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

Statements of Fact in Wittgenstein's Logic of Language

Although all sentences have the form 'statement of fact', there are basically two kinds of sentences, namely statements of fact and grammatical rules (or definitions), although there are also tautologies (some of which are far from idle).

Examples of statements of fact

Very basically indeed. Because the form alone of the following sentences does not explain their use in the language: 'The book is on the table' | 'The earth goes round the sun' | "But, grandmother, what big eyes you have!" 'It appears that Little Red Riding Hood's grandmother has very big eyes' | 'It is good to help one's friends and harm one's enemies' | 'A kitten is a baby cat' | 'Cats don't smile' | 'Cats can't smile' | 'You can't see another person's soul' | 'Nine minus five is four' | 'God is merciful' | 'Things don't just disappear [vanish into thin air]'. The meaning of each of these sentences is learned and taught differently, and these are only a few examples of the different kinds of things we call 'statements of fact', for there are many others.

Preface: what are we calling 'logic'?

Wittgenstein in the Tractatus Logico-Philosophicus:

The exploration of logic means the exploration of everything that is subject to law [in the sense of "laws of nature" or "regularity (that which cannot be otherwise)"]. (TLP 6.3, tr. Pears, McGuinness)

Wittgenstein after the Tractatus Logico-Philosophicus:

Logic is the exploration of everything subject to conventions, in the sense of more or less arbitrary rules. Everything descriptive of the use of words in the language is part of logic (OC § 56).

What counts as an adequate test of a statement belongs to logic ... to the description of the language and the circumstances in which the language is used. (ibid. § 82)

Are the later remarks historically justified? It seems not because:

Aristotle, the Greek Stoics, Mathematical Logic: Logic as form

The "term logic" of Aristotle, if I understand it, has to do with words considered as class and sub-class words. It belongs to an explanation of meaning (i.e. to grammar, in Wittgenstein's revision of the concept 'grammar') that e.g. "All cats are animals" and "Some animals are cats" and "Some animals are not cats" and "Not all animals are cats". In this context these are not statements of fact (empirical propositions), unless they are intended to state facts about the English language.

If "All men are mortal", then "Some men are mortal". That is the grammatical connection between 'ALL' and 'SOME'. In contrast, if "All men are mortal", and "Socrates is a man", then "Socrates is mortal". Those are three statements of fact, if indeed it is a fact that all men are mortal (and that Socrates is a man). So we cannot say that term logic is all grammatical propositions, as e.g. 'All cats are felines' is -- but only that the relationships of the logical operators 'ALL', 'SOME', 'NONE' and 'NOT-ALL' belong to grammar. And that is the essential element of term logic. The propositions conjoined by the logical operators may be grammatical propositions or factual propositions indifferently.

Aristotle regarded his Term Logic as a "computation" (A.E. Taylor, Aristotle [Rev. ed. 1919], (1955) ii, p. 27). Form logic is a calculus, like arithmetic or a game played according to strict (unambiguous) rules, like chess.

Mathematical logic

The Greek Stoics (and the Megarian school) invented the earliest propositional logic (Diog. L. vii, 65-83). The modern "symbolic logic" of the mathematical logicians is also propositional. Propositional logic has to do with the grammar -- as more or less arbitrarily assigned by philosophers -- of the terms 'AND', 'OR', and 'IF...THEN...' in cases where philosophers use those signs to "connect propositions". The "propositions" may state facts -- or they may even be nonsense. The propositional calculus of mathematical logic does not need or want to know, which is why it uses the letters 'p', 'q', and 'r' to "stand for" propositions.

That may not appear to be a correct account, but it is. The rule is, If p is a true or false [although 'true' and 'false' are not defined and indeed are often replaced by '1' and '0'] statement of fact [although 'statement of fact' is not defined], then.... But the rules of form logic are not always faithful to natural language; in fact the rules show themselves to be quite arbitrary. For example, there is form logic's law of contradiction: "P and not-P is always false", but as Wittgenstein pointed out the proposition 'This is beautiful, and this is not beautiful' (pointing at different objects) is a contradiction but it has a use in our language; it is not nonsense. In natural language a contradiction may be true or false, or nonsense, and not only false as in form logic.

In the case of "If P, then Q", Wittgenstein again pointed out that in natural language: (not-P) does not verify (If P, then Q); he gave as an example "If I strike a match (P), there is a flame (Q)", and said: we do not regard it as confirmation of "If P, then Q" if I don't strike a match. But in form logic's definition of 'conditional proposition', the only case in which the conditional is "false" rather than "true" is when the conditional's antecedent (P) is true and its consequent (Q) false. (The logic of counter-factual conditionals.)

But on the other hand - Logic as meaning: Socrates, the Greek Stoics, and later Wittgenstein

The word 'logic' did not originally mean the study of form -- i.e. of reasoning that derives one proposition from other propositions, deducing conclusions from premisses -- rather than a study of meaning -- (if indeed it is possible for logic to be a study of form alone and yet remain part of philosophy). The meaning of language is what Socrates means by the word 'logic' according to Plato and Aristotle and the Memorabilia of Xenophon.

When Plato distinguishes between nouns and verbs (Sophist 261e-262a) and says that a proposition always contains both, that is not a remark only about syntax but also about meaning. Aristotle talks about types of definition, namely "nominal and real definition" [Note 1a]. The Stoics make a distinction between (1) a sign and (2) a sign's meaning (or lekton), namely "what the Greek understands but the barbarian does not although they both hear the sound when Greek is spoken" (Sextus Empiricus, Adv. Math. viii, 11-12; cf. Plato, Theaetetus 163b). And Ockham distinguishes between bare spoken and written signs and the terminus conceptus (or concept) in logic.

The study of form is also not what Wittgenstein means by the word 'logic', when in his later investigations he identifies logic with (in his jargon) grammar = rules of sense [language meaning] and nonsense rather than with syntax (form).

Even logic as it was taught when I was at school included the informal -- i.e. not determined by form -- fallacies of reasoning (such as post hoc: "B came after A; therefore A was the cause of B"). And if logic is the study of form, and those fallacies are not formal, then why call them logic?

Is logic part of philosophy (along with ethics and metaphysics) or is logic only philosophy's tool and not part of philosophy (which appears to have been Aristotle's view)? Is that a philosophical question? (What would it mean to say that it was or it wasn't?) What sets the limits of the concept 'logic'? The historical origin of the English word 'logic' is in the Greek word logos, a word the Greek philosophers used in as many ways as English speakers use the word 'meaning' -- i.e. the words 'meaning' and logos have no defined essence -- nor has the word 'logic' if e.g. "the Fallacy of the Artisans" and "the Fallacy of SOME therefore ALL", which are not fallacies of form, belong to logic.

Is the general definition 'the study of rules' broad enough for the word 'logic'? [Note 1] It is certainly vague enough.

The topic of this page is the grammar or "semantic logic" (But that form of expression is a pleonasm in Wittgenstein's later conception of logic) of sentences that have the form 'statement of fact', many or most of which are not used to state facts.

[Related: a simple criterion by which to make the distinction between the concepts 'fact' and 'opinion' clear.]

Outline of this page ...

What are we calling 'verification'?

The following remarks are over forty years old. They were fresh at that time and not at all subtle, and I have only slightly revised them. They are not the way I would write now, but they do nonetheless make important grammatical points, e.g. that To verify is to act.

But the word 'grammar' is used to the point of abuse. Sometimes it means a description of a use of language, at others a description of the rules of a use of language, while at others times the rules of a use of language themselves. By 'grammar' in "The Grammar of the Greek gods I seem to have meant "a description of how language is used to talk about the gods" (but if there is no Zeus, then what is being talked about? How can we talk about what is not?), "a description of the use of language in Greek mythology".

