Our attempts to deal with “the problems of philosophy” go characteristically wrong because we don’t understand “the logic of our language.” There are limits to language and these delegitimize the endeavor to advance philosophical theories. If we are to resolve our philosophical problems, we must go about it in some other manner, Wittgenstein writes in the preface of his Tractatus, and thus he concludes: “Whereof one cannot speak, thereof one must be silent.” How did he come to his extraordinary call to philosophical silence? How did he seek to justify it? Should we agree with it? We need a critical assessment of his critique of philosophy.
Wittgenstein’s critique of philosophy evolved over the course of his philosophical life. His reasons for questioning the possibility of philosophical theorizing changed from being based on considerations concerning the logical structure of fully analyzed propositions to ones drawing on “the grammar” of ordinary language. Instead of speaking of language in the singular, he came to talk of diverse languages and their varying uses and limitations. To map all this is a major undertaking. The place to begin with is the Tractatus and Wittgenstein’s contest with his mentor and friend Bertrand Russell over what can be said in philosophy. A short look at Wittgenstein’s later way of conceiving the limits of language will indicate how far his thinking eventually came to diverge from the Tractatus.
1.
It was in late June of 1919 that Bertrand Russell got his first look at Wittgenstein’s Tractatus. A few weeks later he wrote to the author that he had now read the book twice and carefully at that. There were, however, important points in it he did not understand and the book was obscure in places because of its brevity. To mitigate such criticisms, he added in carefully nuanced words: “I am sure you are right in thinking the book of first-class importance.” The work had, in fact, convinced him that logical propositions were tautologies and not true in the way in which substantial propositions are. That was a major concession on Russell’s part who, a year earlier, had still spoken of “logical facts” on a par with the empirical ones.
But that was not enough to mitigate Wittgenstein’s displeasure. He wrote back that it was “VERY hard not to be understood by a single soul!” Frege had already told him that he did not understand a single word of the Tractatus. And now Russell was showing that he had failed to grasp the “main contention” of the book “to which the whole business of logical prop[osition]s is only a corollary.” Russell’s concession on the questions of logical propositions was, thus, by itself from Wittgenstein’s point of view only of minor significance. The main point of the book was “the theory of what can be expressed (gesagt) by prop[osition]s – i.e. by language – (and, which comes to the same, what can be thought) and what can not be expressed by prop[osition]s but only be shown (gezeigt); which, I believe, is the cardinal problem of philosophy.” How could Russell have missed this point, given the explicit words of Wittgenstein’s preface?
Russell’s silence was due, in fact, to something else: his lack of confidence in the project of drawing limits to language. He subsequently wrote in the “Introduction” the publishers wanted for the Tractatus: “What causes hesitation is the fact that, after all, Mr Wittgenstein manages to say a good deal about what cannot be said, thus suggesting to the sceptical reader that possibly there may be some loophole through a hierarchy of languages, or by some other exit.” That comment left Wittgenstein, in turn, so dismayed that he initially refused to publish his book with Russell’s introductory essay. The whole point of the Tractatus had been, after all, to show that there was no loophole of the sort that Russell envisaged. And this, he had concluded, made manifest that there could be no metaphysical, logical, ethical, or aesthetic theories. The philosophers were trying to construct theories on the basis of something that could only be shown and there was inevitably a discrepancy between what showed itself and what the theorists made of it. Russell was clearly open to this kind of criticism.
Wittgenstein’s “main contention” in the Tractatus consisted, in fact, of three separate, though related points. The first was the need to distinguish between what signs say and what they show. The second was the claim that whatever is shown by a sign cannot be said either by that sign or in any other language. And the third was that this limit to language prevented any kind of philosophical theorizing. Each one of those points needs attention.
2.
Wittgenstein had engaged with Russell in vigorous debate on questions of logic and metaphysics since his arrival in Cambridge in 1911. At Russell’s request, he had finally compiled a set of notes in the fall of 1913 in which he laid out his considered views on these matters. “Philosophy […] consists of logic and metaphysics,” he had written, “the former its basis.” And this was so because “[p]hilosophy is the doctrine of the logical form of scientific propositions.” The 1913 notes are an important document for the development of Wittgenstein’s thinking. But they contain as yet few signs of what was to become “the main contention” of Tractatus. Wittgenstein did not yet speak in them of a distinction between saying and showing and hence also did not speak of their supposed incompatibility, nor did he deny outright the possibility of logical and metaphysical theorizing.
