Home - Wittgenstein's Logic of Language | Bibliography | Site Search | Site Map

Ignored or Forgotten

In industry he [Chrysippus of Soli or Tarsus (about 282-206 B.C.)] surpassed everyone, as the list of his writings shows; for there are more than 705 of them. He increased their number by arguing repeatedly on the same subject, setting down anything that occurred to him, making many corrections and citing numerous authorities. (Diog. L. vii, 180, tr. Hicks)

I have the time and I will try to tell you the whole story, for nothing gives me more pleasure than to call Socrates to mind, whether talking about him myself or listening to someone else do so. (Phaedo 58d, tr. Grube)

Were it not that most writers say stupid things about Socrates, as well as about Wittgenstein's work in logic of language, and so I will talk to myself instead, for nothing gives me greater pleasure than sorting out my own stupid thoughts in order to "know thyself." And so know, stranger, that this was not written to develop your thoughts but my own (Phaedrus 276d). I think philosophy is this way.

A search for "the truth, which no man knows"

Working in philosophy is really more a working on oneself. On one's own interpretation [conception]. On one's way of seeing things. (CV p. 16 [(1998 rev. ed.)], remark from 1931)

Yet on the other hand, "When two go together, one sees before the other" (Iliad 10.224, tr. Jowett, quoted by Plato in Protagoras 348c-d). Of course that doesn't mean that the other will see it the same way (or indeed see it at all). That is why I study the work of the philosophers, certainly not for its own sake.

Temples dedicated to Truth, which no man knows, and to Reason, which never dried a tear. (Chateaubriand)

Topics on this page ...

Immanuel Kant and Gestalt shift: the conceived percepts are without foundation if they themselves are the foundation

Preliminary

The world "with which common sense and common sensation acquaint us" (Jastrow, "Preface", Fact and Fable in Psychology (1900), p. v). On the one hand, this seems a proposition no one will contest -- i.e. the contrast between the normal and law-abiding (as in "general laws of nature") and the abnormal and erratic. But, on the other hand, there are metaphysical presumptions in that proposition, as e.g. that there are common sensations, for what do I know about anyone else's sensations (sense perceptions, feelings)? and how do we determine whether what passes for sound judgment ("common sense") is sound?

Following Wittgenstein I think, I would say that: common sensation is agreement in the language we use (PI § 241). And common sense is the judgments of a "community of ideas" made using the concepts that are its current common currency (and for the most part have been in circulation for as long as any one knows, such e.g. as 'fear', 'sorrow', 'anger', as well as 'object' and 'space', 'man' and 'woman'; language use is a public institution).

But then what becomes of objectivity?

"Kant was the first philosopher to locate objectivity entirely in the mind," we were told by a professor who taught us about Kant and Fichte (and whose name I regret to say I cannot remember, but believe was Dennis Bradley). But if it is "only in the mind" -- and not a priori (i.e. verifiable independently of experience, as logic and maths are) -- then it is not objective, as we normally use the word 'objective'.

Distinguendum est inter et inter: we must distinguish between cases. "... to locate objectivity entirely in the mind." But that the mind imposes a frame of reference on man, as Kant says, is different from saying that any particular proposition within the framed-picture's truth or falsity is determined by the mind. No, but the "conceived facts" -- i.e. the concepts available to man -- may be, in Kant's metaphysics, entirely determined by the mind, in which case what can or cannot be true or false may be determined by the mind. Nonetheless, within those limits, there is objective truth and falsity.

"Entirely in the mind" -- that is, in the mind's "conceived percepts" -- i.e. percepts (sense impressions) with concepts and therefore not blind ("percepts without concepts are blind") -- and innate concepts (i.e. categories of thought); these two things determine objectivity.

But if, as Kant says, the thing in itself -- i.e. any thing independent of conceived percepts, is unknowable -- then all perception of change is like a perception of Gestalt shift.

