Home - Wittgenstein's Logic of Language | Bibliography | Outline of this page

Remarks about Wittgenstein's Tractatus Logico-Philosophicus

Introduction: the meaning of the title of Wittgenstein's book. Wittgenstein's own title for the TLP was Logisch-philosophische Abhandlung ("Logical-philosophical Treatise"). G.E. Moore suggested the Latin title (in the tradition, although not following the pattern, of the titles of Whitehead and Russell's Principia Mathematica and Moore's own Principia Ethica). And so by 'logical' what does Wittgenstein mean -- and by 'philosophical' (Surely it is important that the book is not titled simply Principia Logica), and by 'treatise' what does he mean?

By nature not convention

Logical research means the investigation of all regularity. And outside logic all is accident.

The exploration of logic means that exploration of everything that is subject to law. And outside logic everything is accidental. (TLP 6.3)

Two English translations of Wittgenstein's German language text, the first by Ogden (1922), which Wittgenstein proof-read before it was published, the second by Pears and McGuinness (1971), published twenty years after Wittgenstein's death.

Logical. By the word 'logic' the TLP means 'the study of laws of nature', as in expressions such as "laws of thought" and "the law of contradiction". The "theory of logic" of that book presumes that there are laws of thought, and that philosophy can discover those laws, and, indeed, that Wittgenstein has discovered them. Thus the TLP is a work of metaphysics.

That earlier definition of 'logic' contrasts with 'logic' = 'the study of rules, conventions', which was Wittgenstein's later (1933-34) use of that word, when he came to view all metaphysical speculation, including his own earlier, as conceptual confusion rather than an investigation of reality.

Although Ogden's translation does not use the word 'law', yet its second sentence ("And outside logic all is accident", meaning that it is not as it is by necessity; cf. 6.37: "The only kind of necessity is logical necessity") suggests that word.

That Wittgenstein had "laws of thought" in mind is shown by this, that "the limits of language" -- which are "the limits of thought" as well (Letter to Russell, 19 August 1919) -- are set by laws of nature, because what else could set those limits? They cannot be conventional, set by man, and therefore subject to flux -- not if they belong to the Wesen der Welt [5.4711] (the "essence of the world").

In the metaphysical world-picture of Wittgenstein's Tractatus, "language is a cage": Natural law determines what can be put into words that are not "nonsense", and what cannot.

In contrast to Aristotle, who classifies logic ("Analytics") as a mere tool of philosophy, Wittgenstein in the TLP says that logic is the essence of philosophy. This is because "the limits of the world are also the limits of logic" [5.61], and the aim of philosophy is to outline the limits of "the world" (the discussable part of reality).

So much then for the TLP's use of the word 'logic' -- i.e. for 'logic' = 'natural law', and 'logic' = 'the study of natural laws'.

But as Wittgenstein later used the word 'logic' -- and as I will use that word in what follows -- logic DEF.= the rules of sense and nonsense, which are conventions, as well as their study. And so, in my view, logic remains the basis of philosophy -- if by 'logic' is meant 'the distinction between sense and nonsense' (which I have christened 'logic of language' in these pages); it is not merely the tool of critical thinking, although it is that too.

Philosophical. What does Wittgenstein mean by the word 'philosophical' in his title? In "Notes on Logic" (1913) he wrote, "Philosophy consists of logic and metaphysics: logic is its basis" (Notebooks 1914-1916, 2nd ed., tr. Anscombe), and originally the work was to be about logic rather than about "the world" (metaphysics) -- it was to be a Tractatus Logicus only. But on 2 August 1916 Wittgenstein wrote in his notebooks, "My work has extended from the foundations of logic to the nature of the world."

And so I think the present book could have been titled "Logical-metaphysical Treatise", because the TLP is a work of metaphysics: its aim is to distinguish reality from misconceived reality ("the [natural law] logic of our language is misunderstood"). That is very different from the aim of Wittgenstein's later work, which is concerned only with the facts in plain sight, not with a reality imagined to underlie and explain those facts.

Treatise. The TLP is a thoroughgoing exposition of both the principles and the conclusions of its topic (The "principle" of the TLP is the relationship between logic (necessity) and the world (accident), and the "conclusion" is the consequent limits of language and thought which follow from that principle). That meaning of 'treatise' is implied by the book's Preface: "... the truth of the thoughts communicated here seems to me unassailable and definitive. I am, therefore, of the opinion that the problems have in essentials been finally solved" (tr. Ogden).

*

Revision of the concepts 'logic', 'nonsense'. Wittgenstein uses the word 'nonsense' strangely in the TLP, for he defines -- not the way we normally use the word 'nonsense' (which is a description) -- but "what nonsense really is" (which is a theory), which is the characteristic method of metaphysics, namely to seek real definitions of abstract terms (or "abstractions", as if these were the names of natural phenomena).

But does Wittgenstein also use the word 'logic' strangely? The TLP's use of the word 'logic' can be historically justified, because 'logic' = 'natural laws of thought' is the traditional use of that word; indeed, Kant's categories of thought and the three laws of thought of the Aristoteleans assume it. (It is Wittgenstein's later revision of the concept 'logic' that is non-traditional and thereby strange.)

Logical necessity. According to Wittgenstein's later view, logic is the study -- not of nature-imposed limits -- but of man-made rules (conventions), which are more or less (but sometimes far less) arbitrary. That is "the logic of our language" or how Wittgenstein defines the word 'meaning' to make the distinction between sense and nonsense objective. Rules of "grammar" (in Wittgenstein's jargon) are the only limits of language.

What is the source of logical necessity -- radical naturalism or radical conventionalism? The word 'radical' = 'thoroughgoing', and both views seem mistaken. Because that there is some vital relationship between general facts of nature and concept-formation (PI II, xii, p. 230), or the possibility of a language game, seems clear. Language both is and is not a cage: it consists of conventions, some mere, but others apparently not (It does seem that some perception-and-conception and consequent rule-making is limited by man's nature).

In the TLP our concepts are the contours of a relief map (natural). Later Wittgenstein will compare our concepts to the lines of a political map (conventional), describing without trying to explain their existence (concept-formation).

The riddles of poets

It was a riddling definition ... that Simonides gave after the manner of poets ... (Plato, Republic 332b-c, tr. Shorey)

I think I summed up my attitude to philosophy when I said: philosophy ought really to be written only as a poetic composition. (CV p. 24)

To write about Wittgenstein's earlier way of thinking, I have to use the tools of his later way of thinking, because I would be philosophically lost (PI § 123) without them. But why? because wouldn't I still have Socrates' method of asking for an account of what you know?

But that method -- namely to seek contradictions in any account of what anyone claims to know and thereby to accept or refute claims to knowledge -- does not by itself alone serve as a path out of the vagueness and confusion that surround us. But to it must be added: (1) Wittgenstein's distinction between a conceptual and a factual investigation, which cannot be understood without (2) Wittgenstein's revised concept 'grammar', essential or defining elements of which are: the distinction between a sign and the meaning of a sign; the comparison of language to tools and the work done with tools; the comparison of the meaning of language to rules of a game (as in "language game"); and [very often] "the meaning of a word is its use in the language" (PI § 43). That contrasts with the unrevised account of the logic of our language (of how our language works), which basically is "Socratic definition" applied to all words, because, as in the TLP, all words are names and the meaning of a name is the object (whether physical or "abstract") the name stands for. What Wittgenstein did was to set aside that way of looking at language.

So elementary has Wittgenstein's distinction between a conceptual and a factual investigation (i.e. between an investigation of language meaning and an investigation of phenomena) become to my own way of thinking that without Wittgenstein's later logic of language, I would still be utterly lost, perplexed, confused, mystified by language in the way that I was for so many years of my youth. Although language meaning isn't always clear now, I do at least now have many tools to use towards making it clear, to "finding my way about" (ibid. § 123) philosophical problems. (The Synopsis of Wittgenstein's Logic of Language describes these.)

Outline of this page ...

Related pages: Wittgenstein at Cassino, an interlinear translation of selections from Franz Parak's Wittgenstein prigioniero a Cassino: Wittgenstein was held at the prisoner of war camp from November 1918 to August 1919; when he was taken prisoner he had in his knapsack the manuscript of his "Logical-philosophical Treatise", which was later published as the Tractatus Logico-Philosophicus. | If "everything speaks in its favor, nothing against it" (OC § 4), must the proposition be true? What did Wittgenstein mean by the word 'know' in the TLP: Don't we know that the sun will rise tomorrow [6.36311]?

The limits of language according to the Tractatus Logico-Philosophicus - "Language is not a cage"

Anrennen gegen die Grenze der Sprache? Die Sprache ist ja kein Käfig. ("Running against" -- i.e. trying to go beyond -- "the limits of language? Language is not a cage.") (LE/Notes p. 14, 17 December 1930)

If the propositions of natural science are the limit of language -- i.e. the only type of language that is not nonsense [TLP 4.06, 6.53], then any attempt to say what is not a proposition of natural science is to try to go beyond that limit -- i.e. it is nonsense. And indeed that is what the TLP claims to be -- a book of nonsense: "My propositions are elucidatory in this way: he who understands me finally recognizes them as senseless" (ibid. 6.54, tr. Ogden).

The propositions of the TLP are "senseless", but they are apparently not meaningless, for they are "elucidatory" and can be "understood". ('Senseless' means they are not of the form: This is how things stand in the world.) Thus 'meaningless' would mean what we normally mean, namely language that does not convey meaning.