Those three claims made their first appearance, however, six months later in a series of notes he dictated to G. E. Moore in April of 1914. Wittgenstein had by then withdrawn to Norway and Russell was in the US. There was thus enough distance between them for Wittgenstein to follow his own independent course of thinking. But, on Russell’s return from the US in June 1914, he was quick to alert him to the new work he had done. “Many things in it are new,” he wrote, which Moore would be able to explain. Around Christmas of 1914, Wittgenstein, now a soldier in Austrian army, wrote once more to Russell expressing his perplexity at Moore’s apparent inability to explain those notes. “Were you able to get anything at all out of his notes?” he asked Russell; he feared that the answer was, No. Perhaps he would have to return to England after the war to explain his work in person. Given Wittgenstein’s conviction that Russell had failed to understand his new way of thinking in 1914, he was probably not altogether surprised to discover, five years later, that he also failed to get hold of the main contention of the Tractatus.
The first point Wittgenstein had been eager to communicate in the Moore notes in 1914 was his rejection of Russell’s conception of logic and “logical propositions”. Russell had argued in his 1913 Theory of Knowledge, as Wittgenstein knew, that logic had its own subject-matter in such notions as “particulars, universals, relations, dual complexes, predicates.” These were the “data” of logic with which we can be said to have some kind of direct acquaintance. Every logical notion, Russell added, “involves a summum genus, and results from a generalization which has been carried to its utmost limit.” This was, indeed, “a touchstone by which logical propositions may be distinguished from all others.” In sharp contrast, Wittgenstein now advanced the bold claim that logical propositions were, in fact, empty tautologies. They were really only “so-called propositions” because they actually said nothing. This did not, however, make them entirely useless. For while they said nothing, they exhibited logical properties of the language and, insofar as language mirrors the world, they also manifested its logical features. The characterization of logical propositions as tautologies called thus for a distinction between saying and showing, and for the realization that these were incompatible functions of language. The Moore notes begin accordingly with the words: “Logical so-called propositions shew [the] logical properties of language and therefore of [the] Universe, but say nothing.” Wittgenstein went on to argue in them that we can “see” that a proposition is a tautology and can therefore also “see” the validity of logical inferences. “E.g., take φa, φa ⊃ ψa, ψa. By merely looking at these three, I can see that 3 follows from 1 and 2; i. e. I can see what is called the truth of a logical proposition, namely of [the] proposition φa . φa ⊃ ψa : ⊃ : ψa.”
The seeing in question is, of course, more than plain visual perception. In order to “see” that the proposition Wittgenstein mentions is a tautology one must understand its syntax (e. g., the role of the punctuation marks). One must also know the function of the letters “φ”, “ψ”, and “a” and one must know that the hook stands for the relation of material implication. One must, in a nutshell, recognize “the logical form” of the proposition (to use Wittgenstein’s term) to see that it is a tautology. In the Tractatus he was to write: “It is the characteristic mark of logical propositions that one can perceive in the symbol alone that they are true, and this fact contains in itself the whole philosophy of logic.” What is not required is knowledge of the truth or falsity of its constituent propositions, and it is also not necessary to appeal to experience to determine the truth of the proposition as a whole.
That we can “see” the truth of logical propositions has two important implications for Wittgenstein. The first is that such propositions “say nothing”; their truth does not depend on any substantive fact. They are true whatever is the case. Russell had treated this as a sign that such propositions deal with distinctive universal facts. Wittgenstein took it to show that there are no logical facts. The second implication is that there is no need to axiomatize logic in the way Frege and Russell had done. “Naturally this way of showing that its propositions are tautologies is quite unessential to logic.” Every logical proposition rests, in fact, on its own feet. Wittgenstein did therefore set out to design a test by which one could easily determine that a proposition was a tautology. He did not make any effort to give his logic a deductive, axiomatic form. “All propositions of logic are of equal rank […]. Every tautology itself shows that it is a tautology.”
3.
The Moore notes reveal that Wittgenstein’s distinction between saying and showing and the resulting claim that there are limits to language are intimately linked to his thesis that logical propositions are tautologies. But from where had Wittgenstein taken that thesis? He had certainly not got it from Frege and Russell who both held that logical propositions express substantive truths. The most likely source is Fritz Mauthner, the eminence grise of Austrian philosophy, for whom the thesis that logic deals with empty tautologies was an integral part of his skeptical view of philosophy. It formed, in fact, the backbone of the entire argument of his Contributions to a Critique of Language. Mauthner put the point most succinctly in an entry on “Tautology” in his Dictionary of Philosophy where he argues that analytic judgments in Kant’s sense are, in fact, tautologies. “The more general the analytic judgments become, the more the propositions reveal that they are tautologies. The highest laws of thought from which one tries to spin out all of school logic are nothing but tautologies.”
