Home | Bibliography and abbreviations | Site Search | Site Map

Wittgenstein's Logic of Language

Chapters 11-13

Note: there are 13 chapters, divided over three Web pages. There is a Table of Contents and an Introduction which explains the nature of the first principles of the later philosophy.

Outline of this page ...


11. 'Nonsense'. And Contradiction

The distinction between sense and nonsense is the foundation of Wittgenstein's logic of language. But what does it mean -- to say that a sign (marks on paper or in sand, spoken sounds, gestures, the physical part of language in contrast to the sign's meaning or use in the language) is nonsense?

When a sentence is called senseless, it is not as it were its sense [meaning] that is senseless [meaningless]. But a combination of words is being excluded from the language, withdrawn from circulation. (PI § 500) [Note 10]

... it's nonsense to say that the colors green and red could be in a single place at the same time. But if what gives a sentence sense is its agreement with grammatical rules then let's make just this rule, to permit the sentence 'red and green are both at this point at the same time'. Very well; but that doesn't fix the grammar of the expression. Further stipulations have yet to be made about how such a sentence is to be used; e.g. how it is to be verified. (PG i § 82, p. 127)

If a sign is nonsense (senseless or meaningless), it is only because we have not given the sign a meaning by defining rules for the sign's use ("fixed the grammar"). So by the word 'nonsense' here we mean 'an undefined sign' or 'an undefined combination of words or sounds'. And this is different from when by the word 'nonsense' we mean 'foolishness' or 'gross error' (Hobbes).

If a sentence is unverifiable, it is because we ourselves have made it unverifiable -- by failing or refusing to set rules for how it is to be verified (Z § 259). Language is our tool; we make the rules of the game. [Note 11]

Philosophical Scribbling

Philosophers often are like small children who, thinking to imitate their elders, scribble some marks on paper and then ask the grown-up, "What does this mean?" (CV p. 17; cf. PG vii § 43, p. 483)

'There are round squares' - false or language without meaning?

The sentence 'There is a square circle' (or 'There is a round square') is meaningless (i.e. an undefined combination of words). And yet if we describe how we actually speak, we shall not call that sentence meaningless -- we shall call it false. In a sense, the sentence is not meaningless: it is composed of English words, each of which we know how to do something with; in this it is not like the sounds an infant makes. But we don't know how to do anything with the combination of words 'square circle', and in that sense it is meaningless. And the sentence 'There is a square circle' is false, but only in the sense that there is nothing we apply that combination of words to, nor do we describe anything we might apply it to.

Must a contradiction be nonsense?

But we may want to say that the sign 'square circle' is not arbitrarily excluded from our language: i.e. that it is excluded because it is a contradiction. The sign 'This is a chair, and this is not a chair' is also a contradiction -- but, when pointing at different things, it has a use in the language.

Or should we say that the sentence is not a contradiction -- because the word 'this' does not have the same meaning in both instances?

The words 'here' and 'there'

No; the two words 'this' have the same meaning. 'Today' has the same meaning today as it had yesterday, 'here' the same meaning here and there. It is not here as with the sentence 'Mr. White turned white'. (RPP i § 37)

But then, what is the meaning of the word 'this'? Its meaning is not given by pointing to the chair that 'this' was used to point to; for if it were, then 'this' would be the name of the chair or a synonym for 'chair'. [Note 11a]

... the word 'name' is used to characterize many different kinds of use of a word ... -- but the kind of use 'this' has is not among them.

... it is precisely characteristic of a name that it is defined by means of the demonstrative expression 'That is N' (or 'That is called 'N'').

... do we also give the definitions: 'That is called 'this'' or 'This is called 'this''? (PI § 38)

The part of speech of the word 'this' may be said to be pointing-word or pointing-out-word. That is its use in the language ("grammar"); and so that is its meaning.

The part of speech of the word 'this' is demonstrative-pronoun. -- But note that a "pronoun" is not a name; e.g. 'he' is not another name for Mr. N[ame] N[ame], even if we use that word to refer to him. The unclear antecedent of a pronoun: we ask: who do you mean by 'him'? But if we answer: 'By 'him' we mean Mr. N.N. -- we are not giving a definition of the word 'him'. Just as if we say 'This is Mr. N.N.' -- we are not giving a definition of the word 'this'.

The meaning is not the bearer

The proposition 'The meaning of a word is the thing the word stands for' is false, because the meaning of a name is not the bearer of the name.

When Mr. N.N. dies one says that the bearer of the name dies, not that the meaning dies. And it would be nonsense to say that, for if the name ceased to have meaning it would make no sense to say 'Mr. N.N. is dead.' (PI § 40)

If the bearer were the meaning, then the name of a nonexistent bearer, e.g. 'fairy', would be meaningless. (I am being very literal because I want to be very literal. If 'thing' means 'anything and everything', then it has no clear meaning. In philosophy vague metaphors are stupifying, barriers to clarity and understanding.)

'Square circle'

A contradictory form does not stop a sign from having a meaning, but if we have no use for -- or if we assign no use to -- a sign, then the sign is meaningless. If we leave the combination of words 'square circle' undefined, we exclude it from our language. In other words, we ourselves make it nonsense (ibid. § 500).

Round and round Trafalgar Square the buses went like circus horses. (Graham Greene)

"How can a bus go round a square, because a square is not round?" Trafalgar Square can be called a square circle, a square roundabout. "But by 'circle' you only mean that traffic goes around this public square."

What we want to say is that this use for the combination of words 'square circle' is not the sense of 'square circle' that is excluded from the language. And so we describe that "sense": we draw a circle on the blackboard and then we draw a square next to it; and then we ask: well, how can a single figure be both this way and that? But what use of the word 'can' is this? What we are pointing out is a grammatical impossibility -- i.e. the absence of rules for the use of a sign, a definition; and the only way to eliminate this impossibility is to invent some use for the combination of words 'square circle'.

Or 'round square'? Suppose we round off the edges of a square but leave the sides somewhat straight --. "You have given the contradiction-sign a new (different) meaning." But then what was the old (original) meaning? The "meaningless meaning" or "senseless sense"? A sense that cannot be described is no sense. (Cf. Z § 249)

Again, you must not forget that 'A contradiction doesn't make sense' does not mean that the sense of a contradiction is nonsense. -- We exclude contradictions from language; we have no clear-cut use for them, and we don't want to use them. (RPP ii § 290)

Because the principle of contradiction is the foundation of all rationality, our naive response to contradictions seems a defense of our reason (sanity), but it shows our ignorance of the distinction between contradictions in form and contradictions in sense.

The Mistake of Moore and Russell

Russell in his "theory of descriptions" says that the "philosophical grammar" of the combination of words 'The round square does not exist' is 'There is no figure both round and square'. Wittgenstein (later, but not earlier (TLP 4.0031)) criticized this way of looking at language.