What is the duck-rabbit in itself? But if someone does not have the concept 'rabbit', he sees no rabbit there -- and, indeed, is there a rabbit there independent of the concept 'rabbit' to be seen at all? (Compare the words 'drawing', 'figure', 'duck')

Can we say, The thing in itself only appears to change? But if there are only appearances -- i.e. if reality (the thing independent of conceived percepts) is unknowable -- then reality only seems to exist (PI § 258) -- i.e. that there is any such independent thing, is nothing more than a postulate (i.e. innate category of thought), which Fichte says we can set aside (the "speculative position" of idealism).

But then, what is 'objective' to mean -- if all perception is subjective (i.e. the work of the agent (subject), not of the object) -- because the word 'objective' is given meaning only when a proposition's truth is determined, not by the subject, but by an object independent of the subject?

The conceived percepts are without foundation if they themselves are the foundation of knowledge. "The objective has its foundation in the unobjective." Surely, but that logical proposition does not mean that objectivity belongs entirely to the subject rather than the object, for if it did, it would simply mean that the word 'objective' had been rendered obsolete, a discarded concept. (I am not seeing what is needed to make all this clear. Unless I did see that with the statement "Nonetheless, within those limits, there is objective truth and falsity" above.)

Rather than place this discussion in the context of the metaphysics of the mind, it might simply be said that human nature imposes some limits on our investigations (the five senses being an obvious example).

Logical construct = conceived percept = fact

Is calculating in the imagination in some sense less real than calculating on paper? It is real -- calculation-in-the-head. -- Is it like calculation on paper? -- I don't know whether to call it like. Is a bit of white paper with black lines on it like a human body? (PI § 364)

Wittgenstein's drawing of a circle-shaped faced with eyes, eyebrows, nose and mouth

Here it is useful to introduce the idea of a picture-object. For instance [Wittgenstein's drawing of circle-shaped face] would be a "picture-face". In some respects I stand towards it as I do towards a human face. I can study its expression, can react to it as to the expression of the human face. A child can talk to picture-men or picture-animals, can treat them as it treats dolls. (ibid. II, xi, p. 194)

Is a sum in the head less real than a sum on paper? -- Perhaps one is inclined to say some such thing; but one can get oneself to think the opposite as well by telling oneself: paper, ink, etc. are only logical constructions out of our sense-data. (ibid. § 366)

Logical construction = conceived percept = fact ... And of course I want to say: Well, but if that's all facts are! Why -- what do you want them to be? Reality as it really is? And what's that when it's at home? (Does reality exist? Does Bishop Berkeley reason that it does, or does not?) The proposition 'God sees what is really real' is a rule of grammar or nonsense (as is absolute point of view = the point of view of God). Is the proposition 'Reality is defined relative to an assigned point of reference' a rule of grammar? or metaphysics? or both -- i.e. is it a "tautology" that is not idle? (Can grammar be instructive?)

Further question: do "the conceived facts" belong to a community of ideas? in contrast to all mankind? in contrast to an individual? if we by 'a conceived fact' we mean 'a percept given light by a concept' (as if to amend Kant's "percepts without concepts are blind" with "are in darkness")?

[More off-target remarks: Goethe and Kant, 'theory' and 'conceived-fact'.]

"Like a universal Gestalt shift"

Vase-profiles Gestalt from Edgar Rubin (1886-1951) of Denmark

Learning Wittgenstein's later logic of language is not like learning a new language by translating everything back into the old. The terms of the new language do not correspond to the old language. The vocabulary and point of view are incommensurable: they are measuring rods, standards of measurement that cannot express one another, as e.g. "What shape is it?" in contrast to "What color is it?"

That is why I likened it to a Gestalt switch: the duck and the rabbit are independent aspects -- but a more apt example is the vase-profiles image, which has the aspects (1) vase and (2) profiles of two faces, because there is no question of both aspects having eyes or ears of mouths (although ducks do not have noticeable ears).

