions of metaphysics from the realm of sense.  The really important matters about which metaphysicians futilely attempt to speak take up residence in an extralinguistic domain.

          Due, no doubt, to the enlightening effect of the passage of time, there now seems to be a more subtle awareness of the complexities and ambivalences of the Tractatus' attitude toward metaphysics than that demonstrated by the logical positivists in the decade and a half following the first appearance of the book.  However, in the mainstream of Wittgenstein interpretation, there is no similar recognition of the subtleties of Wittgenstein's view of metaphysics in his later work.  It is generally held to be the case that, during the period which culminates in the Philosophical Investigations, Wittgenstein conceives of the legitimate role of philosophy as purely therapeutic in character.  Genuine philosophy dispels the illusions of metaphysics by showing the ways in which they stem from confusion about the grammatical rules which govern language in its concrete employment.  Metaphysics has its origin in the "bewitchment of our intelligence by language," and the problems of metaphysics vanish once the spell of language is broken.  But this interpretation ignores the positive function which Wittgenstein allows to metaphysics in The Blue Book, a function which nothing in the Investigations denies, and much confirms.  In The Blue Book, Wittgenstein suggests that the "mental cramp" experienced by metaphysicians can be legitimately removed by those changes in notation in which metaphysical language consists.

          In this article, I will examine Wittgenstein's view of the function of metaphysics in both his early and his later work.  As I have already suggested, I will argue that Wittgenstein never summarily rejects metaphysics, although he is critical about what traditional metaphysicians conceive of themselves as doing.  The therapeutic regimen of Wittgensteinian philosophy is directed toward, not the elimination of metaphysics, but its reformation.  This reformist attitude is best illustrated by Wittgenstein's life-long obsession with what is for him the most salient of metaphysical issues--the problem of solipsism.  I will argue that Wittgenstein holds a solipsistic position in the Tractatus and the Notebooks of 1915-16, and that he does not abandon this position in either The Blue Book or the Philosophical Investigations.  I hope to show that, in each of these works, what most interpreters have taken to be a refutation of solipsism is in actuality an attempt to reformulate the position in accordance with revisionist principles.  Finally, I will try to evaluate the new conception of metaphysics that emerges from Wittgenstein's writings. 

 

1.

 

          In the Tractatus, Wittgenstein provides a sketch of what, in his view, metaphysicians traditionally attempt to accomplish.  He then proceeds to argue that, once we become clear about the nature of language, we will see that the usual purposes of metaphysics are incapable of being fulfilled.

          Wittgenstein contends that traditional metaphysicians attempt to talk about formal or internal properties of objects and states of affairs.  Formal or internal properties are necessary structures which constitute the essence of things and without which things would be inconceivable (TLP 4.123).  The totality of these essential structures comprises the a priori framework of the world.  Essential structures are not themselves mundane matters of fact, but they are required if there are to be such matters of fact at all.  Metaphysicians attempt to give linguistic expression to these transmundane features through the use of formal concepts such as "substance," "object," "fact," "number," and so on.  Wittgenstein tries to show that these formal concepts cannot be admitted into a logically correct notation, and therefore that metaphysicians believe themselves to be doing something that cannot be done.

          Before we consider the arguments which Wittgenstein employs in order to demonstrate the inadmissibility of formal concepts into an adequate notation, it may be helpful first to understand what he means by "form."  According to the Tractatus, the world is made up of facts--existent states of affairs--which are concatenations of simple objects (TLP 2.072).  Simple objects constitute the substance of the world in the sense that, without them, there would be no mundane matters of fact at all.  Although objects are necessary features of the world, the matters of fact into which they enter are thoroughly contingent.  The forms of an object are what Wittgenstein calls its internal properties, i.e. those properties without which the object would not be the sort of thing that it is.  Examples of the forms of objects in general are space and time, while being-colored is an example of the form of a visual object (TLP 2.0251).  In addition to its formal or internal properties, an object also possesses material or external properties.  Material or external properties are dependent upon the contingent concatenations into which objects enter in comprising matters of fact (TLP 2.0231).

          According to Wittgenstein's general principle that there is a structural isomorphism between language and the world to which language is related, formal or internal properties on the one hand, and material or external properties on the other have representatives on the linguistic level.  The material dimension of language is its ability to depict contingent states of affairs through the formation of empirical propositions.  The names which occur when these propositions are adequately analyzed are correlated with simple objects, while the syntactical relations between names isomorphically reflect the contingent, fact-making relations between objects.  However, syntax not only expresses material properties; it also has a formal dimension.  By showing us the ways in which the names of objects are formed, the syntactical rules of language also show what sorts of objects there are, in other words, their ontological types.  The formal concepts of metaphysics are intended to express these essential ontological structures.

          But do formal concepts meet the conditions which all language must meet in order to say something meaningfully?  In order to answer this question, it is first necessary to uncover the conditions of linguistic meaningfulness.  Wittgenstein tells us that they are perspicuously evident in a properly formulated Begriffschrift, a logically correct notation governed by perfectly definite syntactical rules.  Wittgenstein's notion of a Begriffschrift differs from that of Frege and Russell in that he does not intend for it to take the place of ordinary language for special logico-mathematical purposes.  Rather, in his view, such a purified notation constitutes the inner structure of any meaningful language whatsoever; purported linguistic expressions that do not accord with it are not cases of meaningful language at all.

          The perfect notation which lies below the surface of all meaningful language is constructed in such a fashion as to meet the fundamental condition of determinacy of sense.  Linguistic meaning is either determinate or it does not exist: "...what can be said at all can be said clearly..." (TLP, Preface).  According to Wittgenstein, the demand for determinacy of sense is equivalent to the demand for completeness in the introduction of the indefinable simples of language into our logical notation.  The analysis of signs must terminate in the discovery of simple signs not subject to further analysis, for otherwise it would be impossible unambiguously to correlate any sign, simple or complex, with the object or state of affairs meant by the sign (TLP 3.23).  What is more, the syntactical rules which govern the employment of signs must be exhaustively established in advance, for apart from the rule that describes the way in which the sign may combine with other signs, the simple sign is incapable of depicting anything about the world (TLP 3.327).  Signs and rules of syntax function hand-in-glove in enabling language to refer to states of affairs.  For this reason, whenever a simple sign is introduced into our notation, the syntactical rule which oversees its employment must be introduced concurrently.

          For Wittgenstein, only an extensional logic is capable of fulfilling the criterion of determinacy of sense.  This becomes especially evident when we consider Wittgenstein's view of the nature of logical quantification.  Variables can occur in logical notation only insofar as their meaning is completely definite.  Since a variable stands for all of those signs which are its possible substitution instances, in the case of quantification, the demand for determinacy of sense is the demand that the values of the variable be exhaustively specified (TLP 3.316).  There is no chance of discovering a value for a variable at some time subsequent to that in which the variable is lexically introduced.  If we know what a variable means, then we also know all of the values which satisfy it.

          The variable can be understood as a syntactical rule for the formation of names.  It can also be seen as a logical category common to all of its values.  Insofar as the variable is satisfied only by a certain stipulated range of values, the limits of appropriate value substitution show us what sort of object the values name, in other words, what the ontological type of the object is (TLP 4.126).  That is to say, the variable is a formal concept.  However, it is impossible to make any assertions about this concept.  The attempt to do so would necessarily mistake the variable for the name of an object.  Names are designated, not by variables, but by values of variables.  It is true that the variable is nothing other than the range of values which can be substituted for it.  Nevertheless, the range of values must obviously be distinguished from any particular value.  The variable exhibits a logico-syntactical category common to all of its values, and an ontological type common to all of the objects named by its values.  But it is not itself the name of any object.