If I had to say what is the main mistake made by philosophers of the present generation, including Moore, I would say that it is that when language is looked at, what is looked at is the form of words and not the use made of the form of words. (LC p. 2)

"Is existence a predicate? It seems not, for we say 'Fluffy has paws', but not 'Fluffy has existence'." That is an example of looking at the form rather than the use of words. (From the latter point of view, the question is nonsense. What I am pointing out is the absence of rules of grammar.)

The Investigation's Conclusion rather than its Presumption

In Wittgenstein's later logic of language, meaning is not a matter of form (or syntax), but of use. But that "for a large class of cases the meaning of a word is its use in the language" (PI § 43) is the result of Wittgenstein's investigations, not their presumption (ibid. § 107). This contrasts with Wittgenstein's earlier work where it is presumed that language meaning is a function of form (TLP 4.5).

Thus although Wittgenstein selects a rule-based meaning of 'meaning' for his work in philosophy, that does not by itself determine which kind of rules language meaning is determined by. Instead, we must "Look to see" (PI § 66) whether rules of form (syntax) or rules of use in the language determine meaning. [Note 12] Wittgenstein did not presume to know. (The conclusion comes at the end of an investigation, not at its beginning.)

If the word 'logic' or 'the art of reasoning' is selectively defined as 'the study of rules', then the subject of logic of language is the rules of language, and this is why Wittgenstein uses the words 'grammar' and 'logic' indifferently. But implicit in this is that 'logic' = 'logic of language', and that Wittgenstein uses the word 'grammar' as jargon, and in that jargon not only syntax but also the rules of sense and nonsense are the subject of grammar.

The question of language meaning is the base of Wittgenstein's work in philosophy (ibid. § 111). This is why he talks about the logic (or grammar) of language (ibid. § 120).

The problems arising through a misinterpretation of our forms of language are as deep in us as the forms of our language and their significance is as great as the importance of our language. (PI § 111) Your questions refer to words; so I have to talk about words. (ibid. § 120)


Endnotes Chapter 11

Note 10: In what I am calling "Wittgenstein's logic of language" the word 'nonsense' is never synonymous with the words 'absurd' or 'foolish', although the word 'nonsense' is often normally used that way. [That statement needs to be qualified.] According to G.E. Moore's notes Wittgenstein said that:

where we say 'This makes no sense' we always mean 'This makes nonsense in this particular [language] game; and in answer to the question "Why do we call it 'nonsense'? what does it mean to call it so?" [Wittgenstein] said that when we call a sentence 'nonsense', it is "because of some similarity to sentences which have sense". (PP ii, p. 273)]

Examples of "similarity" nonsense are the absurd constructions the rules of syntax allow, e.g. 'Coffee takes me with cream'.

[Wittgenstein used the word 'nonsense' or 'senseless' very differently in the Tractatus Logico-Philosophicus, where a word may be "senseless" but nonetheless convey meaning. The TLP is a metaphysical theory about what language meaning "really is", not a "logic of language" (in my jargon), a theory which has the singular consequence that language may be "senseless" without being meaningless.] [BACK]

Note 11: Only, note that there is a relation between grammar and certain general facts of nature, and recurrent patterns of behavior (sorrow and other emotions).

Aristotle's definition for 'nonsense' was "mere sound without sense" (Posterior Analytics 83a30-33). Cf. Plato's Cratylus 430a). Thomas Hobbes' definition was "words whereby we conceive nothing but the sound". By 'words' Hobbes means combinations of words, given their normal use in the language.

If a man should talk to me of a round quadrangle or immaterial substances [incorporeal body or an eternal-now] I should not say that he was in an error, but that his words were without meaning (Hobbes, Leviathan, in Copleston, History, Volume V, I, 2).

By 'grammatical rules' or 'grammar' Wittgenstein means rules which distinguish sense from nonsense. [BACK]

Note 11a: If I gesture when I say the word 'this', I can use the same gesture each time I say that word. So that the gesture will be the same sign each time I point. And the gesture's meaning will also be the same, regardless of which object I point to. (A gesture is a language sign.) [BACK]

Note 12: G.E. Moore read a paper in which he considered the statement 'There is a fire in this room, but I don't believe there is a fire in this room'. Wittgenstein thought this statement, which he referred to as "Moore's Paradox" (CV p. 76, remark from 1948; also PI II, x, p. 190-191, but looked at very differently there), important because the statement is "something similar to a contradiction" but is not a contradiction in form:

it makes no sense to assert 'p is the case and I don't believe that p is the case'. This assertion has to be ruled out and is ruled out ... just as a contradiction is. And this shows that ... contradiction isn't the unique thing philosophers think it is. It isn't the only inadmissible form and it is, under certain circumstances, admissible [as it is for example in the assertion] 'Let's suppose: p is the case and I don't believe that p is the case'. (Wittgenstein's Letter to Moore, October 1944, M.42)

Here Wittgenstein is looking at the question solely from the point of view of form. He is treating language as if its meaning were a function of syntax, as if language's meaning were independent of its use. And so he makes the categorical remarks "it makes no sense" and "has to be ruled out". But looked at from the point of view of "meaning is use", Wittgenstein instead describes what we do with contradictory forms of language: "we have no clear-cut use for them, and we don't want to use them" -- but not as if the logic of language compelled us not to use them.

If someone told us that there was a fire but that he did not believe it, we would not respond "It's nonsense to assert p and do what amounts to denying p in the same breath! to assert that you believe a proposition is true while asserting that you don't believe the proposition is true!" Nor would we say "You mustn't utter combinations of words without meaning!" Instead we would ask him what he meant: e.g. did he mean that after the trouble he had taken over fire prevention, he was shocked that there was a fire? We would not pass a formal declaration of "nonsense"; rather we would ask for clarification, for an explanation of meaning. Not What is the form of this language? but What is this language -- regardless of its form -- used to do? [BACK]


12. 'Language games' in Wittgenstein's jargon (General definition)

Any description of the use of a sign (a mark or sound in contrast to its meaning) in the language is called 'grammar' in Wittgenstein's jargon. By 'grammar' Wittgenstein means both syntactic and semantic rules for the use of signs. But grammatical descriptions may take various forms.

For instance, we may describe the use of a sign as if it were a move in a game. In Wittgenstein's jargon this description is called a 'language game'.

By 'language game' Wittgenstein means "language and the actions into which it is woven, regarded as a whole", made up of:

(1) signs (spoken sounds, marks on paper, gestures -- i.e. words, phrases and sentences considered solely as physical objects) -- and

(2) the human activities (e.g. greeting, asking, describing, measuring) and the circumstances (meeting a friend, asking for directions to the bank, describing a giraffe to a small child, weighing a block of cheese) in which the signs are used to do some work in our life.

By 'language game' Wittgenstein also means "those games by means of which children learn their native language". (PI §§ 7, 23)

Examples as rules of grammar (explanations of meaning, definitions)

However, nothing serves better here as an explanation of meaning, making meaning clear, than examples. The following examples are "rules of grammar" or definitions -- i.e. descriptions of the use of a word in the language, namely the word 'language game' in Wittgenstein's logic of language. Although language games are as varied as what are normally called games, what defines all these "games" is their rules.