To define the word 'senseless' or 'nonsense' in such a way as to make (i.e. classify) most discourse as nonsense -- i.e. all discourse other than that of the form This is how things stand in the world -- declaring that to be the logic of our language, is the strange project in philosophy of the Tractatus Logico-Philosophicus.

(TLP 4.5): "The general form of propositions is: This is how things stand." -- That is the kind of proposition one repeats to oneself countless times. One thinks that one is tracing the outline of the thing's nature over and over again, and one is merely tracing round the frame through which we look at it. (PI § 114)

That frame is the only "cage". And therefore there is no cage, unless it is one created by our own way of looking at language, by our own way of thinking about language meaning. Language itself is not of logical necessity a cage. (In the TLP language is by "real" necessity (cf. the notion "law of nature") a cage.)

Within the TLP's frame of reference, propositions about what is "the important part" of our life are nonsense. But only within that frame. (Note that the combination of words 'absolute frame of reference' is undefined -- notwithstanding that "absolute reality" is what metaphysics seeks (cf. Plato, Parmenides 135b-c), and a metaphysical view of language and reality is what the TLP is.)

But if the limit of language meaning truly is -- concept-formation = frame-formation, wouldn't it be possible to give nonsense -- i.e. normally undefined language -- meaning by changing -- not the agreed upon facts but -- the frame of reference? For despite our imagining otherwise, what we are looking at language meaning through is not clear glass but a conceptual lens, and things no more need conform to our concepts than the world must have the shape of the window frame through which we look at it.

It is like a pair of glasses on our nose through which we see whatever we look at. It never occurs to us to take them off. (PI § 103)

Couldn't what is nonsense in one conceptual system (i.e. within one frame of reference) not be nonsense in another, and vice versa? The difficulty, however, is to invent an objective system (which is what I think Wittgenstein does or tries to do with his game metaphor). (This discussion is too vague by far.)

Wittgenstein later revised his view of the logic of our language, namely by saying that there is no essence ("general form") of language, but that language has many forms (which he describes as "language games"): "Instead of producing something common to all that we call language, I am saying that these phenomena have no one thing in common which makes us use the same word for all, -- but that they are related to one another in many different ways. And it is because of this relationship, or these relationships, that we call them all 'language'" (ibid. § 65). What distinguishes sense from nonsense varies with the rules of the particular language game: the TLP's statement of the essence of nonsense is discarded along with its statement of the essence (general form) of language.

One might dismiss the TLP with the observation that, "Well, if you are going to define the word 'nonsense' that way ....", but Wittgenstein did not think that he was defining the word (spoken sounds, marks on paper) 'nonsense' -- i.e. more or less arbitrarily assigning limits to our concept 'nonsense' -- but saying instead what nonsense really is. In the TLP nonsense is, like logic, dictated by nature (essence belongs to nature, not to the conventions of "grammar").

[Kant also spoke of a natural inclination, not to try to go beyond the limits of language, but to try to go beyond the limits of experience, but for both philosophers God and "value" are beyond those limits. (Copleston, A History of Philosophy, Volume VI, xiii, 6)]

"The structure of the world and the structure of the proposition are the same"

According to Wittgenstein at that time, even in an ordinary sense, a proposition -- and language that is not nonsense, according to the TLP, consists essentially and exclusively of propositions (statements of fact) -- is essentially a picture of objects standing in relationships to one another [3.1431] ["the case" or "how things stand" is a constellation of simple (atomic) objects]. Wittgenstein's example for Parak: "If a book is on the table ..." states the relation between objects: the proposition is a picture of the relationship.

Words replace objects in the proposition

The essential nature of the propositional sign becomes very clear when we imagine it made up of spatial objects (such as tables, chairs, books) instead of written signs. The mutual spatial position of these things then expresses the sense of the proposition. (3.1431, tr. Ogden)

Cartesian Geometry

What language and the world have in common is a structure. We could picture that structure to be a Cartesian grid, a grid with addresses. A set of atomic objects and their addresses would be a "fact", a constellation of atomic objects, (in the TLP's jargon). And if we changed either the objects or their addresses, etc., that would be a different fact.

The grid is what is constant; the grid is the structure that the world and language share (Language mirrors the world in this way -- that a proposition of language has (in not always a clear way) the same structure as a fact of the world: the proposition is a picture of a fact). That is the TLP's "logic of language" (in the Tractatus's sense of that expression, not in mine).

The relation of language to the world

Wittgenstein was reading in a magazine about a lawsuit in Paris concerning an automobile accident. At the trial a miniature model of the accident was presented before the court. [The physical model was a picture of the facts; it served as a proposition stating how things stood at the time of the accident.] It has this function owing to the correspondence between the parts of the model (miniature-houses, -cars, -people) and things (houses, cars, people) in reality. [The parts of the model and the parts (i.e. words) of the proposition somehow correspond to one another. That is the essence of the relation between language and the world.] (von Wright, "Biographical Sketch", in Malcolm, Memoir 2e, p. 8)

But "essence belongs to grammar" (PI § 371) -- i.e. essence belongs to the frame, not to what is seen through the frame (ibid. § 103). A contradiction would arise if the miniature-car were parked to the right of the bank while the real car was parked in front of the bank; that would be a contradiction in "structure" (TLP 4.1211), making the picture a false proposition.

Every picture is also a logical picture. (On the other hand, for example, not every picture is spatial.) (ibid. 2.182, tr. Ogden)

But does Wittgenstein give an example of a "non-spatial picture"? He speaks of "the gramophone record, the musical thought, the score, the waves of sound" (4.014): the musical score is a language but so is the gramophone record (4.0141) (But there is no picture of the music -- i.e. of what the human being hears but the cat does not, because the concepts 'music' and 'sound' are not equivalent). Are there any examples to give, because isn't 'spatial picture' a tautology (pleonasm), as the propositional form 'This is how things stand' suggests it is? You cannot point to the grooves of a gramophone record and say 'This is how things stand', not as we normally use our language.

A musical score may have the same logical form -- although, note this is a definition of 'same' and 'logical form' in this context: one-to-one correlation between sound and score (notes on paper) as well as between order of the sounds and order of (the notes in) the score -- but its musical score is not a picture of the music, not as we normally use the word 'picture'. Even if the score consisted of drawings of a piano's keyboard, the score would say which keys to strike and in which order, but it would not be what we mean by the word 'music', which is a non-visual phenomenon, regardless of how it is represented visually. Or in other words --

Where are the limits of this simile?

In order to understand the essence of the proposition, consider hieroglyphic writing, which pictures the facts it describes. And from it came the alphabet without the essence of the representation being lost. (4.016, tr. Ogden)

The world of the Tractatus as described is static: "This is how things stand" in the world. (What puzzled Aristotle was motion, why things change (causality). Aristotle's metaphysics, or "first philosophy", is about what does not change, physics about what does. For the pre-Socratics and for Plato, what does not change is the only thing real, the rest is mere appearance; but that is not the common view of things.)

*

Selected definitions and discretion

Although it defines the word 'nonsense' differently from the TLP, Wittgenstein's first later philosophy, too, is based on a selected definition of 'meaning': "Let's only bother about what's called the explanation of meaning, and let's not bother about meaning in any other sense. An explanation of meaning is ... a rule, a convention" (PG i § 32, p. 68-69). Wittgenstein's first later meaning of 'meaning' is based on the relation between "rules of grammar" (in Wittgenstein's jargon) and sense and nonsense. If you follow the rules of a game, you play the game correctly. If you follow the rules of a language game (i.e. comparison of using language to playing a game by following rules), your language has meaning; if you break the rules, your language is nonsense.

With this comparison Wittgenstein wanted to make the meaning (or absence of meaning (PI § 464)) of language in philosophical problems clear. But because selection requires discretion -- i.e. choice -- the question is whether Wittgenstein's selected meaning of 'meaning' does make the distinction between sense and nonsense clear. Certainly not always, because what if a particular "language game" doesn't normally have strict rules, (as normal games have)?

So in Wittgenstein's second later philosophy -- or "logic of language" in my jargon -- he offers an account of our normal concept 'language meaning', namely that in most cases the meaning of a word is its use in the language, and that the meaning of a name is sometimes explained by pointing to its bearer (ibid. § 43). But is this later account too vague to do the job of verifiably distinguishing sense from nonsense (as the TLP's account is too narrow an account of the logic of our language)?

[There may be a third later philosophy, for in On Certainty (§ 37) Wittgenstein seems to acknowledge that a metaphysical use (PI § 116) of language is not necessarily nonsense. But if that is so, how is language with sense to be distinguished from nonsense? Certainly not by reference to "the language game that is its original home" (ibid.).]

At all times, in my view, Wittgenstein's master question was: what gives language (spoken sounds, ink marks) its meaning? For the TLP, the answer is that language that is not nonsense is a picture of an atomic fact (or reducible to pictures of atomic facts), i.e. an arrangement (string) of names of logical atoms, as that book vaguely says the propositions of natural science are. That is how the TLP marks off sense from nonsense. That is its "theory of meaning" or better its "metaphysics of meaning", for it answers or tries to answer the question: what is the reality behind the appearances?

