Home - Wittgenstein's Logic of Language | Bibliography

Bertrand Russell in Alan Wood's The Passionate Skeptic

Selections with comments from Wood's biography of Russell. The "Theory of Descriptions". Russell's disgust with Wittgenstein's "linguistic philosophy". G.E. Moore about Russell. Russell's definition of 'morally bad'. Logical form.

My comments are "logic of language" remarks, which are based on Wittgenstein's writings about the relationship between "grammar and sense and nonsense" (i.e. the relation of rules to the distinction between language-with-meaning and language-without-meaning, especially in the context of philosophical problems).

Why is there jargon in philosophy? Words are tools, and sometimes tools must be modified -- i.e. concepts revised -- before they can be used to do the work a particular philosopher wants them for. It was not that Wittgenstein liked jargon but that he wanted to make distinctions.

Outline of this page ...

"The Theory of Descriptions"

G.E. Moore's historical account (as Wittgenstein's) does not appear to be quite correct. In Plato's Sophist a distinction is made between grammatical [i.e. syntax] and logical [i.e. meaning] form (Russell called the latter "philosophical grammar"). That distinction or notion did not originate with Russell, although it may have been original to him.

"The Theory of Descriptions," said Moore, "was something quite new. It was Russell's greatest philosophical discovery, more important than anything he said later. It was his own work, and not influenced by anyone whatever." (Alan Wood, Bertrand Russell: the Passionate Skeptic (1957), p. 63)

[Wood gives a basic example: Russell] pointed out that instead of saying 'The golden mountain does not exist', you could say 'There is no entity which is both golden and a mountain'. Any "analysis" of this kind removes the phrase 'the golden mountain' from any sentence, and removes any cause for belief that it has some kind of being. (ibid. p. 64)

In other words, 'There is no x such that x is both a mountain and golden'. That form of expression does not [-- or is intended not to --] in any way suggest that something named 'golden mountain' has some kind (who knows what kind) of real, although invisible, existence such as Plato thought his Forms to have. (Real existence, that is, as opposed to mere logical existence -- The distinction between "logical and real possibility" -- e.g. as something that might be described or drawn in a child's fairy tale book about a golden mountain.) What Russell wanted to deny is that there must be a realm where of non-existent objects, such as golden mountains, have real existence. [Russell himself had once believed in Frege's "heaven" where geometric objects and numbers had real existence.]

Etienne Gilson, who ignores the above distinction between real and logical possibility, says that I cannot even deny that something exists without first "positing" that it does exist. Gilson's metaphysics is a scholastic example of philosophy as self-mystification, of language bewitchment at work.

Whenever philosophers say that something "must" exist, they mistake their own picture of reality for reality itself ... although in the following example, the "picture" is not even logically possible, because 'undefined language' = 'nonsense', and not even a god can describe nonsense -- i.e. picture the denotation of an undefined combination of words.

'There is no figure both round and square'

Another example where this issue arises is the combination of words 'The round square does not exist' (p. 63), which might be "logically analyzed" as 'There is no x such that x is both round and square', or maybe 'There is no x such that x is both four-sided and not four-sided', or simply 'There is no figure which is both round and square'. The word 'figure' alludes to geometric forms such as triangles, squares, circles -- all of which are defined (by "the rules of the game" or "grammar") to be as they are. That there are no golden mountains, on the other hand, is an empirical (as opposed to a logical) proposition. There is no logical reason why golden mountains do not exist -- i.e. the combination of words 'golden mountain' is not nonsense. (It is logically impossible for there to be a round square -- i.e. the combination of words 'round square' is undefined and thus what it would be like for a round square to exist cannot be described. Golden mountains, on the other hand, are not logically impossible.)

The theory was put forward in an article called On Denoting, first published in 1905 in [the philosophical journal] Mind ... according to G.E. Moore, when the article was first published "Nobody could make head or tail of it". Moore told me that he himself never understood the Theory of Descriptions until Russell made a clearer statement of it in the Introduction to Principia Mathematica [1910]. (ibid. p. 65)

It had been natural to suppose that the grammatical structure of the sentence was the same as its logical structure. It had been assumed ... that a sentence about golden mountains was saying something about golden mountains, and that therefore such things must have being, or they could not be talked about. Russell's analysis proved this assumption wrong: and it also suggested that there may be many other ways in which we can be mislead by words and the form of sentences. (ibid. p. 66)

Wood then gives another application of Russell's "Theory of Descriptions" which is similar to Wittgenstein's account, in which he gives credit to Russell, of [the "grammar" = "logic" = "meaning" of] the name 'Moses' in the Bible (PI § 79), and which is applicable to any proper name considered as history (if that contrasts with as defined in the present ostensively).

Query: What does "grammar-laden" mean?

That expression could be applied both to Russell's "Theory of Descriptions", where some propositions "must" be rewritten to display their "philosophical form". But that only means that if we regard the form of a proposition as confusing, we can rewrite it in a different form which we do not regard as confusing. And also to Wittgenstein's jargon, where language meaning is also "grammar laden", for there are many meanings of 'meaning'; Wittgenstein chose one, in order to make an objective distinction between sense and nonsense -- not to invent a theory about what the meaning of language "really" is. (Note that Russell calls his invention a theory, indicating that he believes that with it he is saying something about the nature of reality, as if he were talking about what were "really the true" grammatical form of propositions.)

Both Wittgenstein's "grammar" and Russell's "philosophical grammar" are examples of jargon in philosophy: or why philosophy cannot be "a simple story in words of one syllable" without redefining those words. (Philosophers revise the concepts that are our common currency; thus some philosophers produce jargon, and other philosophers produce nonsense.)

Simply putting the word 'Why' in front of a combination of words does not turn that combination of words into a question. In Russell's jargon it may have the correct grammatical form, or structure, or syntax -- i.e. "correct" when by 'grammar' is meant combinations of the categories 'noun', 'verb', 'adjective', etc. -- of a question without having the "philosophical" form or structure of a question. (But I owe you an example of this.)