The importance of Mauthner for Wittgenstein’s thinking is still not fully recognized. He is, however, one of the few, select individuals named in the Tractatus, and his Contributions to the Critique of Language is the only book explicitly mentioned in it. But the reference to him and his book needs unpacking. We read: “All philosophy is “Critique of Language” (though not in Mauthner’s sense). Russell’s merit is to have shown that the apparent logical form of the proposition need not be its real one.” The remark seems dismissive at first sight, but it provides, in fact, a key for understanding the entire philosophical project of the Tractatus. It addresses, first of all, the question what philosophy is to do if it cannot advance theories. In agreement with Mauthner, Wittgenstein holds that philosophy must then turn into a critique of language. He adopts moreover Mauthner’s broadly Kantian understanding of the notion of critique. The critique of language means for both Mauthner and Wittgenstein an assessment of the power and limits of language. All this corresponds, of course, to what Wittgenstein proposes in the preface of the Tractatus. The philosophical program spelled out in it is Mauthnerian in its focus on language, its limits, and the impossibility of metaphysical theorizing. Two things are new in it which take us beyond Russell and his way of doing philosophy. The first is that it puts language at the center of philosophy, not reason, nor logic, nor metaphysics. And the second is that it is concerned with language not as end in itself but for the specific purpose of assessing and limiting the claims of philosophy.
This program has an ancestry that goes back beyond Mauthner as he freely acknowledges in his writings. Mauthner identifies, in fact, two forerunners: Johann Georg Hamann, the contemporary and acquaintance of Kant, and Otto Friedrich Gruppe, an early 19th century German philosopher and philologist. Where Hamann had chided Kant’s critique of reason for its lack of concern with a critique of language, Gruppe had set out to critique Hegel’s philosophy with the help of critical attention to language. Mauthner’s own views display, in fact, a great deal of affinity with Gruppe’s and he actually republished some of Gruppe’s work. There is no evidence that Wittgenstein ever paid attention to Gruppe’s writings, but he did become familiar with some of Hamann’s work, plausibly because of Mauthner’s references to him. We can, in any case, trace one major strand of 20th century philosophy of language from the Tractatus back to Mauthner and from there to the early 19th century and its discovery of language as a germane and, indeed, central topic for philosophy.
While Wittgenstein followed Mauthner in wanting to reconceive philosophy as a critique of language, he remained sufficiently attached to Frege and Russell’s program of logical analysis to diverge from Mauthner’s view of how such a critique was to be carried out. Mauthner, in agreement with Hamann and Gruppe, focused entirely on natural language and its grammatical, philological, and etymological analysis. Wittgenstein, on the other hand, continued to look at language from a logical and formalistic perspective. He adopted, in particular, Russell’s distinction between the (surface and apparent) grammatical form of propositions and their (deep and real) logical form. Hence his conclusion that the critique of language had to proceed on Russell’s terms and not on Mauthner’s. It was only in his later work that he came closer to Mauthner’s shifting from a logical to a “grammatical” analysis of language.
Given the different ways they thought about language, it followed that Mauthner’s reasons for questioning the possibility of philosophical theorizing had to be different from Wittgenstein’s. For Mauthner, language was an instrument for practical orientation in the world and as such ill-equipped for abstract and philosophical theorizing. Wittgenstein, on the other hand, sought to hold on to the representational account of language that he had learned from Russell. And he remained attached to the Russellian thought that in order to see how a proposition represents a fact one had to work out its logical analysis. Every proposition must, indeed, have a logical form in order to represent a state of affairs. To understand the meaning of a proposition required a grasp of this logical form. But that logical form was not something the proposition represented. It was present in the proposition in another way. It showed itself in the proposition as one grasped what it represented. One had to distinguish thus what the proposition says and what it shows. That contrast became particular vivid in the case of logical propositions. Wittgenstein agreed with Mauthner that they were tautologies and thus said nothing. But even in that case, the propositional sign would have a logical form which showed itself. Saying and showing were thus separate functions of propositions. It was in this way that the thesis that logical propositions are tautologies and the say-show distinction came to Wittgenstein at the same time.
4.
The distinction between saying and showing is a common and commonsensical one. We use it to contrast verbal expressions and pictures. A sentence, we say, says something whereas the picture shows us something. But the sentence and the picture may represent the same state of affairs. The sentence says, for instance, that Roosevelt, Churchill, and Stalin met at Yalta. The photograph shows us that they did. Sentences and pictures are different media of representation.