[Was the Tractatus written to Wittgenstein's perennial thesis, namely that the task of the philosopher is to banish metaphysics (and ignore ethics)? Was that book's picturing theory the result of an investigation -- something discovered in the course of Wittgenstein's investigation -- or was it the presumption, not the result of an investigation, a foregone conclusion? It seems to have been (what Wittgenstein regarded as) an insight or realization that became a presumption. It would be harder to say that about the Philosophical Investigations, which, in my view, finds only houses of cards (PI § 118) because it only looks for houses of cards; if the great questions of philosophy are ignored in that book, it can only be because Wittgenstein chose to ignore them.]

*

What is 'unsayable' versus What is 'undefined'

In our normal, everyday language, the expressions 'what cannot be put into words', and 'inexpressible' or 'ineffable', e.g. "nameless emotions" or "what I experience when I listen to music", does not contrast with 'whatever is not picturable', but that is the meaning of "unsayable" in the TLP.

By 'unsayable' in the Tractatus, Wittgenstein means any language other than the proposition type picturing-propositions -- i.e. "propositions" (strings of names) which are isomorphic pictures of "facts" (constellations of absolutely simple, atomic objects): This is the relation in which these objects stand to one another is the "general form of a proposition" [4.5] and the essence of language, although the form of our everyday language may hide this from us [4.002 (But it was Plato, not Russell (4.0031), who invented "logical form")].

To break through the walls of this cage is "absolutely hopeless" (LE p. 12), and philosophers should give up trying to state propositions about aesthetics and ethics, God and the riddle of life, because such propositions do not belong to the language of natural science [4.06] and therefore they are nonsense. They try to put into words what cannot be put into words -- i.e. to say what is not "sayable" but "unsayable".

[Saying and Showing in Wittgenstein's Tractatus Logico-Philosophicus: Uhland's "Count Eberhard's Hawthorn".]

*

Note: in the following discussions, I imply (1) that 'names of atomic objects' are names of absolutely simple things (PI § 48: if 'R', 'G', 'B' are the names of absolutely simple objects (logical atoms), then the proposition 'RGB' names an atomic fact, and 'BRG' another) perceived by the senses as likewise absolutely simple sense data, or at least (2) that constellations ("atomic facts") of these are. But the TLP gives no examples of either atomic objects or atomic facts, nor of absolutely simple sense data. (The proposition Wittgenstein gave Franz Parak as an illustration does not seem to state an "atomic fact" [TLP 2.0, 2.01] nor a book to be an "atomic object" [2.02].)

But Wittgenstein does not say that those "things" ("unknowns" really), or constellations of things, are perceived by the senses, nor does he ask whether they are. He thought at the time (as he told Malcolm) that this question is not for philosophy but for the science of experimental psychology to answer.

And Wittgenstein does not say that aesthetic and ethical value, and God, lie outside the world of atomic facts because they are not perceived by the senses to be among those facts. Is that they lie outside the world a deduction from his earlier definitions and premisses: for if the TLP's account of logic and its relationship to the world is independent of verification by sense perception, then that aesthetic and ethical value, and God, do not show themselves within the limits of the world must be knowable prior to experience (a priori), surely? Where does the TLP's category "what is higher" come from if not from experience of looking in the world ("what is lower") and not finding what is higher there?

If "the world is all that is the case" [1] and "what is the case is the existence of atomic facts" [2], then the sense of the world must lie outside the world [6.41] -- indeed, the sense of everything must lie outside "the world". Because Wittgenstein's "atomic facts" -- if they are known to man as sense data -- do they have meaning (sense) in themselves -- or are they percepts without concepts? Are concepts, then, the sense of the things in "the world"?

In his Notebooks 1914-1916, Wittgenstein wrote, "How things stand, is God. God is, how things stand" (1 August 1916) -- now, is that the "undefined combination of words" type of nonsense, or the TLP type of nonsense (It is not a picture drawn using the language of classical mechanics)? Is 'God' a concept applied to the-world-as-percept [cf. "the world as a limited whole" 6.45]? But God cannot both be the world and what is higher than the world.

[Russell's and Frege's geometric heaven is "on the other side of the sky", with Plato's Forms -- i.e. not in "the world".]

Frege's response to the TLP ("unsayable" because incomprehensible)

How is the expression 'the world' normally used in philosophy? Responding to the TLP Frege wrote to Wittgenstein: "You see, from the very beginning [from the very first page] I find myself tangled in doubt as to what it is you want to say and can make no headway with it" (Letter of 28 June 1919, in Waugh (2008), p. 145, 301).

Wittgenstein wrote to Russell [R.37 dated 19 August 1919] "I also sent my M.S. to Frege. He wrote to me a week ago and I gather that he doesn't understand a word of it at all".

*

Although he told Frank Ramsey (20 September 1923) that it was the result of seven years of thought, Wittgenstein's Tractatus is a youthful work built on a non-existent natural sciences foundation. In the world-picture of the TLP, "the world" is for the natural sciences to investigate, because logic (philosophy) is not concerned with empirical questions, because what is subject to law is not accidental (random) [6.3], which is what empirical facts ("whatever happens to be the case") are.

"A proposition is an arrangement of names"

A proposition (assertion of an atomic fact) is a concatenation [4.22] -- that is, a string (like boxcars on a railway track) -- of names of objects ["The essence of language is the putting together of names of objects" is the idea, an idea which Plato criticised in Sophist 261e-262a in which he says that a proposition consists of names and verbs, never just one or the other]. But Wittgenstein never says what these objects are. 'A book is on the table' is an empirical proposition and seems to be a "proposition of natural science", but 'book' in the TLP's jargon does not mean an instance of what everyone calls by the common name-of-object word 'book', not if the word 'book' names an atomic object -- and if it names an atomic fact, an atomic fact is only a relationship among atomic objects (which stand like stars in a constellation) -- and does anyone know what an atomic object "looks like"?

An "atomic fact" -- from the point of view of perception -- is a relationship among absolutely simple sense-impressions. But what these sense data are, like what atomic objects are, is left for natural science to determine, as is what the absolutely simple sense datum correlated to any particular atomic object is. Because questions about perceptions of the world are empirical questions -- i.e. questions about "what is the case" [1.21, 6.37].

An "absurd" presumption

And that is the TLP's non-existent foundation: the atomic objects and absolutely simple sense data which it presumes but of which natural science knows nothing. Wittgenstein later thought that presumption absurd (Malcolm, Memoir (1984), p. 70).

[What do the Tractatus' "propositions of natural science" look like? Are they what we normally call statements of fact? No, because in the TLP, Wittgenstein does not use the word 'fact' as we normally do, for we do not normally define the word 'fact' as 'a concatenation of atomic objects'. Questions such as "By 'proposition' does Wittgenstein mean an assertion verifiable by means of direct sense perception?" cannot therefore arise, because we know neither how nor even if atomic objects are perceived. (What might the TLP's "propositions of natural science" be? One possibility, as I think Wittgenstein explains in PI § 48.)]

*

On the one hand, only what "can be put into words" can be thought. But on the other hand, Wittgenstein says in the Preface to his book that "thoughts are expressed" in it. But those thoughts are "nonsense" thoughts [6.54] and should not have been thought at all [7], indeed, given that they are "nonsense", not put-able into words, they could not have been thought. (Is that merely a caricature of Wittgenstein's book -- am I wrong in seeing this as the way the TLP abuses language?)

Most propositions and questions, that have been written about philosophical matters, are not false, but senseless. We cannot, therefore, answer questions of this kind at all, but only state their senselessness. Most questions and propositions of the philosophers result from the fact that we do not understand the logic of our language. (TLP 4.003, tr. Ogden)

Wittgenstein writes that someone who understands him "finally recognizes" the propositions of his book as "senseless" (6.54) -- i.e. that its language is nonsense. But does that mean that the TLP's author himself does "not understand the logic of our language"? Or is it because of his understanding it that he is able to use "senseless" language to show others what he understands, namely, what the logic of our language is?

If nonsense is not nonsense, then there is no nonsense

Again, as countless times before: To call language that conveys meaning 'nonsense' or 'senseless' is utterly eccentric, if not itself nonsense. Because if language that conveys meaning is nonsense, then what is language that does not convey meaning?

In TLP 4.112: Wittgenstein sets out his view of what philosophy is, but the TLP does not seem to be an instance of that project in philosophy. The Tractatus seems to me disjointed at points like this. But maybe that is only because I don't know it well.

*

Pythagoras and the TLP

The Pythagoreans defined 'point' as: a unit with position (a "position" is defined relative to other positions). Those "points" are Wittgenstein's "atomic objects", and the Pythagoreans' "objects" (solid bodies composed of points alone) are Wittgenstein's "facts". [See reality is number, in Philosophy of Geometry.]

If "the world" consists of "things" (atomic objects) standing in relation to other "things", like stars in a constellation, then the world must be mapped, as with a Cartesian grid, if it is to be intelligible. And then the question becomes: Where is the origin of that grid? and the answer is found in this, that "the world is my world" [5.641] -- that is, I, "the metaphysical-I" [5.633], am positioned as the origin of the grid.

The simplest proposition, the elementary proposition, asserts the existence of an atomic fact [i.e. of "a combination of objects" (2.01); "The object is simple." (2.02)].... The elementary proposition consists of names. It is a connection, a concatenation, of names [of objects]." (4.21, 4.22; tr. Ogden)

In an ideal language, which our everyday language isn't, each atomic object would be identified by a unique name, i.e. it seems that an ideal language would consist of nothing but proper names. [There is elsewhere discussion (or explanation of meaning) of the topics sign versus symbol, rules of grammar, ideal language and reform of language, in the context of Wittgenstein's earlier and later philosophies.]