The questions asked of a statement of fact are (1) Is the statement true? and (2) "How do you know?" or the question of verification. These are sometimes, but not always the same question, because it is not logically possible for some true statements of fact to be verified. Both questions presume that the meaning of the statement is clear, although the answers to both or either might belong to an explanation of the statement's meaning. [Note 2]

Grammatical in contrast to Factual

Whether or not it is at present raining does not belong to the grammar of the statement 'It is raining', but to the facts of the weather. But that we can using this method -- not as a matter of fact, but as a way of life (knowledge -- i.e. the method by which something can be known -- belongs to the community) -- determine whether it is raining by looking out the window does belong to grammar. (Cf. the meaning is given -- e.g. of the word 'time or of the word 'force' (Heinrich Hertz) -- when the method of measurement of time or force is given, i.e. defined by grammar. Otherwise those words are without meaning, and likewise for 'verification'.)

It is only when we are under cover that there can be a verification of 'It is raining'; e.g. when we hear tapping on the roof and make the hypothesis 'That will be rain'. Then the method of verification is opening the door or looking out the window. But for someone standing under the open sky, or for someone who is looking out the window already, there can be no verification of 'It is raining'. The 'can' here means: grammatical possibility -- i.e. what language is or is not an undefined combination of words (i.e. nonsense).

By 'verification' of the proposition (statement of fact) 'It is raining' we do not simply mean: seeing it is raining. (If we did, we would be using the word 'verification' to make a "distinction without a difference".) By 'verification' we mean: checking whether or not it is raining, e.g. getting up to have a look at something that we do not already see (or having a closer look of what we do see, or seem to see).

'I know it is raining'

If I see it is raining, then the statement 'I know it is raining' is nonsense; unless its sense is simply the same as 'I see it is raining'. The question-and-answer grammar of 'How do you know?' belongs to the grammar of 'know'. [Note 3] To the question 'How do you know it is raining?' the answer 'I know it is raining because I am watching it rain' makes sense as a grammatical remark because the answer gives a defined ground for saying 'I know it is raining'. ("The sentence has a sense." -- "What sense?" We answer by pointing to a use in the language that we have for the sentence. The sentence 'I know it is raining' is nothing more than spoken sounds or marks on paper, and therefore the question is: what gives the sentence meaning?) In contrast the answer 'I know it is raining because today is the second Tuesday of January' is "ungrammatical" -- i.e. it is not a rule of grammar -- but it is also not false; it is nonsense (because "Today is the second Tuesday of January" is not defined grounds for knowing that it is raining).

If I answer 'I know it is raining because I am looking at the sky', then I do not give a verification as my grounds for the assertion 'It is raining'. But if I answer '... because I just went to look out the window', I do give a verification as my grounds. (Here I am describing the language facts in plain view, because they are so plainly in view that we don't see them.)

In the statement 'We can verify 'It is raining' by looking out the window -- because if we see the sky, we see whether or not it is raining', the word 'because' indicates: what follows is a rule of grammar.

'Rain is water' (Hypothesis or grammar)

We learn to use the word 'rain' the way we learn to use the word 'thunder', i.e. in the presence of rain, and that is as far as our instruction in this kind of language goes. But we also learn to say 'rainwater'. And that word may prepare the ground for confusion.

'Rain is always water' is an induced hypothesis; it belongs to meteorology. But when in the course of day to day life we use the word 'rainwater', we are not making reference to that hypothesis. Rather, our naive, normal way of using language does not reveal an hypothesis about rain -- but only a concept 'rain', or a concept 'water', i.e. a rule for using a word (Z § 223).

"But don't we only have this rule because rain is always water? If various liquids besides water rained from the sky, then we would not have the concept 'rain' that we do have." Is that statement a hypothesis about the facts that can explain concept-formation (PI II, xii, p. 230)? Descriptions of counter-factual language games -- depicting anomalies -- may the concepts that we do have clearer to us (CV p. 72). But by 'logic' Wittgenstein meant the study of grammar, i.e. of rules, not of the possible causes of grammar. (OC § 617-8; RPP i § 46-48)

'Rain is always water' may be used to state an hypothesis or to make a rule of grammar (i.e. it can be made a defining quality rain). Compare 'Rain is always accompanied by clouds'. We cannot tell simply by noting the form of what is said which way a sentence is being used (whether as a proposition of fact or a proposition of grammar).

It is simply not our normal way of thinking to distinguish definitions from statements of fact. And the result of this is that we often knot up grammar and fact, verbal definitions and hypotheses, in a way difficult to untangle (Z § 452). Do we know what we are saying all the same? I would not say that.

There are blind alleys in our forms of speech, and when we discover that we have entered on one, we just have to give up on it, go back to the beginning and try a different way.

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)

Examples uncover the truth

Pointing to a method of verifying the statement 'It is raining' (e.g. 'Go look at the sky') is an explanation of what we mean by the words 'It is raining' --. And at this point we write "and so on" In philosophy we always work with few -- far too few -- examples (cf. OC § 555: "The language game is different every time").

Why don't we want to take the trouble of thinking up examples. Partly because we presume that what we are searching for is something general, namely an essential meaning, and so we try to guess at it. This is not a stupid presumption (PI § 340), because we deeply feel that we know the essential meanings of words -- even if we cannot say (ibid. 210) what those meanings are. Nonetheless it is a false presumption.

One cannot guess a word's meaning. One has to look at its use and learn from that. Let its use teach you its meaning. (cf. PI § 340, and II, xi, p. 212)

Although the presumption that one can guess may not be "conceited ignorance", it certainly isn't Socratic wisdom either: "And that which we know we must surely be able to tell?" (Plato, Laches 190c). We have, in my judgment, to follow Wittgenstein in language meaning and Socrates in standard for knowing in philosophy. Otherwise philosophy is lost as babble of words in an endless night of seeming.

Given the way we think, what we presume about language meaning, we are naturally blind to the place of examples in logic of language, not seeing that they are a foundation stone of logic. And we are wrong -- Why? Because most words of natural language do not have essential definitions, and therefore examples are the only way to explain the meaning of these words. We are always more or less dependent on models and the grammatical rule 'and things like this' (PI §§ 292, 139). In other words, this is not an anomaly, not an exception; it is the nature of natural language.

Examples are the true masters to follow in philosophy.

Measurement and verification

The general definition of the word 'verification' is 'test of truth or falsity'. To verify DEF.= to test for truth or falsity. But compare: a length of time is not measured the way a length of cloth is (with a yardstick), nor the length of the growing season, the temperature of water, the air speed, the population of Iceland, the cost of wheat production, the result of an referendum, the loudness of an airplane engine, the weight of a sack of flour, a student's knowledge of algebra, or a man's character, and so on is ... (Is there no general definition of 'to measure'?)

And the concept 'verification' is akin to the concept 'measurement': a general definition, despite being too general to tell us what to do in the particular case, deceives us into presuming we know the essential meaning of a word. But if we did know the essence of verification, i.e. if 'verification' indeed had an essential meaning, we would never be in doubt about how to verify a particular proposition. But now, how do we verify -- i.e. put to the test the truth or falsity -- Russell's proposition 'It snowed on Manhattan Island on the 1st January in the year 1 A.D.'?

We are lazy bones when it comes to using our imagination to do this kind of work -- because we don't expect that examples will yield philosophical insights. But when we do trouble ourselves to think, we discover that examples reveal things that we have overlooked, showing us that we did not know what we thought we knew. Wittgenstein told Drury: "The remark You'd be surprised wouldn't be a bad motto [for my book]" (Recollections p. 157).