A picture represents something by means of its sensual qualities. The green in the picture represents the green of the trees depicted. The two-dimensional arrangements of the shapes in the picture represent the three-dimensional order of the depicted objects. (But in an Expressionist painting the green leaves may also be represented in shades of red. And in a Cubist painting two adjoining objects may appear as merging into each other. In such cases, we may need to know the special language or style of the painting in order to see what it represents.) A sentence, on the other hand, represents things through the conventional meaning of its words. It represents the green of the trees through the word “green” or another equivalent word of possibly another language. And the grammatical arrangement of the words in their linear order in that sentence represents a variety of possible arrangements of the objects of which the sentence is speaking. Pictures and sentences communicate thus in different ways. Much of the time I can perceive with the naked eye what the picture shows, but in order to know what the sentence says I always need to know the conventions and grammar of a language. Different as they are, pictures and sentences are translatable into each other but only partially so. We can’t verbally communicate the full sensual qualities of a picture and thus of the depicted objects. And we can’t fully communicate the high abstractions of which language is capable in pictures. The distinction between saying and showing is thus a useful one.
But Wittgenstein doesn’t employ it as we commonly do and this calls for examination. This becomes obvious from the fact that he considers sentences to be pictures. From the commonsensical point of view they are not pictures at all. Wittgenstein would say that this is because we have an insufficient understanding of how our language works. According to Wittgenstein the sentence says something, but it also shows something and is thus a picture. (What would he say about actual pictures? He does not address that topic, but should we assume that they say something in addition to showing something? If not, then what is the source of the asymmetry?) As a picture, the sentence shows something in addition to what it says. It does so in that it has a logical form which represents the logical form of the state of affairs of which it speaks. Sentences thus have a dual function; they say something but are at the same time pictures that show. But they do not picture like ordinary images through their sensual properties, they picture through their logical form. We can nonetheless “see” the logical form of the sentence just as we can perceive the sensual qualities of the image. The sentence thus “shows”, in fact, two things at once. It shows its logical form and it shows thereby also the form of the state of affairs of which it speaks. Wittgenstein says, on the one hand, that the logical form of the sentence “shows itself” – at least to the properly trained logical eye. But the sentence also shows something about the world. We read in the Tractatus: “That the propositions of logic are tautologies shows [zeigt] the formal – logical – properties of language, of the world.”
The language throughout is allusive and even seductive. But it depends on the uncertain assumption that sentences have a logical form (and not just a grammatical one) and that this form reveals itself intuitively, and that what shows itself in this way depicts and thus shows at the same time something about the world. It appears as if the thought of the sentence as a logical picture was for Wittgenstein both an assumption and something to be argued for. To consider this problematic still leaves, of course, many questions about the similarities and differences and most of all the relation of pictorial and verbal representation. It is true that we can make statements with pictures and draw pictures with statements. But this is something on which the Wittgenstein of the Tractatus offers us little help. As it stands, we may have to take a skeptical view of its distinction between saying and showing.
5.
In the notes he dictated to Moore, Wittgenstein supplements the claim that we need to distinguish between saying and showing with the additional claim that these two functions of language are incompatible. The latter claim is actually independent of the former. For it may turn out that language has indeed two incompatible functions which we try to capture in the words “saying” and “showing” but, for the reasons adduced, there may be no absolute distinction between the perceptual and the conceptual, between sensual forms of depiction and conventional forms of verbal representation.
Are showing and saying mutually exclusive? Is Wittgenstein right in thinking of two functions of language such that the function the first performs cannot, in principle, be performed by the second? Take the photograph of Roosevelt, Churchill, and Stalin at Yalta. The photograph shows that they met. But a history book can also describe that event in words. So, is it true that what can be shown cannot be said? An image can, admittedly, communicate a great deal of information that is not easily put into words. And a verbal account can tell you things that are not easily represented in an image. The two media serve different and complementary purposes. A history book may therefore supplement its verbal account with illustrations. And an art book will provide its reproduction of paintings with verbal explanations. But nothing in this suggests the general conclusion that what can be shown cannot be said.
In the notes he dictated to Moore, Wittgenstein says that every language must have certain properties and “that it has them can no longer be said in that language or any language.” It is not obvious why that should be so. That the tautological character of a proposition shows itself once we grasp its logical form, gives us no immediate reason for the conclusion that it is impossible to say that the proposition is a tautology and that it has such and such a logical form. The incompatibility claim becomes easier understand, however, when Wittgenstein turns to his critique of Russell’s theory of types. Wittgenstein had made a first critical remark about that theory in the notes for Russell in the fall of 1913. He writes there: “Types can never be distinguished from each other by saying (as is currently done) that one has these but the other has those properties, for this presupposes that there is a meaning in asserting all these propositions of both types.” He does not yet, at this point, conclude that a theory of types is as such impossible and he also does not yet mobilize the distinction between saying and showing in order to explain how we can nevertheless grasp type-distinctions. What is also missing as yet is the idea that there are things (such as a theory of types) that are in principle impossible to put into words and that there are therefore limits to language. In his notes for Russell he merely says: “No proposition can say anything about itself, because the symbol of the proposition cannot be contained in itself; this must be the basis of the theory of logical types.” This seems to allow that what cannot be said in one proposition may still be sayable in another one. And the remark also still allows for the possibility of a theory of types.