Again, contrast the vase-profiles Gestalt with the above-and-below table-glass, which is a table-glass regardless of which aspect is seen, as well as with the absolute image and its two aspects of the TLP's cube example, which remains a cube in either aspect.

Table-glass Gestalt

This represents an ordinary table-glass, -- the bottom of the glass and the entire rear side, except the upper portion, being seen through the transparent nearer side, and the rear apparently projecting above the front.

But it fluctuates in appearance between this and a view of the glass in which the bottom is seen directly, partly from underneath, the whole of the rear side is seen through the transparent front, and the front projects above the back. (Joseph Jastrow. Fact and Fable in Psychology (1900), "The Mind's Eye", p. 288)

Rather than simply that it fluctuates, Jastrow writes that the image fluctuates in appearance. Why -- isn't that the correct way to characterize this phenomenon? No, because 'appearance' contrasts with 'reality', and there is no defined test to determine whether which aspect of these drawings is really the image of -- in fact it is nonsense to say that it is really one or the other.

This is not a example of Plato's cave metaphor in Republic 515c: e.g. the rabbit and duck of the duck-rabbit are not mere shadows cast by an unknown or unknowable reality.

But contrast all those examples with the photographic Gestalt shift, where this is a reality: the photograph really is of a cat, not a rat, despite any appearance to the contrary.

A drawn real-and-illusion Gestalt

This drawing is really of a man rather than a piglet, even if you can only see the piglet ("aspect blindness"). In this way the drawing is like the rat-cat Gestalt photograph, which is really of a cat, not a rat.

Drawn protrait of 19th century man, 9 KB

The drawing can be seen now as a piglet's head, with a long snout and large dark ears in a very large collar, and now as the head of a man of the 19th century. It is an example of a real-and-illusion drawn-image Gestalt shift.

The man's hair is the piglet's right ear, the man's moustache the end of the piglet's long snout, the man's chin the piglet's mouth. The man's right ear lobe is the piglet's right eye, and the man's eyes are the piglet's left ear. (If you are having difficulty seeing the piglet, lean your head to the right and focus on the man's ear lobe and the man's neck after the shaded double-chin. The unshaded part of the neck is the piglet's neck. Try looking sideways or half-closing your eyes (squinting). The man is looking to his left; the piglet is looking to its right, its snout projecting out of the frame.)

In this particular case, although the image is a drawing, it is really one thing rather than another, namely the portrait of a man, not the drawing of a piglet dressed in a man's suit.

Not Commensurable, not Translatable

... he is learning a new language and translates everything back into the old. (Jastrow, ibid., "The Dreams of the Blind", p. 361)

We can't understand the Philosophical Investigations by recasting it in the old language -- i.e. the language of the Tractatus Logico-Philosophicus. Wittgenstein's Philosophical Investigations is a new language [sometimes redefinitions of old words, as e.g. 'grammar', sometimes inventions of new, as e.g. 'language-game'] not translatable into the old -- as is shown by the word 'nonsense', which is defined differently by those two works. The definition of 'nonsense' used by the Investigations has no equivalent in the TLP. "Whatever is not a picture of the actual or possible facts is nonsense", the TLP says, whereas the Philosophical Investigations defines 'nonsense' as "any word or combination of words that does not have a use in the language", an idea which cannot be restated in the TLP's language. The contrast between these works resembles the vase-profile, not the table-glass Gestalt.

Of course, one is a "theory of nonsense" (or rather of sense) whereas the other is a selected definition of a word (PI § 500). Nonetheless the definition cannot be fitted into the theory.

Gestalt shift is not Hegelian

There is a thesis and an anti-thesis, but in this case there is no synthesis [Hegel's example of his thesis of development: Parmenides is contradicted by Heraclitus, and their (1) Reality is un-changing and (2) Reality is ever-changing (3) are synthesized by the Atomists, whose Reality's atoms are un-changing, but the atoms' relationships are ever-changing]. Instead, in this case the anti-thesis entirely replaces the thesis.