Wittgenstein's criticism, however, even of Moore, was that both he and Russell focused on the form of language [on word combinations] rather than on the use made of the form of language (LC p. 2). For why talk about form [the structure of language] when what we are concerned with in philosophy is instead the meaning of language, which is not a matter of form (or certainly not a matter of form alone: form does not play the determinant part and indeed in some cases may not play any part at all).

In his later use of the word 'logic' Wittgenstein is concerned with sense and nonsense rather than with form. But what Russell means by 'logic' -- I don't know its extension, how he defines the word 'logic', where he sets that concept's limits.

The Theory of Descriptions and the Meaning of Proper Names (Principia Mathematica *50).

Semantics (The fly-bottle)

Query: against Wittgenstein's aim in philosophy.

Do you mean that you want to keep the fly imprisoned in the fly-bottle (PI § 309), to keep it stuck on the fly paper of a naive view of the way our language works? "The logic of our language is misunderstood" (although not in the TLP's sense). Do you have then an alternative "logic of language" to replace Wittgenstein's with?

Rather than Wittgenstein's aim in philosophy, what that needs to be queried is his conception of philosophy -- of why he set philosophy's limits where he did -- and whether he was justified in doing this. Are the problems of ethics and "physics" [What is now called 'metaphysics' was called 'physics' by the Stoics, who called our physics 'mathematics'] indeed not problems at all, but only disguised nonsense?

What I have called Wittgenstein's logic of language does put an end to philosophy as "x is really" statements -- not necessarily by showing that they are meaningless -- some are -- but that none of them is a statement about reality. Some are, however, new ways of looking at reality, re-conceptions that allow one to see things in a new way, although not in the "real" way, not in the "perspective of God" way philosophers want.

"A new way of philosophizing"

Query: where did Wittgenstein say that, if there is no understanding, then there cannot be thinking?

Anyone thinking in the old way -- i.e. anyone confusing factual with conceptual investigations (Z § 458) -- cannot (for how will they understand his work without knowing its foundation) understand Wittgenstein's work. The query suggests that Wittgenstein, if he had said that, which he did not, was stating facts about things named 'understanding' and 'thinking' and saying that there is no metaphysical possibility of thinking without understanding, as if he were trying to say "what thinking really is".

But when Wittgenstein speaks of "cannot" he is talking about logical, not "real" or empirical, possibility. Here he would have to be saying that the concepts -- i.e. the "grammar", or, rules for using words, of -- 'understanding' and 'thinking' are related in such a way as to make the combination of words 'thinking without understanding' nonsense (i.e. an undefined combination of words). Which is something there would be no justification for saying, because -- if we use the words 'thinking' and 'understanding' the way we normally do (and why create jargon by defining them in a different way) -- we do call confused thinking 'thinking'; we don't say that someone is not "really" thinking if he is confused, although we do say that he does not understand what he is thinking about.

And Wittgenstein did not talk in the old way either about "the essence of things" (Plato's metaphysics, Socrates' ethics) or "the essence of concepts" (Analytic philosophy). He only talked about rules for using language -- i.e. "signs": Sounds and ink marks are in themselves without meaning -- and hence what gives them meaning? Their use according to rules (conventions) in our language. And now we can describe that use, which is public and therefore objective. And when we are clear about how words are used, we stop confusing factual with conceptual investigations, and stop imagining that playing with words can reveal to us what things "really" are (which is what the method Wittgenstein called 'metaphysics' does).

The query, taken as it stands, would show that it was impossible for the person who wrote it to have been thinking at the time they wrote it, because that person did not understand Wittgenstein's new way of thinking.

However, even if the Wittgenstein is correct about "really is" metaphysics, and I think he is in some (indeed, very many) cases, so that it is mistaken to think that philosophy can "understand the world as well as may be" (Russell's aim in philosophy) -- even then would that make philosophy "so small a thing and worthless"? For Bertrand Russell it would, but not to me. Even if "moral virtue is knowledge" is only a way, one way, of looking at things (and not what virtue "really is", whatever that would be when it's at home), it may nonetheless be a way of looking at moral virtue that is useful to ethics --

In what way a way?

[I cannot now remember, years later, what I meant by calling Socrates' thesis a way of looking at things. Before I spoke of the need to adopt a point of reference, because no point is absolute. Was that what I meant -- that it is logically possible to look at ethics from the point of view of Wittgenstein's notion of "absolute value"? which is a point of view that Socrates takes no account of because it is irrational (indeed, it is nonsense (Memorabilia iii, 8, 3)). Does Socrates' thesis confuse a conceptual with a factual investigation, as Wittgenstein says all philosophical problems do (Z § 458)? I would not say that. Once the good for anything is identified as the specific excellence that is proper to it, the question of what that excellence is, is a factual investigation. Once Aristotle's classification scheme is adopted, it becomes a question of fact whether man is a rational animal or not.]

-- it is certainly a very different way from the irrational convictions Kant called "categorical imperatives", or the notion of "weakness of the will", or of depending on the elusive grace of God to change your life (Samuel Johnson). Using the tests of reason and experience to discover what is useful to [what works in] ethics is certainly not an idle thing to do -- and it is philosophy. Just as studying how to think -- i.e. to reason, which is logic -- is also philosophy; as is to ask what is real, which is "metaphysics" -- even if that were only -- not merely -- a conceptual investigation. [Metaphysics, logic, and ethics, the three parts of philosophy.]

Philosophy is also an investigation of the foundations of maths and sciences; it asks, for example: do these subjects, in their own ways, discover the reality of the world as it is in itself -- i.e. independently of man? [The historical branches of philosophy.]

Wittgenstein came down from the mountain to proclaim that "Philosophy is dead" only to find that philosophers would not stop worshipping their golden calf, that is to say, abandon the linguistic self-mystification that is philosophy, according to him. He said that no one understood his early work, the TLP, not even Russell and Moore (Wood, p. 156). Apollo's oracle might have said the same -- but what the god said, like what philosophers say, was said, not to directly answer your query, but to make you think for yourself (just as the inscription at Delphi "Know thyself" was). "A philosopher says: Look at things this way! But that doesn't mean that anyone will." .... "I would prefer a change in the way people live that would make all these questions superfluous" .... "As if you could change the way people dress by talking about it!" (CV p. 61, 62)

Why did Wittgenstein always set the limits of philosophical discourse where he did? Why did he refuse to write for publication about ethics and philosophy of religion, as if he held, long after he had rejected the TLP's account of the logic of our language which justified it, that these are subjects "we must pass over in silence"? Because we know that he was willing to talk about "what cannot be put into words", and he was only willing to do that because apparently it can be talked about: nonsense that can convey meaning is not, after all, nonsense.