*

Or maybe "language is a cage"

"Language is not a cage", Wittgenstein told Waismann (LE/Notes), 17 December 1930. From which it should follow that if anything can't be said, it is only because we are unwilling to make rules that would allow it to be said. As for example Wittgenstein's statement about "absolute value":

I would reject a priori any explanation of meaning on the very grounds that it was an explanation of meaning. (Cf. LE p. 11)

If the eternal questions without answers can't be answered (i.e. are "unanswerable"), why can't they be answered? Is it because we choose not to make them answerable -- or is it because, given the nature of our existence, we simply can't answer them?

Some propositions are vague in meaning because we ourselves have chosen not to make rules that would make their meaning clearer, having uses for indefinite concepts as well as sharp ones. For example, how much exactly is "a handful of sand" or how many exactly is "a forest full of trees"? But the eternal questions of philosophy are not of this kind.

If language is not a cage, then as to trying to know "what can't be known" -- why "can't" it be known? "If a proposition is unverifiable, it is because we ourselves made it unverifiable" (Z § 259) -- e.g. by not setting grammatical criteria for how it is to be verified.

If the limits of language are where we place them, then there are no limits except those of the "grammar" (an explanation of the meaning of the word 'grammar' in Wittgenstein's extended use of that word) we use or can invent (i.e. describe the practice of following). Imagination is the only limit of concept-formation = logical possibility.

On the other hand, the following remark seems to suggest that maybe in Wittgenstein's view, language is indeed a cage:

The results of philosophy are the uncovering of one or another piece of plain [in contrast to disguised (PI § 464)] nonsense and of bumps that the understanding has got by running its head up against the limits of language. (ibid. § 119)

Here Wittgenstein certainly doesn't appear to be saying that mere conventions, rather than facts of nature, form the walls of the cage which is "the limits of language".

But, on the other hand, the remark that "different concepts would seem natural if some very general facts of nature were different" and "we can always invent fictitious language games" (cf. ibid. II, xii, p. 230) appears to suggest that language is not a cage. The question remains unresolved, nonetheless, whether philosophy's unanswerable questions are unanswerable because we ourselves make them unanswerable -- or because of the nature of our existence.

The facts that can account for concept-formation

That language is a cage -- i.e. that the limits of language are imposed by general facts of nature -- is the older, the TLP's view. But what is Wittgenstein's later view? "Let's only bother about explanations of meaning" (PG i § 32, p. 69) -- that is to say: with describing the conventions of language, its grammar or rules of use -- i.e. with the limits of language in that sense of 'limits' only -- without regard to whether those conventions are imposed on us by general facts of nature or not. (The grammatical versus the natural limits of language.)

If the formation of concepts can be explained by facts of nature, shouldn't we be interested, not in grammar, but rather in that in nature which is the basis of grammar? (PI II, xii, p. 230) [But] this description gets its light, that is to say its purpose, from the philosophical problems. These are, of course, not empirical problems; they are solved, rather, by looking into the workings of our language ... (ibid. § 109) [But] we are not doing ... natural history -- since we can also invent fictitious natural history for our purposes. (ibid. p. 230)

Accounting for concept-formation is not Wittgenstein's aim in his later work in philosophy. Should it have been, as it was for philosophers from Plato to Kant? What is the relation between our concepts and the natural world? But that is not a question in Wittgenstein's later work of clarification of meaning in philosophy ( CV p. 19 [MS 154 15v: 1931], TLP 4.112), although it is not entirely unnoticed either (PI § 142, OC § 617).

The meaning of 'nonsense' in the Tractatus Logico-Philosophicus

Preliminary: Wittgenstein does not say that only propositions that can be verified by the senses have sense. But it is implied -- because surely that is exactly what the propositions of natural science ultimately are: propositions that can be objectively verified by experience, propositions that can be verified at some point by the five senses:

"Every scientific hypothesis, if it is to be meaningful, must be begotten of observation and give birth to verifiable predictions. And these initial observations and subsequent verifications must be capable of being described in terms of immediate sensory perceptions." (M. O'C. Drury, "Fact and Hypothesis")

So that even if verification is not spoken of in the context of sense and nonsense, the presumption that it gives language sense is present in the background of the TLP's picture of how our language works -- i.e. of what the logic of our language is.

Wittgenstein told Malcolm that when he wrote the TLP he thought it was the task -- not of logic-philosophy -- but of the natural sciences to determine what the TLP's "objects" ("logical atoms") are. But how could natural science do that except through objective experience, which is to say verifiable by the senses experience? Or what would Wittgenstein have meant by 'the language of natural science' -- for is not the language-that-is-not-nonsense model of the TLP the language of classical mechanics? (I am asking, not telling.)

Question: can there be sense without reference in the TLP's account of "the logic of our language"? or is sense without reference nonsense? (The title of Frege's essay: "Sense and Reference" (1892))

*

What is the meaning of the word 'nonsense' in Wittgenstein's book? Here is one possibility, and if it were correct, it would make Wittgenstein's meaning clearer. In the TLP, 'nonsense' means 'sense-less' -- i.e. 'independent of (not tethered to, but floating free of) the five senses'.

And that there are important things that cannot be perceived by the five senses is shown by all that is "non-sense" -- e.g. ethical and aesthetic "value" and God -- in our life. (This also shows that neither language nor "the world" is what is important: "The limits of my language mean the limits of my world" [5.6], but the limits neither of what is important in reality nor of in life, e.g. "... all that music has meant in my life.")

The "mystical" is non-sense. It "cannot be put into words" [6.522] that are not nonsense, because words with sense are names of atomic objects (or atomic facts) that are correlated to sense data. And any word that is not that type of name is nonsense.

"... what lies on the other side of the limit will be simply nonsense." ("Preface", tr. Ogden; that limit would, I think, be the limit set by the five senses)

By the word 'nonsense', then, the TLP would mean any language sign [6.53] that is "mere sound without sense" -- with the words 'without sense' here given the meaning 'not dependent on the five senses'.

Because by the word 'nonsense', the TLP cannot mean language that is "an unmeaning sound like the noise of hammering at a brazen pot" (Plato, Cratylus 430a, tr. Jowett). Because if it did, then the Tractatus Logico-Philosophicus itself would be mere noise -- which it is not, for indeed Wittgenstein compares his work to a ladder [6.54] that can be climbed, whereas nothing whatever is communicated by undefined combinations of words. And thus the observation that "nonsense" that can convey meaning is after all not nonsense. That is the fundamental flaw in the TLP's account of the logic of our language. (Eccentric or "fantastically arbitrary".)

*

The Recovery of Philosophy

Philosophy consists of logic and metaphysics ... (Notebooks 1914-1916, 1913)

We feel that even if all possible scientific questions can be answered, the problems of life have still not been touched at all. Of course then there is no question left, and just this is the answer. (TLP 6.52, tr. Ogden; see also TLP 4.003)

There is no question left because the only language that is not nonsense is "the language of natural science". And that is the only "of course" about this -- i.e. one proposition is derived from other propositions (OC § 1). And thus nothing "metaphysical" [6.53] or "ethical" [6.42] can be said -- i.e. put into words that are not nonsense -- and thus all philosophy except logic is impossible, were it not that language when used by logic is also nonsense. (Another word for 'metaphysical' (broadly taken) in the TLP is 'mystical'.)

Wittgenstein notwithstanding, the questions asked in philosophy are not the questions asked by natural science, and vice versa. Philosophy and science have different remits (fields of investigation).

The primacy of the proposition

Philosophy's concern is above all for propositions, because only a proposition can be true or false, and philosophy seeks to know the truth (in logic, ethics, and metaphysics). The difficulty is that propositions are not all of a kind: "the kind of verification is the kind of language game" (cf. PI II, xi, p. 224): there are in the grammar of our language many proposition-types, not only the natural science type.

"Because all words are names of perceptible objects, language can only talk about what can be perceived, and because value and God are not perceptible, language cannot talk about them." That does not seem terribly profound, even if it were true (much less if it weren't nonsense). That is, there may be insightful remarks in the TLP, but its picture of language is not one of them.

[The declarative sentence and logical form]

Socratic Ethics

And yet, die Wunder sind geblieben as well as the language to express it in, as Wittgenstein later acknowledged. (Cf. Frank Ramsey's notes apropos, about a thinker Russell called "a very singular man".)

In other words, as part of "The riddle [that] doesn't exist" [6.5] but does, ethics, also remains. And it remains as part of philosophy, if ethics is rational, which it is. For if ethics begins, as Socrates does, with the Delphic precept "Know thyself" and seeks to know the excellence (above all the "moral virtue") that is proper to man, it discovers what the good for man is, and thus how man should live his life (i.e. in accord with that good) -- then Wittgenstein's question in (non-philosophical) ethics (non-philosophical because it is not a thoroughgoing use of reason) about "absolute value" doesn't even arise.

Or again, if ethics begins, as Plato does in the Republic, by contrasting what the good man does with what the evil man does, and asks rhetorically, "If the good man harms his enemies, then what does the evil man do", it sees that the good man harms no one -- then Wittgenstein's question about "absolute value" (which is, I think, a willful, misological, rather than a philosophical, notion) doesn't even arise. [About 'tautology' as defined in the TLP.]