          In Wittgenstein's view, metaphysicians characteristically attempt to make propositional assertions about formal concepts.  By so doing, they mistakenly regard such concepts as names capable of serving as subjects of predication.  For example, they may try to say, "objects are infinite in number."  However, the psuedo-concept "object" is not the name of something.  It is the form that all values which satisfy the "object-variable" possess.

          The inadmissibility of formal concepts into a proper notation becomes evident in another regard when we consider the fact that formal concepts have no special role to play in the context of such a notation.  A formal concept is given as soon as any object falling under the concept is given.  It is impossible to introduce into our notation names of objects belonging to a formal concept and a separate expression for the formal concept itself (TLP 4.1272).  In this case, there is no expressible difference between concept and object.  When language is completely analyzed in accordance with the dictates of determinacy of sense, no set of unique formal concepts is to be found.  An extensional logic, which is the real substructure of all meaningful language, has no room for the language of metaphysics.

          According to Wittgenstein, many of the aspects of the surface grammar of ordinary language entice metaphysicians into making categorial statements about formal concepts.  For example, the same word sometimes has different modes of signification, as is the case with the "is" of identity, existence, and predication.  This superficial feature of our language may lead metaphysicians to speak about a unitary Being which includes within itself each of the functions of the word "is."  On the other hand, two words that have different modes of signification sometimes are employed in propositions in what appears to be the same way.  Since the word "identical" is employed as an adjective like other adjectives, metaphysicians are liable erroneously to interpret it as a predicate (TLP 3.323).  Such confusions are no longer possible when we use a sign-language that perspicuously obeys the logical syntax beneath the surface grammar of ordinary language.  In using a sign-language, we would not be employing the same sign for different symbols and not be using signs that have different modes of signification in a superficially similar way (TLP 3.325).  Once we become clear about the way in which language actually signifies, then we will also see that the use of formal concepts violates the logical syntax of language.  Since this syntax is the condition of linguistic meaningfulness, the formal concepts which fail to accord with it are meaningless.

          No interpreter of the Tractatus can avoid this conclusion: the young Wittgenstein holds that metaphysical statements are meaningless.  But this does not imply that he also believes them to be without any use.  According to the Tractatus, metaphysical propositions, while meaningless in the sense that they violate the logical syntax of language, nevertheless have a function to perform which, in a way, is more exalted than the role played by empirical propositions: they point to that which cannot be said but only shown.

          Wittgenstein's difficult doctrine of showing and saying directly rests on his extensional interpretation of formal concepts.  Through the use of a formal concept, the metaphysician intends to make a categorial statement about the essential constituents of reality.  However, given a correct notation, we see that the formal concept is a variable.  That something belongs to a specific essential category is shown by the fact that its value is a substitution-instance for a variable that can take any value within the range of the relevant category.  For example, that A is a number is shown by the fact that the sign "A" can be substituted for the variable whose range is numbers.  But the attempt to say that A is a number would mistake the variable for the name of a categorial object.  This becomes clear if we rephrase "A is a number" as "The class of numbers has A as a member."  In Wittgenstein's words: "What can be shown, cannot be said" (TLP 4.1212).  And yet, while the metaphysical statement, "A is a number," is nonsensical in that it violates logical syntax, it nevertheless gestures toward the fact that our notation shows that the sign for A is a value of the number-variable.  The statement is useful, illuminating nonsense, as paradoxical as this formulation may seem.

          The refusal of some interpreters of the Tractatus to accept the notion of "illuminating nonsense" stems from their failure to recognize the full consequences of the fact that, in his early work, Wittgenstein does not yet identify the meaning of a sentence with its use.  Of course, by demonstrating that metaphysical propositions are meaningless, the Wittgensteinian dialectician intends to convince the metaphysician that what the latter wishes to say cannot be meaningfully said.  But this dialectical refutation rests on a theory of language which is itself laced with unabashedly metaphysical elements.  In particular, it is impossible to justify the fundamental meaning criterion of determinacy of sense unless we suppose that there are simple objects with which the simple signs of language may be directly correlated.  But this is not something that can be said, since it too violates logical syntax.  Still, the assertion that there are simple objects is illuminating for two reasons: it points toward a situation which is shown by the fact that our notation consists of simple signs; and, when the theory of language that it implies is fully understood, it generates a recognition of its own nonsensical character.  Although it is meaningless, the statement has a two-fold use.

          With reference to the self-transcending character of metaphysical discourse, Wittgenstein writes in the famous paragraph 6.54: "My propositions serve as elucidations in the following way: anyone who understands me eventually recognizes them as nonsensical, when he has used them--as steps--to climb up beyond them.  (He must, so to speak, throw away the ladder after he has climbed up it.)  He must transcend these propositions, and then he will see the world aright."  It is significant that this paragraph clearly distinguishes between the meaningfulness of statements and their use; the metaphysical statements of the Tractatus are nonsensical, and yet they can be used by us to climb up beyond them.  True, the paragraph emphasizes that we must relinquish metaphysical discourse in order to attain the correct point of view on the world.  But it also implies that this point of view is not attainable unless metaphysical discourse is there in order to be relinquished.  Continuing Wittgenstein's metaphor, we may say that the ladder can be thrown away only after it has first been climbed.  The seeing of what shows itself in reality and in language occurs beyond the limits of metaphysical talk.  But the language of metaphysics gestures toward what shows itself even though what shows itself is not something within its province.

          And what metaphysician has ever denied this?  What metaphysician has ever contended that his language had anything other than an indicative function whose purpose is to lead us to the place from which we can see that which shows itself independently of our saying?  Wittgenstein does argue that traditional metaphysicians misunderstand the nature of their activity insofar as they interpret it as the meaningful construction of categorial statements.  But his iconoclasm goes only this deep.  Wittgensteinian metaphysics has a purpose which the great metaphysicians of the Western tradition would not hesitate to recognize as their own.  To traditional metaphysicians, the young Wittgenstein says, not "abandon your goal," but rather "be clear about what it is that you are doing."  In his view, the statements of metaphysics are meaningless in that they cannot occur in a purified logical sign-language.  But they have a function nevertheless.

 

2.

 

          It is somewhat surprising to find a discussion of solipsism in the later paragraphs of the Tractatus, for, at first glance, this discussion has very little to do with the philosophical treatment of logic that occupies most of the book.  How is what is now called the egocentric predicament or the problem of other minds related to the logical theory of language that Wittgenstein develops in his early work?  That there is indeed a relation between these two topics is indicated by the formulation of the solipsistic position with which Wittgenstein begins his treatment of solipsism in the 5.6's: "The limits of my language mean the limits of my world" (TLP 5.6).  Wittgenstein develops this remark in 5.62: "For what the solipsist means is quite correct; only it cannot be said, but makes itself manifest.  The world is my world: this is manifest in the fact that the limits of language (of that language which alone I understand) mean the limits of my world."

          These enigmatic lines suggest the general direction in which interpretation ought to proceed.  In these sentences, Wittgenstein clearly affirms the truth of the solipsistic position.  However, he intimates that, when the solipsist attempts to articulate what he wishes to say, he must do so in the form of a metaphysical statement which violates logical syntax.  The metaphysical stance of the solipsist is correct, but it cannot be expressed in language.  Instead, it "makes itself manifest."  Wittgenstein thus supplies a connection between the treatment of solipsism and the theory of language by means of his doctrine of showing and saying.  In so doing, he establishes solipsism as an illustrative case for the conception of metaphysics that emerges from this doctrine.  In this regard, it is important to note that Wittgenstein does not reject solipsistic metaphysics on the basis of the syntactical meaninglessness of solipsistic assertions.  On the contrary, he embraces such metaphysics by transferring the truth of solipsism from the realm of what can be said to the realm of what shows itself.  In revisionist fashion, Wittgenstein reformulates the metaphysical views of the solipsist in accordance with the theory of meaning and the doctrine of showing and saying developed in the Tractatus.