Jacob Bronowski shook his head in Platonic wonder when he came to Isaac Newton's statement "I don't make hypotheses" -- because he knew that Newton did make "hypotheses" in his private writings: he was a devoted cosmological-theological speculator, even an alchemist.

[What is the key to why Wittgenstein limited philosophy to logic of language? One possibility, but it no more than a possibility, might be his limited aim for himself in his own work in philosophy, if even to speak of "his limited aim for himself" is a correct account.]

Query: Philosophy is semantics. Wittgenstein.

Reply: "Wittgenstein was a very singular man" (Russell). If by 'semantics' is meant "grammar" in Wittgenstein's jargon, then you might want to say that (But it would be a caricature). The caution is that it suggests something far too simple: If only we define words precisely, then we won't get into these muddles (cf. G.E. Moore). Which may be a useful idea -- but it does not work in philosophy unless we first get clear about how to make an objective distinction between sense and nonsense (and what the consequences will be for our work in philosophy if we treat language-meaning as if were subjective) -- i.e. if we do not first get clear about how our language works (its "logic" in Wittgenstein's jargon). Because if we are not clear about that, then how shall we distinguish between a definition that clears up a muddle from a definition that only substitutes one muddle for another. (Many philosophers, although not all, want to set objective criteria for correct answers, but not all succeed at that.)

"... a linguistic philosophy"

The habit of seeing both sides of an issue applied to Russell, at least when he was young, as a philosopher: "Whitehead once called him a Socratic dialogue in himself." (Wood, p. 75) But in old age Russell could be unable (or unwilling) to see the other side, and he was capable of revising the historical record as well:

"I found Wittgenstein's "Tractatus" very earnest and this implied a genuine philosophical outlook in its author," he was to write. "I did not appreciate that his work implied a linguistic philosophy. When I did we parted company ..." (Letter of 20 December 1968, quoted in Clark, Life of Bertrand Russell (1976), p. 370 [p. 703])

But then he must not have originally read that work with much care -- Wittgenstein called the Introduction Russell wrote for it "superficiality and misunderstanding" (Letter to Engelmann of 8 May 1920, quoted by Clark, p. 371 [p. 703]). For although the TLP is a work of "really is" metaphysics [see above], Wittgenstein had nonetheless explicitly said [4.112] in it that philosophy is about clarification, not about trying to "understand the world as well as may be". When he later carried through his clarifications in a non-metaphysical way, Wittgenstein's rejection of Russell's belief that philosophy consists of philosophical propositions (theses) could become clear to Russell. What I don't know is if Russell ever understood that Wittgenstein's fundamental idea had always been that the distinction between sense and nonsense must be [made] objective -- or it does not exist at all. In 1909 G.E. Moore had written in his diary that Russell was "too confident of insufficient explanations as to meaning of words ..." (Clark, p. 106 [p. 673]). Did Russell think that Wittgenstein's point was obvious? I don't know, but giving mouth honor to the problem does not solve it.

The question of whether it was only because of the meaning of the word 'meaning' Wittgenstein had selected (because it made the objective distinction he sought) that his work put an end -- an end Wittgenstein seems to have always wanted -- to philosophy as Russell and his predecessors conceived it, or misconceived it, is neither here nor there -- unless someone offers an alternative definition of 'meaning' that both makes the sense and nonsense distinction objective and makes "philosophical propositions" not nonsense or idle pictures.

Background is needed to understand the remarks on this page; that is why there are so many links. I was quite correct when I wrote that my pages will not be understood without first understanding Wittgenstein's logic of language, his new way of doing philosophy.

Russell's muddle with 'good' and 'bad'

"I find it quite intolerable," he wrote, "to suppose that when I say 'Cruelty is bad' I am merely saying that I dislike cruelty." (Wood, p. 229)

"But 'glory' doesn't mean 'a nice knock-down argument'," Alice objected.

'Cruelty is bad' is not equivalent in meaning to 'I dislike cruelty'. 'x is bad' does not = 'I dislike x'. We use those propositions differently. For example, the first deduction below is valid, but the second is not: (1) Bullfighting is cruel, and what is cruel is bad, and therefore bullfighting is bad. (2) I dislike cruelty, and bullfighting is cruel, and therefore bullfighting is bad. -- That is an invalid inference. You cannot substitute the one proposition for the other ('x is cruel' ≠ 'I dislike x'), whereas if they had the same meaning you could. (If Russell is not talking about meaning here, then it is not clear what he is saying.)

Further, I can say [i.e. it is not nonsense], that although bullfighting is cruel and that although cruelty is bad, I nonetheless like bullfighting -- i.e. I am not logically compelled to say: and therefore I dislike bullfighting. Because what I like and dislike belongs to "the world as I found it" -- i.e. it is not a matter of choice; it is simply part of the given; it is not part of ethics.

Again, we use these two signs differently, and if the use of a sign is [as it is in this particular case] its meaning, then these two signs do not have the same meaning. We do not use the words 'good' [as in 'x is good'] and 'like' [as in 'I like x'] the same way. For example, if I say that school is good, I am not necessarily saying that I like school -- i.e. the one does not logically entail the other -- which on Russell's account it would have to.

I can certainly say that something is bad for me, but that I nonetheless like it. If 'It is bad' had the same meaning as 'I dislike it', then I could not say 'Smoking tobacco is bad, but I like it'. That would not be the same as saying that 'Smoking tobacco is good'. 'I like it' does not mean 'It is good'.

Everything is what it is and not "really" some other thing. (Bishop Butler) Does Russell try to reduce: 'x is good' to 'I like x' -- i.e. does he try to say that the first statement can be "reduced" to the second? Does a "logical analysis" of 'x is good' show that the "philosophical grammar" of 'x is good' is really 'I like x'?