The question I cannot answer, but would like to know the answer to, is what the antithesis (i.e. the new thesis that contradicts -- i.e. denies the correctness of -- the old) to Wittgenstein's later thesis (i.e. logic of language) will be. Or whether there will be an antithesis (an alternative logic of language, a different selected definition of 'meaning' for philosophy) -- because philosophers don't always reply directly to their predecessors.

And so on (philosophically)

To know is to be able to tell

Query: how do you know if you cannot define philosophy?

Because if you can "define philosophy", that is knowledge, and if you know something you can explain what you know to others; and if you can't explain it, then you don't know what you think you know. (To 'explain' means to 'put into words that can be tested by Socratic cross-questioning for their clarity and truth'.) For Socrates, not thinking you know what you don't know is akin to not being in the state of madness.

Does philosophy talk about things or about words?

Query: meaning of meaning by the Socrates.

For Socrates the meaning of word is the thing it names, which would be a particular thing if the word is a proper name, and if it is a common name, then the word's meaning is the common nature (essence) of the things it names. And if we cannot say what that thing is, then we don't know what we are talking about -- and if we talk about things we don't know, our talk is without meaning.

But Socrates' definition of 'meaning', and the standard it sets for knowing the meaning of a word, effectively puts an end to philosophical discourse (Parmenides 132a-135c), whereas, as Wittgenstein writes, in fact -- i.e. if we look and if we are willing to accept what is in plain view -- we see that few common names name common natures, and therefore that most common names are meaningless. However, if we use the word 'meaning' as we normally do, those common names are not "talk without meaning".

The question is then: is Socrates' logic of language a false account of our language's logic, or is Wittgenstein's account merely an alternative to it? Question: is the answer that Socrates is eccentrically assigning a meaning to the word 'meaning' (as Wittgenstein does to 'nonsense' in the TLP) -- or is Socrates stating an hypothesis that in most cases upon looking shows itself to be false?

I think Socrates, if he never presumed to know what he did not know (and had no reason to use language eccentrically), thought that the correctness of a method -- whether of the method of induction leading to generalization (common nature definition) or of the Socratic method of cross-questioning -- was to be judged by the tests of reason and experience to see whether it was clear and to see whether it was serviceable. (For Plato, by contrast, the proposition that the meaning of common name is the common nature it names was an axiom; in other words, rather than the result of an investigation (PI § 107), it was a requirement Plato imposed on language.)

Plato's picture of language: the meaning of a word is the thing the word names, and therefore the meaning of word is independent of the context in which it is used and thus philosophy can talk about things rather than about words (PI § 120). [Others, not Plato, thought we grasp the essence of a thing by mysterious abstraction, which even if we cannot tell others what that essence is -- and therefore Socrates would say we don't know what we are talking about -- our talk is not meaningless.] But, Wittgenstein said, as a public fact most words are not names but have a different use in our language than names have (ibid. § 43) -- and when the meaning of a word is the use that is made of it, its meaning is relative to the context in which it is used.

When Wittgenstein says that "For a large class of cases ..." (PI § 43) is he assigning a meaning to 'meaning', namely 'use in the language', or is he stating an hypothesis (i.e. a testable statement of fact) about how we normally use the word 'meaning'? Well, it is a report of what is public and therefore objective about language use. Objection: but 'The meaning of a word is its use in the language' would be only one meaning of the word 'meaning' (one thing we normally call 'meaning'). Reply: but it is one way we do actually use the word 'meaning' (and Wittgenstein says -- or, maybe, I say -- it is the way most useful to philosophy).

What is philosophy?

Query: wonder is believed to the source of philosophy.

As if this were an hypothesis about a natural-human phenomenon? Well, why not, for Wittgenstein wants "the view from outside" (CV p. 37)? Mightn't the anthropologist say, when studying some strange tribe, "It appears these people ask these questions because they are perplexed and wanting to know what they are ignorant of" (The tribe of philosophers).

Query: contradictions in defining philosophy.