          When applying the doctrine of showing and saying to the problem of solipsism, Wittgenstein expands this doctrine beyond its earlier formulation in terms of the manifestation of ontological type in the relation between value and variable.  Such an expansion is needed because the solipsistic I is not a special ontological kind, but that without which there would be no ontological kinds at all.  There can be no I variable in logical notation in the way in which there can be a number variable, because all variables are variables for the I.  This is why, in 5.6 and 5.62 Wittgenstein speaks of the mineness of language and world, not as an item occurring in either of these domains, but rather as their limit.  If the solipsistic subject is capable of showing itself, then it does not do so as a mundane object or an essential pattern for mundane objects, nor does it do so as a linguistic variable or value of a variable.  Instead, it shows itself as a limit of the world which, nevertheless, characterizes the world as a whole insofar as it is my world; as a limit of language which characterizes language as a whole insofar as it is my language.

          The concept of the solipsistic subject as a limit is central to the remainder of Wittgenstein's treatment of solipsism in the 5.6's.  In these paragraphs, he identifies the solipsistic subject with what he calls the "metaphysical subject."  The metaphysical subject cannot be found in the world any more than the eye which is the source of the visual field can be found within the field.  The eye that we see in the mirror is not the source of the visual field, but an object in the field.  But the source of the field cannot be seen, and nothing in the field allows us to infer that it is seen by an eye (5.633).  Neither can the metaphysical subject apprehend itself as an object in the world or infer its presence from any object in the world.  Two statements which occur in the context of Wittgenstein's treatment of the eye metaphor in his Notebook entries of 1916, but which were not included in the Tractatus, help to shed some light upon the conception of the metaphysical subject as a limit: "The I is not an object" (7.8.16); "I objectively confront every object. But not the I" (11.8.16).  If the metaphysical I attempted to confront itself as an object, then it would still be the one who was doing the confronting and not the object confronted.  The I that I apprehend is not the I that does the apprehending.  The metaphysical I is nonobjectifiable, just as the seeing eye is not an item in the visual field.  And the metaphor can be extended further.  Just as the seeing eye sees not itself but the objective eye when it looks in the mirror, so does the metaphysical subject apprehend not itself but the psychophysical subject when it executes an act of reflection.  Thus, Wittgenstein warns against confusing the solipsistic subject with bogus candidates for this title: "The philosophical self is not the human being, not the human body, or the human soul, with which psychology deals, but rather the metaphysical subject, the limit of the world--not a part of it" (5.641).

          So far, it is not clear whether or not the solipsistic position embraced by Wittgenstein claims, not only that experience is always the experience of an I, but also that the "owner" of experience is unique.  For Wittgenstein's remarks about the nonobjectifiability of the I and the mineness of language and world could be interpreted as expressing a transcendental idealism in which the transcendental I is individuated in the form of a plurality of experiencing beings.  Given such an interpretation, Wittgenstein's assertion that language and world are always my language and my world would simply mean that linguistic and mundane experiences are necessarily owned by some subject, but not by the same subject.  The statement would be true in the sense that it could be correctly asserted by any of a plurality of existing subjects.  But Wittgenstein's solipsism is more radical than the position of the run-of-the-mill transcendental idealist.  Some of the entries in the Notebooks of 1915-16 clearly show that Wittgenstein believes the metaphysical subject to be, not only transcendental, but unique as well.

          The strongest evidence for Wittgenstein's belief in the uniqueness of the metaphysical subject occurs in the Notebook entries of 2.9.16.  Here, in three consecutive lines, Wittgenstein writes: "What has history to do with me?  Mine is the first and only world."  "I want to report how I found the world."  "What others in the world have told me about the world is a very small and incidental part of my experience of the world."  When taken in conjunction with Wittgenstein's contention that the world is necessarily a world for a subject, the second sentence of the first line ("Mine is the first and only world") entails the uniqueness of the solipsistic I.  For it would be impossible for two or more subjects truly to claim that its world was the only world.  Since the special character of the world, which consists in its being mine, derives from its relation to the subject whose world it is, the uniqueness of the world implies the uniqueness of the subject.  The other sentences in the above group merely underscore the uniqueness of the world and of the subject who is attached to it as its transcendental limit.  From an unequivocally solipsistic point of view, natural and human history have very little significance.  The world exists only insofar as I and I alone experience it.  Therefore it has no past which predates my arrival on the scene.  What I know of what supposedly occurred before my arrival, or of what has recently occurred outside of the sphere of my immediate awareness, comes to me by way of the reports of others.  But this information has significance only insofar as it is itself an item in my own experience.  It provides no evidential warrant for the independent existence of its subject matter.  What is important is how I find the world, with the "I" underlined in order to emphasize its uniqueness.

          Wittgenstein's remarks on history place us in a position to understand the ambiguous paragraph 6.43 of the Tractatus: "So too at death the world does not alter, but comes to an end."  Understood in one way, no one would challenge this assertion.  At death, the world ends for me in the sense that it is no longer the object of my awareness.  But Wittgenstein intends something more by this apparent truism.  Just as there can be no historical time which precedes the solipsist's birth, so can there be no historical time which succeeds his death.  The unique world vanishes without remnant with the vanishing of the unique self who is its source.  This is why Wittgenstein says in the final entry in his Notebooks dated 10.1.17: "If suicide is allowed then everything is allowed.  If anything is not allowed then suicide is not allowed.  This throws a light on the nature of ethics, for suicide is, so to speak, the elementary sin."  Suicide is the elementary sin because it is equivalent to world annihilation.

          Wittgenstein's conception of the solipsistic subject as a limit of the world results in his curious identification of solipsism with realism.  In paragraph 5.64, he writes: "Here it can be seen that solipsism, when its implications are followed out strictly, coincides with pure realism.  The self of solipsism shrinks to a point without extension, and there remains the reality co-ordinated with it."  The solipsistic subject is not a mundane object or configuration of objects.  Were we to enumerate everything that the world contains, the solipsistic subject would not be an item in this enumeration.  Neither can the self of solipsism function as the topic of an empirical proposition, since the only role of an empirical proposition is to depict mundane states of affairs.  And any metaphysical proposition about this self is nonsensical in that it violates the logical syntax of language.  The world and its contingent constituents, including the psychophysical I, remain the sole topics of meaningful discourse.  Wittgensteinian realism, however, is a realism of a peculiar sort.  It does not contend that the world exists independently of the knowing subject.  For Wittgenstein, precisely the opposite is true.  Wittgenstein's realism simply says that the subject, apart from whom the world has no existential foundation, cannot be spoken about meaningfully.

          After the self of solipsism shrinks to a point, the unique world remains.  But the uniqueness of the world is still a function of the fact that it is my world.  This is what the solipsistic metaphysician attempts to say.  Only, it cannot be meaningfully said.  Rather, it makes itself manifest.  And it does so in language insofar as the language with which the world is logically isomorphic is my language.  Admittedly, these contentions are somewhat obscure.  Nevertheless, they illustrate the fact that Wittgenstein's belief that metaphysical language is meaningless does not prevent him from embracing a metaphysical position.  For solipsistic metaphysics has a function; namely, to direct one's attention to the mineness of the world which shows itself in language, even though such mineness is incapable of serving as the topic of a meaningful proposition.

 

3.

