Part II of the Investigations has a different format than part I. The remarks in part II are not numbered and groups of remarks are collected in chapters, that is, a set of remarks that seem to be more or less connected. Speculation as to the reason for the difference and relationship between the two parts is a topic for a different work. For our purposes, as I mentioned above, part II appears to continue and amplify the discussion on psychological concepts in the last section of part I. Since the sections or “chapters” seem to form a unit, we should consider them individually.
Let us look at the opening text of section i:
One can imagine an animal angry, frightened, unhappy, happy, startled. But hopeful? And why not? A dog believes his master is at the door. But can he also believe his master will come the day after tomorrow?—And what can he not do here?—How do I do it?—How am I supposed to answer this? Can only those hope who can talk? Only those who have mastered the use of a language. That is to say, the phenomena of hope are modes of this complicated form of life. (If a concept refers to human handwriting, it has no application to beings that do not write.) “Grief” describes a pattern which recurs, with different variations in the weave of our life. If a man’s bodily expression of sorrow and of joy alternated, say with the ticking of a clock, here we should have the characteristic formation of the pattern of sorrow or the pattern of joy.1
Here we see many of the themes from the end of part I. Considering the question of whether an animal can hope reestablishes one of the key points from part I: hoping implies a complicated form of life. There are characteristics and activities that are essential to hoping. We want to lodge hope in the inner life of a person or an animal (what faculty or interior apparatus is the animal missing?) and so we are stuck with a number of conundrums (What apparatus is it that I have? How does it work and how can I find this out?). In many instances, this results in a continual skepticism as to whether someone or something has these emotions. We feel we should have to look inside the soul or nervous system to establish the truth here. Since this proves impossible, we are left wondering, not just about animals, but human beings as well. Our emotions become mysterious phenomena.
To counter this tendency we must look to the grammar of hope—or what Wittgenstein calls this complicated pattern of activities and circumstances that make up this form of life. This may make it seem as if we are ignoring the psychological—the mental—and only talking about a word. It may seem as if we are trying to reduce the psychological to language, and if language is merely a series of signs than connecting the psychological to language appears ridiculous. But I think we must take it as by now established that language is part of our lives—it is an instrument with a use. We must note that we do something with our words. We can think of, e.g., complaining as a state of mind, but it is far more complicated than that. Think of teaching the word complain to someone. We would probably have to explain many factors in order to teach someone this word. For example, we would probably use a typical situation of complaining: what led up to the complaint—we bought something and it didn’t work—then we would explain what we do—we write a letter or call customer service or tell others not to buy the product and so on. There is a whole pattern of essential activities here, without which it makes no sense to talk of complaining and would be necessary in order to complain. It is the same with hope.
Wittgenstein wants us to realize that the question “What is hope?” is not a psychological but a grammatical question. It is about the use of the word hope. For Wittgenstein an actual psychological investigation would be an empirical or scientific one—perhaps something trying to uncover the cause of someone’s neurosis. To understand the essence of the psychological, e.g., hope, we must see what the concept entails. Hoping includes a range and pattern of activities and circumstances.
We often look to explain psychological concepts such as grief or hope in terms of sensations—not like seeing or hearing, but as a type of feeling. But let us, for example, compare grief and hope to similar sensations. Grief is often associated with pain in a certain sense. We very often talk about the pain and grief associated with a loss. Of course pain is a sensation and to begin our analysis let us note that it makes sense to say someone felt pain for a second. But what about grief? Here this doesn’t seem to work—momentary grief doesn’t seem to make any sense. But if pain and grief are similar sensations, then what is the problem? Is grief perhaps so fine a sensation or too slippery that we can’t get hold of it? Or perhaps our words are lacking—our psychological vocabulary is inadequate. But if this were the case why don’t we just invent new words? Or fine-tune our senses by introspection.
Again, these strategies are doomed to failure because the concept is wrong to begin with. Grief cannot be reduced grammatically simply to a sensation. Rather, it is a complicated form of life. That there is a form of life to “pain” mostly escapes our notice because pain can be such an overwhelming and omnipresent sensation that we forget the importance of a pain behavior and so on for the concept pain. But, as we saw above, Wittgenstein was very careful to help us see past this limited view of the concept pain. Here in order to isolate the sensation of grief we might wish to “point” to the sensation: “Aren’t you feeling grief now?” But of course the temporal question—”how are you feeling? “—does not answer the question as to what grief is—what grief means. As we have seen, the response “What I am feeling now is grief” could not settle the question as to the nature of grief. This is as useless as a botanist claiming, “What I think counts for the species ‘oak tree’ determines what an oak tree is.” We would instantly reject unverifiable introspection as a model for scientific method in botany, physics, and chemistry. However, we seem to allow it when it comes to human nature, psychology, and so on. We cannot “get hold of” the sensation of grief because we have got our concepts mixed up. We are incorrectly modeling our concept of the supposed sensation of grief on the concept for the sensation of pain. Our failure to speak about the “sensation” of grief is not because of the inadequacy of our psychological vocabulary. Rather, it is a result of trying to force psychological language into an incorrect mold.
In section ii p. 149e, Wittgenstein continues to distinguish what is essential from what is inessential in meaning, focusing on the difficulties created by using models of meaning that incorporate ideas such as introspection, epistemology, and psychology. Here he amplifies an earlier discussion, the theme of which we might call “experiencing the meaning of a word.” Again, the idea under scrutiny is that there is something that I can locate in my mind that occurs as I “mean” something—and this is what counts as meaning. In other words, I may have many experiences as I say a word, and if I want to use a psychological or epistemological model of meaning, I might tend to look in this direction for the psychological or mental “content” of meaning.
First, Wittgenstein wants us to note that a combination of words has a different sense than each of the words individually. Where this sense comes from, if we look outside the use, can seem like a bit of a mystery. It is tempting to think that when I say a sentence with a particular expression that it is the feeling behind the expression that imparts the meaning—that this is the psychological content of meaning. Wittgenstein will return to this topic, but here we can say that while the expression behind a sentence is important, it is not a separable atmosphere that accounts for meaning. Let us look at the following text:
Experiencing a meaning and experiencing a mental image. “In both cases,” we should like to say, “we are experiencing something, only something different. A different content is—proffered—is present—to consciousness.”—What is the content of the experience of imagining? The answer is a picture, or a description. And what is the content of the experience of meaning? I don’t know what I am supposed to say to this.—If there is any sense in the above remark, it is that the two concepts are related like those of “red” and “blue”; and that is wrong.2
In other words, there is a use for “the experience of a mental image.” You could reproduce a mental image by drawing it or describing it. But now we want to “transfer” this language to the “experience” of meaning. This transfer seems possible because we think experience always means the same thing regardless of the context. But this assumes way too much. We would likely have a difficulty “putting our finger on” the experience of the meaning of a word. It may seem as if we have a problem with the faculty of experience—as if this faculty needs to be adjusted somehow. But the problem is, in fact, grammatical. To illustrate this, think of finding a use for “the experience of the meaning of a word.” And now what would that be? No doubt when we say a word we may find that it is familiar or expressive, and we might have a particular feeling when we say a word, as when we tell someone that we love them. There may be many accompaniments for meaning. But why should these feelings be the meaning or even an experience of the meaning? If these feelings constitute an experience of meaning it should have something to do with meaning. One should be able to produce something verifiable to connect these feelings with what is meant. If I try to give this kind of content to this supposed experience of meaning, I might describe this experience in a variety of ways, but if I try to compare this experience to the experience of a mental image the comparison breaks down. We will find that the “experience of a mental image” and “the experience of meaning a word” mean very different things. I could draw a mental image on paper and this would represent my experience—I could say to someone “This is what I was thinking about”—or I could describe my image. Now consider what I would do if I wanted to describe the experience I had when I heard a word or phrase. Here I could tell someone the word or explain the meaning of the word and say, “I find this particular word very exciting,” or whatnot. But what parts would the feeling and so forth play in explaining the meaning of the word here? What sense does “experiencing the meaning of a word” have? Again, saying I had a particular feeling as I pronounce a word does little to explain the use of a word—does little to explain what I mean by it. Hence this idea runs aground. I might have a particular experience when I say a word, but it is difficult to make sense out of the idea that this is an experience of the meaning. As we shall see, there are conflicting ideas of meaning at work here.