If man's only wisdom is to know that he is not wise (Plato, Apology 23a-b), then how can philosophy be the pursuit of wisdom? If metaphysics seeks knowledge of reality, but can only speculate (speculation is not knowledge) not prove, then how can metaphysics be "love of wisdom" rather than "love of speculation"? Things like this maybe.

Query: the concepts of philosophy. The tools of philosophy, language and logic.

Philosophy's conceptual tools: not only logic (artless reasoning), but language as well (language supplies concepts, categories, e.g.): not only new concepts (as e.g. Wittgenstein's revision of the concept 'grammar'), but also new categories which are used to make distinctions: being poor in distinctions = being limited in philosophical understanding. (Thus the words distinguendum est inter et inter: we must distinguish between and between -- i.e. note the differences, not only the similarities, between cases. If A is only similar to B in some ways, then A must also be different from B in some ways.)

Voluntarily seeing-as versus seeing involuntarily

We can imagine this illustration "appearing in several places in ... a text-book" where "something different is in question every time: [1] here a glass cube, [2] there an inverted open box, [3] there a wire frame of that shape, [4] there three boards forming a solid angle. Each time the text supplies the interpretation of the illustration." (PI II, xi, p. 193)

Text-book illustration given interpretations

I can see the schematic cube as a box; -- but can I also see it now as a paper, now as a tin, box? (ibid. p. 208)

The question is: what is 'can' to mean here? How is the tin and paper example different from the glass cube, etc. example?

When might the text supply that particular interpretation? It might be concerned with what conditions and at what speed a paper in contrast to a tin box of that shape would collapse.

RPP i 70 duck-rabbit Gestalt

But seeing what "the text supplies [as] the interpretation of the illustration" is not seeing (i.e. experiencing) a visual snap. And that is what interests me here, that those are not examples of what I have called 'Gestalt shift': because although there is an absolute, namely the schematic-box, there is no visual snap between aspects as, for comparison, there is to seeing the duck-rabbit (Philosophical Investigations) as now a rabbit, now a duck, or the labeled-cube (TLP) as now convex, now concave.

"Only in the mind"

Suppose we said, "In a Gestalt shift, the change is not in the thing [absolute figure], but in the mind." Maybe very well, but the word 'mind' is not a place name: "the mind" is not a location. Indeed all that saying "The change takes place in the mind rather than in the thing" amounts to is contrasting: the thing changes with the thing does not change, but our perception of it does ('perception' = 'seeing', may be either visual change or change in understanding; 'perception' here is also equivalent to 'mind'). The words 'in the mind' -- a verbal formula that contributes more obscurity than insight (It might be good to discard the word 'mind' altogether).

Now, what would Bishop Berkeley say? If to be is to be perceived, does all perceived change happen in the mind: is all perception Gestalt shift? "Reality is the duck-rabbit, as a conceived fact/percept."

Normally if we say 'It's only in your mind', we continue with 'not in reality'. It's only in your mind, not in independent reality. Is a Gestalt shift then only "imaginary"? That's not the way we normally use the word 'imaginary'. (The point isn't the specific words we use here, but instead the explanation we give of their meaning.) It is nonsense to say 'the object does not change' unless it is also possible to say 'the object changes' -- i.e. if there is no object independent of the perceiver. Yes, but that is only a rule of grammar, not a theory of reality. How do we normally use the expression 'the object has changed'? That question is defeated by the picture: "But metaphysics questions at a deeper level."

And if there is no essence?

I can set a limit to the concept here. (PI II, xi, p. 208)

And this limit will be set more or less arbitrarily, because we don't in practice use the expression 'Gestalt shift' in any but a very broad way. "What is a Gestalt shift?" That question is defeated by concept fluidity, i.e. the vague limits of the concept (i.e. the absence of strict rules = the absence of an essence).