... shortly before 1914, Russell had been convinced by Santayana that there are no such things as objective ethical values. "Good" and "bad" merely express subjective likes and dislikes.... All rational argument [in ethics] could only be hypothetical, of the form "If you want such and such a result, then you must so-and-so". (Wood, p. 97-98)

[But basing your discussion, your thinking about ethics on the concept 'value' is not the only way to think about ethics. Socratic ethics is not based on the concept 'value'. (Is the expression 'objective moral value' nonsense, i.e. an undefined combination of words?)]

Good habits and Self-control

... said Russell, the secret of modern moral education was to make good behavior a matter of habit, not self-control.... even the acquisition of good habits was a painless process. (Wood, p. 158)

(That view of moral education belonged to Aristotle.) Russell deprecated the time that had been dedicated in his youth to studying Latin and Greek. "... he criticized too much emphasis on the classics." (ibid. p. 165) But more attention to the classics, especially in this case Plato's Sophist, would have shown Moore (who had had a classical education) and Wittgenstein (who had not, having learned only Latin, and who was never in his life a scholar) that the Theory of Descriptions was not entirely a new notion, not entirely a new way of looking at language.

Likewise when Wood talks about Russell and good-habits. This is found in Epictetus ("Choose the life that is noblest, for habit will make it sweet to you"). My reply to Russell would be that self-discipline is a good habit. [Good-habit formation.] One which one develops and consequently comes to love the practice of. This was the case with Socrates "growing daily in goodness".

What is the Most Useful Way to look at Ethics?

Note: this continues earlier discussions of an aspect of the ethics of Xenophon's Socrates: "care of the soul", as well as the discussion of Socratic ethics.

If the good is the useful -- that is -- is it? For if to be good is to be "good for something", then what is a Schubert sonata good for? I think that must be seen in the context of "care of the soul". A man is somehow (I don't know how) made better by the experience of the combination of the good and the beautiful found in all serious music. It has the quality of depth, but whether or how it can share this depth with the listener, I don't know. Am I made better -- does it make me more ethical? Obviously that question cannot -- logically cannot (i.e. this is a matter of definitions or "grammar") -- be answered because I have set no criterion for a correct answer. Is "the peace which surpasses all understanding" of any ethical worth?

The identification of the good with the useful is not a theory of goodness (It does not say what goodness "really is"). It is a way of looking at ethics. (And the same is the case with "Virtue is knowledge" -- it does not say what virtue "really" is.) But is it a useful way of looking at ethics in the case of Schubert's music? What if any, for it may not be ethical at all -- i.e. it may not help one to "grow in goodness" -- is the place of the appreciation of beauty in the good human life?

A way of looking at things is a tool. Every [each] way of looking at things is a tool. Different conceptual tools [in this case, different frames of reference] are useful for different types of work, to accomplish various tasks. Ethics has many such tools, ways of looking at things that are useful to accomplish its practical ends. Sometimes Jesus' answer to the question "Who is my neighbor?" and the petitions of the Lord's Prayer are useful, sometimes the Platonic Socrates' Virtue is knowledge is useful, sometimes the sayings of Epictetus are useful: various ways of looking at things are useful in our efforts to become good men by replacing through reasoned self-control the bad habits acquired in the time of ignorance with good habits (because "what we do from habit is sweet to us"), keeping watch over ourselves in order to one day find ourselves, like Socrates, "growing daily in goodness". These are not theories about reality, but ways of looking at human experience and the place of reason in our lives.

Seeking what is "higher" (in Wittgenstein's sense) is an act of worship, of "Love God with thy whole heart". But what has that to do with "... and love thy neighbor as thyself"? I don't know. But is not piety -- i.e. correct behavior towards the gods -- a moral virtue -- i.e. does it not belong to the good human life? What is the ethical value of looking up into the "starry sky above" (Kant)? Is it not a question of the spirit in which the thing is done? (And is the same the case with Schubert's music? I don't know.)

Looking at your life from the point of view of eternity in contrast to worldliness. Does this make you better as an ethical personality? It belongs to "the unexamined life is not worth living" to look at your life from outside: "We are discussing no small matter, but how to live". Maynard Keynes contrasts "being good" with "doing good". Does it belong to the "grammar" (definition) of 'good' that "doing good" is the only form of goodness, that "doing good" is "being good" but not vice versa? Surely one cannot be good without doing good ("... and love thy neighbor as thyself") -- but is that the end of it? When Socrates in Xenophon says that he is "growing daily in goodness", what does that include?

Examining "how to live" -- seeking the truth about ethics is itself "doing good", according to Socrates, for what is more useful than knowing how to live our life in a way that is dedicated to the good, to knowing the the excellence that is appropriate to man (in contrast to other forms of life). If we know what is good, we will do what is good ... unless we are lying to ourselves: "I say I know, but I think I know something better."

Hodie mihi, cras tibi

"It'll be your turn to die one day." (Solzhenitsyn, Cancer Ward, tr. Bethell, Burg, i. 15, "To Each Man His Own")

"You're going to die one day too. Today me, but tomorrow you." Seeing your life "from the point of view of eternity" rather than from the point of view of worldly concerns. (Mons. Venditti used to remind the people that: Siamo nati, e dovete morire.)

"He was always talking"

Sixty years afterwards I was pressing G.E. Moore for any recollections he might have of Russell as an undergraduate.... there was one thing he remembered quite definitely about Russell. "He was always talking," said Moore. (p. 28)

There is a saying that no one ever learned anything with his mouth open. But that "never" is incorrect. I have frequently learned two things from my own talking: (1) that I do not know what I am talking about, because I am unable to explain it to others, and (2) I actually do have some insight into something I supposed that I had no insight into. In sum: I learn that I have ideas that I need to develop. If you are always talking, however, you are never listening to what others have to say; and that is why, Zeno the Stoic said, you have two ears but only one mouth, so that you may listen and learn more and talk less. Pythagoras did not allow his students to talk for at least the first two years they were at his school. (The story may be a legend, but it is not told for nothing.)