Verification of picture-signs and Meaning

By the word 'grammar' Wittgenstein meant any 'description of the use of language', but particularly its comparison to the rules of a game (as in language game). Where it is grammatically possible to compare a picture with what it is a picture of, verification is possible, and how the comparison is made belongs to the grammar of the statement of fact -- which I will call here "the statement of fact as a picture" or "picture-sign" (using the word 'picture' strictly as in the Paris court case model origin of Wittgenstein's TLP, the word 'picture' not meant figuratively). And the comparison is what is meant by the word 'verification' in this case, as e.g. 'The book is on the table' is a picture-sign that can be compared with what it is a picture of; if the book is indeed on the table, then the picture-sign is true.

But not every picture-sign can be verified (Fairy tale pictures)

The following remarks are very old, and very abbreviated; they seemed an important summary to me at the time. Note that 'grammatical possibility' is synonymous with 'logical possibility' here.

Some picture-signs cannot be true or false -- because verifying them is grammatically ruled out. For example, 'There are elves in the forest'. It belongs to the grammar of the word 'elf' that nothing can count for, nothing against, the existence of elves (in the forest or elsewhere).

Fairy tales and e.g. the Greek myths present us with pictures that are not statements of fact. Although they are not pictures of the world, they are pictures in the way that an artist's illustrations of fairy tales are pictures.

But whereas there can be picture-signs to illustrate the gods of the myths, there cannot be picture-signs of God (cf. LC ii, p. 63). There are three points to make here:

The question isn't whether in everyday practice we verify the proposition, but whether the proposition can -- logically and really -- be verified. Is it agreed that I am in a position to know that it is raining (All verification is public)? [Verification is knowledge. Knowledge belongs to the community.]

Ways of Looking at Things (One kind of Tautology)

There are statements of fact which -- despite looking as if they might be tested for truth or falsity -- are defined in a way that makes them unfalsifiable. Such statements are called 'tautologies': they are "true" under any and all circumstances, which is to say that 'true' and 'false' do not apply to them: what cannot be false cannot be true either. (These are grammatical remarks.)

Interpretations that cannot be false

An interpretation is a way of looking at some thing, not the only way, and certainly not a necessary way.

Unfalsifiable statements sometimes appear at the border of science. For example, the grammar Freud assigns to his hypothesis of psychoanalysis that 'All dreams are wish-fulfillment' makes it an hypothesis impossible to falsify: its grammar is so unlimited that it can be extended to explain any and every possible dream. In other words, the hypothesis is anomaly proof, a perfect bomb shelter. [Note 6]

That is one kind of tautology, the tautology as a way of looking at things (PI § 401). It says: Look and see dreams this way! (CV p. 61)

The tautology as ethical rhetoric

Plato's method of tautologies in ethics is a different kind of tautology, an investigation of rules of grammar, rhetorical as 'If the good man does evil, then what does the bad man do! If the good man harms his enemies (i.e. makes them worse rather than better), then what does the evil man do!' and 'If a god would do that, then what would a demon do!' In our language we make a distinction between good and evil, and what Plato shows is how that distinction is made, e.g. 'The good man harms no one' is a rule of grammar, but it is also a rule of ethics (an answer to the question of "no small matter, but how to live").

Foundational Propositions

There are statements of fact that can be called 'foundational propositions'. These are statements that serve as grounds for other statements, but are themselves held without grounds. They are sometimes called 'principles' (though that word is also used countless other ways) or 'first principles'. [Note 7]

That I have two hands is, under normal circumstances, as certain as anything I could produce as evidence for it.

That is why I am not in a position to take the sight of my hand as evidence for it. (OC § 250)

For one statement to serve as grounds for another, the former must be more certain than the latter. (But, note, where grounds can be given, they also must be given; only the absence of logically possible grounds makes a statement foundational.)

Foundational beliefs, as well as foundational statements, are neither true nor false, and they are neither knowable nor unknowable -- because grounds determine what is true, what is knowable, and foundational beliefs are without grounds. They can only seem to be true --

... whatever is going to seem right to me is right. And that only means that here we can't talk about 'right'. (PI § 258)

Where there are no grounds, neither is there right or mistaken. There is only seeming to be one or the other; there is neither truth nor knowledge. (In this respect, foundational beliefs grammatically resemble rules [as do the pictures of myths].)

A statement that under normal circumstances serves as a foundation may in other circumstances be a testable statement of fact, of course. 'I see my hands' may be grounds for my belief that 'I have two hands' if I awaken in a hospital after a traffic accident. (It is the particular use of a sign only that gives the sign its meaning. The form of expression {statement of fact} or {proposition} has no essential meaning (definition, use in the language): only its form is constant from use to use.)

"My belief cannot be mistaken"

My foundational beliefs are "the foundation of all [my] judging" (OC § 614). And therefore I cannot be mistaken about them, even if they are false. For example, were I to assert 'I am a giraffe' or 'My name is Anselmus' -- these statements would not be mistakes. They might show confusion to the point of madness, but a mistake is grammatically excluded (ibid. § 674).

As my life shows (ibid. § 7), I live in accord with my foundational beliefs. And, as there are no grounds for foundational beliefs, that is the only sense the words 'I believe' might have here, which contrasts with 'I believe for such-and-such reasons'. (This view always stood for Wittgenstein: "What do we do with a statement? ... that is how one must decide" (ibid. § 230). The use, the act or deed, determines the meaning.)

"-- even if it is false"

Nonetheless, foundational beliefs are statements of fact. I am a man (not a giraffe), I do have two hands (not one or none), etc. -- But they are not statements that I myself can test for truth or falsity.

And my condition is the same as everyone else's in this respect, of course.

It is part of the language game with people's names that everyone knows his name with the greatest certainty. (OC § 579, emphasis added)

[Related question: What is the relationship between the concepts 'belief' and 'disbelief' ('doubt'); if there no grammatical need for grounds in the case of belief -- is there such a thing as a foundational disbelief (doubt), for example? Is there a point beyond which belief and disbelief are grammatically impossible?]

The groundlessness of belief

All foundational beliefs are "absolute": the foundations of our lives do not themselves have foundations (the logical impossibility of doubt does not imply the presence of grounds; there is no bedrock beneath the bedrock). We walk upon the air with respect to our foundational beliefs. [Note 8]

At the foundation of well-grounded belief lies belief that is not grounded. (OC § 253; cf. PI § 482)

Nothing one does can be defended absolutely and finally. But only by reference to something else that is not questioned. (CV p. 16)

The difficulty is to realize the groundlessness of our believing. (OC § 166; cf. § 235: "And that something stands fast for me is not grounded in my stupidity or credulity.")

But foundational beliefs are only subjectively -- i.e. not-objectively -- certain. Because objective certainty, like objective doubt, must have grounds.

And therefore the combination of words 'absolute knowledge' is undefined, because to 'know' means to 'have sufficient grounds', and what is absolute is independent of any grounds: absolute DEF.= foundational. (These are grammatical remarks.) (There is a further grammatical reason why the expression 'absolute knowledge' is nonsense.)

It is true that we can compare a picture that is firmly rooted in us to a superstition, but it is equally true that we always eventually have to reach some firm ground, either a picture or something else; so that a picture which is at the root of all our thinking is to be respected and not treated as a superstition. (CV p. 83, remark from 1949; cf. p. 86 [MS 174 1v: 1950 §§ 1b-2])

And that something stands fast for me is not grounded in my stupidity or credulity. (OC § 235)

*

There are statements of fact in the Philosophy of Psychology about an individual, which though unverifiable by that individual, are nonetheless true or false. For example, it is nonsense to say that I verify whether or not I have a headache. So I cannot know whether or not I do either. Nor can I believe I do. (PI § 246)

Real and Logical Possibility

The word 'fact' does not have an essential meaning. And neither have the words 'statement of fact' and 'hypothesis'. That is why when we try to remind ourselves of their meanings, we want to say that they are vague notions. That is a mistake, but one that we make again and again -- so long as we do not break with the notion that a word names an essence, a constant in all contexts, which is the meaning of the word.