Which is one reason essential definitions are wanted, because how are we to think sharply with unsharp tools? But wanting sharpness, and there being sharpness in practice, aren't the same thing. And we would have to make rules to set sharp borders ourselves, were that what we really wanted -- but it isn't: what we want is for our concepts in practice to be the way they are not.

Not to use the word 'see' equivocally

If we say that someone, looking at the duck-rabbit figure, sees now a rabbit, now a duck, that is the word 'see' in its visual sense. But if we say that someone sees that the astronomical data can be rearranged -- is that 'see' in the visual sense? "Copernicus sees another way to arrange the sun, the moon, the earth and planets relative to one another." Suppose we imagine Copernicus rearranging discs of paper, moving the sun to the center of concentric rings and the earth to one of the rings (The geocentric map is much more complicated than that, incorporating e.g. the epicycles of Mars), that would a visual shift in position, but it is not a visual snap or jump (which is part of what I am using to define 'Gestalt shift').

Kant and Copernicus

... just what Copernicus did in attempting to explain the celestial movements. When he found that he could make no progress by assuming that all the heavenly bodies revolved around the spectator, he reversed the process, and tried the experiment of assuming that the spectator revolved, while the stars remained at rest. (Kant, Critique of Pure Reason, Preface to the 2nd Edition, tr. Meiklejohn)

What does Kant (or his translator) mean by 'progress' here: progress towards what -- towards finding the simplest explanation, i.e. model? In the geocentric model, the earth is [chosen to be] the fixed point of the system, whereas in the heliocentric model, the sun is [chosen to be] the fixed point. All motion is then described relative to the designated fixed point. That was the metaphorical "shift" Copernicus made. (Tycho Brahe's celestial model is a variation of the earth as fixed point of the system.)

Kant said that he would follow Copernicus' shift in point of view, by focusing (i.e. designating as the fixed point), not on the nature of the object ("the thing in itself"), but on the subject ("the human mind"): not on what is perceived but on the perceiver. (As if to ask, I'd say, What are the qualities of the telescope rather than the qualities of the stars? e.g. the magnification and curvature of the lens.)

Neither Copernicus nor Kant is the visual snap or jump I mean by 'Gestalt shift'; instead they are only Gestalt shifts metaphorically. Otherwise basically I think the word 'see' is being used equivocally here. 'Looking at the outlined cube, Copernicus sees that the outline can be based on face b b b b rather than on face a a a a.' What we can't say is that the outline cube corresponds to the sky we see when we look up. Or can we? Can't Copernicus, when he looks at the sky, see the earth going round the sun rather than the sun going round the earth? But what can he point to in order to indicate the difference? With the outline cube we can use labels, and with the duck-rabbit we can draw arrows to indicate e.g. the duck's bill or the rabbit's mouth.

"Concept-blindness" and "aspect-blindness"

We can speak of "aspect-blindness" or "Gestalt-blindness" -- i.e. being unable to see an aspect of a Gestalt -- as a subset of "concept-blindness". Wittgenstein asks what someone should be called who "cannot see how a reasonable man may use [the word 'God'] seriously", whether this person is blind to something, as we can say that someone who cannot see the rabbit-aspect is blind to something. But when Wittgenstein asks about the word 'God', he is not using the word 'see' in its visual sense but in its sense of understanding: 'I see' = 'I understand'. And that seems to be the way Kant and Copernicus would be using it, too.

"Tell me, was anything ever done?"

Let me wipe it first; it smells of mortality. (King Lear iv, 6)

I don't think the king's torso-less legs in the desert should be regarded as consoling to the futility of my own thinking or the modesty of my unlived life, that both kind hearts and Ozymandias lie in graveyards soon enough forgotten.

We never go forward in philosophy (knowledge of life's meaning), just to some other place.

Site copyright © September 1998. Send Internet correspondence to Robert Wesley Angelo with corrections or criticism. Last revised: 1 October 2019 : 2019-10-01 (Original version 9 July 2018)

The URL of this Web page:
https://www.roangelo.net/logwitt/ignored-or-forgotten.html

Back to top of page

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