"... that mathematics was mere rules for using signs"

At one time he even came to believe that ... both mathematics and logic were merely conventions about the use of symbols and words. "Two and two are four" was like saying that "there are three feet to a yard". (Wood, p. 55)

There is a difference between saying that something is founded on rules and that something consists of rules. Mathematics is founded on rules, but it does not consist only of rules -- but also, and mostly, of what may be deduced from rules. [The following of rules may play no necessary part in the creation of new maths, but all mathematics must be justifiable by rules, even if only after the fact of its creation. If that is not a correct account of maths, then I do not understand mathematics at all, for does not 'mathematics' = 'calculi'?] Does chess then consist entirely of rules; what if anything does consist entirely of rules? Traffic laws e.g. consist entirely of rules; the laws are not founded on rules [although, of course, the laws must be self-consistent if they are to do their work]. You cannot begin with "the axioms of the rules of the road" and deduce all the rest of the rules of the road from these; the objective of safety e.g. can be achieved in many different ways, as can the objective of maintaining traffic flow. Mathematics, on the other hand, does not have that kind of arbitrariness; e.g. axioms are the foundation of geometry, and what may be deduced from them follows without discretion.

The difference between "Two and two are four" and "There are three feet to a yard" [p. 55] is shown by this: that the meanings of these propositions are explained differently. [That these propositions are different in kind, if they are different in kind, is shown by the difference in their explanations of meaning: their meanings are explained is distinct ways.]

It is not a rule of mathematics that if any two numbers sum together to equal a multiple of three, then together they make up a number evenly divisible by three. 1 + 5 = 6, and 6 is evenly divisible by 3, and therefore -- but there is no "therefore" about it; it just happens to be this way -- 15 is also evenly divisible by 3. (cf. 5 + 1 = 6, and 51 divided by 3 equals 17.) That is not the case -- it is not the same with the number 5; for example, 2 + 5 = 7, which is not evenly divisible by 5, although 25 is evenly divisible by 5. The situation with the number 3 is an observation; it is not a rule: it is not something that was intended, and it is not something that is essential to mathematics [It is not necessary to mathematics that it be this way]. Mathematics would not be damaged were it otherwise than it is with the number 3.

Russell's absence of method, rather Platonic, way of thinking

[Russell wrote to F.H. Bradley in 1914 that] I don't know how other people philosophize, but what happens with me is, first, a logical instinct [Comment: what's that when it's at home?] that the truth must lie in a certain region ... If I do not hit the exact point in the region, contradictions and difficulties still beset me, but though I know I must be more or less wrong, I don't think I am in the wrong region.

The only thing I should [ = would (conditional), not 'should' = 'morally obliged'] ever, in my inmost thoughts, claim for any view of mine, would be that it is in the direction in which one can reach the truth -- never that it is the truth. (Wood, p. 49-50)

Wittgenstein told Moore that "a method had been found ... it did not matter whether his results were correct or not". Both Russell's and Wittgenstein's statements suggest Plato's "I stick to my principles, and if experience seems at variance with these, then I don't pay it any mind." For how could Russell be certain about his "region" or Wittgenstein about his "method" -- without determining whether the results they yielded were correct or not? They did unintentionally what Plato had done intentionally.

My own testimony, for what that is worth, is that I have often found that I was in the wrong region, that the solution lay in a very different direction from the one I had first thought it did, and that the particular method I had tried to employ was not the correct one, but that a different method was needed.

The anti-Platonic view: knowing reality is not like following the axioms of geometry: it is not for geometry-like theorems to decide what is real. That the truth about reality is geometry-like is the Platonic view, a view that was later taken up by Descartes and the other Rationalists who followed the deductive method of philosophizing.

Russell and Moore and Wittgenstein

Note: this continues the discussion Character and Truth. Wittgenstein seemed to believe that a man did not really choose his philosophy, but that the way in which a man philosophized was an expression of the man's moral character. It was like saying that there is no philosophical truth (Philosophy is clarification only: there are no philosophical propositions to be true or false), and therefore, in that way, philosophy itself is not so important. What matters is not so much the results of a man's thinking, but whether or not the man is an honest thinker, the spirit in which his work is done. The conscientiousness with which Wittgenstein did his own philosophical work was a conscious expression of his character. The same, he judged, was the case with Russell and G.E. Moore.

When Wittgenstein could not respect a man's character -- as 'character' is measured by a man's values and ideals and whether he is faithful to them -- he did not want to discuss philosophy with him. As if to say that philosophy could only be good if the man who discussed it was also good. Alexander said that "The measure of a man is what he occupies himself with" (Diog. L.), meaning that a man of noble character occupies himself with noble things, a man of low character with low things. Wittgenstein wrote in English to Russell from Norway in 1913 (R.22): "As to Wiener I can only say that, if he is good at Math, Math isn't much good." As to mathematics I don't know, but if someone's work in philosophy isn't done conscientiously (which is one measure of 'character'), then I doubt it will be any good, because mistakes in philosophy get passed over when work is not done conscientiously, as if a man's own character might blind him to the truth, which, apparently, he does not really care about anyway.

I have long wondered, which is to say I don't know, whether Wittgenstein could have said what Etienne Gilson said (my paraphrase): "I never confused philosophy with religion. If a man's religion is at stake then his whole life is at stake, which is not the case with philosophy." His relationship to God (i.e. the place of the concept 'God' in his life) was vitally important to Wittgenstein, as was his ethics -- but he did not regard either of these as philosophical topics. Philosophy was logic, and logic was logic of language studies (although language is not so important in aesthetics and religion, and these topics must be studied using other methods). And logic is important, but maybe it does not put "a man's whole life" at stake, as his religion and ethics do. (Socrates, with his trust in thoroughgoing reason, could not have said that, but maybe Wittgenstein could.)

Would Wittgenstein have said (in contrast to Plato, Gorgias 506c) that he would rather remain in his mistakes than have his mistakes corrected through discussions with a man of not the best character? I don't know. I think that he came to have that view of Russell. Moore felt that way about Russell too; he tried to avoid seeing Russell when possible (Clark, p. 97).

Once they [Wittgenstein, Russell and Moore] were having coffee and talking together when Russell suddenly turned to Moore and said: "You don't like me, Moore, do you?"

Moore thought carefully and replied: "No".