Wittgenstein asked: What is the essence of games, the one defining thing that all games have in common, which is their meaning? And his investigation showed that there is no such thing. Most words do not have "meanings" if 'meaning' = 'essence' or 'defining common nature' -- and among the words without "meaning" are the words 'fact' and 'statement of fact', as the examples of statements of fact above show. (The only thing all statements of fact have in common is a form and a name, namely 'statement of fact' or 'proposition'.)

Logically possible = describable = described

A distinction can be made between logical and real possibility. 'Logically possible' means: what can be described, 'can' because it is described (What is logically possible is also grammatically possible: 'possible' here means 'defined' as in 'this language has a use in the language; it is not merely an undefined combination of words') -- e.g. 'The cat plays a fiddle' (as in the nursery rhyme "The cow jumped over the moon") states a logical possibility. But 'The cat plays a fiddle' does not state a real possibility; because it has never been known to happen (Cats do not play violins). A 'real possibility' is what can be observed, 'can' because it sometimes is observed to happen. [Note 9]

What is really possible is also logically possible, but not every logical possibility is a real possibility. And the type of possibility that a statement expresses can be changed by a change in circumstances (i.e. changed facts), e.g. 'Man walks on the moon'.

Both logical and real possibilities have the form statement-of-fact, but for the sake of clarity, I will make this rule -- which is more limiting than our language's normal grammar -- namely: 'statement of fact' DEF.= 'a statement that is verified or falsified by facts'.

Thus a statement that expresses only a logical, and not a real, possibility is either false, or neither true nor false, but it is never true -- which is why it is not a "real" possibility. And therefore, a statement of what is only logically possible cannot be an hypothesis. Because by 'hypothesis' I mean a statement that can be determined to be true or false -- 'can' because a method of verification has been defined. So, for example, 'There is an afterlife (life after death)' and 'There is no afterlife' are not hypotheses; they state only logical, not real, possibilities. They are pictures (like a cat playing a fiddle), but not pictures of the facts.

[Note that sometimes 'logically possible' = 'not self-contradictory'.]

Statements of fact that are hypotheses

The meaning of an hypothesis includes the criterion for verifying it. Without this criterion, there is no hypothesis (because there is no serviceable meaning, only nebulosity). That belongs to a possible definition for the word 'hypothesis' (the rest of the general definition being: an 'hypothesis' is a 'statement of fact that has yet to be tested for its truth or falsity'). That is in fact the way I use that word, but I am not the only one who uses that word, of course.

'I have cracked ribs.' Is that an hypothesis -- and if it is an hypothesis, what does it hypothesize (i.e. what is true if it is true)? If I am unwilling to set a standard of verification, then the statement is not an hypothesis. But the criterion I might set is most varied. I might answer the question "How do you know?" by saying "I asked my older sister", for that is a possible criterion (although if my sister is only seven years old, it might not be the best). Or I might set other criteria, for example, "That is what the doctor said", or "I looked in a textbook", or "I was shown an x-ray of my chest".

If someone said that the method for verifying or falsifying the proposition 'There is an afterlife' or 'There will be a last judgment' would be to die, that proposition would not be an hypothesis -- because that "method" is no criterion (verification is an event in this world). Not every imagining (logical possibility) is allowed by the rules of this game, namely the game of hypothesis formation. A criterion must state a real possibility.

"I take it, in the first place, that neither of us is prepared to admit diabolical intrusions into the affairs of men. Let us begin by ruling that entirely out of our minds." (The Adventure of the Devil's Foot)

Holmes states "a rule of the game" of hypothesis formation, by ruling out supernatural explanations for natural events. (That rule was Thales' project, later called 'Philosophy'.)

If someone said that the method for verifying the proposition 'There is life on another planet' would be a spaceship that could travel at the speed of thought, I would say that proposition would not be an hypothesis. But then I ask: by the rules of what game? by the rules of natural science, or by the rules of logic (i.e. sense and nonsense)?

The unverifiable pictures of logical possibility, the dreams of imagination. ("A thought experiment is not an experiment." How the statement is being used -- what work it is being used to do -- shows whether the statement is an hypothesis or not.) About the last two examples, I would say that as hypotheses, these statements are meaningless. As pictures, on the other hand, of course they are not meaningless: e.g. we can make a drawing of a spaceship or of a ghostly man awakening after death. But I am not willing to call just any picture an 'hypothesis', because the distinction between logical and real possibility is important to philosophy (We don't want either fantasy or conceptual confusion for our philosophy).

Philosophy makes distinctions we don't normally make

The word 'hypothesis' as it is normally used marks an concept, which like most concepts, has vague = indefinite borders (There is no essential definition of 'hypothesis'). So that if we want to use that word as a tool for thinking philosophically, we may have to assign rules for its use -- that is, to choose a meaning for it, set limits. That is what Isaac Newton did when he invented the Rules for Reasoning in natural Philosophy: he made rules: This is what I am calling a 'Law of Physics' in contrast to 'an hypothesis, whether physical or metaphysical'. Newton made distinctions. Wittgenstein also did this when he chose the meaning of 'meaning' he used for his work in philosophy, and when he revised our concept 'grammar'. When I use the word 'grammar' in logic of language studies, I am using his jargon.

The danger from generalization in philosophy. And contrariwise.

That is the way I choose to limit the grammar of 'statement of fact' and 'hypothesis'. In other words, I am not stating the "true" meanings of those signs, nor am I stating how anyone must be using those signs if uttering them with sense (Z § 467).

But then, is what I am doing wise? For aren't I now predetermining how I am going to look at sentences? Aren't I laying the foundation for blindness to differences by closing concepts that in our normal way of speaking are open-bordered rather than closed?

Concepts may alleviate mischief or they may make it worse; encourage it [foster it] or stop it. (CV p. 55)

The "mischief" in a generalization is that it may put an end to our inquiry if we mistake a generalization for knowledge, and so stop looking for knowledge in the particular case, in further examples. On the other hand, calling every sentence a 'statement of fact' as we do in the language of everyday makes all sentences seem the same. But they are not. I want to alert myself to a certain difference among the cases I am familiar with; e.g. I do not want to mistake just any picture for an hypothesis, just any possibility for a real possibility. [Note 10]

Grammar, Myths and Fairy Tales

We are entering a new region of language use (LC i, 2-3, p. 1), sentences of the form of "statement of fact" in fairy tales and myths, sentences that appear to state facts but do not. For example: 'Elves are capricious.' What is the grammar -- i.e. the use in the language -- of these sentence-signs?

Statements about elves express logical, not real, possibility. Because the grammar of the word 'elf' sets no observational criterion for that word's use: Nothing we observe or fail to observe directs our use of the word 'elf'. Fairy tales do.

The word 'elf' is not the name of an object or phenomenon; it is not a word we use to talk about the world, and therefore our use of that word is not answerable to the state of the world. To what then is it answerable?

Question: does anything we say about elves belong to the grammar of the word 'elf' (PP iii, p. 312)? This question presupposes that we know what 'rule of grammar' is to mean in this case. In all other cases we have contrasted 'description of the use of a sign in the language' with 'the use of the sign in the language' -- but what use has the word 'elf' in the language?

We contrast sentences about the facts with the facts themselves. But the word 'elf' is not used in sentences that state facts about the world. On the one hand the word 'elf' is a non-name word, but on the other hand it is the name of a fairy tale object.

To understand the language of fairy tales we can make analogies from the language of facts to the language of fairy tales, but analogy only goes so far. (Verification and fairy tales.)