 

          There is no doubt that the theory of language that Wittgenstein develops in his later work differs in important ways from the one set forth in the Tractatus.  In particular, The Blue and Brown Books and the Philosophical Investigations abandon the search for a crystalline world of logical structure deep below the surface of ordinary language.  In these works, Wittgenstein no longer insists upon determinacy of sense, with its demand for simple signs and simple objects, as a criterion of linguistic meaningfulness.  He now holds that linguistic meaning consists, not in the ability of language to depict complexes of objects, but rather in the ways language is used in the concrete context of actual forms of life.  Wittgenstein's contention that there is an indefinite multiplicity of language games which employ language for various purposes, especially underscores his new position that there is no single correct logical analysis of language.  Language functions in a variety of ways.  The task of philosophy is to describe this variety rather than to limit it in accordance with some artificial conception of what language really is beneath its deceptive surface.

          This much is indisputable.  But what should be a matter of dispute is the way in which Wittgenstein's later theory of language affects his view of metaphysics.  The orthodox opinion has it that the mature Wittgenstein resolves the ambivalences of the Tractatus' conception of metaphysics in favor of an unabashedly anti-metaphysical point of view.  Metaphysical statements are the result of confusions about the grammar which governs language in its actual use.  According to this interpretation, the job of the philosopher is to cure the metaphysician of his compulsion to make outlandish statements by showing him the grammatical sources of his confusion.  This cure is effected by returning metaphysical expressions to the everyday language games from which they originate, thereby demonstrating that such expressions are never used in the way in which the metaphysician wishes to use them.  The only legitimate function of philosophy is entirely therapeutic in character.  Or, to vary the metaphor, the philosopher is a sorcerer who breaks the spell that the apparent grammar of language casts over the metaphysician.

          There is no point in denying that Wittgenstein's later works contain much prima facie evidence in support of this interpretation.  In fact, it is possible to cull from these works a virtual catalogue of the sources of metaphysical confusion.  For example, Wittgenstein tells us that the occurrence in ordinary language of a substantive expression, such as "number," "length," "meaning," and so on, which does not refer to a physical object, induces the metaphysician to search in Platonizing fashion for a thing which corresponds to the substantive.  But the everyday use of such substantives has nothing to do with the supposition of corresponding things.  Or the occurrence of a generic concept may lead the metaphysician to look for a common essence shared by all instances falling under the concept, whereas the concept, in its actual employment, may only designate a loose network of overlapping resemblances among the multiplicity of instances.  Or the metaphysician may be led into error by superficial grammatical analogies between certain expressions; for example, since the expressions "writing" and "thinking" appear to be similar, he may look for a place in which thinking occurs analogous to the physical location in which writing occurs.  The list, of course, could be expanded.  But there is no need to do so, since I do not wish to argue against the contention that Wittgenstein enumerates several categories of metaphysical error.  Nor do I wish to deny that Wittgenstein holds that a primary task of philosophy is to uncover the sources of metaphysical error in order to eliminate them.  However, I will attempt to show that neither of these positions is equivalent to a rejection of the metaphysical enterprise.  Rather, they are warnings against doing metaphysics in a particular way.  They challenge, not the possibility of metaphysics, but its traditional self-understanding.

          In the preliminary studies for the Philosophical Investigations, posthumously published under the title of The Blue Book, Wittgenstein develops the metaphor of the "mental cramp" in attempting to elucidate the origin of metaphysical language.  The first appearance of the metaphor at the very outset of the book seems to provide supporting evidence for the anti-metaphysical interpretation of the later Wittgenstein's work.  The passage, already alluded to, reads: "The questions `What is length?', `What is meaning?', `What is the number one?' etc. produce in us a mental cramp.  We feel that we can't point to anything in reply to them and yet ought to point to something.  (We are up against one of the great sources of philosophical bewilderment: a substantive makes us look for a thing that corresponds to it.)"  According to the orthodox position, this passage aptly demonstrates Wittgenstein's therapeutic approach to metaphysical issues.  The mental cramp experienced by the metaphysician derives from a confusing aspect of the apparent grammar of ordinary language: the role that the substantive plays in the construction of a perceptible sentence as opposed to the function it performs in the context of its actual use.  This confusion is the source of the mental cramp which leads the metaphysician to make nonsensical statements.  When the therapist makes the metaphysician aware of the grammatical source of his cramp, the latter is cured of his desire to make metaphysical statements, and metaphysical error is thereby eliminated.  Against this interpretation, we should note first of all that, in the passage, Wittgenstein does not speak of metaphysical error, but rather of "philosophical bewilderment."  It is certainly true that, in Wittgenstein's view, philosophical bewilderment can be resolved in a way which produces metaphysical error; for example, the metaphysician may erroneously suppose that number, length, and so on really are thing-like entities.  But it is at least possible that Wittgenstein also believes that philosophical bewilderment can be resolved through the generation of a form of metaphysical language which is not embroiled in grammatical confusion.  Secondly, we should note that the orthodox interpretation does not really explain the rationale behind Wittgenstein's use of the mental cramp metaphor, except insofar as it incorporates it into the expanded metaphor of philosophical therapy.  But why does Wittgenstein choose specifically to identify the basic metaphysical ailment as a kind of cramp?

          The answer to this question is to be found further on in The Blue Book.  Since it occurs in the context of a discussion about solipsism, its full treatment must be postponed until we deal with Wittgenstein's mature view of solipsism in the next section of this article.  Nevertheless, at this point it is appropriate to consider the answer in general fashion.  Wittgenstein begins by telling us that there can be no common sense solution to a metaphysical problem.  For example, when the solipsist tells us that only his experiences are real, it is not to the point to answer him: "Then why are you telling us this since you don't believe that we really hear it?"  The solipsistic metaphysician is not out of his mind; he does not fail to see what everybody else sees.  Neither is his dispute with the common sense point of view similar to the dispute of the scientist who challenges ordinary opinion on the basis of a more subtle understanding of matters of fact.  The source of the strange statement of the solipsist is a kind of mental discomfort which differs toto coelo from the discomfort experienced when our curiosity about certain facts is not satisfied or when we cannot find a natural law capable of explaining our experience.  Rather, the discomfort experienced by the solipsist metaphysician, and, indeed, by all metaphysicians, has its origin in a fundamental dissatisfaction with the notation of ordinary language.  He experiences a mental cramp because he is constrained by the boundaries of that language in which he has grown up.  Wittgenstein writes:

Our ordinary language, which of all possible notations is the one which pervades all our life, holds our mind rigidly in one position, as it were, and in this position sometimes it feels cramped, having a desire for other positions as well.  Thus we sometimes wish for a notation which stresses a difference more strongly, makes it more obvious, than ordinary language does, or which in a particular case uses more closely similar forms of expression than our ordinary language.  Our mental cramp is loosened when we are shown the notations which fulfill these needs.  These needs can be of the greatest variety.

          What is striking about this passage is that its therapeutic answer to the mental cramp of the metaphysician does not consist in uncovering any source of metaphysical error.  On the contrary, what loosens the cramp is precisely the construction of a metaphysical notation.  What is more, the use of such a notation can be supported with many sorts of reasons depending upon the multiple needs which the notation satisfies.  Whether or not such reasons constitute a decisive justification of metaphysical discourse is a question which will concern us presently.  However this may be, according to the passage quoted above, metaphysical language is itself a form of therapy.

          In spite of the fact that Wittgenstein's elaboration upon the mental cramp metaphor expresses a positive view of the character of metaphysical language, it also contains a critique of a certain way of understanding what sort of enterprise the metaphysician is engaged in.  Wittgenstein contends that traditional metaphysicians misunderstand the nature of their own activity insofar as they interpret this activity in analogy with the enterprise of the natural sciences.  In the past, the metaphysician has believed himself to be making assertions about matters of fact or about a peculiar realm of occult objects which exist independently of the statements made about them.  For example, the dispute between mentalism and materialism has been taken to be about whether or not there really is a queer sort of entity called the "mind," or about what the relationship is between this entity and physical objects.  But all that the mentalist and materialist are really doing are offering alternative kinds of notation.  The dispute is about which form of notation ought to be adopted and not about what is or is not the case.