They went on chatting about other things, and Wittgenstein was left puzzled and upset as to how Moore and Russell could still seek and enjoy each other's company. It was a little incident very typical of all three men. (Wood, p. 87-88)

Wittgenstein may have taken Moore to mean that he did not respect Russell as an ethical personality and as such it was incomprehensible to Wittgenstein why Moore would wish to share Russell's company. To Wittgenstein, character, as he measured it, in his "black and white" judgments of men (Drury's thought), was an obstacle to his continued relationship with Russell at the time they were discussing the TLP, as Russell wrote to Lady Ottoline (but I don't remember where), treating this humorously, which Wittgenstein did not; in his letters to Russell (R.26 and R.27) written in 1914, he cites a difference in ideals as a reason to discontinue their relationship, although he does agree to continue it -- for Russell had been his teacher, and Wittgenstein still felt respect and affection towards him -- so long as they avoided all questions of value-judgment.

Letters to Russell, Keynes and Moore 1912-1935, tr. McGuinness (1974), R.27 dated 3.3.14. from Norway: 'value-judgment' (Werturteil: "judgment as to value"); "our ideals [Ideale] could not be more different". Russell wrote that "Wittgenstein was a very singular man, and I doubt whether his disciples knew what manner of man he was". But there is reason to doubt whether Russell himself knew, for Wittgenstein had no tolerance whatever for either glibness or irreverence, both of which were very characteristic of Russell.

The impression of pre-war Cambridge given by Keynes' Two Memoirs is that the one outstanding figure was G.E. Moore, while Russell is barely mentioned except by way of criticism. There is other testimony ... that Moore was the only man who could cope with Russell in argument, with his persistent "Do you really mean that?" And his unanswerable way of wagging his head sorrowfully to express incredulous reproach. Some of those who heard them arguing during this period thought that Moore was solely concerned with searching for the truth, whereas Russell liked to score debating points. (Wood, p. 88)

So far as Keynes and Russell were concerned, there seems to have been a certain latent hostility on both sides.... According to Russell, Keynes had the "sharpest intellect and clearest" intellect he had ever come across ... "When I argued with him I felt that I took my life in my hands, and I seldom emerged without feeling something of a fool." (ibid. p. 89)

Given how sharp Russell's own intellect was, that is a most remarkable compliment. But maybe Wittgenstein would have said that this was not a question of natural endowment, but that Keynes' sharpness was a matter of character and that Russell would have been just as sharp himself were it not for that. This is all conjecture.

Russell and Ockham's razor

Occam's own words were "It is vain to do with more what can be done with fewer." (Wood, p. 91)

Here 'vain' means 'pointless'. It is "done in vain" means that: it serves no useful purpose; indeed, it accomplishes nothing.

The following, I think, is an example of Russell's use of Ockham's principle, but also it is very like Aquinas' first principles in that it treats the net cast into the sea as if the net were part of the sea itself: different sized nets will catch different size fishes (The nets are frames of reference, of course [cf. Tractatus Logico-Philosophicus 6.341]). It is not reality that is caught, but only what our way of looking at reality catches. And it presumes that we can catch the whole -- i.e. all -- of reality with the net of our five senses (with or without instruments, microscope, telescope e.g.).

It is also related to Russell's Realism directly after he broke with Idealism:

In the first exuberance of liberation, I became a naive realist and rejoiced in the thought that grass is really green, in spite of the adverse opinion of all philosophers from Locke onwards. (My Philosophical Development p. 61, quoted in Clark, p. 71)

And is it really green in the dark as well? If man has the right [i.e. the grammatical right -- what other kind would this be -- empirical?] to say that grass is really green, then the dog has the right to say that grass is really grayscale. 'Objects exist independently of me', which (is a rule of grammar) is all that Russell's "naive realism" could amount to. Were it an experiential question we would be left in skepticism about how objects really exist, for we perceive them only within some frame of reference or other. However, grammatically objects exist in this mode: these concepts apply to them: 'mass', 'color', 'dimension', 'shape', 'size', 'density', 'fluidity', etc. -- But why does anyone regard a rule of grammar as if it were a revelation? Because that was our pre-WII (Russell's name for the later Wittgenstein of the Philosophical Investigations) state -- one that is easy to see now -- i.e. after it has been shown to us, but no one saw it before Wittgenstein.

Realism was not, as he thought, an escape from Idealism for Russell, but simply a prolonged failure to understand the logic of our language (or, in any case, what I have called Wittgenstein's logic of language). At best Realism and Idealism are "incomparable pictures" -- i.e. pictures for which there is no defined technique for comparing with the reality they are said to picture, pictures drawn as it were from the perspective of a god who can see what we cannot see (In other words, they do not describe reality but are idle) -- and at worst nonsense.

Russell's 'The world I perceive is independent of me' (Realism) is either a rule of grammar [which, if we are to describe our common usage, as a matter of fact it is] or nonsense, for it is certainly not an empirical proposition (statement of fact).

["But doesn't an infant need to learn where its own body stops and "the external world" begins?" I don't know. No one can remember what he experienced as a baby. And how would a baby tell us, for babies do not use language, language being a socially acquired instrument. And we only apply the word 'learn' to a baby by analogy (comparison of what we know to what we don't know).]

Russell uses the expression "naive realism"; however, it is not clear how that differs from "sophisticated realism". Maybe it would be clearer to say "uncritical realism" versus "critical realism". But if anyone were critical -- in Wittgenstein's sense -- then he would not be any kind of Realist (or Idealist).

Back to Russell's use of Ockham's razor:

Russell had ... been led to emphasize the word 'similar' by the technique of "minimum vocabularies". This was the surviving relic of his belief that we can get knowledge about the structure of reality by studying the structure of sentences. His idea, put in simple language, was this: Try to find how few words you need to describe the Universe; if you cannot describe it without using some particular word, then there must be something in the Universe corresponding to that word.

Yes, but this "must" may be no more than the requirement of a particular frame of reference, or even a way of perceiving that is peculiar to our human form of life, like the concept 'green', for example. Lichtenberg: "Do not call it a theory, but only a way of presenting it to the human mind." (A theory, in metaphysics, and that is what Russell is doing, attempts to say what reality "really" is.)