Primitive Language games

Wittgenstein described a very simple language game of the kind St. Augustine pictures: "As a child I learned the names of things" (PI §§ 1, 3). In that picture of language use: "words are names and the meaning of a name is explained by pointing to the object, or class of objects, the name stands for" (ibid. § 43). For example:

Givers and Takers

We can picture a community of givers and takers whose entire language consists of two words only: 'apple' and 'pear'. They use these words when they give and take their goods (of which there are only two kinds). That is all they ever do with language. When a taker comes into a giver's warehouse and calls out 'Pear!', the giver gives him a pear; and likewise an apple, whichever the taker asks for. That is an example of a "language game", a primitive but at the same time complete language game or language.

The meaning of this language is shown by what the giver and the taker do. And what their actions -- i.e. the game they play with words (or "language game") -- shows is the rules they follow for using the words of their language. That is the relation between the concepts 'rule of grammar' and 'meaning'. [Note 14]

With Counting

The two words of that language are both names of objects, and that is a language such as Augustine pictures. But now we can picture a slightly less simple language. We can picture that the giver and the taker count by reciting the numbers 'one' to 'three' -- which are all the numbers in their language. Then when the taker calls out '2 pear!', the giver gives the taker one pear followed by another pear as they recite the words 'one', 'two'.

Numbers are not names of objects here. Indeed numbers are not names at all, but they have a different use in this community's language -- as they do in our own. That is, we have introduced not only new words (sounds) here, but also a new use of language (a new part of speech), namely counting-words, to add to the above use of language name-of-object words.

We now have two examples of language games: the giver-taker game, and the giver-taker-with-counting game.

With Pointing

We might picture a further language game with the words 'here' and 'there' added to the giver-taker game. The taker places baskets before the giver and points and calls out 'Apple here! Pear there!' This introduces pointing-words to the community's language. These words are not the names of the locations pointed to -- indeed, they are not names at all, but like numbers, they have a different use in the language.

From simple to more complicated games

And we can go further, combining all three parts of speech. Now not only 'Pear!' or 'Apple here!', but '3 apple here! 2 pear there!' And we can add judgment-words and negation-words, e.g. 'not correct' if the giver gives apples instead of pears, and so on.

Continuing in this way we can build up more and more complex language games. [Note 14a]. In these games name-of-object will be only one part of speech among many others. To see this is one point of describing or inventing language games. [Note 14b]

The purpose of these inventions

When we look at such simple forms of language the mental mist which seems to enshroud our ordinary use [or meaning] of language disappears. We see activities, reactions, which are clear-cut and transparent. (BB p. 17)

Through these "activities and reactions" -- e.g. throwing a ball and catching it (Wittgenstein's original example) or giving and taking pears and apples -- the meaning [or one meaning] of language is made clear. The "mental mist" clears: language meaning no longer seems a nebulous something floating above our eyes, but a system of rules, public, verifiable, concrete, kept or broken (thus resulting in sense (kept) or nonsense (broken)).

Rules of the game, mistakes and nonsense

In the Givers and Takers game, if the giver gives an apple when the taker calls for a pear, the giver plays the game wrong, i.e. breaks the rules of the game. But if the giver gives a child's block when the taker calls for a pear, the giver does not play the game at all (OC § 446), because giving anything but an apple or a pear is not a move in this game; it is the equivalent of talking nonsense (undefined language).

Childhood's language games

These simple language games "are more or less akin to what in ordinary language we call games. Children are taught their native language by means of such games" (BB p. 81; cf. p. 17). Think of how a mother teaches her child to count or to add and subtract ("adding and take-away"): we call this teaching the child arithmetic, but we could with as much grammatical right call it teaching the child a game; indeed, from the child's point of view it is a game. (Cf. teaching the child to tell time with a paper clock face and paper hands.)


Parts of Speech as Language games

Wittgenstein also used the expression 'language game' to point out various kinds or "families" of sign use. (These are commonly called "parts of speech" or "word families", but there are far more parts of speech in Wittgenstein's grammar than in elementary school grammar.) Signs belonging to the same language game family -- e.g. the family of number-words, the family of psychological-words, the family of name-of-object words, and so on -- have the same general grammar. For example, it belongs to the grammar of name-of-object words that these signs are defined ostensively or demonstratively (i.e. that their meaning is explained by pointing to the bearer of a name), and that questions about e.g. size, shape, color, weight, and location, can be asked and answered. If we are told that an unfamiliar sign's part of speech is name-of-object, we already know much of the sign's grammar; we know what type of use to expect.

Consider the definitions 'A circle is a shape' and 'Red is a color':

The words 'shape' and 'color' in the definitions determine the kind of use of the word, and therefore what one may call the part of speech. And in ordinary grammar one might well distinguish 'shape words', 'color words', 'sound words', 'substance words' and so on as different parts of speech.

There wouldn't be the same reason for distinguishing 'metal words', 'poison words', 'predator words'. It makes sense to say [i.e. these are statements of fact] 'iron is a metal', 'phosphorus is a poison', etc. but not [i.e. these are definitions] 'red is a color', 'a circle is a shape' and so on. (PG i § 25, p. 61)

Nonsense is produced by trying to apply to a sign the grammar of a different language family from its own (PP iii, p. 312). E.g. with an apparent question like 'What color is the number 3?' (BB p. 47) which tries to treat the number-word '3' as if it were a name-of-object word. The [syntax-allowed] question 'How much does the number 3 weigh?' is nonsense. [Note 15]

Proposition Types as Language games

Wittgenstein writes "The kind of certainty is the kind of language game" (PI II, xi, p. 224; cf. § 481) and he contrasts three kinds of propositions: psychological certainty ("He is much depressed"), mathematical certainty ("25 x 25 = 625"), and foundational certainty ("I am sixty years old") -- i.e. here he is calling different kinds of propositions different kinds of language games.

Historical propositions such as 'Columbus discovered America in 1492' would make up another proposition type with its own type of certainty (and method of verification). And there will be many other proposition types each with its own type of certainty and method of verification (in cases where there is a method of verification; if there is no method, that is an important part of a proposition's grammar (Malcolm, Memoir (1984), p. 55)).

Ethical propositions

We can contrast the ethical proposition 'The good man harms no one' (Plato, Republic 331e-335e) with the experiential proposition 'The book is on the table' (Wittgenstein's example given to Parak: "If a book is on the table, then this is a fact") by saying that the type of verification is the type of language game, and therefore that ethical propositions and experiential propositions (statements of fact) are different language games. Plato's proposition, which is tautological, is verified by reason alone, whereas the statement of fact is verified by experience.