          Now, Wittgenstein clearly believes that the adoption of metaphysical notations can be supported with reasons.  By referring to the variety of needs which are left unsatisfied by ordinary language, we can defend our attachment to metaphysical modes of expression.  However, he just as clearly believes that the assertion of metaphysical statements cannot be justified in the way in which the assertion of scientific statements about lawful regularities can be justified insofar as they adequately explain a certain range of experiences of fact.  Although we can give reasons, and even good reasons, for engaging in metaphysical discourse, we cannot justify such discourse as though we were supplying a description of the world.  We will provide a more complete account of the notion of justification in the following section of this article.  At this point, it is sufficient to indicate that Wittgenstein's contention that metaphysics is unjustifiable in his special sense does not mean that it must be eliminated from the domain of significant language.  It only means that the metaphysician ought not to think that he is accomplishing something that, to use one of Wittgenstein's favorite metaphors, is not a permissible move in the metaphysical language game.

          Wittgenstein's later view of metaphysics differs from the one at work in the Tractatus even though, in neither case, does he reject metaphysics out of hand.  As I have argued, in the Tractatus, Wittgenstein holds that, while metaphysical statements are meaningless, they nevertheless have an important use.  This use consists in indicating what shows itself independently of language.  In his later work, however, Wittgenstein identifies the meaning of statements with their use.  Insofar as metaphysical statements have a use, he now considers them to be meaningful.  However, their use no longer consists in their ability to indicate what lies beyond the domain of language.  Metaphysical language does not, even indirectly, point us toward any feature of the world.  Metaphysical language has many different sorts of use which have their foundation in the needs expressed by our demand for metaphysical notations.  But these needs are overlooked if we conceive of metaphysics as a form of world description, or as an indirect allusion to what manifests itself either within the world or at its limit.  For the young Wittgenstein, metaphysics is meaningless, but illuminates reality.  For the mature Wittgenstein, metaphysics is meaningful, but tells us nothing about the nature of the real.

 

4.

 

          Just as the orthodox interpretation holds that the later Wittgenstein espouses an unequivocally anti-metaphysical position, so does it hold that he exhibits this anti-metaphysics by means of a refutation of solipsism.  According to the mainstream of Wittgensteinian discipleship, if Wittgenstein was indeed a solipsist in the period of the Tractatus, then at least he recognizes and corrects his youthful error in The Blue Book and the Philosophical Investigations.  I will try to show that this interpretation rests on a misunderstanding of the argument structure of Wittgenstein's later treatment of solipsism.  What is taken to be a refutation of solipsism is, in actuality, an attempt to restate the solipsistic position in such a way that it does not violate the parameters of significant discourse as indicated by Wittgenstein's mature theory of meaning.  In order to demonstrate that this is so, it will be necessary to follow carefully the actual progression of the only piece of continuous argumentation on the topic of solipsism by Wittgenstein that we possess, one which is to be found in The Blue Book.  The need for a careful consideration of the unbroken line of argument must be stressed because it is so seldom recognized.  For the most part, the orthodox interpretation goes astray by neglecting the special character of the progressive development of Wittgenstein's argument and by reducing this argument to a series of unconnected and purportedly anti-solipsistic positions.  Once they are situated in the purposive context of Wittgenstein's entire treatment, his remarks on solipsism can no longer be construed as a refutation of this metaphysical position.

          Wittgenstein's treatment begins by considering the common sense response, the response of naive realism, to an initial statement of the solipsistic position: "I can only know that I have personal experiences, not that anyone else has."  To this statement, the realist might reply that, although we don't know that the other person has a pain, for example, we certainly believe that he does when we pity him.  But this answer does not help the solipsist since he is not concerned with prephilosophical or premetaphysical belief.  The fact that a solipsist pities does not count as evidence against his philosophical views since his pity, as it were, occurs on a different level than his solipsism.

          Recognizing the futility of his first response, the common sense person might adopt a more philosophical approach, and assert that there is no difficulty in imagining that I have what someone else has.  But, according to Wittgenstein, this new tact confuses the grammar involved in saying that I have an experience with the grammar involved in saying that I have a physical object.  In order to exhibit the syntactical confusion of the second common sense response, Wittgenstein considers the grammatical difference between the two statements: "A has a gold tooth," and "A has toothache."  "A has a gold tooth" means that the gold tooth is in A's mouth, which accounts for the fact that I am not able to see it.  This case is not analogous to the one in which I say that I am not able to feel A's toothache because it is in his mouth.  According to Wittgenstein, it is quite conceivable that I might feel pain in a tooth in another person's mouth.  The solipsist does not wish to deny this.  What he does wish to deny is that I can feel another person's pain in his tooth.  Unless we become familiar with the idea of having pain in another person's body, we are liable to confuse the metaphysical proposition, "I can't feel his pain," with the experiential proposition, "I can't (haven't as a rule) pains in another person's tooth," a proposition which is similar to the experiential proposition, "I can't know whether there is a gold tooth in A's mouth, because his mouth is closed."

          Wittgenstein's strange contention that it is conceivable that we might experience pain in another person's body rests on his theory that there are a multiplicity of independent criteria that count as evidence for the locality of pain.  Various tactile, kinaesthetic, and visual sensations, as well as the statements of others, all serve as evidence for the locality of pain, while no single bit of evidence allows us to locate a pain in a way which is not subject to error.  For example, when I feel pain in my tooth, the pain has its own peculiar tactile and kinaesthetic neighborhood.  When I attempt to locate the pain with my finger by touching my tooth, the tactile and kinaesthetic sensations involved in moving my finger the appropriate distance and in the appropriate direction, the visual experience of seeing my finger touch my tooth while looking in a mirror, another person's statement that I am touching my tooth, and so on are all criteria for the location of pain in my tooth.  However, by itself, none of these criteria insures the truth of the proposition, "There is pain in my tooth."  According to Wittgenstein, the reason why it is difficult to conceive of experiencing pain in another person's body is that the sense experiences involved in locating pain regularly coincide in such a fashion that we normally locate pain in our own bodies.  But such coincidence is no more than a contingent, experiential nexus of evidence.  It is possible that I might have the sensation of toothache along with the tactile and kinaesthetic experiences normally associated with seeing my finger travel from my chin to my tooth, but correlated with the visual experience of seeing my finger travel from another person's chin to his tooth.  In such a case, it would be perfectly reasonable to say that I have toothache in another person's tooth.  The solipsist can admit this possibility without abandoning his position.  For he would then say, "I may have toothache in another person's tooth, but not his toothache."

          As I have already suggested, the basic point that Wittgenstein wishes to make in establishing the possibility of having pain in another person's body is that metaphysical propositions differ fundamentally from empirical propositions.  The common sense realist errs in that he confuses the solipsist's assertion of metaphysical necessity with the assertion of empirical "necessity," that is to say, regularity of experiential coincidence.  For example, the realist might say, "I grant that you can't know when A has pain, you can only conjecture it."  But, Wittgenstein tells us, when the realist admits the impossibility of knowing when A has pain, he thinks of a case analogous to the one in which we can't know that another person has a gold tooth in his mouth because his mouth is shut.  With respect to the case of the gold tooth, it makes sense to speak of conjecture because it also makes sense to speak of seeing the tooth when the mouth is open.  But when we say that we can't know whether the other person has pain, we don't mean that as a matter of fact we don't know, but rather that it doesn't make sense to say that we know.  Since it doesn't make sense to say that we know, it also doesn't make sense to say that we conjecture.  It isn't that knowing that another has pain is a goal that we cannot reach so that we must be satisfied with conjecturing, but rather that "there is no goal in this game."  This case is like the one in which we say that "You can't reach the end of the series of cardinal numbers."  In both instances, we are not making a statement about a fact of human frailty, but rather about a linguistic convention. 