On Wittgenstein's account, the subject of ethics is "absolute value", but as the above discussion of Socratic ethics shows, that need not be the case. The concept 'value' is not essential to ethics. It is not necessary to think in those terms (i.e. use those concepts). Socratic ethics doesn't need a world-view ("absolute value", God) as its foundation; it can stand on its own.

Hume's empirical propositions and ethics

When two principles really do meet which cannot be reconciled with one another, then each man declares the other a fool and heretic. (OC § 611)

Collisions of non-rational principles, as of religious doctrines, aside, that is not the result of a public investigation. Wittgenstein simply asserts that "there is no disputing values" (Seneca).

Socrates versus Hume

"Can an ought be derived from an is?" Socrates shows that ethics can be derived from man's nature and condition. And further that the question misconceives what ethics is, because the good for man is moral nobility: and moral nobility consists of justice, piety, courage and self-restraint (the contrary claim lands one in endless absurdities). Yet from Hume to Wittgenstein philosophers have asked that question and answered: no, because what is and what is not of "ultimate value" is not a question of empirical fact that can be disputed, but of irrational instinct, inclination and conviction. (Hume's account of ethics is poor in categories (proposition types).)

... misconceives what ethics is, that is, looks at ethics the wrong way. "A philosopher says, Look at things this way!" But must anyone look at ethics one way rather than another, such that anyone who does not look at the conduct of our life from the viewpoint of Socrates is not looking at ethics philosophically, but instead going against reason? We can place different conceptions side by side. And we can note that with respect to "value" what Wittgenstein says: "Here I do not use reason."

Incomparable ≠ Senseless

That metaphysical propositions are not verifiable -- i.e. are not propositions of natural science -- does not strip them of their use in our language (Malcolm, Memoir (1984), p. 55). An incomparable picture is nevertheless a picture.

All of which has the result that, contrary to Wittgenstein's view, Philosophy consists of Logic, Ethics, and Metaphysics, and "logic of language" is its tool (and "a mind, too", which I doubt more and more that I have).

The depth of philosophy

And so it is not to be wondered at that the deepest problems are really no problems. (4.003, tr. Ogden)

Where science must come to a halt, where its progress is barred by unsurmountable barriers, there begins the realm of art which knows how to express that which will ever remain a closed book to scientific knowledge. I Rektor magnificus of the University of Vienna, bow humbly before the former assistant teacher of Windhaag. (Adolf Exner, January 1892, quoted in Hans-Hubert Schönzeler's Bruckner, rev. ed. 1978, p. 99)

At Windhaag in 1841-1843, beyond teaching in the school and playing the organ in church, Bruckner had been made to work in the fields after school, fetch and carry and spread manure. (ibid. p. 25)

But art and music do not come in at the limit of scientific knowledge, as a sort of supplement. Nor is their relationship reversed. Music and science do not have a linear relationship (It is not a question of which comes before the other, as with Aristotle's meta-physics), nor have they a parallel relationship. In fact they are discontinuous: they have no relationship at all: "... will ever remain a closed book". (Remits: the limits of an investigation.)

*

"The limits of my world"

Question: why does Wittgenstein say "the limits of my world" rather than "the limits of our world"? It suggests that every man lives in his own world, perhaps unable to know the world. Is the world, then, the totality of propositions of natural science? Since "the world" of the TLP is only the subject studied by natural science, it appears that 'the world' ≠ 'reality' in that book, for otherwise there could be no "the mystical". (In a Venn diagram, the sets {the world} and {reality} are not coterminous, but rather {the world} is a subset of {reality}, an unimportant subset in Wittgenstein's view.)

In fact what solipsism means, is quite correct, only it cannot be said, but shows itself ["makes itself manifest"]. (5.62, tr. Ogden [tr. Pears, McGuinness])

But if it cannot be said, then how is it that Wittgenstein is saying it? Because in the jargon the Tractatus, the word 'said' = 'put into the language of natural science', and 'Solipsism is correct' is not of that proposition type. It is "nonsense" that apparently is not nonsense -- i.e. not meaningless.

And if solipsism is correct, then why isn't the TLP's first proposition: "My world is all that is the case"? Because if solipsism is correct, then what do I know beyond the existence of my world -- indeed, I have no reason to suppose there is anything beyond it. What do I know of "the world" -- unless 'the world' simply means 'my world'? (An hypothesis to account for the text: the world is the totality of facts; my world is the facts I know. Or, in other words, 'the world' means the all the propositions of natural science which are true, whereas 'my world' means the subset of those facts which I myself have language for.)

That the world is my world, shows itself in the fact that the limits of the language (the language which I understand ["of that language which alone I understand"]) means the limits of my world. (ibid.)

But what is this language that is the only language I understand? If it is not the language of natural science, then it is nonsense. So, then, the more natural science someone knows the more extended the limits of his thought, language and world are. [TLP 3.01: "The totality of true thoughts is a picture of the world" (tr. Ogden), 4.001: "The totality of propositions is the language" (tr. Ogden).] Is that Wittgenstein's meaning here? (The word 'world' in the TLP.)

But Wittgenstein says knowledge of "the world" is knowledge of "what is lower" [ibid. 6.432], meaning that it is not "the important part" to know. Indeed, the part that is important to know is the part that cannot be "said" -- i.e. the language which is "nonsense" and "cannot be thought".

What is the relation between "the mystical", "the world", and reality -- that is, among those concepts in the TLP?

The World ≠ Reality

Question: if the limits of language are the limits of the world, then does "the world" include God, ethics, aesthetics (music, the appreciation of art), love? and everything else which according to the TLP we cannot talk about, cannot put into words, but must pass over in silence?

Are all those things not part of the world? Then 'world' apparently doesn't equal 'reality'. The limits of my language are not the limits of my reality.

The TLP concerns itself only with the limits of language ("the logic of our language is misunderstood") -- i.e. with the island that is the "world" (Engelmann's metaphor), not with the vast ocean beyond it (but that ocean is just as real -- i.e. just as much part of reality; indeed, that ocean, "the mystical", is the important part of reality, in Wittgenstein's view) --, which is determined by language's relation to the world. What "cannot be put into words" = "what cannot be pictured" belongs neither to the world nor to my language (the essence of language being to picture: we make for ourselves picture-models of the facts [Engelmann, Memoir p. 100; TLP 2.1]).

What was important to Wittgenstein was not "the world", not "what can be put into words', but the inexpressible (i.e. that which cannot be put into words) "the mystical". The limits of my world are the limits of the unimportant (but also the beginning of the important).

["The limits of my language mean the limits of my world." -- The press gives these words the opposite of Wittgenstein's meaning: 'nonsense', 'the world', and 'reality' in the TLP's jargon.]

What gives the TLP's "nonsense" its meaning?

How can I understand nonsense -- i.e. what gives "nonsense" language its meaning? and what is its meaning, because it does not consist of names of atomic objects or atomic facts? For it must be that I am able to understand it -- because the TLP is itself a book of "nonsense" -- and we can neither construct nor climb a ladder [ibid. 6.54] -- i.e. derive meaning from -- undefined combinations of words.

That there are different proposition types is an essential logic of language insight, but to call all proposition types except the natural-science type 'nonsense' is simply to rename the name-of-object versus name-of-abstract-object distinction, calling the latter "nonsense language" rather than "abstract language". And so the question becomes, What gives "nonsense" = "abstract language" its meaning? And now we have arrived at the first sentence of Wittgenstein's The Blue Book and at what I called "Wittgenstein's logic of language".

"... it must be that I am able to understand it." But what is the word 'understand' to mean here? Because as we normally use the word 'nonsense', language that can be understood is not nonsense. But just the contrary.

Is there an escape from the mind?

In fact what solipsism means, is quite correct, only it cannot be said, but shows itself. (TLP 5.62)

Compare Descartes' "New Way of Ideas" -- If philosophical thinking begins in the mind with the axiom that "The direct object of perception is an idea in the mind", then it is impossible for philosophy to ever get out of the mind, because it can know nothing about "the world" outside itself, not even that there is such place. And neither can the solipsist get from "my world" to "the world" -- what begins in the mind or in "my world" ends in the mind or "my world".

Does Wittgenstein use the word 'world' ambiguously in the TLP?

If good or bad willing changes the world, it can only change the limits of the world, not the facts; not the things that can be expressed in language.... The world of the happy is quite another than that of the unhappy. (6.43, tr. Ogden)

I find this incoherent. Aren't "the facts" the limits of the world -- for remember "The world is everything that is the case. The world is the totality of facts" (1.0, 1.1)? But man's attitude towards the facts does not change the facts, and so how can it change the limit of the facts -- namely "the world"? And "The world and life are one" [5.621] -- but what happened, then, to "all that music has meant in my life" -- as well as all that aesthetic and ethical value and God "have meant in my life"?

But if A shows itself, then isn't what A shows the meaning of A?

In the Tractatus Logico-Philosophicus, a "nonsense" combination of words such as 'ethical and aesthetic value' [6.42, 6.421], 'the higher' [6.432] or 'that the world exists' [6.44] must have meaning in some sense of the word 'meaning'. But, according to that book, it is meaning that cannot be "put into words" that are not nonsense -- meaning that no one can "give an account" of that meaning to be put to the test in Socratic dialectic. But instead it is a meaning which mysteriously, unaccountably, shows itself (As if emphasizing the word 'shows' made anything clearer).