But even within ethics there is another language game or proposition type, namely those propositions which are answers to questions such as 'What is the good for man?', answers which are philosophical theses to be put to the tests of Socratic cross-questioning (namely reason and experience), to be agreed to as true or refuted as unclear or false. The proposition 'The good for man is life in accord with the excellence that is proper and unique to man, namely rational moral virtue' is tested in a quite distinct way (experience is a test) from the propositions of Plato's grammatical method in ethics (reason alone). And that proposition is distinct from the purely experiential proposition language game type, because it is based on an axiom ("The good for a thing is existence in accord with the excellence proper to that thing").

Religious propositions

And we can contrast ethical propositions with religious propositions such as 'God is the father', where experiential verification is not logically impossible (ways of verification could be invented: "If the gods do evil, they are not gods" (Euripides)), but instead forbidden by the rules of the game -- but not "game (as in language game)" -- because here the prohibition does not belong to grammar. Because what piety prohibits is not nonsense; indeed, if it were nonsense , it could not be prohibited. Not every game in which language is used is a "language game"; not every rule is a rule of grammar.

That would be one possible way to define the proposition type 'religious-proposition': as a statement and act of faith.

Other religious propositions, e.g. 'There will be a Last Judgment' and 'Life can educate one to belief in God' (CV p. 86, remark from 1950), are unverifiable, but unverifiability alone cannot serve as their defining quality -- because the absence of a method of verification applies to other types of propositions as well, e.g. to many foundational propositions (e.g. if I say 'Here is a hand' (Moore) in normal circumstances) and to first-person sensation propositions ('I have a headache'), which are not religious propositions.

Note that rules of grammar are also a proposition type. E.g. the proposition 'Elves do not exist' is not a statement about elves. Which proposition type is 'God exists'? Which shows that a proposition may be more than one type, e.g. both a grammatical proposition and a profession of faith.

Defining proposition types (rules)

Therefore we must not talk blithely of "proposition types as language games". We must set criteria to distinguish between types. And here a most useful criterion is one Socrates set for definitions, namely to state, not only what all propositions of one type have in common, but also what it is that differentiates propositions of that type from propositions of all other types.

Propositions of ambiguous use

Are some propositions essentially ambiguous? Because which is the proposition 'Logic is the art of reasoning' -- a rule of grammar (i.e. a rule for using the word 'logic') or a statement of fact (about the subject Logic, i.e. an hypothesis, e.g. historical, about the nature of that phenomenon), or is it both at the same time?

Or is it rather that the same object (marks on paper, spoken sounds) can be looked at from different points of view? that the same tool (language-sign) can be used to do more than one job -- but that it needn't be clear which job the tool is being used to do?

Should it be said that I am using [language] whose meaning I don't know, and so am talking nonsense? Say what you choose, so long as it doesn't prevent you from seeing the facts. (PI § 79)

Does Wittgenstein's logic of language break down here? If it points out the absence of a rule to resolve an ambiguity, logic has done its work.

Language games as forms of life

A further explanation of meaning Wittgenstein gave of his expression 'language game' is: A language game is a "way or form of life" (PI §§ 23, 19), a description of a way human beings live or might be imagined to live.


13. Meaning is not a Matter of Form, but of Use

Wittgenstein defined 'logic' as 'the study of rules', and the rules of language meaning are called 'grammar' (in Wittgenstein's jargon). But that does not say which kind of rules determine language's meaning -- for there are rules of form (syntax) but there are also rules of meaning (i.e. of use in the language or in a language game) that are independent of form.

Wittgenstein's first general explanation of the meaning of 'meaning' is that the meaning a word is the object the word names (or stands for), and the meaning of a proposition is the order (syntax) in which those names are arranged. That is the TLP's account of language meaning. [Note 13]

Wittgenstein's second explanation of the meaning of 'meaning' is that the meaning of a word, or combination of words, is its use in the language (PI § 43), and that use is most often not to be the name of anything -- but instead language is a collection of tools with many different uses. (I am calling naming a use of language, although Wittgenstein did not.)

Wittgenstein's first definition of 'meaning' is a very restricted variation of the traditional explanation of language meaning, although, in contrast, Wittgenstein's variation makes metaphysics (philosophy) impossible.

The written word in contrast to the word's use in the language (The dictionary)

The power language has to make everything look the same ... is most glaringly evident in the dictionary ... (CV p. 22, remark from 1931)

The differences between signs (i.e. patterns of ink) is shown on the printed page (This is language's syntactic grammar), but the differences between the signs' logical grammars (semantic grammar) is not shown there. (PI §§ 11, 122)

Three Syntactically Identical Sentences

The formulas of syntax (e.g. noun + verb = sentence) may make a language easier to learn, but a valid syntactic formula by itself does not guarantee that language has meaning. Indeed, syntactic analogies often suggest nonsense, i.e. combinations of words with no normal use in the language.

From the point of view of elementary school grammar -- which like mathematical logic is concerned with form (syntax) rather than with substance (meaning) -- the following three sentences are identical, because they have the same parts of speech arranged the same way: definite-article + noun + verb + preposition + possessive-pronoun + object-of-preposition (noun).

(1) The book is in my desk.
(2) The pain is in my shoulder.
(3) The plan is in my head.

That parsing of the sentence may help someone who is learning English. But it tells that person no more than it tells us about the meanings of those sentences -- i.e. it does not tell us how to use those sentences to talk (and think) sense rather than nonsense.

It does, however, tell us something about the meanings of the words that make up the sentences: e.g. 'noun' is defined as the name of a person, place or [any]thing [else]. That is, it tells us just enough to foster confusion, and even more confusion if the word 'substantive' is substituted for 'noun'.

When syntax does not produce nonsense, it can of course play a role in language meaning. For example, to someone who knows English, there is an obvious difference in meaning (i.e. normal use in the language) between 'The mouse is in the grain' and 'The grain is in the mouse'. (But compare 'The mouse is in the house' and 'The house is in the mouse'. Syntax does not tell you that the latter is normally nonsense.)

False Paths

We are all taught grammar at school. We are all led up the garden path:

Language contains the same traps for everyone; the immense network of well-kept false paths ... Therefore wherever false paths branch off I should put up sign-posts which help people get by the dangerous places (CV p. 18, remark from 1931) ... Human beings are deeply mired in philosophical, i.e. grammatical, confusions. [And to pull them out we] must, so to speak, regroup their entire language [into semantic parts of speech].

Teaching philosophy [i.e. philosophical grammar] involves the same immense difficulty as instruction in geography would have if a pupil brought with him a mass of false and far too simple ideas .... (Wittgenstein's chapter "Philosophy" § 90, in the "Big Typescript" [ca. 1933], tr. Luckhardt and Aue)

Being poor in categories: having only a single (and false) picture (model) of how our language works, namely that words are names and the meaning of a name is the thing the name stands for (whether that thing is concrete or abstract). Poverty in categories applies also to grammatical categories.

The Three Syntactically Identical Sentences, in Context. Philosophical Grammar.

Because most signs do not have essential meanings, it is only in the particular context that a sentence has not only syntax, but also meaning. It is only when we assume that "the meaning of a sign is the essence of the thing it names", which is a false assumption, that it appears otherwise to us.