          According to Wittgenstein, by understanding what is wrong with the common sense response to solipsism, we are brought to an important recognition: the use of modal expressions such as "must," "can," or "can't" in metaphysical propositions exhibits a grammatical rule which has no more than a conventional status.  Recognizing the grammatical character of metaphysical modality destroys the apparent similarity between metaphysical and empirical propositions.  Metaphysical ways of speaking grow out of discontentment with our ordinary grammar.  The metaphysician, therefore, does not engage in any dispute about matters of fact, and, to the extent that he believes himself to be so engaged, he fails to understand what metaphysics is.  Instead, he offers a new form of expression which has absolutely nothing to do with what is or is not the case.  For example, when I say, "I can't feel the Other's pain because my pain is my pain and his is his," I am simply making a grammatical statement about the use of the phrase "the same pain."  I am saying that I don't wish to apply the phrase "I have his pain," or "He and I have the same pain," and instead, perhaps, will apply a phrase such as "His pain is exactly like mine."  In so saying, I am merely expressing a determination to use the phrase "the same pain" in a certain way, and am not asserting anything about the nature of pain.

          The common sense realist fails to understand the nonempirical character of solipsism when he thinks that solipsism can be refuted by adducing falsifying matters of fact.  To the solipsist's assertion, "Only my pain is real," the realist might reply, "Surely there is evidence that other people are not pretending when they say that they have pain."  But the solipsist does not claim to have discovered that others who said they had pain were cheating.  What he objects to is the application of this expression to the case of pretending.  That is to say, he objects to a feature of our ordinary notation.  To be sure, the solipsist may not be aware of the fact that he is proposing a convention.  According to Wittgenstein, this lack of awareness is precisely what is wrong with traditional solipsism.  He criticizes the traditional self-understanding of solipsism by means of an intriguing metaphor:

[The solipsist] sees a way of dividing the country different from the one used on the ordinary map.  He feels tempted, say, to use the name "Devonshire" not for the county with its conventional boundary, but for a region differently bounded.  He could express this by saying: "Isn't it absurd to make this a county, to draw the boundaries here?"  But what he says is: "The real Devonshire is this."  We could answer: "What you want is only a new notation, and by a new notation no facts of geography are changed.

To the extent that the traditional solipsist sees himself as finding the real Devonshire, he becomes involved in a futile cartographical dispute with the common sense realist.  But, Wittgenstein suggests, if the solipsist understands the purely conventional and grammatical character of his own position, then he places himself beyond the polemical reach of the naive realism of common sense.

          Once we recognize that a properly formulated solipsism does not disagree with common sense realism about any practical issue concerning our everyday experience of the world, then we are also led to recognize that there is no ground for denying the solipsist the right to employ his peculiar notation.  Solipsistic language can be used in such a fashion that it is able to perform all of the functions performed by the ordinary language of common sense realism.  Since Wittgenstein now identifies the meaning of an expression with its use, the unhindered functioning of solipsistic language entails its meaningfulness.  The solipsist recommends the adoption of a notation in which only his own experiences count as real.  Expressions such as "A has real toothache" where A is not the solipsist are to be excluded from the notation just as the rules of chess exclude the possibility of a pawn making a knight's move.  Since the experiences of others are not considered to be real experiences in this symbolism, reference to the solipsist as one experiencer among others is superfluous.  Instead of saying, "Smith (the solipsist) has toothache," we might as well say, "There is real toothache."  We can even imagine others acquiescing to the solipsist's linguistic recommendation.  Instead of saying, "L.W. sees so-and-so," they will say, "So-and-so is really seen."  Whether or not others can be gotten to adopt this despotic symbolism is a political, not a philosophical question.  On purely philosophical grounds, Wittgenstein asks: "And why shouldn't we grant [the solipsist] this notation?"  Further on in The Blue Book, he answers: "There is...no objection to adopting a symbolism in which a certain person always or temporarily holds an exceptional place."

          The solipsist offers his notation because he is dissatisfied with our ordinary language which, as Wittgenstein says, tends to hold the mind rigidly in one position.  The desire to move into another position is the desire to satisfy needs left unsatisfied by ordinary language.  By elucidating these needs and showing how they are fulfilled by a change of symbolism, the solipsist can defend the notation recommended by him over the objections of common sense.  In The Blue Book, Wittgenstein hints at one need taken care of by the language of solipsism.  He suggests that one of the solipsist's quarrels with ordinary language is that it fails to stress the distinction between expressions involving the possession of physical objects and expressions involving the "possession" of experiences.  As we have seen, the common sense realist is led into philosophical error by the apparent analogy between the statements, "A has a gold tooth," and "A has toothache."  His contention that it is possible to conjecture that A has toothache but not to know that A has toothache rests on the mistaken supposition that toothache is like the gold tooth which cannot be seen because the mouth is closed.  According to this position, what prohibits us from knowing that A has toothache is a sort of quasi-empirical, which is to say, metaphysical barrier that prevents us from achieving direct contact with A's toothache, but does not prevent us from conjecturing the presence of A's experience of toothache say, perhaps, from A's swollen cheek.  The solipsist agrees that a swollen cheek is evidence for the medical condition of toothache.  What he denies is that a swollen cheek is evidence for A's toothache-experience, and he does so on the perfectly reasonable linguistic grounds that where knowledge in the sense of direct contact with an experience is excluded, conjecturing must also be excluded.  Now, Wittgenstein certainly does not provide us with a systematic defense of solipsism in The Blue Book.  But, from this example, it is easy to see how such a defense would proceed.  The solipsist would begin by enumerating the misleading analogies, inappropriate associations, and failures to stress important grammatical differences implicit in ordinary language, and would show that no such problems arise in a solipsistic notation.

          In Wittgenstein's view, then, solipsism is both a meaningful and a defensible use of language.  But it is not a justifiable use of language.  What Wittgenstein means by this can be seen from the context in which he makes the pronouncement.  After telling us that there can be no objection to adopting a notation in which a certain person is conferred with the honor that only his experiences are counted as real, Wittgenstein says: "What, however, is wrong, is to think that I can justify this choice of notation."  In an oblique reference to certain of the doctrines of the Tractatus, Wittgenstein argues that when I say that only what I experience is real, I do not intend this "I" to refer to the psychophysical subject.  For, even if my behavior, my character, and my body were to change abruptly, I would want to say that this has nothing to do with the I who is the subject of experiences.  I may try to justify my solipsism by contending that the psychophysical subject that others know, for example, as L.W. is the "vessel of life," the contingent "seat of that which really lives," and which is the true subject of experiences.  But, in arguing in this fashion, I have given a quasi-empirical description of the world.  I have said that, in the world, in this body with its associated collection of behavior patterns and characterological traits, there dwells an occult entity who is the genuine and unique subject.  However, a solipsist who understands what his metaphysical position is all about doesn't wish to say this at all.  He recognizes that he is not engaging in a form of world description, but rather is recommending a notation.  When Wittgenstein says that solipsism cannot be justified, he means that it cannot be argued for on the grounds that it is a superior account of matters of fact.  Solipsism does not tell us anything new about the way the world is, nor does it tell us that our ordinary way of speaking about the world is false.  The position of the solipsist can be defended, but it must be defended on purely linguistic grounds.