The topic recurs in the Moore notes but now in a sharpened version which relies on the say-show distinction. “Th[e] same distinction between what can be shewn by the language but not said, explains the difficulty that is felt about types”. That M is a thing and thus of the lowest type cannot be said but shows itself by the symbol, or more precisely by the way the use of the symbol. And the same holds for relations and other truth functions. According to Russell himself, a predicate applying to entities of one type cannot meaningfully be applied to entities of another type. We can therefore not use a predicate “thing” or the predicate “relation” to distinguish between things and relations since that would require us to say that this is a thing but that is not; or this is a relation and that is not. The radical conclusion has to be that “a THEORY of types is impossible.”
Wittgenstein concludes from this that words like “thing” and “relation” will have to disappear from our language when we analyze propositions in which they occur. The distinction between things and relations will manifest itself in there being different kinds of signs for them and that those signs cannot be substituted for each other. Wittgenstein adds to this that the same must hold for the term “fact” though he does not offer us an explicit reason for this in the Moore notes. When we analyze our propositions, “the words ‘thing’, ‘fact’, etc. will disappear,” and for each of them a new symbol will appear instead which is “of the same form as the one of which we are speaking.” His implicit assumption seems to be that the sign standing for a fact, i.e., a proposition, cannot be substituted for the sign of a thing, i.e., a name, or for the sign of a relation, i.e., a relational predicate. That means, of course, that a proposition such as “The world is the totality of facts, not of things” can’t stand up under the analysis. What are we then to make of that sentence at the beginning of the Tractatus? It is far from obvious how it is to be taken. Does it just stand there as an exemplar of metaphysical nonsense? Could it be meant to show something, as the Moore notes suggest? But what then it does it show?
It is striking that Wittgenstein critiques the idea of a theory of types in the Moore notes but makes no general statement about the possibility or impossibility of metaphysical propositions. He also says nothing about the possibility or impossibility of ethical and aesthetic theories. His comprehensive claim that “the problems of philosophy” cannot be resolved in theoretical terms that he makes in the preface of the Tractatus is, in fact, a late addition to the work. It is not a thought that had guided him from the beginning.
6.
Peter Geach has argued that Wittgenstein’s distinction of saying and showing and the claim that what can be shown cannot be said derive from Frege’s account of the difference between objects and concepts. These are, according to Frege, categorially distinct in the sense that there is no predicate that can meaningfully be applied to both. No literally correct description of the distinction between objects and concepts is thus possible. Our attempts to give such a description amount only to giving hints as to what kind of distinction is intended. Our attempted descriptions are just practical devices for familiarizing ourselves with the logical notation. Once we are comfortable in using that notation, we do not need them anymore. We will know how to handle the different signs for objects and concepts in the right way.
Geach’s hypothesis sounds attractive. Wittgenstein’s reasoning appears, indeed, to parallel Frege’s. But the Moore notes undermine that assumption since Wittgenstein does not mention Frege’s treatment of the distinction between objects and concepts in those notes. He moves, instead, from a discussion of the claim that logical propositions are tautologies to a critique of Russell’s theory of types. The comparison of what Frege writes about objects and concepts and what Wittgenstein has to say about types is nevertheless illuminating. It shows that they are concerned with what we may call categorial distinctions.
We can explain that notion by saying that two entities are categorially distinct if there are no predicates that can meaningfully be predicated of both. Frege’s objects and concepts are meant to be categorially distinct in this sense and so are Russell’s logical types. Nothing can therefore meaningfully be said about both objects and predicates and their difference. Nothing can be said meaningfully about entities of different types. It must be said at this point that this is true even of the notion of categorial difference itself. It must be illegitimate to say of two categorially distinct entities that they are because that would come to an attempt to apply predicates of categorial distinctness to both of them. Consider the predicate “categorially distinct from y.” In order for this to serve to distinguish between categorially distinct entities it must be meaningful to apply the predicate to both entities of the category of y and to those that are not of that category. But that is, of course, ruled out. It should be added that all talk of “entities” of different categories must also be illegitimate. If we nevertheless use such language or even feel forced to use it, the resulting propositions will all be strictly nonsensical. Frege would say that they can give us “hints” as to how to use our language; Wittgenstein that they may show something about the logical form of our language.