By 'grammar' we mean a description of the use of words, but what does the word 'grammar' contrast with in the case of fairy tales? (Are there statements of fact in fairy tales)?

If someone says 'Elves are capricious', what is he saying? We ask him: what do you mean -- how do you know (PI § 353)? And he answers that he has read this in E.T.A. Hoffmann's tales. And how did Hoffmann know? Did he find this in Paracelsus? Let us imagine that once upon a time someone first stated that elves are capricious. Was that person stating a rule of grammar (i.e. defining or explaining the meaning) of the word 'elf'?

What do we mean by 'rule of grammar'? A rule that gives a sign (a word or phrase or sentence) a use in the language. Does 'Elves are capricious' give a rule for using the word 'elf'? It tells us what not to say. But is 'Elves are capricious' an explanation of the meaning of the word 'elf'? If we say 'Elves are not capricious', are we giving a false account of the grammar of the word 'elf' -- or are we making a "false statement about elves"?

If we don't know what to say here, is it because we have entered a new region of language, and what work Wittgenstein's tool of logic, namely 'grammar', is to do here is not clear?

The Language Game with the Word 'elf'

A "language game" is a use of language governed by rules, as a game is governed by rules. This is Wittgenstein's simile, and jargon.

We are treating 'elf' as if it were the name of an object -- but an invisible one. We are trying to apply the grammar of the part of speech name-of-object to the word 'elf', even though names of objects are ostensively -- i.e. visually (in most cases, but the scent of oranges and sound of a violin are also defined ostensively) -- defined. And so we ask if we can make a false statement about elves, because we can make false statements using names of objects. (But the part of speech of 'elf' is fairy-tale-sign.)

This is a rule of grammar: 'In fairy tales, 'elves' are invisible, beautiful, young and of a capricious temperament.' But if we wrote a story in which we said 'Elves are not really capricious', would we be writing nonsense? The answer that suggests itself is that, because we would be changing the rules (a definition is a rule of grammar), we would be "playing a different game" (Z § 320).

Changing the rules of the game mid-play

If part way through a game of chess, we decided to allow "Bishop's Option" -- would we still be playing be chess? There is no rule -- where would we look for a rule? If in every story that had been told about elves, elves were by nature capricious, and now someone told a story in which elves were not capricious, would he still be telling a story about elves?

Compare statements about the Olympian gods that were made by the early Greeks. Can a rule for using a word also be a statement of belief? And why shouldn't a picture-sign serve both as a rule of grammar and as a description of what is believed? (Cf. Plato's tautologies: rules in ethics which are also rules of grammar.)

Why should it not do the same service ...? And it is the service which is the point. (PI II, iv, p. 178)

Why shouldn't a statement do dual-service? Indeed, isn't that just what one does here?

The Grammar of the Greek gods

Is every statement about the gods a rule of grammar? 'Zeus is the cloud-compeller' -- is that a rule of grammar? Homer calls Zeus the "Cloud-gatherer", and we cannot dispute observed fact about this with Homer, because there are no observed facts about the gods (Herodotus, History 7.129). But does the statement 'Zeus is the cloud-compeller' explain the meaning of the word 'Zeus'? It belongs to the grammar of the word 'Zeus' that 'Zeus' is the name of an Olympian god -- but the question is: where does the definition of the word 'Zeus' end -- and where do "statements about Zeus" begin ... if there are non-grammar (non-definition) statements about Zeus?

Is 'Zeus is the Earth-shaker' a false statement (Poseidon, not Zeus, is the Earth-shaker) or is it nonsense? It isn't nonsense, if by 'nonsense' we mean 'an undefined combination of words', because any god might depicted playing the part of Earth-shaker. But is the proposition 'Zeus is the Earth-shaker' false because it is a false account of the grammar of that word -- or because it is inconsistent with Homer (verification by literature)? Is there a difference here? -- Well, but is Homer a dictionary then? Is the statement 'Poseidon is the Earth-shaker; Zeus, the Cloud-gatherer; Apollo, the Truth-teller' a rule of grammar? But mightn't Homer's work be used that way, i.e. to give explanations of word meaning? Does the question-sign 'Who is Zeus?' contrast with 'What is the meaning of the word 'Zeus'?' or are their meanings, their uses in the language, identical?

Theory of Descriptions (Definition in Russell)

Wittgenstein refers to Russell's Theory of Descriptions (PI § 79). If we are unsure who (or what) someone is talking about, we may ask questions like: By 'Zeus' do you mean the god who chained Prometheus to the rocks? and Do you mean the god who turned himself into a swan? About Zeus there are many poets telling many stories -- as well as philosophers denouncing the poets for misrepresenting the gods -- and so it seems that various things might or might not be meant by the word 'Zeus' depending on which poet or philosopher is used as the authority ("dictionary"). Is there an essence of Zeus?

... and when you do see the facts, there's a lot you won't say. (ibid.)

The expression 'rule of grammar' seems suddenly too vague to work with. Where there is no rule for how we are to go on, we have to make a rule. We have to decide on a criterion for what does and does not belong to grammar in fairy tale language -- that is, if we want to make distinctions here with the word 'grammar' (words are tools).

Statements of fact that do not state facts (Form versus use)

In fairy tales we find statements with the form 'statement of fact' that do not even pretend to state facts. It seems that a fairy tale is a "world unto itself", a world with its own (un)natural laws. And that the writer of a fairy tale may follow rules that he has taken from other writers, or he may make up the rules to please himself. Our question is: which of these rules are grammatical rules?

We dream that we fly through the sky. Now, is 'We fly through the sky' a rule of grammar, or is 'We can fly through the sky' the rule of grammar? And then what is 'We fly though the sky' -- Does it too belong to grammar -- or does it belong to the facts ... of our dream? As if to say, "In my dream I flew through the sky, not underwater" -- but, of course, dream facts would not be verifiable. "Dream facts" -- as if the word 'fact' made the dream any more real than a fairy tale.

Is 'We fly through the air' a statement of fact or a rule of grammar? Or is that question nonsense -- i.e. an undefined combination of words? We haven't defined the word 'grammar' with respect to language as it appears in fairy tales and dreams (which are themselves fairy tales).

The "facts to see" (PI § 79) here is how strange the use of language in fairy tales is: statements of fact that do not state facts, untethered donkeys. (Are fairy tales counter-factual? Only if that means unfactual.)

*

A world of logical possibilities only

Fairy tale language appears to derive from the language of everyday life (OC §§ 617-618) -- but that the real possibilities of everyday life are of no concern to the fairy tale writer. In the fairy tale, what(ever) can be described can happen. I.e. in fairy tales there is no difference between real and logical possibility, or, in other words, in the fairy tale real possibility does not exist.

Absolute reality belongs to dreams and not to life. (Greene, "Under the Garden" ii, 6)

That is the reality of the fairy tale: a world existing in its own shell, independent of verification or falsification by the world of our experience, and therefore beyond skepticism. (In this it is unlike tales of history.)

In the fairy tale, "what can be described can happen too" [TLP 6.362]. "But what can be described belongs to grammar" (logical possibility). But what does 'grammar' mean here? Is there any grammatical difference between what can happen in a fairy tale -- (anything can happen in a fairy tale) -- and what does happen in the tale?

Interconnecting grammars creatively

In a scene in Hoffmann's The Golden Pot the old woman Lise recounts to Veronica something that happened to Veronica when the old woman was not there. "But how did you know?" a terrified Veronica asks. Lise answers: "But I was there the whole time; I was the coffeepot; didn't you recognize me?" The objects of everyday life are talked about -- but the writer does not bother about real possibility. Does it belong to the grammar of the words 'old woman' and 'coffeepot' that old women can metamorphose into coffeepots? Or does that possibility belong to the grammar of words belonging to the part of speech "fairy tale word" (The meaning of a word is context determined) -- if there such a part of speech? Or do fairy tale writers simply use the part of speech 'name of object word' to connect the grammars of familiar words, e.g. 'coffeepot' and 'old woman', in absurd (i.e. un-real) ways?