If God can show itself without the use of language, then Wm. James's deaf-mute Mr. Ballard (Z § 342; cf. ibid. § 109) could have had thoughts about the origin of existence and God before he learned language. But can the origin of existence and God be "show" itself without language? (Philosophy is discourse of reason: Mr. Ballard "language-less thoughts" about the origin of existence and God have to be into words if we are to understand them. But is it logically possible -- i.e. describable -- for the ideas 'God' and 'the origin of existence' to even exist without language?)

But how, then, do we know -- if we can know -- whether something does or does not show a meaning? For instance, is it possible for "what is higher" to show a meaning to some people only and not to others -- and how do we know whether someone sees what-is-shown rather than has an illusion? (As with Wittgenstein's metaphor "family resemblances" that leaves the word 'family' undefined, there is no defined difference between 'showing' and 'not-showing', i.e. all showing is seeming to show (PI § 258).)

The "connection between grammar and sense and nonsense" (BB p. 65) in Wittgenstein's later logic of language (which might be characterized as prescriptive: "Look at language this way!", because there are many meanings of 'meaning') is essentially public: any grammatical "account" is public and therefore objective (and if it were not rule-based and objective, would there be any reason to call it a logic of language?) -- and thus explanations of meaning can be put to the tests of Socratic dialectic. But the TLP simply ignores any question of verification, of How do you know? It makes no distinction between mysticism and self-mystification.

In contrast to the TLP's meaning of 'nonsense', Wittgenstein's later definition of the word 'nonsense' (PI § 500) corresponds to our normal use of that word when by 'nonsense' we mean 'an undefined word' or 'an undefined combination of words', which is Aristotle's "mere sound without sense" (like the "music" of a weather harp). There are as many meanings of the word 'meaningless' as there are of the word 'meaning' -- but not all are useful to the philosophical understanding.

Note that Wittgenstein's later meaning of 'nonsense' is utterly different from that word's meaning when by 'nonsense' we mean 'foolishness' or 'absurdity' and by 'senseless' we mean 'foolish' or 'absurd'. For instance, the "nonsense verse" of Lewis Carroll's Alice books (PI §§ 13, 282), however absurd that verse may be, is not what is meant by the word 'nonsense' in Wittgenstein's later account of the logic of our language (at least according to my account of it).

[The German words sinnlos and Unsinn are used the same way as the English words 'senseless' and 'nonsense'.]

*

Final characterization of the TLP

The relation between Wittgenstein's Tractatus and his later work can be described in Wittgenstein's own words: the early work in philosophy is an example of the metaphysical speculation of those philosophers who mistake conceptual investigations for factual investigations (RPP i § 949). For the use of the word 'nonsense' in the TLP is nothing more than Wittgenstein's jargon: it is not, as he thought it was at the time, an insight into the essence of language and language's connection to "the world".

The Tractatus Logico-Philosophicus is a book to be read by the light of what followed it -- namely, Wittgenstein's later work in philosophy. (Wittgenstein wrote in his "Preface" that the Philosophical Investigations "could be seen in the right light only by contrast with and against the background of my old way of thinking". The old way is that "the meaning of a word is the object the word names".)

"All that music has meant in my life"

What is my understanding of Sraffa's (if my memory doesn't fail me about whose words these are, and it does, because they are Frank Ramsey's words and they do not appear to be apropos of Wittgenstein) criticism of the Tractatus, that "If you can't say it, you can't whistle it either" (Wittgenstein was a talented whistler)? Question: Is music nonsense, according to the TLP's definition of the word 'nonsense'?

There is a science of sound because sound is "in the world". But it "is possible to be interested in a phenomenon in a variety of ways" (PI § 108), and what interests our discussion is not the physics of sound, but music (what man hears but the machine does not when music is played) as an example of "what is higher" making itself manifest. Is this aspect of music "nonsense" that can show what words cannot say? But in this case, we would not say: "But "nonsense" that can convey meaning is not after all what anyone calls 'nonsense' -- and if you can whistle a meaning, then you can also put that meaning into words." No, not in this case. And so it may be strange that Wittgenstein lists God and ethical and aesthetic value, but not music, as things that "show" themselves.

It is impossible for me to say one word in my book about all that music has meant in my life; how then can I possibly make myself understood? (DW p. xiv; cf. Recollections p. 160: "How then can I hope to be understood?")

Natural science studies many things about sound, but not "all that music has meant in my life" (and were it to study it, it would not be from the point of view that interests us here). Nonetheless, what does "all music has meant in my life" have to do with understanding Wittgenstein's work in philosophy -- because it is that work we want to understand, not the man Wittgenstein himself. What has the "spirit in which it is written" [CV (1998 rev. ed.) [MS 109 204: 6-7.11.1930] -- although clearly Wittgenstein is talking here about the aim of the work, rather than its "spirit" (a cloudy notion) -- ] have to do with that work unless the work was written to deceive or mislead (Xenophon, Memorabilia iv, 6, 1)? Philosophical work is done in the spirit of wanting and seeking to know the truth, and the only test of its truth is how it stands against the tests of reason and experience -- which has naught to do with its author's character. (Could Wittgenstein really have said, as Socrates, Goethe and Albert Schweitzer could: "my life is my argument"?)

Wittgenstein and "everything great and important"

Is the spirit of the Philosophical Investigations shown by its "passing over in silence" what is, in Wittgenstein's view, most important in our life? But does Wittgenstein say that -- or does he say instead:

Where does our investigation get its importance from, since it seems only to destroy everything ... that is great and important? .... What we are destroying is nothing but houses of cards ... (§ 118)

By continuing to be silent about (refusing to discuss in his book) the philosophical questions that are "great and important" in our life, as the meaning of existence and ethics and why he "cannot say one word about all that music has meant", what does Wittgenstein show about his view of what philosophy is, can be, should be? Russell's "trivial" and "tea table activity" remarks apply in my view. Wisdom is, as Plato does, to discuss the "great and important" philosophical questions -- even if it is only, as Socrates himself did, to point out the limits of man's knowledge (and in some cases that man can know more than it is presumed by some philosophers that he can). Philosophy has to be used as a cure -- especially for Wittgenstein's overreaching (some ≠ all).

"Wittgenstein's Fundamental Idea did not Change"

One day we discussed the development of his thought and he said to me ... "My fundamental ideas came to me very early in life" ... I think perhaps the remark that Wittgenstein made, that after his conversations with Sraffa he felt like a tree with all its branches lopped off, has been misinterpreted. Wittgenstein chose his metaphors with great care, and here he says nothing about the roots or the main trunk of the tree, these -- his fundamental ideas -- remain I believe unchanged. (DW p. ix)

But how to decide which ideas were fundamental, whether Drury's belief that Wittgenstein's idea that (1) there are natural (not man-imposed) limits to language [4.115], and that (2) beyond those limits lies what cannot be put into words [(clearly) 4.116], but that (3) nonetheless shows itself: "it is the mystical" [6.522] -- was one unchanged fundamental idea for Wittgenstein or not?

The meaning (definition) of the word 'nonsense' in the Tractatus Logico-Philosophicus versus the meaning of nonsense language in the Tractatus Logico-Philosophicus. And this is it, that in the TLP, nonsense -- i.e. nonsensical language -- has meaning. To be nonsense -- to be a declarative sentence "without sense" -- means not to be a picture of how things stand in the world. Nothing more, nothing else.

To mark the limit of language (TLP's task, purpose)

Query: the Tractatus identifies the relationship between language and reality and defines the limit of knowledge.

(The limit of language being the limit of knowledge and thought.) That seems to be the task it sets for itself. The "picture theory" (inspired by the scale-model used in a French court case where there is a one-to-one correspondence between the word and the thing the words names) is its answer to the question of what the essential relationship between language and the world is, its key insight; the essence of that relationship marks the limit of knowledge = the limit of all that can put into words. (Whether the identification of the relationship Wittgenstein makes is correct is the question of whether "words are names and the meaning of name is the thing the name stands for".)

Query: what is the purpose of the Tractatus Logico-Philosophicus?

To identify the essential form of the proposition -- namely, "This is how things stand" (TLP 4.5). The scale-model in the Paris court case is a picture of how things stand in the world, just as a proposition is a picture of how things stand in the world. The proposition (statement of fact) with sense is one that corresponds (pictures) the objects in the world. And to identify thus the true limits of sense and nonsense is to demonstrate the impossibility of metaphysics, because all propositions that are not nonsense belong to physics (natural science). The limit of philosophy is what can be put into words -- and what philosophy wants to say (to talk about God, value, all that is higher) cannot be put into words. (To state facts is the essence of language.)

"Religious language is not metaphorical"

Die Reden der Religion sind auch kein Gleichnis; denn sonst müßte man es auch in Prosa sagen können. ("Religious language is not metaphorical; for you must be able to restate metaphors in prose.") (LE/Notes, 17 December 1930, p. 14)

... in ethical and religious language we seem constantly to be using similes. But a simile must be a simile for something. And if I can describe a fact by means of a simile I must also be able to drop the simile and to describe the facts without it. Now in [the case of ethical and religious language] as soon as we try to drop the simile and simply to state the facts which stand behind it, we find that there are no such facts. And so, what at first appeared to be a simile now seems to be mere nonsense. (LE, p. 10)

For example, is it true that the proposition 'God is like a father who loves his children' is a combination of words that has the form of a simile -- but does not have the use (meaning) of one -- because we cannot "drop the simile and simply ... state the facts which stand behind it"?