(1) My sister is searching through the books on the bookshelf. She asks me where N.N.'s book is. I answer: 'The book is in my desk.'

(2) I have fallen down the steps. I am crying. A friend asks me where it hurts. I answer: 'The pain is in my shoulder.'

(3) I am building an oven out of bricks. A friend asks to see the plan. I answer: 'The plan is in my head.'

A "false and far too simple" view of language expects all nouns ("substantives") to do the work of the word 'book' in sentence (1: 'The book is in my desk'). In Wittgenstein's logic, the part of speech of the word 'book', in this particular example, is name-of-object; it is also the part of speech of the words 'desk' and 'shoulder'.

However, name-of-object is not the part of speech of the words 'head' (in this example) and 'pain'. The part of speech of the word 'pain' here is psychological-word. The phrases 'in my desk' and 'in my shoulder' are statements-of-location, but 'in my head' is not. The grammar of sentence (3: 'The plan is in my head') is similar to the grammar of 'The plan is not written down', despite a misleading grammatical analogy suggesting otherwise.

Non-name words

The word 'plan' is sometimes a name-of-object word (e.g. a printed blueprint), and sometimes not (which shows that the word 'book' is also sometimes not a name-of-object word: the content versus the manuscript). "And when it is not a name-of-object -- what is its part of speech?" Compare the words 'idea', 'concept', 'meaning' ... Wittgenstein never wrote an "entire regrouping" of our language into parts of speech. What we should not do is call these 'name-of-idea-words' (in contrast to 'name-of-object-words'), because that perpetuates the false picture "all nouns are names of things", nor call these words, as has traditionally been done, 'abstractions'. First, because that would be too varied a class (like noun, abstraction would include 'mind', 'time', 'five', 'beauty', 'elf'). And second, because the word 'abstraction' --

makes the difference between the meanings [grammars] look too slight. (It is like saying: numerals are actual, and numbers non-actual, objects.) An unsuitable type of expression is a sure means of remaining in a state of confusion. It as it were bars the way out. (PI § 339)

Abstractions would be like numbers "non-actual objects", ghostly things "on the other side of the sky" (Plato), were it not that 'abstraction' = 'abstract term', not "abstract object" (which is an unsuitable type of expression, a false analogy or comparison). [Note 16]

A "term" is a word, and if we define 'concept' as 'rules for using a word', then 'abstraction' = 'concept'. Why? Because there is nothing else for it to mean, not if meaning is a public event.

A form of expression may lead up the "false path" Wittgenstein wanted to put a sign-post at the entrance to; it would negate the purpose of his grammatical "regrouping". The word 'abstraction' "makes the difference look too slight", because it suggests "abstract object" in contrast to "physical object". But the words 'time', 'five', 'elf' and 'concept' are not names of anything: they have entirely different uses in our language. If we wanted to call that huge class of words something, 'non-name-words' (or 'non-name-nouns') might be clearest.

Sign-posts for false paths

But Wittgenstein did not need to regroup our entire language into semantic parts of speech. For his purpose it was enough to put up sign-posts -- i.e. assemble grammatical reminders. Wittgenstein's book Philosophical Grammar was not intended to be a textbook (PG, June 1931 margin note, quoted p. 487).

I am not aiming at some kind of completeness, some classification ... [My examples] are only meant to enable the reader to shift for himself when he encounters conceptual difficulties. (PI II, xi, p. 206)

Sign-posts help us to avoid going up "false paths" because if we see that name of something is not a particular sign's part of speech, we see that we have to look for a different part of speech, a different grammar.

Other examples of sign-posts are: Ask for the use rather than for the meaning -- because the word 'meaning' suggests the name of some thing whereas 'use in the language' suggests a tool. And: Ask for the use in the particular context because meaning is defined by examples not essences. In this way:

What the sign slurs over, its use makes clear. (TLP 3.262)

There is no thoughtless way to determine whether a proposition is a philosophical thesis or nonsense. We have to examine each particular case. Wittgenstein has given us some tools for doing this, but those tools have their limits. The solution to philosophy's and life's problems is not found in primitive language games, which is the only place thoroughgoing conventions are found.


Endnotes Chapter 12-13

Note 13:

Summary of the TLP's account of the logic of our language

The "world" (i.e. everything that both (1) exists and (2) can be put into words with sense) consists of simple things (atoms) standing in relation to one another. The words of language name these simple things, and propositions (declarative sentences of the form: This is how things stand) are strings of names that mirror the relations (arrangement) of those things to one another in the world. Thus the proposition 'The car is parked in front of the bank' has a different sense from the proposition 'The car is parked at the side of the bank', etc. That is the essence of language, of the only language with sense, i.e. the only language that is not nonsense.

Or rather it would be, except that nonsense that conveys meaning is not nonsense, as we normally use the word 'nonsense' (which is to mean either foolishness or undefined language). According to the TLP the paragraph above, indeed this entire Synopsis, is nonsense, as is the TLP itself. And therefore the TLP is not a logic of language in my sense (jargon) -- because it does not make a distinction between "nonsense with meaning" and "nonsense without meaning". [BACK]

'True' and 'false' in the Tractatus Logico-Philosophicus

The TLP's criterion of truth: "This is how things stand in the world if the proposition is true": if the elements (names of things) and their organization (syntax) in the proposition mirrors the elements (objects) and their organization (arrangement) in the world, the proposition is true (from which it follows that a proposition can have sense even if it is false). How is it that the proposition 'The paper is in tatters' has sense but 'The tatters is in paper' has none? It is because one proposition (sentence) depicts a possible arrangement of things in the world, whereas the other does not.

In the TLP, sense and nonsense follow from the possibility of being true or false, which is what the proposition alone can be. The result is that according to the TLP most language is senseless.

If the only propositions with sense are propositions made up of words that name objects in the world (although note that Wittgenstein doesn't say how atomic facts are translated into propositions made up of the names of perceptible objects), then any proposition containing non-name words ("abstractions") will be senseless [TLP 6.53]. And therefore philosophy will be impossible. (The language of the TLP is a language with a very limited vocabulary.)

Note 14: If the custom is for the taker to say 'Thank you' and the giver to reply 'Not at all', is that part of the language game, or is politeness like the paper hat (for it seems to do no work in the game: the transaction does not require it), or is it itself a language game? From the point of view of Wittgenstein's logic of language, it does seem pointless. Nevertheless it is a use of language (cf. greeting: 'Hello', 'Good-bye'). [BACK]

Note 14a:

Complicated games can be built by adding new elements to primitive ones

[Language games] are ways of using signs [spoken sounds, marks on paper, the physical aspect of language as opposed to its use] simpler than those in which we use the signs of our highly complicated everyday language. Language games are the forms of language with which a child begins to make use of words.... When we look at such simple [primitive] forms of language ... [we] see activities, reactions, which are clear cut and transparent. On the other hand we recognize in these simple processes forms of language not separated by a break from our more complicated ones. We see that we can build up complicated forms from the primitive ones by gradually adding new forms. (BB p. 17)

[BACK]

Note 14b: Note that it is not necessary for a language game (or language) to contain any names at all. For example:

A Language without names: A Narrow Pass

Picture a community who must cross a narrow path wrapped around a mountain, a path too narrow for more than one person to take at a time, and too rounded for anyone crossing it to see both ends of. Opposite the path, on an opposing mountain, stands a watchman who overlooks the entire path and directs traffic across it. If no one is crossing from the opposite direction he shouts 'Pass!', and if someone is crossing from the opposite direction, the watchman shouts 'Wait!' Those are the only words in this language, and that is the only use that is ever made of those words. The words' part of speech is "imperatives" or command-words.