So, we may say that the sense of a sentence does not depend on the feelings, etc., that accompany it. True, the sense of a sentence is more than the sum of its parts, but the sense does not accompany a sentence. Rather, it is found in use.
Turning to the text in section iii, we see that Wittgenstein focuses on a similar idea. But instead of examining the idea that we can experience the meaning of a word, which leads to the incorrect notion that we can (or must) introspectively ascertain meaning, here Wittgenstein briefly looks at the notion that an image is the same as an idea.
Let us look at the following text from section iii:
What makes my image of him into an image of him? Not its looking like him. The same question applies to the expression “I see him now vividly before me” as to the image. What makes this utterance into an utterance about him? Nothing in it or simultaneous with it (“behind it”). If you want to know whom he meant, ask him.3
The opening two sentences of the text perhaps at first seem paradoxical because it is precisely resemblance that we think is at the base of x being an image of y. Something is an image because it looks like that which it represents. The key idea here, though, is what is it that makes this image my image. In other words, if I imagine someone what do I need to have or do in order to determine that I have imagined that person? Clearly nothing further is required to make this image mine—nothing behind or betwixt or between the image makes it mine. Wittgenstein is again focusing on those tendencies that lead us into inventing psychological processes where none are required. I think of a friend or my mother or my child. Does this only become my image of that person if it meets certain criteria or passes certain tests? I am not here referring to the phenomenon of trying to remember, e.g., an address, phone number, or a song lyric—that is, of trying to remember something correctly. Rather, we want to test the idea that a mental image needs something else to identify it as one’s own. If it were the case that some external or extra factor is needed for me to identify an image as my own then, as strange as it may sound, even the memory mistakes would be impossible to identify as yours. For what would identify the mistake as your mistake—some other external factor? And how would you know whether this was correct? Clearly we become involved in an infinite regress.
It would be strange to say, “First I think of something, then I have to find out whether I thought it.” If this were the case then of course we would be open to the most horrendous skepticism. Again we would never know what we thought because we would first have a thought, then need another thought to check the first, and another thought to check the second, and so on into an infinite number of thoughts with nothing ever being decided. But this can be avoided by noting that “image” here does not depend on resemblance. The meaning of the phrase “I am imagining so and so” cannot be subjected to the same critique an art professor would make of a portrait: “That’s not a very good likeness.” We cannot say here “Well, draw your image and we will see whether you have imagined so and so.”
Thus a picture cannot substitute for the idea here. Resemblance, or whatever, in an image of so and so cannot make the image into my image. The mistake is thinking the picture on its own must accomplish something—that the image in the mind carries something with it that makes it my image of x or my representation of x. Thinking this way makes my images, thoughts, and so on a source of skepticism, for I am now in the awkward position of having to justify the fact that this is my image. Paradoxically, with this model the image is unimportant and my imagining something is insufficient to make sense out of “I imagine x.” For I must “get at” what “lies behind” the image—whatever it is that justifies it as an image. I have made this psychological object—whatever it is—the determining factor in this image being mine.
We can see that so much of what Wittgenstein has said before comes into play here. When I look for something adjacent to the image that identifies it as mine or that causes it to be an image, I am inventing a “myth of meaning” for the word image and bypassing the actual meaning of image. We think that by introspection we can see what identifies this as my image of x. We then look in vain for that which extends meaning to the image, and then begin to wrap the image in theoretical entities.
Clearly when I say that I just thought of x, all this “checking” did not occur—I did not identify an image and now I say, “I really thought of it.”
Such a model literally cuts us off from ourselves. It makes us think that our thoughts or our imagination require criteria of identity. It seems as if what thoughts are mine depends on some as yet undiscovered other factors—although it is unclear who or what gets to determine which characteristics count and which don’t. Presumably only the psychologist gets to tell which of my thoughts are genuinely mine and which come from “elsewhere.” This model also cuts us from others, as Wittgenstein indicates in section iv:
“I believe that he is suffering.”—Do I also believe that he isn’t an automaton? it would go against the grain to use the word in both connexions. Suppose I say of a friend: “He isn’t an automaton.”—What information is conveyed by this, and to whom would it be information? To a human being who meets him in ordinary circumstances? What information could it give him?4
To say of someone, “I believe he is suffering” might be a statement about his mental state when perhaps the person’s depression or anguish is not obvious. There are circumstances where this would make sense. Perhaps at a funeral a person is putting on a good face, but inside he is feeling deep grief. Or perhaps you read an account of some tragedy in the paper, and you imagine that the people involved must feel awful. So a statement of this type might be considered an assumption or a guess about someone’s feelings or emotions at a particular time. We would likely go up to the person at the funeral and say, “You must feel terrible—is there anything I can do?”
But if I say “I believe that my friend is not a robot,” belief cannot mean the same thing. Clearly the circumstances for the question are not ordinarily present. We are not “of the opinion” that other people are human beings. What would make us doubt it? Under what circumstances would rendering an opinion on this matter be required?
Of course the planet could be taken over by aliens who are replacing people with robots. Then I suppose the question might make sense. Or we might be doing philosophy. If I think my statements about another person are of a nature that they always require justification, that there is something behind my words that I must perceive so that they are meaningful, then the result is a skepticism about “other minds,” other’s interior states, and so on. Because there is no limit or end to the need for justification here I can never be sure that I am right. What would tell me that I have perceived the correct something that lurks behind my words? The above model does not tell us. It is only if we accept the above model that the statement “I believe he is a human being” makes sense.
Let us suppose, as in the model above, that my ideas about someone are images or pictures and that which makes these images meaningful is something “behind” these images. The same problem we saw previously occurs: that which I want to know or speak about becomes irrelevant—it is logically excluded from my language. It is not the case that I simply think about or relate to the individual—”Here comes my friend. He looks depressed. I’ll ask him what’s the matter.” If the above model is correct then this makes no sense. The meaning of my words is seen as something behind or surrounding my mental images. According to this model, the feelings, or whatever, behind my thoughts make them meaningful. Hence, really, what the person is feeling is not important, but my feelings are. I must identify my own feelings or “inner whatevers” to see if he’s depressed. He could be sitting alone crying his eyes out and I have to search myself introspectively in order to determine whether he is depressed. How would I ever know anything about him on this model? To counter this tendency we must note that the phrase “He is depressed” says nothing about the observer. I don’t need to ascertain anything about myself to use the word depression correctly. Likewise, as we have seen, depression does not solely refer to something going on “within” someone. When I note someone is depressed, this is not the result of peering into his or her mind or central nervous system. Here again, if we adopt this approach to understanding the psychological, skepticism would be the result.
However, as Wittgenstein says in section v, he is not completely denying the “inner.” Psychology is not reduced simply to treating human behavior.