          The remainder of Wittgenstein's treatment of solipsism in The Blue Book is occupied with demonstrating that certain traditional ways of expressing solipsism are inadequate insofar as they are misled by ordinary grammar into attempting a quasi-empirical self-justification.  I will consider only one of these demonstrations since the general principle involved is the same for all of them.  The solipsist may attempt to express his position by saying, "This is what's really seen," while gesturing in an expansive fashion so as to indicate the entire visual field, but not attempting to indicate any particular item in the field.  He could just as well perform this gesture mentally since the statement is supposed to have a primitive meaning for him, but to be incapable of being understood by anyone else.  But, according to Wittgenstein, when the statement, "This is what's really seen," is used in conjunction with a putative pointing gesture toward the entire visual field, the statement is meaningless.  The solipsist's statement does not serve to indicate one thing in contrast with another and, therefore, indicates nothing.  It makes no sense to say that I am pointing at something when what I am pointing at cannot be distinguished from anything else.  My solipsistic pointing gesture has no neighbor; since it does not realize one possibility among others, it lacks the articulation which is a condition of the meaningful use of language.  By attempting illegitimately to extend the ordinary grammar of ostension, the solipsist believes himself to be identifying a quasi-empirical structure: the unique visual field.  But, in so doing, he robs his statement of any concrete use.

          Let us summarize the results of our interpretation of Wittgenstein's treatment of solipsism in The Blue Book.  The treatment may be broadly analyzed into three sequential, but somewhat overlapping, phases: 1) it defends the solipsistic position against the anti-solipsistic arguments of common sense; 2) it tries to show that there can be no objection to adopting a solipsistic notation, but that such an adoption is unjustifiable; and 3) it argues that traditional ways of expressing the solipsistic viewpoint violate identifiable criteria of meaningfulness.  The net result has a negative and a positive aspect: one the one hand, solipsism is not capable of being justified in Wittgenstein's special sense, or rigorously demonstrated in any traditional sense; on the other hand, a reformulated solipsism can be meaningfully expressed nevertheless, and also supported with reasons insofar as it satisfies certain needs left unsatisfied by our ordinary notation.

          Is there anything in the Philosophical Investigations to suggest that Wittgenstein abandoned the position described above during the ten year period that separates the dictation of The Blue Book to his students at Cambridge from his completion of the Investigations?  On the contrary, evidence in the Investigations indicates that its view of solipsism is identical with that of The Blue Book.  What makes it difficult to find this evidence on a cursory reading is the fragmented, aphoristic way in which Wittgenstein wrote the Investigations.  The remarks on solipsism in this difficult work are not situated in the context of a continuous sequence of argumentation.  In order to understand these remarks, it is necessary to supply their context, and their context is to be found in The Blue Book.  The Blue Book is the real key to understanding Wittgenstein's later view of solipsism.  All that the interpreter can do is show that the paragraphs of the Investigations which deal with solipsism are fully intelligible only against the contextual background supplied by The Blue Book.

          Many of the Investigation's remarks on solipsism belong to the final phase of The Blue Book's treatment of the topic; that is to say, they advance criticisms of traditional attempts to express the solipsistic point of view.  As is the case with their Blue Book counterparts, it is erroneous to interpret these passages as though they were refutations of solipsism.  It will be sufficient to consider only one of the passages.  In paragraph 398 of the Investigations, Wittgenstein writes: "`But when I imagine something, or even actually see objects, I have got something which my neighbor has not.'  -I understand you.  You want to look about you and say: `At any rate only I have got THIS.' -What are these words for?  They serve no purpose."  Anyone familiar with The Blue Book can see that Wittgenstein's argument against global ostensive definition is in the background of this passage.  But he goes on in the same paragraph to apply the principle that expressions have meaning only when they are articulated in relation to possible counter-expressions even more broadly.  To say that, "Only I have got THIS," is meaningless, not only because it fails to indicate any particular item of experience, but also because it fails to indicate any particular subject of experience.  I am not saying that, of all the subjects in the world, only one has got THIS, but rather that there is only one subject in the world capable of possessing THIS.  But if it is meaningless to say that other people have what I have, then it is also meaningless to say that I have what I have.  The concept of unique ownership of experience is as inarticulate, and hence nonsensical, as the attempt to point to the unique visual field.  Now, this argument is a refutation, not of solipsism, but rather of the misleading notion of experiential ownership.  As we have seen, one of the principal motives of a properly formulated solipsism is recognition of the fact that ordinary language suggests an inappropriate analogy between "possession" of experiences and possession of physical objects.  According to Wittgenstein, it is the solipsist, not the common sense realist who understands that one does not have a toothache in the way in which one has a gold tooth.

          More direct support for the thesis that Wittgenstein does not abandon solipsism in the Investigations comes from those passages which correspond to the second phase of The Blue Book argument, that is to say, those which suggest that there can be no objection to adopting a solipsistic notation even though such an adoption cannot be justified.  Consider paragraph 403 of the Investigations:

          ...If I were to reserve the word "pain" solely for what I had hitherto called "my pain," and others "L.W.'s pain," I should do other people no injustice, so long as a notation were provided in which the loss of the word "pain" in other connections were somehow supplied.  Other people would still be pitied, treated by doctors and so on.  It would, of course, be no objection to this mode of expression to say: "But look here, other people have just the same as you!"

          But what should I gain from this new kind of account?  Nothing.  But after all neither does the solipsist want any practical advantage when he advances his view!

Thus, according to Wittgenstein, as long as the language of solipsism is able to perform all of the functions performed by ordinary language, there can be no objection to speaking the solipsist's idiom.  But neither do we gain any practical advantage in so speaking.  We do not discover any new fact capable of expanding the range of our practical possibilities, and so are unable to do anything more with our solipsistic idiom than what ordinary language does.

          In the Investigations as in The Blue Book, Wittgenstein contends that solipsism cannot be justified.  But he also continues to distinguish between justification and those reasons which may be legitimately adduced in support of solipsism.  In paragraph 289 of the Investigations, he writes: "To use a word without a justification does not mean to use it without a right."  If the solipsist attempts to justify his mode of expression, then he mistakes a grammatical convention for a "quasi-physical phenomenon."  In so doing he misses the real significance of his linguistic recommendation, namely that he has discovered "a new way of looking at things." (PI 401)  Solipsism is compatible with our ordinary experience of the world.  It does not challenge any commonly recognized fact about this experience, nor does it claim to have discovered a new, previously unnoticed fact.  But it does represent a shift in our orientation toward experience, a new way of looking at the world which, in Wittgenstein's view, is both intriguing and defensible.  These remarks of the Investigations must be understood in connection with the mental cramp metaphor developed in The Blue Book.  Solipsism as one species of metaphysics is a way of bursting out of the confining perspective imposed upon us by the grammar of ordinary language.  It is a new way of regarding things which does not alter any aspect of the things themselves, but which, nevertheless, is a perfectly legitimate expression of the human desire for freedom of conceptual movement.

 

5.

 

          In examining Wittgenstein's approach to metaphysics, I have challenged the mainstream of Wittgensteinian scholarship by attempting to show that, in The Blue Book and the Philosophical Investigations as well as the Tractatus, Wittgenstein 1) is a revisionist metaphysician rather than an anti-metaphysician, and 2) embraces a solipsistic metaphysical position.  This interpretation has implications that extend beyond the narrow confines of Wittgenstein exegesis.  For, if I am right, then there is a fundamental distortion in the way in which twentieth century philosophy recounts its own history, a distortion which concerns the nature and possibility of discourse about reality.  In both the Anglo-American and Continental traditions, the philosophy of our own period often assumes a hostile stance with respect to the older metaphysical heritage.  The positivistic effort to consign metaphysics to nonsensicality, the distrust of large metaphysical issues in ordinary language philosophy, and the Heideggerian and Derridean "deconstruction" of the Western metaphysical tradition all end by dividing philosophy into a metaphysical past and a post-metaphysical present.  Since the publication of the Tractatus, Wittgenstein's work has been a major resource in our rush toward post-metaphysical self-definition.  However, if it is true that Wittgenstein is himself a revisionist metaphysician, then perhaps what characterizes the contemporary period is not so much its anti-metaphysical posturing as its surreptitious practice of a new mode of metaphysical thinking.  As we shall see, in the context of this broad historical issue, Wittgenstein's mature conventionalism is ultimately more significant than his early predilection for the ineffable.