A fairy tale cat can smile, something it is logically impossible, i.e. undefined language, for a real cat to do (PI § 583). But, then, the cat can also talk and do other human things (cf. ibid. § 282), so that the distinction between a house cat's face and a human face is lost, and in that context it is logically possible, i.e. defined language, for a cat to smile. (Cf. "and the dish ran away with the spoon".)

Robert Schumann said that his titles were "sly hints to the meaning" of his music. So when he called his Opus 132 "Fairy Tales" or his Opus 113 "Fairy-tale Pictures", was he making a "grammatical remark"? Was Schumann giving "an explanation of meaning", stating a "rule of grammar"? And if someone says he finds Schumann's titles "just right" (although he cannot say why), does that mean that he has "understood" the explanation? Again, there is no rule; we have given definitions of 'grammar', but nowhere stated the essential meaning of that word; it does not have one.

Nonsense and Impossibility

"Nonsense" in Wittgenstein's logic of language is always logical -- never real -- impossibility, because by 'nonsense' Wittgenstein means 'undefined words or combinations of words' -- not 'foolishness', not 'absurdity'.

When a word is called meaningless, it is not as it were its meaning that is meaningless. (PI § 500)

The combination of words 'The cow jumped over the moon' is not nonsense in Wittgenstein's logic of language, because a drawing in a children's book of nursery rhymes titled 'The cow jumped over the moon' can serve as a definition of that sentence: the picture gives the sentence's grammar (if one has already learned to use the words 'cow', 'jump', 'moon' and 'jump over'; cf. a drawing of a cow jumping over a fence). A drawing can serve as a definition: "it is the service which is the point" (PI II, iv, p. 178), not what serves.

Nonsense DEF.= logically impossible.

'Logically possible' means describable. (Moritz Schlick)

It is logically possible for a cow to jump over the moon, just as it is logically possible for a cat to play a fiddle, or for a dish to run away with a spoon.

Ostensively 'cow', 'moon', 'jumping'

We define some of the words of these sentences ostensively (i.e. by pointing): This is a cow, This is the moon, This is called 'jumping over', and so on -- and the picture-sign 'The cow jumped over the moon' relies on these established uses of the words. Then it puts them together in a silly ("nonsense" = absurd) way.

But does a drawing of a cow jumping over the moon really make the sentence a tool that we know how to do something with? Well, we can use the picture to illustrate a children's story; the drawing explains the meaning of the sentence. Telling stories, reciting verse, is a use of language.

The meaning is the use -- in the language

It gives us a picture: we picture a cow jumping over the moon -- something cows, in fact, never do. The picture-sign does not state a real possibility, we say. Very well, we have a picture; but what can we do with it?

"We can use it to amuse children." Is that its meaning according to Wittgenstein? First of all, the sentence is not a word; and second, the meaning of a word is its use in the language (PI § 43). That is a description of the grammar of the word 'meaning' as we sometimes use that word, but it is also the meaning of the word 'meaning' Wittgenstein chose for his logic of language (or even, that is his logic of language).

There are many meanings of the word 'meaning', and we could say that one meaning of 'The cow jumped over the moon' -- i.e. what it amounts to in our life, its significance or importance for us -- is to amuse children.

But that meaning of 'meaning' is not what Wittgenstein is talking about; that is not what Wittgenstein means by the word 'use'. Rather than "use in the language", he could have written "use in the language game": the meaning of a word is its use in the language game. And, as noted above, the use in the language of e.g. the word 'cow' is explained by pointing at cows or pictures of cows, that is, ostensively; in logic of language its meaning is no more or less deep than that.

A concept, not a theory

Philosophers have said that some "common sense statements of fact" that belong to our everyday language must be false [Note 12]. School children are taught that the human eye is a receptor organ: it receives, it does not transmit. But as we normally speak, the eye "casts glances, it flashes, radiates, gleams ... When you see the eye you see something going out from it" (Z § 222). But natural science says that 'Something goes out from the eye' is a false statement, that it belongs to a discarded theory of sight.

Is it a metaphor that "the eye flashes"? No, we are not first taught a prose expression and later given a metaphor. We learn to use precisely that form of expression, 'The eye flashes', and we learn it ostensively.

The sentence 'I see a bird flying over a lake' has a use in our language, which we can describe. But if at school we are told an inverted image is focused on the retinas of our eyes, should we therefore say that the statement 'I see a bird flying over a lake' is false because what we really see is a bird flying under a lake?

Our everyday use of the words 'over' and 'under' is guided only by our conventions for using the words 'over' and 'under'. If we say 'I see a bird flying over a lake' in the normal circumstances, we are not denying (or affirming) that the lenses of our eyes invert images; we are not alluding to an alternative theory of sight (we are not alluding to any theory of sight).

The language game ... children are taught needs no justification. (PI II, xi, p. 200b)

For "naive language", that is to say our naive, normal way of expressing ourselves, does not contain any theory of seeing -- does not show you a theory but only a concept of seeing. (Z § 223)

A 'concept' -- that is, 'rules for using words'. This is how Wittgenstein looked at our language.

Nonsense is produced by trying to express by the use of language what ought to [i.e. must] be embodied in the grammar. (PP iii, p. 312)

We could change the rules, switch the grammars of the words 'over' and 'under'. But to say that our present convention is false is to talk nonsense. Conventions are neither true nor false, but only more or less arbitrary.

*

Philosophy and common sense

"Idealism is a speculative position" (Fichte) -- don't try to conjure up a picture of it; it does not affect our normal sight and hearing: the idealist sees what everyone else sees. Is the idea that there are "philosophical positions", in contrast to "common sense" beliefs -- and that the former are expressed in statements that are contrary to our normal way of speaking? But if our normal way of speaking only shows us a concept (rules), then if there is such a thing as a statement of a philosophical position -- must it be either a language rule or nonsense? (The essence of man in Plato and Aristotle does not appear to be a question about rules -- but neither is it expressed in "unreadable sentences" that sound English -- but are not.)

Endnotes

Note 1a: Nominal or descriptive (distinctive or characteristic) definitions (Anal. Post. B 8, B 10) versus Essential or real definitions (ibid. B 13) (Copleston, History of Philosophy, Volume I, xxviii, 3). Topics Book VI begins "the discussion of definitions" with the obscurity caused by ambiguity, metaphor, and the non-literal use of words in definitions (What Aristotle neglects to mention is the obscurity caused by his own speaking in generalities with far too few examples to make his meaning clear). [BACK]

Note 1: When professors of philosophy say 'Logic is ...' as if they were stating what the essence of the thing named by the word 'logic' really is, they are setting limits to what they want to have called 'logic' and persuade others to call 'logic'. These statements aren't hypotheses (theories) about what logic really is, but only assign arbitrary limits to our common concepts. ("Persuasive Definitions", by Charles Leslie Stevenson, in the journal Mind, July 1938)

"The art of reasoning"

According to the OED, "the art of reasoning" is the oldest traditional meaning of the word 'logic' (although that is different from the meaning found in Diogenes Laertius i, 18, namely 'dialectic' or 'question and answer'). And surely reasoning includes trying to define words e.g. as Socrates did, but also as in a different way Wittgenstein does. In Xenophon's Memorabilia logic is referred to as "the art of words" (i, 2, 31). W.K.C. Guthrie writes that for the Greeks the word logos was a "maid-of-all-work". As well as the general definition 'art of reasoning', the word 'logic' could also be defined as 'ways of reasoning'.