Then psychology treats of behavior, not of the mind? What do psychologists record?—What do they observe? Isn’t it the behavior of human beings, in particular their utterances? But these are not about behavior.”I noticed he was out of humor.” Is this a report about his behavior or his state of mind? (“The sky looks threatening”: is this about the present or the future?) Both; not side-by-side, however, but about the one via the other.5
In other words, our psychological language, “He is depressed,” etc., is not about behavior and so neither should our grammar, our description of the use of these words, try to force our psychological language into this mold. When I say that someone is depressed, I am neither referring solely to an inner state nor solely to particular actions and circumstances. But as we have seen it is only within the context of certain characteristic activities and circumstances that our language of psychology operates. To put it in a philosophical context, neither the Cartesian or behaviorist paradigms by themselves provide a correct context for psychological language.
In section vi on page 155, Wittgenstein returns to the notion that meaning is that which accompanies a word. Again, such an idea often causes confusion regarding our psychological language because we wish, understandably, until we consider the issue, to associate some mental event or occurrence with our psychological vocabulary. But, most often when we are doing philosophy, what we choose as a “mental event”—the feelings and so forth that accompany our psychological language—is mistakenly substituted as a referent for the meaning of a word. What is required instead is understanding the actual employment of the word. In section vi Wittgenstein says:
How should we counter someone who told us that with him understanding was an inner process?—How should we counter him if he said that with him knowing how to play chess was an inner process?—We should say that when we want to know whether he can play chess we aren’t interested in anything that goes on inside him.—And if he replies that this is in fact just what we are interested in, that is, we are interested in whether he can play chess—then we shall have to draw his attention to the criteria which would demonstrate his capacity, and on the other hand to the criteria for the “inner states.”
Even if someone had a particular capacity only when, and only as long as, he had a particular feeling, the feeling would not be the capacity.6
This passage gives a good overview of the material in part II. As we have seen, we might want to attach our psychological language to a mental process. The idea of a psychological process as a referent for our psychological vocabulary can be quite attractive. Here we perhaps model the psychological on the mechanical. We watch a machine, like an engine, at work and there are a number of steps or operations that the engine goes through in order to accomplish its result. The same might be said of a computer or a calculator. We think of the steps that the computer program takes and we see a clear analogy to our brains. Hence we assume thinking or understanding must be an analogous process. However, the comparison can be misleading on a number of levels. A mechanical view of thought can make it seem as if thinking is something that goes on in us quite apart from anything else—like a computer program running, only one that we don’t see or know anything about. Thus thinking becomes something “hidden” or “behind” writing, speaking, doing a calculation, or playing chess. Thinking in this model is pictured as a brain or thought process running concurrently with these activities. But when we want to know whether someone can solve a math problem, speak French, or play chess we aren’t interested in his inner feelings, experiences, processes, or what have you. If someone applied for a job as an accountant, would it be sufficient for him to say—”Well, I have that ‘I can balance the books feeling’ quite regularly”? Suppose your surgeon said, “I never studied surgery, but I feel as if I did.”
As established, Wittgenstein clearly acknowledges that words can have a certain feeling attached to them just as a piece of music played in a certain way can give us a particular feeling. But this feeling cannot be separated from the music. It is necessary that you hear or play the musical phrase to get this feeling. This is not an “atmosphere” or an “aura” that accompanies the music; rather, it is how the music affects us. We need not appeal to something “accompanying” the music to explain its pleasing us, making us sad, and so on. So to, the fact that words move us or stimulate us need not be “caused” by something surrounding the word. Why appeal once again to this mysterious and unseen atmosphere, third party, or psychological middle man? Why does the feeling that a word gives have to be mediated by something “surrounding” the word? Again this is probably because we need something to feed or bolster the idea of the mental process—or the mechanical picture of thinking, understanding, and so on. Whether and what is happening in the brain when one thinks or understands is an empirical question. For Wittgenstein, the philosophical issue is that this model for our concepts cooks up a misleading picture of what is going on here. If we think our psychological language is meaningful because of some process going on behind it, and then we attempt an empirical investigation into this process, we are only chasing a linguistic dream.
We can create many captivating pictures with our language—many such dreams (and nightmares). Consider this text from section vii: p. 157e.
The evolution of the higher animals and of man, and the awakening of consciousness at a particular level. The picture is something like this: Though the ether is filled with vibrations the world is dark. But one day man opens his seeing eye, and there is light.
What this language describes is a picture. What is to be done with the picture, how it is to be used, is still obscure. Quite clearly, however, it must be explored if we want to understand the sense of what we are saying. But the picture seems to spare us this work: it already points to a particular use. This is how it takes us in.7
The above picture is powerful because of the importance of the topic and the picture’s generality; it seems scientific and explanatory in a very fundamental way. We seem to have captured very general facts of nature. This picture seems to be the encapsulation of the theory of evolution and the creation of human consciousness. The passage even echoes Genesis, but puts a more modern and therefore more “acceptable” spin on, “Let there be light.” In this way the picture may become impervious to challenge and actually seeps into our everyday parlance—our everyday way of looking at human beings and our place in the world. The picture is not the gateway to investigation but seems to answer all our questions or, at least, to dictate the path of future inquiry.
But I think Wittgenstein would say that the investigation is only just beginning. We must establish a use for the picture. What does it describe? What does it teach us? How is this picture to be applied, and what fruit will it bear? The picture may in fact say nothing at all and have no use whatsoever. But even so I think it says a great deal about us. Why do we find this idea so appealing, and why are we so inclined to let this picture dictate our future actions, beliefs, and perhaps even our science?
We will return to these questions in the last chapter. For now I think we should note that the above picture has the character of a theory—a foundational statement about, say, human consciousness. However, it remains uninvestigated and so far useless. For a variety of reasons it is simply “received” and accepted. It is this that Wittgenstein wants to guard against.
Empiricism traditionally had been one of those grand unifying theories in philosophy. In empiricism ideas are supposed to be derived from sense experience, sensations, or sense-impressions. In the next few sections (with the exception of section x, which we will treat separately) and particularly in section xi, Wittgenstein seems to be focusing on some of the central claims of empiricism. This is not to say he has never raised this issue previously, but here he seems to be taking the discussion to a deeper level—particularly in his “seeing-as” discussion in section xi.
If there is a relationship between sense experience and ideas, then Wittgenstein would want to explore the relationship between the grammars of, e.g. know and sensation. As we have seen it may not be obvious that our inquiry is or should be grammatical. But again it is not the case, generally speaking, that sensation or knowledge gives us any great problem. We know this, but not that. We feel hot or cold or tired or we see colors and feel textures and so on. It is the concepts that give us trouble. It is when we consider our ideas of knowledge and sensation that our manifold difficulties arise. Most often, the relationship between sensation and knowledge confuses us, and so it is the logic or use of these concepts that must be explored. To accomplish this Wittgenstein begins with an example. In section viii on p. 158e8 Wittgenstein asks us to consider a very slight movement of your finger and then investigate the phrase: “My kinesthetic sensations advise me of the movement and position of my limbs.”
Here Wittgenstein wants us to imagine very slight movements in one of our fingers and the feeling we get during these movements. We feel these movements, and it is tempting to say that these movements tell us how our fingers are moving and so cause us to know, say, the position of our fingers. This might be considered a picture of the relationship between sensation and knowledge. Again, the picture might be one of a mechanical connection between sensations or the nerves in the finger and the brain. Thus we would seem to have an empirical basis for the language of sensation as the empiricist sees it—a direct connection between knowing and seeing. But as we will see, the logic or the use of these words is not found in such a picture—hence the picture leads us astray. Let us look at this text from section viii.