The notion of categorial distinction has a long history in philosophy and can ultimately be traced back to thinkers such as Parmenides and Plato. It gained a more explicit role in medieval theology and its claim of the complete otherness of God. On this account, none of the predicates we use to describe finite and created beings can possibly apply to God who is infinite and uncreated. The term “good” we apply to human beings cannot therefore be literally applied to God. In its most radical version, this form of theology concludes that we cannot even say that God is a being or exists since those are also terms we have taken from our experience with finite and created beings. One of the attempts to escape from the dilemmas so created is the theory of analogy which allows us to conclude that God has characteristics that are analogous to those that we use to describe finite and created beings. But is this sufficient? We may say that God displays a certain quality when he acts that is analogous to the one that humans have when they do good. But that still commits us to the assumption that we can name God in the same way in which we can name human beings, that we can speak of God’s actions in the same way in which we can speak of human actions. And it commits us to the assumption that we can speak the quality of divine action in the same way in which we can speak of the quality of human action. If we really insist on there being a categorial difference between the immanent and the transcendent then it will turn out that we can say nothing that is literally true of God. God then becomes the completely other. Our utterances can only serve some other purpose, such as that of displaying an attitude of reverence.
The assumption of limits to human understanding and of the impossibility of a dogmatic metaphysical theorizing is of Kantian provenance. But how much did Wittgenstein know of Kant’s philosophy when he wrote the Tractatus? He was certainly familiar with some details of that philosophy. In the notes dictated to Moore in 1914 he refers, in passing, to the Kantian idea of space and time as forms of intuition, but without seriously pursuing it. Half a year later he wrote in his war-time notebook of the light the theory of tautologies might throw on Kant’s question of the possibility of mathematics, but again without expanding on the remark. In the Tractatus, finally, he referred to the Kantian problem of the right and left hand glove which cannot be turned into each other, though once more without elaborating the point. All those references were, in fact, marginal and don’t allow us to infer that Wittgenstein was greatly knowledgeable of Kant’s philosophy. Wittgenstein’s knowledge of that philosophy may, in fact, be an indirect one, partly derived from Schopenhauer and partly from Mauthner.
There are certainly reasons for assuming that Wittgenstein was directly or more probably indirectly familiar with Kant’s characterization of the divide between the empirical and the transcendental as a categorial one. In an entry on “Limiting Concepts” in his Dictionary of Philosophy Mauthner draws attention to the fact that Kant considered concepts meant to refer to the thing in itself as such limiting concepts. In edition A of the Critique of Pure Reason Kant had written:
That a concept, although itself neither contained in the concept of possible experience nor consisting of elements of a possible experience, should be produced completely a priori and should relate to an object, is altogether contradictory and impossible. […] An a priori concept which did not relate to experience would be only the logical form of a concept, not the concept itself through which something is thought.
And he had gone on to apply that lesson, in particular to the notion of a noumenon and the distinction between phenomena and noumena. In a passage actually quoted by Mauthner, Kant writes of the concept of a noumenon, “that is, of a thing which is not to be thought as object of the senses but as a thing in itself, solely through a pure understanding”, that it is merely a “limiting concept” and “only of negative employment.” The concept of a noumenon is thus “problematic,” but we need it to “to prevent sensible intuition from being extended to things in themselves.” Nevertheless, we have to grant that “the division of objects into phenomena and noumena and the world into a world of the senses and a world of the understanding, is therefore quite inadmissible in the positive sense”.
Mauthner was ultimately skeptical about Kant’s talk of limits of understanding. A limiting concept, he wrote in his Dictionary, is “what one cannot imagine and is yet forced to think; or what one can’t think and yet can only think.” What he meant was that as soon as we form the concept of a limit we find ourselves forced to think about what is on the other side of that limit. As soon as there is a limit to understanding, we are on the way of assuming something that is beyond understanding. The Wittgenstein of the Tractatus, on the other hand, was drawn to the Kantian language of limits. He is certainly aware of the difficulty inherent in the notion of a limit to which Mauthner draws attention. In the preface to his book he writes that “in order to draw a limit to thinking we should have to be able to think both sides of this limit (we should therefore have to be able to think what cannot be thought)”. But he believes at this point that he can avoid that difficulty by speaking, instead, of the limits of language while this notion has, in fact, the same problem. There are places in the Tractatus where this problem becomes apparent as when he writes of “the metaphysical subject” as “the limit – not a part of the world.” Similarly, when he says that value “must lie outside the world”, as if we could draw a line between its inside and its outside.
7.