But what is Wittgenstein asking be done? "No father gives his son a stone when he asks for bread" (cf. Luke 11.11), and your Father in Heaven is like that. Is that proposition not a simile -- but e.g. a profession of faith instead?

What does Wittgenstein mean by "a statement that simply states the facts" -- does he mean a proposition that is a simile can be verified or falsified by experience, and that Jesus's proposition is not so verifiable and, therefore, it is not a simile?

It seems instead that the proposition 'God is like a Father who loves his son in that He never hands him a rock when asked for bread' is not a "simile" in Wittgenstein's sense of that word, although it is formally a simile: a comparison using the word 'like' or 'as'. It is not an empirical proposition, but it is a metaphor.

If a religious assertion cannot be put to the test, is it not an assertion at all?

A simile can be put to the test; a religious proposition cannot? But Frederick Copleston says of the proposition 'God loves mankind' that it does exclude something -- and, indeed, mustn't it exclude something if it isn't to be nonsense? Or can there be statements that are compatible with any and every set of circumstances? Surely "Either A is, or A is not". But what does e.g. Jesus's proposition exclude -- that man comes to harm rather than benefit? But it doesn't exclude that, does it? Is it true, then, that absent a test of experience, "what at first appears to be a simile shows itself not to be a simile"? (Religious similes are nonsense in the TLP's sense of the word 'nonsense' -- but is e.g. the proposition 'God loves mankind' a combination of words that conveys no meaning?)

"Not everything that has the form of a proposition (A simile is a proposition) is one"

According to Wittgenstein, religious beliefs are in no way hypotheses; they are not comparisons of facts -- i.e. they are not propositions in the TLP's sense of the word 'proposition', although they have that form. They serve instead as life-guiding pictures; for example, if someone believes in a last judgment, that is not belief that a proposition is true; it is belief that he ought to live his life in a particular way and that if he does not he will be punished in an afterlife. (The believer may believe that his belief is a proposition about reality ("how things stand or will stand"), but according to Wittgenstein it isn't "really" because it isn't used the way a proposition is used: e.g. it is not put to a test of experience. 'Religious belief is independent of anything that happens in this world' would be a grammatical rule.) (Likewise, Plato statements about the realm of Archetypes are not propositions, but instead only rules of grammar: a vocabulary and rules for using it; cf. the doctrine of Holy Trinity. All metaphysical statements are this way, according to Wittgenstein. The question is: what is the meaning of a metaphysical picture? The "grammatical meaning", that is. Or what other sense of the word 'meaning' is serviceable for philosophy?)

What is the difference between someone who says he believes in the Resurrection and someone who says he doesn't, if both men live the same way, such that an observer sees no difference in them? What does it "mean" to say that someone believes or doesn't believe, if 'meaning' is defined by way of life, such that words without deeds are "meaningless"? And what of another man who says the issue isn't whether an historical event is true or false or will or will not happen, because the Gospels are books of religious mythology not history, but nonetheless says he believes there will be a last judgment (as Plato does in the Gorgias explicitly on the basis of a myth, or Tacitus when he speaks of a moral posterity), does that mean e.g. that he examines his conscience and holds himself accountable for every misdeed? Can he say, 'I believe in a last judgment, although I don't believe such an event will actually occur'?

Reality of "the Mystical"

Religious belief is not belief about this, the visible world (that is, about perceptible reality), and statements of fact (propositions) according to the TLP are statements about this world only. But this world is not the whole of reality, because there is as well what Wittgenstein calls "the mystical", the part of reality of God and "value".

In contrast, in his later work Wittgenstein talks only about kinds of language use. He does not speak of a mystical reality, but only of a religious view of life. Does the TLP mean a mystical world (in contrast to the perceptible world) -- does Engelmann's metaphor suggest that it does? That distinction is not made in Wittgenstein's post-TLP work.

In the religious context Schweitzer said that the proposition 'God is the father' is "a thought of God's" (cf. Isaiah 55.9) that man can no more understand than a goat can understand a man's thoughts. But where is the difference between incomprehensible and nonsense (words without meaning) here?

Although, is that the only possible test -- i.e. the only possible meaning of Jesus's saying, that man does not come to harm? Or isn't that the question: What is the meaning of Jesus's saying if it is -- as it apparently is -- not the obvious one?

Elsewhere I asked: "Without Jesus, what life would the picture of God as the father have? Without the force of Jesus's personality, it would be dismissed as a falsehood ..." But isn't that a function of which meaning we assign to Jesus's words?

Can you really say that in religion unclarity is not a blunder?

Religious-propositions (Proposition type)

How can we restate 'God is the father' in prose? Well we can't, can we. I say "a comparison, an analogy is made and that analogy defines the statement". But does the definition also state that the picture 'God is the father' is not to be compared with our experience of the world? But does that "is not to be" belong to grammar -- i.e. is a comparison forbidden by the rules of the game? Can we say that verification-falsification is logically impossible (i.e. not describable)?

No, we can't say that (It would not be true to say that). The prohibition is made by piety: "You mustn't put God to the test." Jesus's saying that God is the father is an act of faith ("words are deeds" (CV p. 46)). Trust God, trust me, that it is so. Likewise Job has untestable faith in God: "Even though he kill me, I will trust in his goodness."

"Life has a sense"

As, I think (this is its meaning), Wittgenstein wrote in his Notebooks 1914-1916, "To believe in a God means to understand the question about the meaning of life. To believe in a God means to see that the facts of the world are not the end of the matter. To believe in God means to see that life has a meaning" (8 July 1916, tr. Anscombe), or as, in any case, I wrote: Faith in God is faith that life has a sense, and that it is a good not an evil sense: You could also say that by the word 'God' I mean 'life's meaning', and of course I believe that meaning is good not terrible.

Pregare è pensare al senso della vita. "To pray is to think about the meaning of life." (Notebooks 1914-1916, 11 June 1916)

Then do we want to call 'God is the father' an exemplar of a proposition type, a religious-proposition? What is defining of this type of proposition is not that these propositions are unverifiable-unfalsifiable -- but that their use excludes putting them to the test of experience. They would thus be unlike the "questions without answers" I describe -- because, being philosophical questions, there is no religious prohibition against asking those with the intention that they be open-mindedly examined (put to the tests of reason and experience), or in other words rationally asked.

Does 'use' = 'grammar' here? If any explanation of the use of language, if everything descriptive of the "language game", belongs to grammar, then Yes. Because if you don't follow that rule, then you play the game wrong or not at all [OC §§ 662, 446]. It looks a very strange rule of grammar; nonetheless it is essential to (defining of) the language game. (Grammar and piety.)

Restatements in Prose

Surely 'God created man' is not a restatement in prose of Michelangelo's ceiling fresco (cf. LC p. 63), which would be the "poetry" (metaphor) to be restated. Should the same be said about the side fresco, the Last Judgment, that it is not restated in prose by 'He will come again in glory to judge the living and the dead'? (About that picture I am not clear.)

As to the proposition 'God forgives man's sins', there is no empirical measurement of this (Language of measurement is here undefined) -- but does that make the proposition an undefined combination of words? But why shouldn't that proposition be said not to be a metaphor? For how would it be restated in prose -- Isn't it already prose?

Or should I say that 'God forgives man's sins' = 'God is like the human judge, the merciful judge, who pardons men their wrong-doing'? Is that the proposition's "logical form" (Russell)? A proposition's logical form is the form that makes its grammar (use in the language) clear. It is not as it were what the proposition really is. (The concept 'logical form' is after all a tool invented for our use in seeking clarity, not a metaphysical insight.)

In this case the simile form appears to make the proposition's grammar clear: it shows that we are making a comparison between God and a human judge, just as elsewhere we make a comparison between God and a human father. (Just as when those who call Jesus Christ the Son of the Father, as in the Trinity, are comparing Jesus to the son of a human father. But Jesus does not say that God is like a father to us, but that God is our father. Now, which is the logical form of his speech?) But when is being a simile not being metaphorical ....

If 'God created the world' is neither a statement of fact nor a simile, then how shall its proposition type be defined? We cannot simply call it a life-guiding "picture", because of pictures that guide our life there may be many types (some either are or are akin to statements of fact). About religious pictures, it does seem that they are both like and not like similes -- i.e. that they are defined by making analogies that it is forbidden to try to verify or falsify by projection -- i.e. comparison of the picture to the world of experience.

But we mustn't generalize based on a very few cases only

Because on the other hand, if we define the proposition type 'religious language' that way, then we have some examples in mind -- but not others. Because does that definition fit cases such as 'I believe there will be a Last Judgment', cases where nothing is grammatically forbidden. (If indeed there would be anything to forbid, because it is not logically possible (Just try to describe how) to put the proposition 'There will be a Last Judgment' to the empirical test.)

What we have defined (given a description, an account of) above is only one type of religious proposition -- not the essence of religious propositions.

Belief in God versus Belief that God exists

The proposition 'God exists' can be, as it usually is, falsified by the existence of evil in our world -- if, that is, that is the criterion of verification we set for that proposition, treating it as if it were an empirical hypothesis.

An argument against the thesis 'God exists' takes this form: (1) God must, by definition, be all-good, and (2) because to be all-good is, by definition, to do and make only what is good, (3) whatever is created by God must also be good. (4) But from the ethical point of view, much of the Creation is cruelly evil. (5) Therefore, God does not exist, if by 'God' is meant 'the Creator of all that is', which is its normal meaning.