“But after all, you must feel it, otherwise you wouldn’t know (without looking) how your finger was moving.” But “knowing” it only means being able to describe it.—I may be able to tell the direction from which a sound comes only because it affects one ear more strongly than the other, but I don’t feel this in my ears; yet it has its effect: I know the direction from which the sound comes; for instance, I look in that direction.9
While it is true that I certainly feel a sensation in my finger, to say that I know how my finger is moving means that I can describe its movements. If someone asked, “In what direction is your finger moving?” it would be useless here to say simply “I have a feeling in my finger.” This statement could mean a variety of things, but it certainly doesn’t tell us what we know about the motion of the finger—direction, rate, etc. This example illustrates the logical or grammatical disconnect between the question “what do you know?” and the statement “I have a feeling.” To the question of the direction of the motion of my finger a description of the motion would be an answer. Here again this shows us something about the concept know. The upshot is that know and sense are very different concepts and so have different uses. We cannot simply blindly connect them.
Just as the relationship between our “external” sensations, such as touch, and our ideas of these sensations can be confusing, so too, our “inner” sensations, such as our emotions, can cause similar difficulties when we come to our ideas or what we know of these feelings. As we noted previously when we discussed pain, it would not be correct to say “I know my pains” as if it is through a mental awareness that we experience pain. I do not know my pains—I simply have them. In a similar way, as Wittgenstein notes in section ix, it would be incorrect to say grief is experienced through observation or through a particular inner sense. Again, it would be difficult to make sense of someone saying “I am observing my grief”—except to mean simply that you are grieving. What use could we make of the phrase? Through what sense would we be observing our grief? Is the grief that you observe somehow altered when you are observing it? Grief clearly does not mean “something that I have observed in myself.” I can see that someone is in mourning—but how would I see it myself? As with pain, grief is something that is felt, talked about, expressed, or observed in the behavior of others.
Again, in general, the problem is thinking that there is a reference for our psychological vocabulary, when really there are different language games and circumstances, different contexts that delineate our psychological language. There is no single, overriding logical structure to our psychological language. Rather, the logic of these concepts is more involved. If I wanted to explain fear to someone I might act it out—but this would be difficult for hope or belief. Thus our attempt to cover the logic of our psychological terms with a blanket category such as “mental state” is bound to cause confusion. Here we can say that just as we must pay attention to the language of our sensation, so too we must understand the language of our emotions. Our emotions as we experience them are not something “inner” to which we are trying to reach with our ideas or faculties of knowing. Rather, we feel our emotions and our concepts of our emotions and our sensations belong to a variety of language games and contexts. They cannot simply be reduced to the concept “mental state.”
Section x focuses on the idea of belief. However, belief seems to be discussed in a way that seems to be inconsistent with the previous sections on psychology, sensation, and so forth. But as we will see the inconsistency is only apparent because Wittgenstein is still discussing the logic of psychological concepts.
The topic of the section is “Moore’s paradox,” which Wittgenstein thought was very important and overlooked.10 This topic actually requires a deeper and far more extensive discussion that would take us beyond the bounds of the present volume. This idea is the subject of Wittgenstein’s On Certainty, which occupied him for the last year and a half of his life. What we will do here is to discuss a rough sketch of the general problem generated by Moore’s paradox and point the reader toward Wittgenstein’s general solution.
Moore’s paradox is that someone can say “I suppose or speculate that something is the case, but I don’t believe it,” but I cannot assert that something is the case and I don’t believe it. Both sentences seem to be about the same thing, but the first sentence is okay and the second sentence makes no sense. In other words, we can imagine that the earth is round and that someone doesn’t believe it. Or we can frame the sentence as a hypothesis “Let us just say: The earth is round but I don’t believe it.” This makes sense. But now take the same sentence as an assertion that someone makes—”The earth is round, but I don’t believe it.” This makes no sense. The person would be declaring something to be a fact, something that he knows to be true, and in the same sentence declaring that he can’t accept it as true. Moore wanted to outlaw this on psychological grounds—but for Wittgenstein the assertion violates the logic of assertion.
A letter Wittgenstein wrote to Moore in 1944 explains his position succinctly:
It seems to me that the most important point was the absurdity of the assertion, “There is a fire in the room and I don’t believe there is.” To call this, as I think you did, “an absurdity for psychological reasons seems to me wrong or highly misleading. (If I ask someone, “Is there a fire in the next room?” and he answers “I believe there is,” I can’t say: “Don’t be irrelevant. I asked you about the fire, not about your state of mind!”) But what I wanted to say is this. Pointing out that “absurdity” which is in fact something similar to a contradiction, though it isn’t one, is so important that I hope you’ll publish your paper. By the way, don’t be shocked at my saying it is something “similar” to a contradiction. This means roughly: it plays a similar role in logic. You have said something about the logic of assertion. Viz.: It makes sense to say “Let us suppose: p is the case and I don’t believe that p is the case,” whereas it makes no sense to assert “P is the case and I don’t believe that p is the case.” This assertion has to be ruled out and is ruled out by “common sense,” just as a contradiction is. And this just shows that logic isn’t as simple as logicians think it is. In particular: that contradiction isn’t the unique thing people think it is. It isn’t the only logically inadmissible form and it is, under certain circumstances, admissible.11
In other words, logically speaking if a statement leads to a contradiction, then that statement is ruled out. If I say, “I have a square circle,” we say that this statement is illogical and makes no sense because a square circle is a contradiction in terms. That the person does not have any such thing is not decided empirically—there is nothing to look for. But we could imagine a child asking, “Is there such a thing as a square circle?” In this context the phrase, in a way, makes sense. I know what the child means, and I can answer his question. We would explain to the child there is no such thing. An oxymoron is a contradiction put to literary use: “That was the loudest silence I have heard. As usual, the chairman’s silence speaks volumes.” Or on a movie set we might say, “That house is not a house—it only a façade.” But in most cases if a person said, “I just bought a house but it’s not a house,” it would be difficult to know how to reply. Generally we would have to conclude the person is not making any sense. We would clearly have to rule out such statements if made in a factual context. “I just gave you a half pound of turkey, but it’s not turkey.” “Here is your book that is not a book.” “I gave you your antibiotic but it is not an antibiotic.” If these statements are admissible then any statement is admissible—nothing would make sense. We would have to move to Wonderland (“Nobody is in the drawing room.”—”Good. Tell him I’ll be in shortly.”). But clearly contradiction does not only have a formal logical status.
Moore points out a similar absurdity. The assertion “x is the case, but I don’t believe it,” has to be ruled out. It says, “I claim that x is a fact, but can’t accept it.” But to claim x is a fact rests on an established language of facts and claims. To make sense of the idea of a claim or an assertion presupposes certain circumstances—in general, it presupposes a body of knowledge. This is something that often escapes notice. Think of telling someone about a play in a baseball game: “The last out was a classic 6-4-3 double play.” If you know nothing of baseball terminology then that would be opaque. Even if you know nothing about baseball you might describe the scene thus: “Well, one guy threw the ball to the other guy who threw it to some other guy while these two guys were running.” But even that description presupposes a lot of language. Let us say you had never seen any games of any sort. Then perhaps you could only say: “There seems to have been a lot of activity just now.” But even that presupposes at the very least the whole idea of making a report or telling what has gone on, and there is a great deal that is presupposed in even that simple language game.
In other words, making a claim makes sense in certain circumstances and presupposes a certain body of knowledge that must be accepted on the part of the person making the claim. We cannot simultaneously make a claim and reject that background knowledge. In our present case doing so would make the word belief meaningless.