Mauthner’s critique of language was motivated by an anti-philosophical bias. Its goal was to revive a Pyrrhonian skepticism about philosophy. To underline that position, Mauthner quoted Sextus Empiricus as saying that we must in the end even throw away the ladder by which we have reached that skeptical conclusion. Wittgenstein was sufficiently impressed by Mauthner’s Pyrrhonianism to borrow from him the metaphor of the ladder and incorporate it into the penultimate section of the Tractatus. That does not mean that he identified completely with Mauthner’s position and that he also subscribed to an undiluted Pyrrhonian view. In contrast to Mauthner, Wittgenstein did not want to deny that there were philosophical problems. The preface to the Tractatus said only that we have been looking at those problems in the wrong way. We have been asking the wrong questions about them. Our hope to resolve those problems by means of a philosophical theory has been mistaken. That still left the problems themselves in their place. Unlike Mauthner, Wittgenstein recognized the power, the draw, and the significance of those problems.
We get an inkling of this from what Wittgenstein writes about the problem of life, i.e., the problem what meaning life has for us or how we can give it meaning. The problem is real enough but it can’t be resolved with a theoretical formula. It resolves itself, rather, when we learn how to live. Once we have come to accommodate ourselves to the conditions of our own existence and to the factuality of the world, when we have learned to be “happy” we will no longer be bothered by the question that has previously tortured us. “The solution of the problem of life is seen in the vanishing of this problem.” Those who have solved that problem won’t afterwards be able to say in what the meaning of life consists. The solution of the problem of life shows itself, instead, in the act of living.
The Wittgenstein of the Tractatus did not think that all the problems of philosophy would find their resolution in the same way. He appears to have thought that the problems of metaphysics will be seen to have arisen largely through a misapprehension of the logic of our language. In this area, he seems to have come closest to Mauthner’s Pyrrhonism. The problems of logic, on the other hand, are resolved by discovering logic embedded in our language in such a way that we can come to see it without having to propound a logical theory. In ethics and aesthetics the solution of our problems is to be found in the practices of our lives.
What distinguished Wittgenstein from Mauthner was his keen attunement to the problems of philosophy. He kept returning to them to mull them over again and again. In a note preserved in Zettel he wrote:
Some philosophers (or whatever you like to call them) suffer from what may be called ‘loss of problems’. Then everything seems quite simple to them, no deep problems seem to exist any more, the world becomes broad and flat and loses all depth, and what they write becomes immeasurably shallow and trivial. Russell and H. G. Wells suffer from this.
The solution of the problem of life, he had said in the Tractatus, consist in its disappearance, but for Wittgenstein himself the problem never disappeared. Life remained a struggle. The problem of life kept haunting him. And that was for him how it had to be. “In philosophizing we may not terminate a disease of thought. It must run its natural course, and slow cure is all important.” Russell’s shortcoming and probably also Mauthner’s, the shortcoming of the philosophical theorists and the Pyrrhonian opponents was, in the end, that they all set out to terminate the problems of philosophy as quickly as possible. The task was, however, to learn to persist in them and let their resolution come to us slowly.
8.
Wittgenstein returned to the theme of the limits of language which he had first announced in the preface of the Tractatus, ten years later in his “Lecture on Ethics.” Now focusing on ethics rather than logic and metaphysics, he argued once more that ethical propositions are strictly nonsensical. They nevertheless serve a function which is to indicate something of profound human importance. “Ethics,” he said in conclusion, “so far as it springs from the desire to say something about the ultimate meaning of life, the absolute good, the absolute valuable, can be no science.” All this was still in line with what he had said in the Tractatus. But the “Lecture on Ethics” adds to this an intriguing further thought. It is that while there are determinate limits to language, there are also contingent limits that we should pay attention to. The lecture begins thus with a reflection on what Wittgenstein thinks he can communicate in it. There is first of all his halting English. Then there is the impossibility of saying something systematic and scientific in a single lecture. And there is finally his reluctance to saying something merely popular. All this may sound trivial, but it points to the fact that the limits of language are more variable than the Tractatus had recognized.
That was a thought Wittgenstein explored in his later writings. There were different languages, he concluded, with different expressive power and thus with different limits to what can be said in them. In contrast to the Tractatus he no longer assumed that there could be one complete language with its one set of ultimate limits in its expressive power. Our language had become richer with the invention of the symbolism of chemistry and the notation of the infinitesimal calculus. Certain things could now be expressed with great precision that could not be said before those innovations. “Our language can be regarded as an ancient city: a maze of little streets and squares, of old and new houses, of houses with extensions from various periods, and all this surrounded by a multitude of new suburbs with straight and regular streets and uniform houses.” Like any such city, our language will have boundaries at any given moment, but those boundaries may shift and expand over time. What could not be said becomes expressible in new forms of language. “There are countless kinds; countless different kinds of all the things we call ‘signs’, ‘words’, ‘sentences’. And this diversity is not something fixed, given once and for all; but new types of language, new language-games, as we may say, come into existence, and others become obsolete and get forgotten.”