Logic is the art of reasoning. If it were the study of form only, it would be the mechanics of reasoning.

A general definition of 'logic' in Wittgenstein's sense would be 'the study of rules' as in rules of the game or "language game" (The comparison of using language to playing games: what defines a game is its rules, although not every use of language is strictly governed by rules, not as we normally use the word 'rule'). Or by 'logic' we (often) mean 'how to think soundly', which Wittgenstein could say means "how to see the logic of our language aright".

Aristotle's Analytics (The Organon)

The name logic was unknown to Aristotle. Alexander of Aphrodisias (200 A.D.) was the first writer to use the word 'logic' to mean "the study of reasoning". (David Ross, Aristotle, 5th ed. (1949) ii, p. 20, 20n5)

Aristotle gave the title 'analytics' to what was later to be called 'logic'. He did not classify logic as one of the parts of philosophy, but instead called it only a tool (or instrument) for work in philosophy. The Peripatetics said that "Logic occupies in philosophy the place of an organon" (Alexander of Aphrodisias [fl. early 3rd Cent. A.D.], In Topics 74.29). Aristotle himself uses the word 'organon' in Topics 163b9-11 to mean 'instrument'. (Guthrie, Aristotle (1981), p. 135, 135n3)

[The twelfth century] more or less follows the Aristotelian classification. Thus Logic is a propaedeutic or preamble to science [knowledge of things] proper and deals with concepts, not with things. It is divided into Grammar and the Ratio Disserendi ["rational discussion"] (Dialectic, Rhetoric and Sophistic). (Copleston, History of Philosophy, Volume II, xvi, 1)

The definition of 'logic' as 'the art of reasoning' comes from Antoine Arnauld, the Port-Royal co-author of "The Art of Thinking" (1662), with whom Blaise Pascal is connected. Rare synonyms for 'logic' are 'noetics' and 'dianoetics' = 'discursive thinking' = 'discourse of reason'.

Question: are all rules by convention, or are some rules dictated by nature ("Laws of thought") as some philosophers have said? Is it by convention that our language contains the concept 'object'? According to Gilson, Thomas Aquinas thought that there are principles (e.g. 'Every whole is greater than its part', which is presumed to be a statement of fact [but of what kind of fact? Not foundational, because it is logically impossible for them to be false]) without which it is not possible to think. To answer this question, if there is no essential definition of the word 'rule', we shall have to look at the particular case rather than give way to an impulse to try to generalize (which is based on a false picture of language meaning).

Wittgenstein's grammar vs. fact distinction -- (Note that 'fact' here means 'statement of fact', not as it were 'artifact') -- is not the same as the analytic-synthetic distinction. A 'grammatical proposition' is a 'rule of grammar', a convention for using a sign, whereas an "analytic proposition" supposedly states what the "essence of a concept" is. Above, the example 'Logic is really ...' would be an "analytic proposition", as if to say: "I have pondered about logic and somehow, I don't know how, found its essence, which can be stated thus ..." (Contrary to "pondering" is Socrates' inductive method of definition, which may discover that a word is with or without a defining essence.)

In other words, there is no "real definition" -- i.e. speculative hypothesis about the "true nature" of logic -- although an account, which may be true or false, and therefore can be called a "historiographical hypothesis", can be given of the use of that word through human history (as is done above). But that is history of philosophy, not philosophy.

There are no real definitions -- and this is very important to see -- of any "abstract terms" -- we can only describe the use of those terms in the language. Earlier I used the formula: "In philosophy we define words, not things". But that is too broad because it is possible to make hypotheses about objects or phenomena (e.g. thought), but making such hypotheses is not the work of logic-philosophy, and therefore the formula should be "Logic does not define things; it only defines words". (Note.--The form of expression 'abstraction', or worse, 'abstract object' -- is extremely hazardous to understanding of the logic of our language. The word 'mind' e.g. is not the name of an object of any kind.)

Aristotle's minor premise

According to W.K.C. Guthrie, "In the standard Aristotelian syllogism, as distinct from the medieval derivative, the subject of the minor premise -- and so of the conclusion -- is a class or species, never an individual" (Aristotle (1981), p. 349). If the minor premise can't be an individual, then we cannot use a premise like 'Socrates is a man' or have as the conclusion 'Socrates is mortal' -- as in "All men are mortal. And ... Therefore ...". The closest standard Aristotelian syllogism is maybe: "All life is mortal. Man is life. Therefore, man is mortal." [BACK]

Note 2: Epistemology, in Wittgenstein's philosophy, is not about investigating mental states, but about describing the use in the language of such words as 'know', 'believe', 'certainty', 'doubt' -- i.e. it is a grammatical investigation.

We are not analyzing a phenomenon (e.g. thought) but a concept (e.g. 'thinking'), and therefore the use of a word. (PI § 383)

Note: that is logic of language's interest: "Only it's possible to be interested in a phenomenon in a variety of ways" (ibid.) -- e.g. the phenomenon of man in contrast to use in the language of the word 'man' -- not only in Wittgenstein's philosophy's way. (Cf. "There are many meanings of 'language meaning', not only the one Wittgenstein chose for the foundation of his philosophy.")

Our investigation is therefore a grammatical one. Such an investigation sheds light on our problem by clearing misunderstandings away. (ibid. § 90)

Epistemology is the philosophy of psychology. ("Notes on Logic" (1913), Appendix I, of Notebooks 1914-1916, 2nd ed., tr. Anscombe, p. 106)

The theory of knowledge is the philosophy of psychology. (TLP 4.1121, tr. Ogden)

There are two parts to the Philosophy of Psychology: the language of feeling (sensations, emotions, dispositions) and the language of mind (psychological versus physiological vocabularies).

And it can be quite complicated. For example the word 'believe' is used in almost as many different ways as the word 'make' in English. 'To make a bed', 'make a mistake', 'make a sound', etc. What do these have in common with one another beyond the word 'make'? They are all activities? -- So are sleeping and murder, but that doesn't make them any more alike. [BACK]

Note 3: But only because I want to make this connection (Z § 325). I.e. this is the way I choose to limit these concepts. I want to mark off a false path here. [BACK]

Note 6: Drury retells a story told by the French psychologist Pierre Janet (1859-1947):

Janet was talking to an enthusiastic pupil of Freud. "Last night," said Janet, "I dreamt that I was standing on a railway station: surely that has no sexual significance." "Oh! indeed it has," said the Freudian; "a railway station is a place where trains go to and fro, to and fro, and all to and fro movements are highly suggestive. And what about a railway signal; it can be either up or down, need I say more?" (The Danger of Words p. 17)

Drury comments:

Now as Janet rightly went on to point out, if you allow yourself such freedom in symbolism, every possible content of any dream whatsoever can be forced into this type of interpretation. The theory has become "fact proof"; it just can't be refuted. (ibid.)

This is an example of what I am calling a 'tautology' = 'way of looking at things'. What Freud invented was a way of looking at dreams, a way that may be useful in certain psychological contexts (I wouldn't know). But because Freud would allow nothing to count against the truth of his "theory", it is not a scientific theory, because a theory, as Drury defines the word 'theory' is -- (although there are exceptions in science) -- falsifiable: a scientific theory is not "anomaly proof" (cf. 'waterproof'), and it is not a statement of fact (although it may be stated in that form).

Indeed all ways of looking at things -- all "tautologies" -- are neither true nor false: they are only more or less attractive, more or less helpful (or unhelpful) from various points of view (CV p. 55).