As to the origin of natural language, language need not have begun with naming things (nor need children acquire language, as Augustine imagines, by learning the names of things). It may have begun instead with command-words as in the narrow-pass language game, as a response to a community's needs: "Everything arose naturally from their special needs ... everything, including their language" (The Country of the Blind). Words are tools. [BACK]

Note 15: The technique of asking "ungrammatical" questions is, however, sometimes used in creative prose to suggest imagery, as e.g. in the question 'Can a sound have blue eyes?' which Johannes Kreisler, alluding to Julia, asks in Kater Murr. Wittgenstein, however, was concerned with "philosophers' prose" -- i.e. with the marks and sounds with which we mystify ourselves because we don't understand the logic of our language.

'What color is the number 3?' This is a category mistake -- but the categories here are grammatical (These categories are the parts of speech), i.e. this is a question about conventions (rules, verbal definitions), not about "the categories of reality" (if that combination of words has any meaning).

It is in this context that can be found Wittgenstein's earliest definition for 'grammar' (i.e. explanation of his jargon; compare the explanation given to G.E. Moore between 1932 and 1933 (PP i, p. 256; ii, p. 276: "any explanation of the use of language") for the great change in Wittgenstein's thinking). Wittgenstein explained his thoughts to Bertrand Russell at the beginning of May 1930; Russell wrote in his Autobiography, Letters to G.E. Moore dated 5 May and 8 May 1930:

He uses the words 'space' and 'grammar' in peculiar senses, which are more or less connected with each other. He holds that if it is significant to say 'This is red', it cannot be significant to say 'This is loud'. There is one "space" of colours and another "space" of sounds.... Mistakes of grammar result from confusing "spaces".

Suppose, for example, a certain patch of wall is blue; it might have been red, or green, or &c. To say that it is any of these colours is false, but not meaningless. On the other hand, to say that it is loud, or shrill, or to apply to it any other adjective appropriate to a sound, would be to talk nonsense. There is thus a collection of possibilities of a certain kind which is concerned in any fact. Such a collection of possibilities Wittgenstein calls a 'space'. Thus there is a "space" of colours, and a "space" of sounds. There are various relations among colours which constitute the geometry of that "space".

Wittgenstein uses the word 'grammar' to cover what corresponds in language to the existence of these various "spaces". Wherever a word denoting a region in a certain "space" occurs, the word denoting another region in that "space" can be substituted without producing nonsense, but a word denoting any region belonging to any other "space" cannot be substituted without bad grammar, i.e. nonsense.

"Spaces" will soon be replaced by "parts of speech" or "language game types", and there will be no more talk about "geometry" either, for that will be recognized for what it is -- sc. metaphysical imagination added to the facts (an "explanation"), a picture that stands in the way of seeing the logic of language in plain view (PI § 305). [BACK]

Note 16: All concepts are "abstractions": we disregard aspects of things when we define -- i.e. set the limits of -- concepts. All concepts are a selection of data. But that applies equally to both name and non-name words. (As an example of abstraction: physics is an abstract science (Arthur Eddington).)

Note that it should follow from the theory of abstraction that by 'abstraction' we mean a common nature (an essence) and nothing else, and that the word 'abstract' is always a verb and never an adjective, that 'abstract term' and 'abstract object' are undefined combinations of words (i.e. nonsense).

However, that is not the only way the word 'abstraction' is normally (misleadingly and confusedly) used, for often by 'abstraction' is meant "name of an intangible something", a picture suggested (1) by the false grammatical rule: words are names and the meaning of a name is the thing the name stands for, regardless of whether that thing is an object, phenomenon or idea or is perceptible or essentially imperceptible, and (2) by superstition. That picture is what these remarks are aimed against. [BACK]


Synopsis - Concluding Remarks

Wittgenstein's revision of our common concept 'grammar', or revised use of the word 'grammar', is jargon (The word 'grammar' doesn't normally suggest definitions, i.e. rules of sense and nonsense). And therefore, ought anyone else to adopt Wittgenstein's "grammar" for the word 'grammar' for his own thinking in philosophy?

My account of Wittgenstein's "logic of language" (a phrase which in my jargon means: a way of distinguishing language with meaning from language without meaning) has been reasons for adopting that logic. But there are reasons against as well, as e.g. it is quite useless to Socrates' quest in ethics and Plato's quest to explain the concept-formation of common names, much less mankind's eternal riddles (which, Wittgenstein not withstanding [TLP 6.5], do exist).

But the reason for not adopting Wittgenstein's logic of language that should concern students of philosophy most is that: its jargon may be poorly understood and then stupidly applied (e.g. "autonomous language-games"). The danger is a careless, doctrinaire attitude towards Wittgenstein's ideas.

Whereas, on the contrary, a philosopher is not someone without doubts about his own work. Wittgenstein told Malcolm that he would like to make a copy of his book for his friends with notes in the margin such as "This is fishy" and "This is not quite right" (Memoir 2e, p. 75), and he thought that he sometimes contradicted himself, writing one thing one day, then something just the opposite another.

Wittgenstein told Moore that "a method had been found" (PP iii, p. 322) -- a method, not a preconceived model to which everything is forced to conform. In philosophy he often compares the use of words to playing games, where what defines a game is its rules. Making that comparison is Wittgenstein's method of language games. (It belongs to his general method of setting aside the question of truth and falsity to look at sense and nonsense instead.)

Logic and Philosophy

The question of whether anyone ought to adopt Wittgenstein's concept 'grammar' -- that is, use Wittgenstein's logic of language to distinguish sense from nonsense -- is a different question from whether anyone ought to adopt Wittgenstein's philosophy, which if it is an heir to past philosophy (BB p. 28), is an heir that either renounces its inheritance or dismisses it as nothing more than an illusion created by language (PI § 118, TLP 4.003). As to Wittgenstein's 'grammar' --

A new or revised concept may lessen confusion -- or only encourage more. (cf. CV p. 55, remark from 1946)

Concepts lead us to make investigations; are the expression of our interest, and direct our interest. (PI § 570) What I am looking for is the grammatical difference. (ibid. II, viii, p. 185)

What should be asked: Do Wittgenstein's methods have limited or universal application to philosophical problems? Are philosophical problems all mere "houses of cards"? e.g. as Socrates and Plato ask: "Is moral virtue knowledge of good and evil?" and "Is there an afterlife?"