And this idea of relativized limits is taken up once more in Wittgenstein’s last notes. There are different world views and in each one only some things will be comprehensible. In On Certainty, he describes the general form of this new way thinking about relativized limits as follows:
But I did not get my picture of the world by satisfying myself of its correctness: nor do I have it because I am satisfied of its correctness. No: it is the inherited background against which I distinguish between true and false. […] The propositions describing this world-picture might be part of a kind of mythology.
Each world picture has its distinctive limitations – and for any given person, it is “[a]bove all […] the substratum of all [their] enquiring and asserting”. But world views change. They can expand or be replaced and thus the limits of what is comprehensible and what we can speak of will change with them. What Wittgenstein now calls the “mythology” of world pictures – the description of the boundaries of our thinking about the world – can thus become “fluid” again:
The mythology may change back into a state of flux, the river-bed of thoughts may shift. But I distinguish between movement of the waters on the river-bed and the shift of the itself; though there is not a sharp division of the one from other.
Wittgenstein’s interest in limits and the possibility or impossibility of drawing them sharply thus remained unbroken until the end of his life, but was radically transformed by his new-found attunement to relativity and the diversity of world pictures.
References
Biesenbach, Hans. Anspielungen und Zitate im Werk Ludwig Wittgensteins. Sofia: Sofia University Press, 2014.
Geach, P. T. “Saying and Showing in Frege and Wittgenstein”. In Essays on Wittgenstein in Honour of G. H. von Wright, edited by Jaakko Hintikka, 54-70. Amsterdam: North-Holland, 1976.
Kant, Immanuel. Critique of Pure Reason. Translated by Norman Kemp Smith. London: MacMillan Press, 1929.
Russell, Bertrand. “Introduction.” In Tractatus Logico-Philosophicus, translated by C. K. Ogden and F. P. Ramsey, 7-23. London: Routledge & Kegan Paul, 1922.
———. “The Philosophy of Logical Atomism.” In Logic and Knowledge: Essays 1901–1950, edited by Robert Charles Marsh, 175-281. London: Allen & Unwin, 1956.
———. Theory of Knowledge: The 1913 Manuscript. Edited by Elizabeth Ramsden Eames, London: Routledge, 1992.
Mauthner, Fritz. Beiträge zu einer Kritik der Sprache. 3rd ed. 3 vols. Stuttgart: Cotta, 1913.
———. Wörterbuch der Philosophie. 2nd ed. 2 vols. Munich: Georg Müller, 1910.
Potter, Michael. Wittgenstein’s Notes on Logic. Oxford: Oxford Unversity Press, 2008.
Sluga, Hans. Gottlob Frege. London: Routledge, 1980.
———. “Wittgenstein and Pyrrhonism.” In Pyrrhonian Skepticism, edited by Walter Sinnott-Armstrong, 99-117. Oxford: Oxford University Press, 2004.
Stern, Alexander. The Fall of Language: Benjamin and Wittgenstein on Meaning. Cambridge, Mass.: Harvard University Press, 2019.
Wittgenstein, Ludwig. Cambridge Letters: Correspondence with Russell, Keynes, Moore, Ramsey and Sraffa. Edited by Brian McGuinness and G. H. von Wright. London: Blackwell, 1995.
———. “A Lecture on Ethics.” In Philosophical Occasions 1912-1951, edited by James C. Klagge and Alfred Nordmann, 37-44. Indianapolis: Hackett, 1993.
———. Notebooks 1914–1916. Translated by G. E. M. Anscombe and edited by G. H. von Wright and G. E. M. Anscombe. Oxford: Blackwell, 1961.
———. On Certainty. Translated by Denis Paul and G. E. M. Anscombe and edited by G. E. M. Anscombe and G. H. von Wright. Oxford: Blackwell, 1969.
———. Philosophical Investigations. 4th ed. Translated by G. E. M. Anscombe, P. M. S. Hacker, and Joachim Schulte. New York: Wiley, 2009.
———. Tractatus Logico-Philosophicus. Translated by C. K. Ogden and F. P. Ramsey. London: Routledge & Kegan Paul, 1922.
———. Zettel. Edited by G. E. M. Anscombe and G. H. von Wright. Berkeley: University of California Press, 1967.