The proposition 'I believe in God' seems to entail belief that God exists. But since the word 'exist' might mean countless things, that remark makes nothing clearer.

There is a distinction between the religious proposition 'I believe in God' and the non-religious proposition of the Natural Theology of Metaphysics (i.e. questions about God without reference to revelation -- whether with regard to "the God of the philosophers" or to Bonhoeffer's "God as a working-hypothesis" or Deus ex machina, which is akin to a pseudo-physics (natural science)) 'There is a God', where disproof of the proposition 'God exists' is logically possible -- i.e. definable. Likewise proofs by empirical evidence are distinguished from proofs by divine theology (i.e. questions about God with reference to revelation).

A false description of practice, a straw-man -- i.e. what no believer means by 'belief in God' -- often stands in the way of understanding religious belief.

For example, as stated above: 'I believe in God' = 'I have faith that life has a sense, that its sense is good rather than evil, although I don't know what its sense is.' That proposition is not put to any test, and indeed "The point is that if there were evidence, this would in fact destroy the whole business" (LC p. 56) -- i.e. there would be no concept 'religion' (and thus no religious propositions).

Belief in supernatural things

What the grammar of propositions of the form 'I believe in ...'? in religion, in contrast to advocacy propositions in ethics, e.g. 'I believe in forgiving wrongdoing" = 'One should forgive wrongdoing' = 'It is good to forgive wrongdoing'.

Is the definition of the religious proposition type tied to consequences for the way the believer lives his life -- i.e. is having a particular type of consequence defining of religious belief (I mean of our concept 'religious belief')?

Is that what we normally mean by saying that someone believes something, namely that he acts in a way that is different from how he would act if he did not believe?

What is the "logical form", the "philosophical grammar", of the proposition 'I believe in God'?

The proposition 'I believe in God' is not a metaphor. Or is there a restatement in prose that would make that proposition's meaning clear?

The concept 'religious belief' -- i.e. the grammar of the expression 'religious belief' -- is interconnected with concepts such as 'piety' (See "definition by related concepts"). Focusing on a particular word or form of expression may be an obstacle to the understanding.

If I want to talk about God, about something essentially imperceptible and to that extent indescribable, I have to use the only means we have at hand, which are anthropomorphic means, because by 'God' we mean something rational.

The declarative sentence and logical form

Bertrand Russell's concept 'philosophical grammar' = "logical form" makes a distinction between syntax and meaning, and rewrites forms of expression in a way that makes their obscured-by-an-inappropriate-syntax meaning clear. An example Wittgenstein gives is the declarative sentence 'It is God's will' rewritten as the command 'Do not grumble!'.

Now why am I so anxious to keep apart these ways of using "declarative sentences"? ... It is simply an attempt to see that every usage [Art = "kind", "type", "sort"] gets its due. Perhaps then a reaction against the overestimation of science. The use of the word "science" for "everything that can be said without nonsense" already betrays this over-estimation.

But of course the words "see that they get their due" & "overestimation" express my point of view.

The philosopher says "Look at things like this!" (CV (1998 rev. ed.) [MS 134 143: 13.-14.4.1947])

A command -- (like a rule of grammar: "And whether they take a form that we would naturally call a 'rule' or not, these remarks do the work of a rule, and so they are rules -- regardless of their form. In logic tools (i.e. signs) are defined by the work they are used to do") -- may have the form of a declarative sentence without having the logical grammar of a declarative sentence (which, according to TLP 4.5 is "This is how things [are perceived to] stand").

About the proposition 'God is the father', would Wittgenstein say that it is not really a declarative sentence? that the proposition 'No father when his son asks for bread ...' is really the command: 'Have faith; trust in God!' But Wittgenstein asks,

Now why am I so anxious to keep apart these ways of using "declarative sentences"? ... Did people in former times really not properly understand what they wanted to do with a sentence? (CV (1998 rev. ed.) op. cit.)

Did Jesus not really intend to tell his hearers about the nature of God and the kingdom of God, but only to issue commands for them to obey? Or the Fourth Gospel's statement "God is love" -- is that not really a declaration about God, about "all things visible and invisible" reality, but only a command to "love one another"?

Isn't Wittgenstein just imposing the materialist presumption, that since nothing exists that cannot be perceived by the senses, there cannot be any claims about "the imperceptible" that are really existential claims rather than fantasy (imagination); it is only false grammatical analogies that make it appear that there are?

Look at declarative sentences in fairy tales -- you would not say they are claims about an imperceptible reality. They are "pictures", but not pictures of a possible reality, which is what metaphysics and religions claim their "incomparable pictures" to be.

Can it be universally said that an incomparable picture can only be a command. Were Plato's statements about Forms, or Frege's view of geometry, really either rules of grammar they invented (rules for talking about things, like the grammar of 'Holy Trinity') or commands?

What is the grammar of the miracle stories?

"Bear and Forbear"

But can't the command 'Do not grumble!' be rewritten as the proposition 'Whatever happens is God's will, and the good for man is to accept it as such', which is what Stoicism's world-view and ethics say? Or can philosophy discover a form of expression's true logical form, i.e. the only possible use the expression can have?

But on the other hand, other logically possible uses must be described, not merely vaguely alluded to. What is a possible use of 'It is God's will' as a declarative sentence? A proposition does not say "how things stand if it is true" -- the users of the proposition do: they must set criteria for how things must stand if the proposition is true. And so far no such criteria have been set -- either by Stoicism or by Christianity -- and as such the words 'It is God's will', as a declarative sentence, is so far only an undefined combination of words.

It seems, therefore, that the logical form of 'It is God's will' -- if logical form = use in the language -- is command rather than statement of fact (proposition, declarative sentence); it is a guidepost, not an existential claim ... Yet it may be hard to accept that grammatical account -- why? is it a "stupid prejudice" that stands in the way of accepting it, namely a false grammatical analogy?

Does this discussion make the grammar of 'God is the father' clearer? Jesus does state how things must stand if that proposition is true ("No father when his son asks for bread ..."). But things don't stand that way, not as we normally judge things. And therefore is 'God is the father' a command (e.g.) rather than a false statement of fact?

The Catholic Profession of Faith, the Apostles, the Nicene Creed -- have these the logical form 'command' rather than 'declarative sentence' (Can they be rewritten as commands)? Can all religious doctrine/dogma be rewritten to its logical form 'command', if that is the logical form doctrine? No Catholic would say that. Nonetheless, even the apparent historical assertion "He was crucified under Pontius Pilate, suffered, died and was buried" is not what it appears to be, because, judged as historical fact is normally judged, namely by secular standards, there is no evidence that it happened or indeed that Jesus ever existed. So what is its logical form, or in other words its meaning if 'meaning' is defined as use in the language? But, on the other hand, if someone did not believe that the resurrection of the dead will be a real event, as real as the rising and setting of the sun ... You do not have to be a Catholic to love your neighbor as yourself, but you do to believe the promises of the Church regarding an afterlife.

God as Creator of the World and Ethics

I ask Wittgenstein: Is the existence of the world connected with the ethical?

Wittgenstein: Men have felt a connection here and have expressed it in this way: God the Father created the world [Cf. TLP 6.44], while God the Son (or the Word proceeding from God) is the ethical. That men have first divided the Godhead and then united it, points to there being a connection here.

(LE/Notes, 17 December 1930, p. 16)

The relationship is not that simple, not for a reasoning human being, that is. (Albert Schweitzer's distinction between God the Creator and God as an Ethical Personality.)

Although the rest of creation is amoral, in man God has created morality, the morality, which in Christianity, is given its highest form by Jesus's ethics of love and his picture of a kingdom of God governed by love rather than power.

And if God the Father created morality in man, then God the Father is not amoral -- although that does not make the how-ness of the world any less puzzling: the non-human creation remains amoral and, from the ethical point of view, so different from our picture of what we would expect it to be. (Nothing about the world from an ethical point of view makes sense to us. What we regard as our best impulses, the natural world appears to despise.)

"God as creator of the world and of ethics." -- There are laws of nature as there are also laws of ethics -- and both are natural law; both are rational. That is Socratic Ethics (and it does not use the word 'law' equivocally here). Had the Greeks pictured God as creator, it would have been as the creator of rationality, of what is rational -- i.e. of law; and, indeed, that is akin to Stoicism's picture of God.

Wittgenstein's picture of God as an arbitrary law-giver who declares what is good and what is evil is not the only possible picture.

Unidirectional comparisons

A comparison says "A is like B in such-and-such way", but if there is no A? For example, if 'A' names an abstraction, which is surely what the word 'God' does, can there be no comparison. Is there such a thing as a one-sided analogy? In some cases, there appears to be: for consider the case of describing something that does not, but logically might, exist. We say A is like B, even though there is no A, as e.g. "The kingdom of God is like ..." (although, note that the grammar of the word 'God' is not like the grammar of 'the kingdom of God').

[See also the discussion: Is Wittgenstein's method of language games -- because a method is what it is -- useful for understanding religion?]

Site copyright © September 1998. Please send corrections and criticism of this page to Robert Wesley Angelo. Last revised: 2 April 2016 : 2016-04-02 and 20 August 2015

The URL of this Web page:
https://www.roangelo.net/logwitt/remarks-tlp.html

Back to top of page

Wittgenstein's Logic of Language - Introduction and Table of Contents | Bibliography | Site Search | Site Map