Clarification: Logic of language's meaning of 'meaning'

Philosophers have sought criteria by which to test whether a proposition (statement of fact) is true or not. However, before a proposition can be true or false, it must not be nonsense. And therefore the question Is the proposition true? depends on the prior question What is the meaning of the proposition? Wittgenstein's own "new way of thinking" reverses the order of his old way, because it asks about sense and nonsense rather than about truth and falsity.

But not every meaning of the word 'meaning' makes the distinction between language with meaning and nonsense objective (verifiable), and thus not every meaning of 'meaning' can be used to find the truth. (The concepts 'truth', 'objective' and 'knowledge' are inseparable.)

Metaphysical ghosts

Whether "outer or inner", the ghosts conjured up by metaphysics are not the meaning of words in logic.

The difficulty in logic-philosophy is to clearly distinguish logic from metaphysics. Metaphysics seeks to explain the facts in plain view: its theories say that what is visible is explained by what is invisible; thus its speculative explanations are essentially unverifiable. Logic, in contrast, goes no farther than the rules, definitions, conventions, which are verifiable facts in plain view. Thus nothing is hidden (invisible) in logic (PI § 435).

Three metaphysical theories of language meaning:

  1. The meaning of a name belongs to the thing the word names: it is the thing named or the essence of the thing named; (1a) Plato: the meaning of a word is the supranatural Archetype (Essence) it imitates.
  2. The meaning of a word belongs to the word itself: it is like an atmosphere the word carries around with it wherever it goes (PI § 117).
  3. The meaning of a word belongs to the mind: it is an idea in the mind of the speaker and hearer of the language.

A metaphysical theory of meaning cannot be affirmed or refuted by the two tests of Socratic cross-questioning, i.e. it is not an hypothesis verifiable or falsifiable by reason and public and therefore objective experience. It cannot be knowledge.

Meaning as a Halo

You say to me: "You understand this expression, don't you? Well then -- I am using it in the sense you are familiar with." -- As if the sense were an atmosphere accompanying the word, which it carried into every kind of application. (PI § 117)

"You understand this expression, don't you? Well, the way you always understand it [in contrast to in this or that particular way] is the way I too am using it." This is to treat meaning as a halo that a word carries around with it and retains in every kind of application. (CV p. 44, remark from 1944)

There are many meanings of the word 'meaning', not only the one Wittgenstein chose to use in philosophy

Now if for an expression to convey a meaning means for it to be accompanied by or to produce certain experiences, our expression may have all sorts of meanings, and I don't wish to say anything about them. (BB p. 65)

As Wittgenstein used the word 'meaning' at one time, 'without rules' DEF.= 'without meaning'

I want to play chess, and a man gives the white king a paper crown, leaving the use of the piece unaltered, but telling me that the crown has a meaning to him in the game, which he can't express by rules. I say: as long as it doesn't alter the use of the piece, it hasn't got what I call a 'meaning'. (BB p. 65)

"Whatever seems right" in contrast to Logic

If we use the general definition 'logic' DEF.= 'the art of reasoning', we can contrast Wittgenstein's logic of language with W.E. Johnson's unobjective "logic" of language: If I say that a proposition has meaning for me, no one has the right to say it's meaningless. Johnson's "meaning" is the king's paper crown (ibid.).

This remark can only become clear if we understand the connection between [rules of] grammar and sense and nonsense.

The meaning of a phrase for us is characterized by the use we make of it. The meaning is not a mental accompaniment to the expression. Therefore the phrase 'I think I mean something by it', or 'I am sure I mean something by it', which we so often hear in philosophical discussions to justify the use of an expression is for us no justification at all. We ask: "What do you mean?", i.e. "How do you use this expression?" (cf. BB p. 65)

There are many meanings of the word 'meaning': This was the meaning Wittgenstein chose, because it makes an objective (verifiable) distinction between sense and nonsense in language. "What I am looking for is the grammatical difference" (PI II, viii, p. 185), the use of language that is governed by public rules, the moves of a language game in plain view. The king's paper crown plays no part in a game if a game (chess) is defined by its rules.

In Wittgenstein's logic of language, meaning is not a question of "whatever seems right" (PI § 258). Perhaps it would be clearer if we said that: for a large class of cases, a meaning of a word is its use in the language (cf. PI § 43).

Wittgenstein's particular interest in the phenomenon of language

It's possible to be interested in a phenomenon in a variety of ways. (PI § 108)

The phenomenon of language, for example. Wittgenstein's interest is not in theories of meaning: logic makes "no hypotheses", whether physical, psychological or metaphysical. It is descriptive of the facts that "are always before our eyes" (ibid. § 415) -- in plain view, only. "A philosopher says, Look at things this way". From the point of view of grammar and sense and nonsense was Wittgenstein's way of looking at language meaning (or logic).

Metaphysical theories can be invented about language, as by Plato, about a hidden reality imagined to explain concept-formation. And Wittgenstein's interest in language was, like Plato's interest, directed by philosophical problems (ibid. § 109), but, unlike Plato, in Wittgenstein's way of looking at things, those problems are not about the phenomena that words are presumed to name, but, rather, they are problems about the grammar of words, problems that are solved by using Wittgenstein's selected meaning of 'meaning' (or logic) -- namely the public conventions for using signs, something verifiable.

"Time is unreal"

When a philosopher asks if time is real, this sounds as if the meanings of the words 'time' and 'real' were already known, and that therefore the question were about time itself. This assumes that the meanings of the words 'time' and 'real' are the same in all contexts in which we use those words.

The view of the logic of language taught at school is that 'word definition' means what is found in dictionaries -- i.e. brief equivalent-word formulas (general definitions) -- where the meaning of a word is a generality whose applications should be clear to any educated person, and that on the basis of such definitions we can leave considerations about language behind and talk about things rather than about their names.

Language follows thought and thought follows things. (Copleston)

In other words, the philosopher "defines things, not words": a definition states what the nature of a thing, e.g. time, is, which assumes that the meaning of a word and the thing the word names are identical.

If that were true, then the question would not be what use we might invent for the combination of words 'Time isn't real', but whether the proposition 'Time is unreal' is true or false, i.e. our question would be about time, not about the concept 'time' (Z § 458). Well, but for a proposition to be true or false, it must not be nonsense. We learn to use the word 'time' by being taught to read a clock; we are not taught what is being measured by the clock (indeed, that question is nonsense suggested by misleading syntactic analogies and hypostatization).

Wittgenstein's logic of language is semantic and verifiable -- i.e. Wittgenstein's meaning of 'meaning' makes the distinction between sense and nonsense objective. It shows that philosophical investigations are investigations of concepts, not of phenomena, by showing that the meaning of a word is not the thing the word stands for (or names, regardless of whether that "thing" is tangible or intangible). To redefine or assign meanings to language (to examine and revise philosophical concepts or abstract ideas) is not the work of a "grammarian" (Kant), but of a philosopher.