In this way, for instance, he tried to see whether he could find a vocabulary which dispensed with words standing for "universals"

Does 'universals' here means 'common names'? On the other hand, however, 'similar' is not a common name. --

and he found that he could not dispense with the word 'similar'. He came to the conclusion that: "The fact that we need the word 'similar' indicates some fact about the world

But the world is our world. (Not in the TLP's sense, of course, but had not Wittgenstein's book "stimulated" Russell to any "thoughts of his own"?) --

not only about language. What fact it indicates about the world, I do not know." (Wood, p. 224-225)

But again, language is our -- i.e. human -- language. What words a language used by a different species "could not dispense with", we might try to imagine, but always only on analogy to ourselves, which seems to limit the possible applicability of anything we might imagine to species very like our own.

But Russell's insight that our language does indicate something about the world -- for if the world were otherwise than it is, if some "very general facts of nature" were "different from what we are used to", we would not have the concepts (i.e. rules for using words) that we in fact do have (cf. PI II, xii, p. 230) -- is very important. But on the other hand, so is the insight that Kant had -- that "concepts without percepts are blind", but that those concepts belong to a human mind, just as those percepts belong to human senses.

"The Scope and Limits of Human Knowledge"

"Indeed," he concluded, "such inadequacies as we have seemed to find in empiricism have been discovered by the strict adherence to a doctrine by which empiricist philosophy has been inspired: that all human knowledge is uncertain, inexact, and partial. To this doctrine we have not found any limitation whatever." (Human Knowledge, Its Scope and Limits (1948), p. 527, quoted by Clark, p. 493)

First, I do not understand what the purpose of the word 'human' is in the title of the book; do we know of some other kind? -- i.e. the concept 'knowledge' is a human concept (and we apply it to other animals only by analogy to ourselves), if, that is, we suppose that there are any other kind of concepts than those of mankind, to whom the concept 'concept' belongs.

Second, if all knowledge is "uncertain, inexact, and partial", then no knowledge is "uncertain, inexact, and partial". Here ALL implies NONE. But we do not use the word 'uncertain', 'inexact', and 'partial' that way; instead, 'uncertain' contrasts with 'certain' and depends on that word for its meaning; and likewise with 'inexact' and 'partial'. Many words have meaning only when they are contrasted with their antithetical words. And the antithesis of 'empirical certainty' is not 'mathematical certainty' (or 'deductive certainty'), but rather 'empirical uncertainty'.

Indeed, mathematics is neither certain nor uncertain; it is a game played according to -- i.e. justified by -- the strictest of rules. But a rule of a game is not an empirical proposition. ('All x is y, and B is x, and therefore B is y' is not an empirical proposition; it is a rule, and as such neither certain nor uncertain.) The notion of "mathematical certainty" is Platonic.

This is also related to what Drury wrote (cf. DW p. 72) -- that in every investigation something must stand firm [cf. Wittgenstein's OC § 337], otherwise we do not call it an 'investigation', which is to say that something must be "certain", "exact" and "complete" in every investigation. In every investigation there must be something that is not investigated -- namely, whatever it is that is used as the standard in the investigation.

Non-deductive inference versus deductive inference (Logically necessary versus merely probable)

I had become increasingly aware of the very limited scope of deductive inference as practiced in logic and pure mathematics ... I realized that all inferences used in both common-sense and in science are of a different sort from those in deductive logic, and are such that, when the premises are true and the reasoning correct, the conclusion is only probable. (My Philosophical Development (1959), p. 190, quoted in Clark, p. 492)

In 1943 Russell's had made plans "to attempt to systematize non-demonstrative inference" (Russell quoted by Clark, ibid.) and the outcome was his book Human Knowledge: Its Scope and Limits (1948). The book is in six parts, and part five concerns the Theory of Probability, and part six is about "Postulates of Scientific Inference" (Wood, p. 219). Russell had set out to say what we needed in addition to empiricism, and "he found the answer in five rather complicated "postulates"" (ibid. p. 221).

Wood says that Russell had dismissed the notion of substance by calling a table (such as Arthur Eddington spoke of) an event (ibid. p. 223). But that is not the way we use the word 'event': a table is not an example of an event; it is instead an example of a solid object. Of course you can look at any given object as if it were an event; you can describe it that way. But you should realize that when you do that, you are not saying what the object, for instance a table, really is, but only the way you are looking at it: you have simply changed the frame of reference.

The mistake is when you imagine, as Russell does, that an object, a table in this case, must really be an event -- indeed that there are no objects at all, only events. Which is the mistake of metaphysics: "trying to express by the use of language what ought to be embodied in the grammar". That is to say: Russell thinks that he is using our word 'event' in our normal way -- but what he tries to say with the word 'event' is not our normal way of using that word (where 'event' gets its meaning only by its connection to such words as 'object' and 'at rest'). So that if Russell is not to talk nonsense, he must state new "rules of grammar", that is, he must redefine the word 'event', which would then become a word belonging to Russell's jargon. But that of course was not what Russell wanted to do nor what he thought of himself as doing -- creating jargon.

Creating jargon was what Wittgenstein had done with the word 'grammar', although he recognized that he had done this and accepted it.

Norman Malcolm criticised the notion of "common sense" propositions, which is the notion that our normal way of expressing ourselves (our language) embodies the point of view of "common sense", as opposed to philosophical-metaphysical propositions which embody the point of view of reality. Russell would be, as it were, rejecting "common sense".

I look at a piece of paper in the half-light and see that it is white. But when I bring it into full-light I see that it is yellow. Now, which color is it really? The word 'really' asks for a standard of judgment or measurement, a criterion of correctness. But which color is the paper in itself? That question is meaningless -- and that tells us something important about the grammar of our word 'color'. It does not tell us, however, anything about "the essence of color, or, of colored objects" -- because that form of expression is undefined. Our examination of language reveals to us concepts -- i.e. the use of words [rules for using words], not metaphysical theories ["What is the really real?"] about reality. (Z § 223)

Is a shadow real? That is a good example of a metaphysical question, if the subject-matter of metaphysics is the "really real" versus the "only apparently real" (which some philosophers call "the view of common sense"). We do have normal uses for the words 'real' and 'apparent', but they have nothing to do with metaphysics' "ultimate reality".