When you say “Suppose I believe . . .” you are presupposing the whole grammar of the word “to believe,” the ordinary use of which you are the master.—You are not supposing some state of affairs which, so to speak, a picture presents unambiguously to you, so that you can tack on to this hypothetical use some assertive use other than the ordinary one.—You would not know at all what you are supposing here (i.e., what, for example, would follow from such a supposition), if you were not already familiar with the use of believe.12
Let us say someone says, “Suppose I believe the earth has existed only since yesterday.” So the sentence is a hypothetical, yet it seems to be saying something important about belief—that it makes sense to say that you believe something that is manifestly false. It looks as if there is this assertion “the earth has existed only since yesterday” and that “belief is perhaps a psychological state that is appended by the individual to the assertion.” But this misunderstands both belief and assertion. We cannot just haphazardly connect words without considering their use—this assumes that they have a meaning apart from their use, and I hope we are sufficiently disabused of this notion by now.
In what context would the statement “the earth has existed only since yesterday” be considered an assertion—that is, a statement of fact? Surely, it is simply nonsense. Certainly this statement presents a picture to us. I can imagine all sorts of things in connection with it—although they would have to be quite fantastic—memories of the past have been falsely implanted in everybody and all the buildings and so forth did not take years to build but were created in an instant and so on. But what actual use would this statement have? How would I insist on this, or begin to try to prove it, offer evidence for it—as I would any other assertion? Hence calling this sentence an assertion would be misleading at best.
But now this statement, since it is ruled out as an assertion, looks like a perfect candidate for belief. Isn’t it the case that we can believe something outlandish or nonsensical, something for which there isn’t a shred of evidence? Yes certainly, but that is not the point at issue. We are not concerned here with what it is permissible to believe but with the logic of belief and assertion itself. We cannot use belief to turn a statement into an assertion, when logically we don’t have an assertion. Belief will not turn “Blimey truggles the wipples had” into a good English sentence. When a sentence such as “the earth has existed only since yesterday” is shown not to be an assertion—what sense can we make out of “I still believe that it is”? Can my “belief” here somehow turn the statement into an assertion? Can we say something like, “Well, for me it is an assertion”? But as we have seen, we can’t privately define assertion.
Belief does not mean a psychological act but presupposes the grammar of belief. To say I believe in a certain statement cannot alter the meaning of the statement, any more than it can change the fact. What you accept or don’t accept is up to you—but this cannot alter logic. To attempt to see belief in this way changes the language of belief in such a way that negates the ordinary use of belief.
Let us return to our original predicament. Someone asserts that “p is the case and I don’t believe that p is the case.” Why does this assertion have to be ruled out? In what way is it similar to a contradiction? On the surface it seems that Moore is correct, and we would have to say this is absurd on psychological grounds. We can’t say on one hand that I am certain x is clearly a fact that I am convinced of—I am asserting it to be true—but on the other hand that I can’t accept it. But Wittgenstein wants to say the absurdity is a logical one, and a fundamental one at that.
If we allowed the above assertion, then assertion as well as belief and certainty would lose their senses. To call a statement an assertion means we have to be able to use it as an assertion. That is, a wide range of conditions and circumstances has to be obtained in order for a statement to make sense as an assertion. We cannot pluck a statement out of the air, call it an assertion simply because it may have the form of an assertion and say I believe it or I don’t believe it—I doubt it or not, etc. Here the conditions and circumstances for belief, doubt, etc., would be lacking.
Hence we must find a use for the statement “I assert p is the case, but I don’t believe p is the case.” Clearly, this is not the assertion that p is this case. If this is a statement about belief, then in what sense is it one? The only sense it might have is something like “I don’t believe a particular fact that I have just asserted.” But how can we make sense of the phrase “don’t believe” in this context? To assert a fact means accepting a series of background conditions and circumstances. Again, a fact cannot be asserted in a vacuum. Assertion, fact, and so on, have sense within particular language games. But “I don’t believe what I have asserted” can only mean rejecting the very background conditions that give the assertion sense. Just as we just can’t assert a statement in a vacuum, neither can we disbelieve or doubt a statement in a vacuum. To doubt or disbelieve a statement means rejecting that which supports a statement as well. In other words, to doubt the statement “the earth has existed for millions of years” would mean removing it from all the circumstance that give the words their meaning. In order to make sense of doubt, here I would have to abstract the statement from its normal language game and declare the statement’s ordinary employment essentially meaningless. I would have to surround the statement with a variety of conditions: we are all insane or deluded—our memories have been implanted, dinosaur fossils are fakes, all geological evidence to the contrary is wrong, and so on.
Although it may not be apparent, there is a strong connection between the previous material on psychology and this section on Moore’s paradox. As we can see here, belief cannot be construed to mean simply a mental state. As with grief or hope or pain, this can be seen through the grammar of the word—the logic of the concept. Clearly any assertion about the psychological presupposes the grammar of assertion—that is, the series of conditions and circumstances regarding any assertion and our psychological language. Deviating from the stricture that meaning is found in the use of the word can lead us into absurdities.
Section xi is quite lengthy and a good deal of it is a (somewhat) uncharacteristically sustained effort on the idea of seeing something as something. Hence it is often known as the “seeing-as” discussion. Certainly, the idea of a “visual impression” figures prominently in this section, and very quickly Wittgenstein focuses on the various ambiguities involved in the concept of “a visual impression.” One of the most perplexing problems occurs when we consider the fact that something may be seen in a variety of ways. For example: what is it about a stop sign that makes me see it as a signal to stop? Is this really something to do with my visual impression of the stop sign? Does a person who knows nothing about traffic signs have a different visual impression? If so, how would we tell? When I look at a page written in Arabic script it means nothing to me, while someone who speaks Arabic finds the writing full of significance. Some people look at modern art and are astounded and see something deep and meaningful, while others just scratch their heads and see only childish doodles. Are both groups seeing something different? What would that be? Again, if I don’t see the beauty in a modern sculpture, am I lacking something in my perceiving faculties?
As usual, Wittgenstein does not outline the source of his difficulty, but much of this section is apparently a response to reading the works of Gestalt psychologists—particularly Wolfgang Kohler, and perhaps Kurt Koffka. Gestalt thinkers were eager to counter what we might call sense-data theories of perception in which a visual perception is constructed out of more discrete bits of perception. For Gestalt thinkers, visual perceptions come to us as organized wholes. They were much exercised by the “ambiguous” figures like the “duck-rabbit,” illustrations of the interplay of figure and ground such as the “double cross,” and so on that play such a prominent part in this section of the Investigations. Gestalt theorists were anxious to counter mechanical theories of sense perception that held that sensations traveled along hardwired neural pathways and were built up into visual impressions by the brain. They used these so-called ambiguous figures to demonstrate their objection to the sense-data idea. Say we see a red book lying on a black table. “Traditional” psychology says that we only see color, so the book obscures part of the table. Since the part of the table the book is actually resting on is out of our view, we then do not actually have a visual experience of what we say we see. To make sense out of “I see the book on the table,” the traditional empiricist psychologists have to resort to, e.g., habit or learning to fill in the perceptual gap here. But the Gestalt thinkers are eager to explain the truth of the statement, “I see the book on the table” in terms of visual experience—in terms of our perceptual faculties.