'Time', 'Mind', 'Will'

To Ellis McTaggart's assertion that "Time is unreal", G.E. Moore could have replied that, although there are temporal phenomena (i.e. events), the word 'time' is not the name of anything; just as although there are mental phenomena, the word 'mind' is not the name of anything; and just as although there are phenomena we call 'voluntary acts', the word 'will' is not the name of anything. Or, in other words, the words 'time' and 'mind' and 'will' are not name-of-object words (nor name-of-phenomenon words). That is not their grammar, their part of speech, their use in our language. (Thus "the mind" is not the cause of mental phenomena, nor is "the will" the cause of voluntary acts. And the proposition 'Time is unreal', which is equivalent to the proposition 'Change is an illusion', is a false grammatical proposition. It is certainly not an experiential proposition.)

In logical grammar not all syntactic nouns are names of objects ("persons, places or things") (PI § 373). The words 'time', 'mind' and 'will' are examples of such nouns. But it would be misleading to say that therefore time, mind and will are not real. There are other roles for nouns to play in the language.

"Time is the measurement of change"

The word 'time' has no meaning apart from a method by which time is measured, i.e. if you want to know how 'time' is used in the language, look for examples of what are called 'measuring time', because that is how the word 'time' is learned and how its meaning is explained (taught). Note that no object named 'time' is involved in the explanation (teaching); the child is not shown an object named 'time'.

"Indefinable signs"

An idea often found in the philosophy of geometry is that there are indefinable signs, as e.g. 'point', 'line', and 'plane'. Pascal wrote that geometry does not stop to define "things clear and understood by all men"; e.g. the word 'space' "to those who understand the language", so naturally points out the thing it signifies "that whatever clarification we might want to give would contribute more obscurity than instruction" (On the Geometrical Mind, tr. Scofield). And from the logician W.E. Johnson: "the indefinable means that which is understood"; and philosophy "may ultimately adopt a term as indefinable only where, because it is understood, it does not require a further process of definition" (Logic I, vii, 3).

The criterion for our understanding a sign's meaning is that we use the sign correctly. But in order to know whether a sign is used correctly, we need to describe the rules by which the sign is used correctly, i.e. meaningfully. This description of the sign's grammar will serve as the objective criterion knowledge requires. (BB p. 65)

Therefore we have to reject the notion of "indefinable signs", if only on the Socratic ground that we so often think we know what we don't know. Socratic wisdom begins with the words 'I don't know', and the assumption in philosophy of one's own ignorance is a sound assumption.

They only err who think they know what they do not know. (Augustine; cf. Plato, Apology 21d, Xenophon, Memorabilia iv, 6, 1)
The difficulty in philosophy is to say no more than we know. (BB p. 45)

And it is only by trying to define signs that we discover whether we know what we are talking about -- i.e. whether we are talking nonsense or not. Johnson gave as examples of indefinable signs the words 'true' and 'false' (I, i, 3), and his remark to Drury shows that the word 'meaning' would be another. Wittgenstein, however, pointed out the confusion about language meaning in philosophy which results from the assumption that language is the mere transparent clothing of thought about things and therefore that its meaning is those things, or Plato's presumption that the meaning of a word is the essence of the thing the word names.

Further explanation of meaning (Kinds of definitions)

If we say that an undefined sign is meaningless -- i.e. if 'undefined' DEF.= 'meaningless -- then we have to show anyone who claims that a given sign "cannot be defined" -- how it can be defined.

Yet, that may seem impossible. "For it is evident," Pascal wrote, "that the first terms we wished to define would presuppose others for their explication ... and thus we should never arrive at the first". There are two parts to the response to this.

First, this much is true -- that no explanation of meaning "stands in need of another -- unless we require it to prevent a misunderstanding" (PI § 87). If the use of a sign is already clear to us -- i.e. if we are not lost but instead already "know how to go on", i.e. how to play the game (ibid. § 123), then what is the expression 'further explanation' going to mean?

Understanding is a Public Event

Remember what we actually do in life (Picture this in the context of the apples and pears language game above). We give a child instructions and if the child does not correctly follow the instructions, then we give the child "further explanations". What the child does -- how it acts on our instructions -- is the criterion for applying words like 'clear' and 'understood'.

Language is not clear in itself. Although all language with sense or nonsense is public, language is only clear to some individual or other -- but that something is clear to someone is demonstrated by what he does: 'to understand' means to act in defined ways in defined circumstances (ibid. § 154); it is a public event, a skill put to the test, not the name of an occult process in the mind. (These are entirely grammatical remarks, as the word 'defined' indicates; they belong to a verbal definition of the word 'understanding'; they are not a metaphysical theory about what understanding "really is".)

And second --

How are we taught to use a word? How do we teach others to use the word?

While it is true that in many cases a sign-for-sign substitution rule will not be of any help to us, that does not stop there being other techniques for giving explanations of meaning. Rules of grammar take various forms -- there are various kinds of definitions. Wittgenstein's logic gives alternative methods of definition.

Thus, for example, we can describe a game children play with blocks of wood and correspondingly shaped holes. "Here," we say as we hand the child a block, "find a square space for this one"; and so on. This is an example of what Wittgenstein called a 'primitive language game'.

This game reminds us of one use we make of the word 'space'. It is an example of a non-substitution technique for giving an explanation of meaning. It is a partial answer to the questions: how do we learn to use a particular sign, and how would we teach someone else to use it? (Note that as with 'time', no object named 'space' plays a role in the teaching.)

One thing we always do when discussing a word is to ask how we were taught it. Doing this on the one hand destroys a variety of misconceptions, on the other hand gives you a primitive language in which the word is used. Although this language is not what you talk when you are twenty, you get a rough approximation to what kind of language game is going to be played. (LC p. 1-2)

That is a very different technique from screwing up one's eyes and somehow (I don't know how) trying to guess (or divine as the theory of abstraction and "analytic truth" entail) the "meaning" of a word when that word has been divorced from our life, as if its meaning were a thing with an independent existence. Losing oneself in a cloud of vagueness, one asks oneself: "What is space really; what is space in itself?" rather than "How are we taught to use the word 'space'? How do we learn to use that word?" and "How do we teach a child to use that word?"

When philosophers use a word ... and try to grasp the essence of the thing, one must always ask oneself: is the word ever used this way in the language game that is its original home? (PI § 116)

By exclusively demanding general sign-for-sign definitions of words, our English teachers mislead their students. A child's natural response when asked for the meaning of a word is to say, "Oh, that's like when ..." -- i.e. the child points to examples from our life. The child, unlike the English teacher, has seen the logic of our language.


Site copyright © September 1998. Send Internet mail to Robert Wesley Angelo. Corrections and criticism are always welcome. Last revised: 13 October 2024 : 2024-10-13

Back to top of page

The URL of this Web page:
https://www.roangelo.net/logwitt/logwitt3.html

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