Query: Wittgenstein, meaning of world.

When I first glanced at this I saw 'word' rather than 'world' ... and that is the difference between the TLP and the Philosophical Investigations: world becomes word. (Which was the reason for Russell's criticism or attack or antipathy towards the latter work.)

When Russell talks about God, it's clear that he isn't talking about God; what he's talking about is Providence. That is clearly what he means (CV p. 50). And he speaks contemptuously of "comfort" (Wood, p. 233). But Dostoyevsky's Kirillov did not object to the word 'comfort': "No, let it be comfort". We don't ridicule someone who puts a cushion on a chair to make it comfortable there, but an imaginary cushion would not make a person comfortable. The same is the case with belief in God; no one would want to believe in an imaginary God, if one believed it were imaginary (There would be no comfort in that). So what is the point of Russell's denigrating the search for comfort -- i.e. peace of mind. Russell says "The universe is unjust" (Wood, p. 236), but how does he know that? Because he treats it as an empirical question, a matter of experience rather than of "religious pictures" (pictures that do not state hypotheses, but are used in a very different way). Which has the result that Russell and the religious person are simply not talking about the same thing. Russell speaks of comfort, but Wittgenstein did not find the picture of a Last Judgment comforting.

Russell is an example of one way men respond to the question of God according to William James: by becoming atheists (in Russell's case, a strident atheist). But Russell had not been a mathematician who somehow wandered into philosophy (He was not like C.D. Broad). He had had an impulse towards the transcendent from childhood, if only as a particular mathematical picture of maths resembling Frege's "geometric heaven", an impulse he later ruthlessly rooted out in himself.

"They are only interested in the questions, not in the answers"

Discussing the fondness of Oxford philosophers at one time for investigating the "common usage" of words, he remarked that "To discuss endlessly what silly people mean when they say silly things may be amusing, but can hardly be important". (Wood, p. 234)

The Oxford professor J.L. Austin's How to Do Things with Words is not philosophy. What Austin does in his discussion of what he calls "performative utterances" is to describe a particular use of language that is a very different use from e.g. stating a fact. And Wittgenstein would call that "collecting curiosities" (PI § 415), because Austin had not undertaken his investigation of language in order to deal with a philosophical problem. Language only matters because philosophy matters, and it is in philosophy we are interested not in language per se. But Austin's work can hardly be described as being about "silly people saying silly things".

Anthony Quinton, who was at Oxford with Austin and Gilbert Ryle, suggested that philosophy begins in boredom, that people only exercise their intellects when they've nothing more interesting to do. Goethe did say that poetry begins in boredom. I don't know; I would not say that (and I don't think Goethe meant it either).

Russell had tried to separate his work in philosophy from his journalism. But it seems not to have worked out that way, as if journalism had corrupted his way of thinking, and polemic became more and more prominent as he grew older. Or maybe this Voltarian streak had always been there. In later years Wittgenstein told Moore that he had seen Russell, and about Russell Wittgenstein then said: "glib and superficial, as always" (quoted in Clark, p. 494).

And he satirized the attitude of some modern philosophers by telling the story of a shopkeeper whom he once asked the shortest way to Winchester:

"He called out to a man in the back premises:
""Gentleman wants to know the shortest way to Winchester."
""Winchester?" an unseen voice replied.
""Way to Winchester?"
""Shortest way?"

"He wanted," said Russell, "to get the nature of the question clear, but took no interest in answering it. This is exactly what modern philosophy does for the earnest seeker after truth." (Wood, ibid.)

But the man in the shop would not have said "Dunno", because 'I don't know' is an answer to the question: it is not an expression of uninterest: it is simply an acknowledgement, and not a boast, of ignorance. No, rather than "Dunno", the man would have said, "I don't know (and I don't care to know). The gentleman knows where he wants to go; isn't that enough for him? It's enough for me." Something like this.

But is philosophy necessarily that way? G.E. Moore's view was that "if only you clearly define your question, then its answer becomes obvious to you". Because philosophy is not about seeking facts (such as the "the shortest way to Winchester"); it is about clarifying our own thoughts. "Philosophical investigations -- conceptual investigations"; that is a later remark, but Russell could have found it already in TLP 4.112. And anyone who disagrees with that remark must demonstrate that it is false -- by producing a "philosophical proposition" that is not either grammar or nonsense. But there is no rabbit in the philosopher's hat.

Philosophy begins in wonder -- i.e. in being perplexed and wanting to be unperplexed. The Platonic Socrates does seek answers to his questions (which is what Russell says we should do in philosophy), but the answers that Plato's Socrates seeks exist only in the imagination of Plato, not in the way of thinking of the historical Socrates, who "did not make hypotheses". The answers to Plato's questions would not be statements of fact (unless they were statements of fact about the way we use language, in which case they would be a false account). They would not be the "real definitions" metaphysics seeks, but only conventional definitions (i.e. rules for using words). Plato did not, in my view of course, understand the logic our language and, arguably, neither did anyone else before the later Wittgenstein. But trying to clarify one's own thinking can hardly be called "endlessly discussing what silly people mean when they say silly things"; instead, it is what the Delphic oracle set Socrates to doing twenty-five hundred years ago: it is seeking wisdom.

He only asked philosophical questions because he genuinely wanted to know answers ... When, at one time, I thought of sub-titling this book "The Great Questioner", he pointed out that he had done something to answer questions too ... (Wood, p. 243)

But we call so many different kinds of things answers that the word 'answer' is not the most useful of tools. To show that there is no answer to a question, for instance, may be to answer that question. As is to show that a question has no clear sense -- however suggestive its combination of words may be -- and that it either must be given a sense or acknowledged to be nonsense.

... Russell himself once wrote that the value of philosophy lies largely in the questions themselves. (ibid.) "There is really more merit in raising a good problem than in solving it." (Letter of 1 June 1913, quoted by Clark, p. 205)

Site copyright © September 1998. Send Internet mail to Robert [Wesley] Angelo. Last revised: 24 June 2010 : 2010-06-24

The URL of this Web page:

Back to top of page

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