*

An exception to these remarks is Plato's method of tautologies in ethics, propositions which are "true" only because they display the interconnections of actual rules of grammar -- and yet are also guides in ethics (Plato's "no small matter, but how to live"). [BACK]

Note 7: Foundational beliefs are not like a metaphysician's absolute axioms ("first principles"). Because although I myself cannot be mistaken about my foundational beliefs, I can be wrong; a foundational statement can be a false statement of fact. Although I myself can have no grounds for doubt, others may have ways to test what I say, e.g. whether I am a giraffe or have two hands are matters of fact. [BACK]

Note 8: "We walk upon the air" means that there are no objectively certain statements of fact to serve as foundations to our life and thought -- i.e. the expression 'objectively certain' is without meaning here, an undefined combination of words (PI § 500) in this context.

Because what makes a proposition 'objectively certain' is that there is compelling evidence of its truth. Truth, like facts and knowledge, belongs to a community (of ideas, way of thinking, way of life; we see this in schools, in law courts). "No reasonable person will deny that the proposition is true." (That is a general definition of 'objective certainty', but it is very general. What counts as compelling evidence varies between particular kinds of cases; cf. the concepts 'verification' and 'measurement'.)

A belief may be justified by other beliefs that are more certain than the belief is itself. But a belief needn't be justified -- and "foundational beliefs" cannot be justified.

There is no bedrock beneath bedrock, or in other words, there is no absolute- or super-bedrock.

Earths flat and spherical (pictures)

B may rest on A, but A cannot rest on A -- or what will it mean to say that a thing rests on itself? There are no self-justifying propositions, i.e. propositions that serve as their own grounds (foundation). The ancient picture of the flat Earth resting on the back a tortoise whose legs "go all the way down" has no application to reality. Or has it?

The meaning of a picture isn't independent of the way the picture's meaning is explained. The early Greeks, that is to say in the 8th century B.C. according to Hesiod, pictured the Earth as a flat disc -- but need the question of what lies at the disc's edge be defined language? "Where were you when I laid the foundation of the Earth? Tell me, if you have understanding" (Job 38.4). Does that question require the picture of a flat Earth?

"We walk on the air" is a metaphor, as indeed are the pictures "grounds", "foundation", "bedrock". Metaphors are important in philosophy, especially when they mislead us. It is important, therefore, not to misunderstand the picture: there is no bedrock beneath bedrock, nor does bedrock "go all the way down". Bedrock is to what we compare the place on which our thought rests; but every metaphor has limits to its application. (The logic of comparison: A is like B in such-in-such way/s, not A is identical with B.)

The picture we are given at school. A building might rest on whatever is beneath it all the way down to the Earth's center (i.e. beyond which a thing cannot fall, a structure collapse; at that point 'fall farther' is an undefined combination of words). And so a foundational belief is like the Earth's center, not like the earth beneath a building, a structure which can collapse if whatever lies directly beneath it shifts (e.g. in an earthquake). But note that there is nothing beneath the Earth's center for the Earth's center to rest on.

Can a foundational proposition fall as a building does in an earthquake, e.g. fall to further reasons or experience of life? No, because if grounds can be given one way or another for a belief, that belief is -- by the definition of 'foundational' above -- not foundational.

As an object of comparison, there is the question: Is there an absolute point of reference (point of view)?

"Mere pictures and metaphors"

"This is a discussion of mere pictures." But pictures in philosophy must be examined to see what their application is, i.e. to if they are pictures of anything (ibid. § 424), as apparent metaphors in philosophy must be examined to see if they really are metaphors or not (LE/Notes p. 14). [BACK]

Note 9: I won't bother about other kinds of possibility here. But Moritz Schlick used the expression 'empirical possibility' ("Meaning and Verification" in The Philosophical Review, July 1936, p. 347) which, I believe, includes both "real" (in my sense above) and "theoretical" (in the sense of "consistent with the laws of nature") -- so that maybe the expression 'real possibility' here is my jargon (although the logical/real possibility distinction is based on the verbal/real definition distinction), i.e. an expression whose meaning I have specified with rules that are different from the normal rules (or absence of rules) of our grammar. An equivalent expression for 'empirical possibility' might be 'natural possibility'. Another kind of possibility would be things that are, or are not at this time, technologically possible.

The antitheses 'dry' and 'wet'

"Is water wet?" Only if it is possible for water to be dry. Otherwise the "question-sign" (i.e. combination of words arranged in the syntax of a question) is mere ink-marks without sense. The question 'Is water wet?' is entirely a question of logical possibility (i.e. defined language). [BACK]

Note 10: A professor of biblical studies and philosophy whose classes I once attended, Paul Trudinger (fl. 1976), studied at Adelaide University with J.J.C. Smart. At their tutorials, Smart took Trudinger again and again over the distinction between a real and a logical possibility. Trudinger would go home and say to his wife, "Why is he doing this? Don't I understand this?"

Professor Trudinger was the first scholar I ever met, a man of many languages and deep learning. When I met him I discovered what I wanted to be: an educated man. Before this I was ignorant, but I did not know that I was ignorant (Xenophon, Memorabilia iii, 9, 6). And so although I did not know him well, he had in this way the greatest influence on my life. It is why I shall always think of him as my teacher. With Professor Trudinger you could have a philosophical discussion where the meaning, if any, of the language used was challenged. It was from him that I first heard of "linguistic analysis".

I was thoroughly perplexed in those days, and although I am still very much puzzled, at least now my thinking has a few foundations, the principal of which I have called "logic of language". The expression is Wittgenstein's, but the jargon is my own. (This idea was the first time in my life when I may have had a "thought of my own".)

Vagueness and confusion and metaphor

Then when I transferred to the university I told the academic advisor (Mario del Carril) there, in answer to the question of why I wanted to study philosophy, that "all my life I have felt surrounded by vagueness and confusion", which he said was a good reason to study philosophy (Plato, Theaetetus 155d), and when I said that I wanted to study "linguistic analysis", he asked me if I had read Russell (which I had not), and sent me to a course about Wittgenstein. The instructor for that course made nothing clearer to me (The TLP is after all a work of metaphysics), but in Wittgenstein's later books I felt that I had found what I was looking for, although it took some ten years for me to come to the understanding of it found in my Wittgenstein's Logic of Language.

Philosophy should make cloudy and blurred thoughts clear and with sharp borders. (TLP 4.112)

In philosophy thought means language: philosophy is discourse of reason. Philosophy begins here with Socrates.

Many years ago I heard that Professor Trudinger was in western Canada, and for more than thirty years I knew nothing more; but he later returned to Australia. Schweitzer said about Socrates, "What would that ancient world have become without him?" And I wonder what I would have become if I had not met Professor Trudinger. Strange it is that people take more care over their choice of doctor than over their choice of teacher -- and that they do this not only for themselves but also for their children -- because a doctor only has care of the body whereas a teacher has "care of the soul" (Plato, Protagoras 313a-c; cf. Lesser Hippias 372b-c).

To end this philosophical-autobiographical note, I will add that although for many of the intervening years Wittgenstein has strongly affected my philosophical thinking -- but that effect has been entirely in "logic of language meaning", not in world-view. From the point of view of the latter, as Russell said, "Wittgenstein was a singular man", and I myself see things very differently (in ethics, thoroughly, in religion, somewhat) from the way he did. For example, to direct a statement (I adapted) of Schweitzer's against Wittgenstein: "However we look at it, existence will remain for us a riddle." My philosophical hero has ever been Socrates: philosophy is not something to be cured of, as Wittgenstein thought, but something to be cured by. [BACK]

Note 12: See Norman Malcolm's account in his lecture "George Edward Moore", in Malcolm's book Knowledge and Certainty (Ithaca: 1963). [BACK]

Site copyright © September 1998. Please send corrections and criticism of this page to Robert Wesley Angelo. Last revised: 23 January 2022 : 2022-01-23

The URL of this Web page:
https://www.roangelo.net/logwitt/logwit11.html

Back to top of page

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