Gestalt holds that the piecemeal explanation offered by empiricist or sense-data psychologists is incorrect. Rather, as we noted, for the Gestalt thinker, the visual experience is organized as a whole from the beginning. The Gestalt theorist maintains that there exists an “isomorphism” between visual impressions and brain processes because he is anxious to explain visual experience solely in terms of the sensory apparatus of the observer. There are, they thought, structures in the brain that correspond to the organization of visual experience. To show this the Gestalt psychologists developed a number of ingenious experiments—generally involving optical illusions, many of which are probably familiar to the reader. For example, if we take two parallel lines on a page one above the other, most people have no problem seeing that they are parallel. But if the top line is intersected by a series of short lines inclined to the right and the bottom line is intersected by lines inclined in the opposite direction, invariably people will say that the lines are not parallel. The Gestalt theorist claims that since this response is automatic or spontaneous it is not the result of habit or learning. Visual experience is subject to the “structural dynamics” of the sensory process. Just as the interplay of physical forces produce a soap bubble as a whole entity—not something constructed out of pieces, so too our visual experience is constructed whole and entire out of the interplay of “forces” in our sensory apparatus. Certainly, it is hard to deny that our visual experience contains certain “spontaneous groupings” that can easily be said not to be learned. We seem to naturally search out patterns and groups in what we see. Does anyone learn to see random configurations of stars as constellations or shapes and figures in clouds or similar patterns in nature—the delta at the mouth of a river, or the s curve in a road or a path? The Gestalt theorists argue that seeing these phenomena occur spontaneously or naturally and are due to the dynamics of our sensory apparatus.
But we must note a series of (by now) familiar philosophical concerns emanating from what Gestalt proposes. Let us say it is the case that our visual perception is organized in exactly the way we experience it thanks to the aforementioned isomorphic structures in the sensory apparatus. Our sensory apparatus mediates then what we experience—what we experience is determined by how we experience it. In a sense, for Gestalt, all seeing is seeing-as.
Now this idea of course has a very modern ring. Even though the Gestalt thinker is at odds with the sense-data theory of perception, most of us will recognize a relationship between the Gestalt idea and the idea that color is created when light strikes the retina or sound is created when vibrating air molecules strike the eardrum. In other words, both the Gestalt theorist and the sense-data theorist agree that our experience is at least partly the creation of our sensory apparatus. The Gestalt idea, of course, sounds somewhat Kantian: we know phenomena, which are partly a creation of the perceiver. And scientists have long become used to the idea that the observation of a system distorts the system. The observer leaves his imprint on what he observes.
But we must realize that as with sense-data theory, the Gestalt theory is all-encompassing as far as our sensory apparatus is concerned. All of what we observe is “filtered” through our perceiving faculties. Hence we can only speculate about the relationship of what we experience to the “external world.” The Gestalt theorist would have to agree with Kant that there is part of our experience—the “external world” source of our experience—that remains “noumenal” or unknowable. We must take it on faith that our experience equals reality. The result of this idea is then a skepticism regarding the existence of an external world.
We can push this even further to a statement of solipsism. For how can I know that what I experience is the same as what everyone else experiences? We assume that everyone’s sensory apparatus is the same, but these “isomorphic structures” have so far proved elusive. Hence it is possible that my world is unique to me.
As we have seen, Wittgenstein has previously gone to great lengths to undermine ideas such as these. In that Gestalt is antiempiricism in psychology, I think there is some common ground between Gestalt and Wittgenstein, but Gestalt falls into many of the same errors that plague empiricism.
Wittgenstein notices a great deal of conceptual confusion in Gestalt ideas. Again, Wittgenstein is clear that he is not carrying on a psychological investigation but a logical one. First we have to consider the concepts see, seeing-as, and visual impressions. As we saw above Gestalt wants to collapse any distinctions between these concepts. Seeing means having a “visual impression,” which has to do with the structural dynamics of our sensory apparatus; hence seeing is a complex operation that might be understood better as seeing-as. Seeing is never “direct” or immediate. Opening your eyes is not like raising the shade on a window to reveal the world—what is seen is always mediated through the sensory apparatus.
We must first attend to the grammar of these expressions; in doing so we will also see the importance of language when considering experiences of this type. In general, for Wittgenstein, it is a mistake to try to understand the concept of seeing in purely visual or psychological terms. If this sounds so strange that it is putting up a roadblock to continuing, think of the meaning of the phrase “I see your point.” Certainly this statement means nothing visual. The tendency might be to think that it means something psychological. But again, if taken quite seriously the visual model or psychological model for the meaning of this statement is hard to apply here. When you say, “I see your point,” you did not “see” or intuit something in the other person’s mind. But did you “see” something in your own mind? I hope we have seen enough at this point so that we see that this strategy is useless. What exactly did you see, and if you saw anything at all, how would it be relevant to what the other person meant? Surely when he was making his point he was not referring to anything going on inside you. This phrase, of course, means something like—”I agree” or “yes that follows.” Again, interpreting seeing here along visual lines only leads to confusion.
Wittgenstein begins the “seeing-as” analysis by introducing and discussing certain ambiguous figures of the type favored by Gestalt thinkers. Again, these are figures that do not change on the page but change for the observer. For example, a drawing of a cube may at one time appear to project out of the page or recede into it; a step may be drawn so that it appears to be concave or convex. On page 165e Wittgenstein introduces the so-called “duck-rabbit.” Turned one way it looks like a duck. But turned so that the “duck’s” bill is facing up the bill looks like the ears of a rabbit, and so the picture may be seen as a rabbit or a duck depending on the orientation of the observer. Although it may not be obvious at first, there is quite a tangle of concepts here.
If someone looks at the “duck-rabbit” he may not notice the dual nature of the picture. He may simply say that he sees a picture of a rabbit. It is important to note that the verification of what he sees is part of a language game, and so we have necessarily entered the realm of grammar. If the person does not notice the dual aspect of the picture, he would not say he has “seen it as” a rabbit. He simply says, “It’s a rabbit,” and when we ask him what he sees, he explains his perception by pointing to rabbits or pictures of rabbits or he describes them. When someone picks up his fork at dinner it would make little sense for him to say, “I am now seeing this as a fork.” Notice, though, if we see the dual aspect of the duck-rabbit and someone else does not and simply says “it’s a rabbit,” we might say of that person, “Now he is seeing it as a rabbit.”
I am shewn a picture-rabbit and asked what it is; I say “It’s a rabbit.” Not “Now it’s a rabbit.” I am reporting my perception.—I am shewn the duck-rabbit and asked what it is; I may say “It’s a duck-rabbit.” But I may also react to the question quite differently.—The answer that it is a duck-rabbit is again the report of a perception; the answer “Now it’s a rabbit” is not. Had I replied “It’s a rabbit” the ambiguity would have escaped me, and I should have been reporting my perception.13
Clearly there are different grammars for a direct perception versus what Wittgenstein calls noticing an aspect or the change of an aspect.
Again the point here is that we must guard against a purely visual understanding of the meaning of all of our language regarding this experience. If we fail to notice the different language games involved here—some sensory, some psychological, some logical, then we may be tempted to try to load all these concepts onto the back of one language game: “visual impression.” This is approximately the gist of the Gestalt theorists’ understanding of seeing. The Gestalt thinker wants to explain seeing in terms of sensory apparatus, hence what is seen has to be explained by some connection to or something happening to our visual impressions, but Wittgenstein wants to show that this is apt to cause confusions.
Let us look at some texts concerning this idea on page p. 167e.