          It is true that, in the Tractatus, Wittgenstein adopts a general orientation toward the problem of metaphysics that is also characteristic of his later work.  Metaphysical issues are expressed in a linguistic medium.  The possibility of expressing such insights is equivalent to the possibility of employing metaphysical language in the way in which the metaphysician wishes to employ it.  Wittgenstein is thus engaged in a quasi-transcendental inquiry into the conditions of the possibility of metaphysics, similar in some respects to the investigation that Kant launches in the first Critique, with the difference that Wittgenstein, unlike Kant, focuses the discussion on the theme of language.  Can metaphysics say what it claims to say, or in the words of the Tractatus, can it gesture toward that which it claims to show?  However, although Wittgenstein's basic question does not change over the course of his philosophical career, there is an important transformation in the framework within which he answers this question.  In the Tractatus, Wittgenstein assumes that language has a single correct logical analysis.  Conditions of linguistic meaningfulness are perspicuously evident in a properly formulated Begriffschrift.  Since a metaphysical way of speaking violates these conditions, metaphysical language is meaningless.  But Wittgenstein also holds that language depicts what is independently the case, and that it shows what is independently essential to what is the case.  Regardless of what we say about them, and regardless of the way in which we say it, the world and its essential structures are what they are.  The extralinguistic autonomy of what is necessarily involved in there being a world establishes the usefulness, and hence, the legitimacy of metaphysical language.  While meaningless, metaphysics is able to point toward that which makes itself manifest independently of language.

          Wittgenstein jettisons both of these theses in his later work.  He quite convincingly demonstrates the absurdity of the first of the Tractatus' presuppositions.  The logician has no privileged standpoint from which to determine which uses of language are acceptable and which are not.  The activity of logical analysis is itself parasitic upon pre-existent modes of expression, pre-existent language games that go on about their business heedless of what the logician has to say about them.  Whatever verdict may come down from the logical court of judgment, a notation is meaningful so long as it has a use.  The anti-dogmatic pluralism which emerges from this recognition is both admirable and philosophically compelling, although we may wish to stress that, for certain purposes, some forms of expression are more adequate than others.  But Wittgenstein's rejection of the second of the Tractatus' presuppositions is both questionable and more deeply, if less visibly, influential.

          In The Blue Book and the Philosophical Investigations just as in the Tractatus, Wittgenstein does not reject metaphysics straightforwardly; rather, he revises it in accordance with his conception of the nature of linguistic meaning.  But, in his later work, Wittgenstein administers the kiss of death in the course of this revision.  More precisely, metaphysics survives, but at the price of relinquishing its claim to illuminate our situation within the world, of detaching its utterances from all contact with reality.  The gears of metaphysical language continue to turn, but they no longer engage anything other than themselves.  It is at this point that Wittgenstein's solipsism reaches it final and subtlest expression: not in the clear assertion of the uniqueness of his own subjectivity, but in the closure of metaphysical speech.  The ancient desire to achieve contact with what lies beyond the limits of talk is radically frustrated.  The young Wittgenstein worried that the world would vanish at his death.  The mature Wittgenstein realized this annihilation in advance of his own disappearance from the scene.

          Wittgenstein remains an intriguing figure in the history of twentieth century philosophy because his most profound and rarefied work has all the dimensions of an intense inner spiritual crisis.  In light of his continuing adherence to solipsism, and the final dissociation of his insights from the goal of articulating an extralinguistic world, Wittgenstein's frenzied vigils to Russell's room in the middle of the night, Moore's belief that Wittgenstein was on the edge of madness following the completion of the Tractatus, and a number of other well-known idiosyncrasies take on a larger meaning.  But what Wittgenstein lived as a personal crisis, we now confront as a philosophical tendency which is becoming increasingly popular on both sides of the Channel.  The withdrawal of the world, the conventionalist interpretation of meaning, and the insular autonomy of language are dominant themes in theoretical currents as geographically disparate as Rorty's critique of the "mirror of nature" and Lyotard's postmodernism.  I hope that the present article contributes to our understanding of the near history of this fashionable philosophical nihilism.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

NOTES

 

 


 

          . I exclude from this category a very atypical student of

Wittgenstein's, J.N. Findlay.  See especially his "My encounters with Wittgenstein" in Studies in the Philosophy of J.N. Findlay, ed. Robert S. Cohen, Richard M. Martin, and Merold Westphal (Albany: State University of New York Press, 1985), pp. 52-69, and his Wittgenstein: A Critique (London: Routledge and Kegan Paul, 1984).  I would like to thank Professor Findlay for commenting upon the present article.

          . For examples, see Garth Hallett, A Companion to Wittgenstein's "Philosophical Investigations" (Ithaca: Cornell University Press, 1977), P.M.S. Hacker, Insight and Illusion (London: Oxford University Press, 1972), and Anthony Kenny, Wittgenstein (Cambridge, Massachusetts: Harvard University Press, 1973).

          . Citations of the Tractatus are from Ludwig Wittgenstein, Tractatus Logico-Philosophicus, trans. D. F. Pears and B.F. McGuinness (London: Routledge and Kegan Paul, 1961).

          . On this point see Hacker, Insight and Illusion, p. 23.  In general, Hacker's treatment of the Tractatus is extremely illuminating.  My own exposition is greatly indebted to it.

          . Citations of the Notebooks are from Ludwig Wittgenstein, Notebooks, 1914-1916, ed. and trans. G.E.M. Anscombe (New York: Harper and Row, 1961).

          . Ludwig Wittgenstein, The Blue Book in The Blue and Brown Books (New York: Harper and Row, 1965), p. 1.

          . Ibid., p. 17.

          . Ibid., p. 7.

          . Ibid.

          . Ibid., pp. 58-60.

          . Ibid., p. 59.

          . Ibid., p. 55.

          . Ibid., p. 66.

          . Ibid., pp. 46-74.

          . Ibid., p. 48.

          . Ibid.

          . Ibid., pp. 49-53.

          . Ibid., p. 54.

          . Ibid.

          . Ibid.

          . Ibid., p. 55.

          . Ibid., p. 54.

          . Ibid., p. 57.

          . Ibid.

          . Ibid.

          . Ibid., p. 66.

          . Ibid., p. 59.

          . Ibid., p. 66.

          . Ibid.

          . Ibid., p. 63.

          . Ibid., p. 65.

          . Ibid., p. 66.

          . Ibid., p. 71.

          . Citations of the Philosophical Investigations are from Ludwig Wittgenstein, Philosophical Investigations, trans. G.E.M. Anscombe (New York: The Macmillan Company, 1958).

          . See Richard Rorty, Philosophy and the Mirror of Nature (Princeton: Princeton University Press, 1979), and Jean-Francois Lyotard, The Postmodern Condition: A Report on Knowledge, trans. Geoff Bennington and Brian Massumi (Minneapolis: University of Minnesota Press, 1984).

          . In the sense in which Nietzsche characterizes "active" nihilism as "the denial of a truthful world" in favor of a perspectivist affirmation of multiple interpretations.  Friedrich Nietzsche, The Will to Power, trans. Walter Kaufmann and R.J. Hollingdale (New York: Vintage Books, 1967), pp. 14-15.