I suddenly see the solution of a puzzle-picture. Before there were branches there; now there is a human shape. My visual impression has changed and now I recognize that it has not only shape and color but also a quite particular “organization.”—My visual impression has changed;—what was it like before and what is it like now?—If I represent it by an exact copy—and isn’t that a good representation of it?—no change is shewn.14
In other words, if we look at the concept of visual impression for a moment we might see that the above experience as a confirmation of the Gestalt theory. The puzzle-picture that Wittgenstein is alluding to is probably of a type we might be familiar with from children’s magazines. Amidst the drawing of, in this case a tree, the artist “hides” a series of other pictures by mingling them with the parts of a tree. A branch is subtly altered so that if you pay attention to it you see a baseball bat—or where the branches become very fine there is a human hand, etc. Thus where there was once color and shape there is now a particular organization that gives rise to a figure. In many ways this is similar to seeing constellations in random grouping of stars. The patterns appear to the observer spontaneously.
But why must this be due to some alteration in my visual impression? Let us assume that “visual impression” means what is literally before my eyes. Certainly it seems natural to say that when I notice something new in what I am looking at then my visual impression is new as well. But we should be wary of locating this change in a spontaneous mechanical manipulation of the sensory apparatus. If my visual impression in that sense has changed, I should be able to demonstrate or show it. What is “before my eyes” should have altered. But the best way of showing this change would be before and after pictures—but of course in this case both pictures would be the same.
The temptation here, in order to maintain the theory, would be to say that the picture we are looking at is not the visual impression, but the visual impression is something “inner”—within the observer. But here, Wittgenstein wants to point out, we have taken a very wrong turn.
And above all do not say “After all my visual impression isn’t the drawing; it is this—which I can’t shew to anyone.”—Of course it is not the drawing, but neither is it anything of the same category which I carry within myself. The concept of the “inner picture” is misleading, for this concept uses the “outer picture” as a model; and yet the uses of the words for these concepts are no more like one another than the uses of “numeral” and “number.” (And if one chose to call numbers “ideal numerals,” one might produce a similar confusion.)15
Let us say a visual experience is an inner picture in the manner suggested above—perhaps we can imagine it to be a picture made by my imagination that duplicates the scene I am looking at. So the inner picture is like a mental copy of what I see. It has the same borders and dimensions as what I am seeing—the same arrangement of shapes and colors—it is as if I took a photograph of the scene with my mind. But can we give the same sense to “inner picture” as we do the “outer” one—the one we hang on the wall? Our language certainly allows us to create this idea of the “inner picture.” But what can I do with it? I can show the outer picture, take it down from the wall, reframe it, retouch it, sell it, give it away, throw it away, paint over it, put it in another room, and so on. None of this applies to the “inner” picture. Who can I even show it to? I can only make a drawing of it and now it is no longer inner, but outer. Thus we try to talk about a visual impression as an “inner picture,” but we are using “outer picture” as a model and the grammars of the two concepts are very different. The circumstances in which the concept “outer picture” works do not obtain with “inner picture.”
If you put the “organization” of a visual impression on a level with colours and shapes, you are proceeding from the idea of the visual impression as an inner object. Of course this makes this object into a chimera; a queerly shifting construction. For the similarity to a picture is now impaired. If I know the schematic cube has various aspects and I want to find out what someone sees, I can get him to make a model of what he sees in addition to a copy, or to point to such a model; even though he has no idea of my purpose in demanding two accounts. But when we have a changing aspect the case is altered. Now the only possible expression of our experience is what before perhaps seemed, or even was, a useless specification when once we had the copy. And this by itself wrecks the comparison of “organization” with colour and shape in visual impressions.16
Clearly colors and shapes in a “visual impression” or what you see can be pointed to and described just as they could in any picture or scene. Organization—the kind alluded to by Gestalt; that is, the spontaneous grouping in the visual field as the result of the structural dynamics, cannot be treated in the same way. It is not a part of what I see but supposedly a necessary part of how I see it—a necessary part of the visual field. If I arrange a series of dots on a page to form a cross, I can say, “These dots form a cross.” This description would be a quite ordinary part of many different language games—guessing shapes, connect the dots, perhaps used as a type of diagram, and so on. But that these dots are seen as a cross due to a certain structure in my sensory apparatus is a psychological hypothesis. There is no logical necessity in saying my visual impression has a particular organization in addition to colors and shape. It makes sense to say the drawing has a pattern or organization. But to insist that this is a feature of my visual impression turns the visual impression into an inner object that has no use—its part in any language game of seeing is hard to see.
If we have a picture that can be seen in various ways we can get someone to copy the picture as is and then draw the aspect that he sees. He can draw the duck-rabbit as it is and then draw it in order to show that he sees the rabbit aspect. But if we have a cube drawn on a page in which the aspects change back and forth—it sometimes appears to recede into the page, sometimes projects out—then the person can only in essence make two copies if I ask him to copy the picture and draw what he sees. The picture must be looked at to see the aspects change. But what we have drawn on the page or copied is restricted to colors and shapes. “Organization”—other than the arrangement of color and shape in what we have drawn—does not appear or enter into the process, yet the process is complete. Hence we cannot see organization as an element equal to colors and shapes in the visual field. “Organization” as used by the Gestalt theorist is redundant—it does not appear in any description of the visual field. I can describe the colors and shapes in what I see and their arrangement, but beyond this how can I describe organization? What role does it play in the language game? Again it is difficult to give a meaning to organization as an integral part of the concept “visual impression.” What is important in understanding “visual experience” is the language game we play with these words. The meaning of the word cross is sufficient to see an arrangement of dots as a cross.
In the next few pages—168e.–173e.17—Wittgenstein continues to examine the concepts of seeing and seeing-as. He takes a number of examples and continues to focus on the various interrelationships between the concepts of seeing, perception, seeing-as, and so on. The general idea is showing that these various concepts and the relationships among concepts cannot be reduced to a single formula.
As we noted we also must be aware of the role that language plays in many of these experiences. Objects can play various roles in language games, and these roles can become quite complex. For example, a triangle drawn in a book may serve many different roles, depending on the context or use. It can be an arrow, a pointer, a mountain, and so on (cf. p. 171e).18 Children may take a large box and use it to play “house.” For them, in a sense, the box disappears, and for all practical purposes it is a house.
This may look like an odd experience. But how is it possible to see an object according to an interpretation?—The question represents it as a queer fact; as if something were being forced into a form it did not really fit. But no squeezing or forcing took place here.19
This experience can seem odd if we try to fit it into a visual mold or if we try to account for it through our sensory apparatus. But it may be that we are not dealing with something purely visual. When children see the box as a house, what does this mean—are we talking about a sensory experience? Generally this would not fall under the category of a hallucination. Rather, the children pretend the box is a house by treating it as a house—surrounding it with similar circumstances and conditions, this is the door, these are the rooms, this is the address, these are the people who live there, and so on. There is no need to try to force this experience into a visual mold. Hence to see something according to an interpretation is quite a usual experience, as long as we see that the difference between this experience and a purely visual experience is conceptual. We are simply making a distinction within the family of cases related to the use of see.
Here, though, we seem to have hit on something with larger repercussions and an important relationship to some earlier material in the Investigations. Consider the following text on p. 173e.
Here it occurs to me that in conversations on aesthetic matters we use the words: “You have to see it like this, this is how it is meant”; “When you see it like this, you see where it goes wrong”; “You have to hear this bar as an introduction”; “You must hear it in this key”; “You must phrase it like this” (which can refer to hearing as well as to playing).20
This is interesting because it is one of the few times Wittgenstein mentions aesthetics in the Investigations. As we noted earlier, Wittgenstein makes a similar statement in part I when he connects understanding a sentence to understanding a theme in music. There he wanted to say that expressing the meaning of a musical phrase means giving that expression a language—e.g., a musical passage may be explained to someone by telling him the passage has to be heard, for example, as an “argument” between the strings and the winds—they are both fighting for dominance—or something of that sort. Clearly, a sentence may have similar patterns of inflection or emphasis—words are often chosen for their color or liveliness, and so on. This adds to the meaning of the sentence in the sense of its significance. Clearly if I say the word “hey” to someone in an even tone of voice, I may just be saying hello. But if I shout it, then things have changed—I might be telling the person to beware of some danger, etc. The important point is that communicating is seen here as something that requires the mastery of certain techniques that vary over a range of circumstances. Hence it seems Wittgenstein wants to talk about meaning or significance in aesthetics in a related way. There is perhaps a relationship between seeing-as and understanding a musical phrase and understanding the meaning of a sentence.
Apparently Wittgenstein sees a similar relationship within the family of concepts of seeing. Seeing something as something often requires a particular interpretation or a mastery of certain concepts. Seeing-as is a particular concept of experience—similar to the experience associated with understanding a piece of music.
“Now he’s seeing it like this,” “now like that” would only be said of someone capable of making certain applications of the figure quite freely. The substratum of this experience is the mastery of a technique. But how queer for this to be the logical condition of someone’s having such-and-such experience! After all you don’t say that only “has toothache” if one is capable of doing such-and-such.—From this it follows that we cannot be dealing with the same concept of experience here. It is a different though related concept. It is only if someone can do, has learnt, is the master of, such-and-such that it makes sense to say he has had this experience. And if this sounds crazy you need to reflect that the concept of seeing is modified here.21
In other words, rather than understanding “visual experience” as something purely sensory and all visual concepts as a part of this mold, I think Wittgenstein wants to say that seeing-as has a language—it is part of a language game and a form of life—like grief or hope, as we saw in the discussion in part I. To understand this experience we need to pay attention to its language. However, Wittgenstein is not trying to completely characterize this experience. Rather, he is attempting to pass on the tools that we need to understand it.
It is also of note that Wittgenstein wants to include an understanding of aesthetic experience along these lines. Again, as we saw in part I, I think he is trying to indicate that we should steer clear of the psychological as an explanation of aesthetic experience. The aesthetic must be understood on conceptual grounds as well. A sensory or psychological theory is not required to explain how and why we phrase a piece of music in a particular way. Rather, we must take into account the technique involved—the language that is used.
Think of the expression “I heard a plaintive melody.” And now the question is: “Does he hear the plaint?” And if I reply: “No, he doesn’t hear it, he merely has a sense of it”—where does that get us? One cannot mention a sense-organ for this “sense.” Some would like to reply here “Of course I hear it!”—Others: “I don’t really hear it.” We can, however, establish differences of concept here.22
Clearly we can hear a sad or plaintive melody, and we can certainly hear the melody. But does it make sense to distinguish the sadness as something that is heard or sensed apart from the melody? Is there something separate here that we can identify as the sadness? Again the problem is that the melody is something musical, while if we want to “separate out” the sadness as something existing separately this can appear to be a certain psychological component to the music or a hearing or understanding of the music. The melody then almost becomes unimportant because the emotional aspect of the music is now something psychological—perhaps imposed on the music from within the listener or an “aura” accompanying the music requiring some sort of as yet unnamed sensory faculty. Interpreting music, then, would be not related to music but to psychology.
But what would this extra sense be? What is it that you are experiencing apart from the way the melody is played or heard? Wittgenstein relates this to the idea of experiencing the meaning of a word. As we have noted, words can be charged with meaning, and we may have certain experiences while hearing them or saying them. We might learn a new language and notice that the words do not have the ring of familiarity to them as yet. Or if we set up a code with someone and say that when we say the word Tuesday we really mean Wednesday. I might feel very unnatural when I’m saying this, and I might feel that Tuesday has not yet taken on its “new” meaning. All these feelings are undeniable, but it is not the psychology but the use that dictates meaning.
But the question now remains why, in connexion with this game of experiencing a word, we also speak of “the meaning” and of “meaning it.”—This is a different kind of question. It is the phenomenon which is characteristic of this language-game that in this situation we use this expression: we say we have pronounced the word with this meaning and take this expression over from the other language game. Call it a dream. It does not change anything.23
Pronouncing a word with a particular feeling does not “make” it “mean” anything. This expression is “on loan” from the other language game, and we can establish a difference in this family of cases. So too with “hear” in the above text. In each case we want to say we can “hear” the plaint. But if we want to interpret this as something heard or sensed apart from the melody we are using “hear” in the sense of hearing a sound, but the language necessary to make sense of this is lacking. But correcting this situation is not simply a matter of inventing new words, but rather requires a conceptual distinction. Just as meaning is not something inner or something that accompanies a word, so too the plaint of a melody is not something separable from the melody. Such an idea could make sense only with, say, psychological theorizing behind it, but this is entirely unfounded. We would have to locate a hitherto unknown sense—and this is unnecessary once the conceptual distinction is made.
The case with seeing-as and Gestalt is similar. The hypothesis that there are structures in the brain that account for certain features of visual experience comes from, in part, trying to understand seeing and the experience of seeing in a purely visual way. This idea ignores the conceptual dimension. Not every case of seeing something in a particular way needs to be explained by appealing to our sensory apparatus or perceptions. Certain types of seeing—when children see a box as a house—require the mastery of certain techniques—a certain language game. There are a wide variety of meanings of to see, and it is a mistake to think this can only be explained by reference to an inner psychological process. It is not necessary to look at something accompanying visual experience to understand the meaning of “visual experience.” Once we turn away from this picture the mental cramp created by trying to force our concepts into this one mold is relieved.
Seeing-as represents the final substantial section of the Investigations. As with the rest of the work we have not highlighted each and every discussion in this long and difficult section, which probably should have its own book-length treatment. Hopefully, from what has been said the reader can sift through the end of this section and the last few pages of the work.
1. PI 148e., PI, Prentice Hall ed., 174e.
2. PI 148e., PI, Prentice Hall ed., 175e.
3. PI 151e., PI, Prentice Hall ed., 177e.
4. PI 152e., PI, Prentice Hall ed., 178e.
5. PI 153e., PI, Prentice Hall ed., 179e.
6. PI 155e., PI, Prentice Hall ed., 181e.
7. PI 157e., PI, Prentice Hall ed., 184e.
8. PI, Prentice Hall ed., 185e.
9. PI 158e., PI, Prentice Hall ed., 185e.
10. Ludwig Wittgenstein, Culture and Value, 76e.
11. Quoted in Garth Hallet, A Companion to Wittgenstein’s “Philosophical Investigations” (Ithaca: Cornell University Press, 1977) 656.
12. PI 164e., PI, Prentice Hall ed., 192e.
13. PI 167e., PI, Prentice Hall ed., 195e.
14. PI 167e., PI, Prentice Hall ed., 196e.
15. PI 167e., PI, Prentice Hall ed., 196e.
16. PI 168e., PI, Prentice Hall ed., 196e.
17. PI, Prentice Hall ed., 196–202e.
18. PI, Prentice Hall ed., 200e.
19. PI 171e., PI, Prentice Hall ed., 200e.
20. PI 173e., PI, Prentice Hall ed., 202e.
21. PI 178e., PI, Prentice Hall ed., 208–209e.
22. PI 178-9e., PI, Prentice Hall ed., 209e.
23. PI 184e., PI, Prentice Hall ed., 216e.