Philosophical Investigations by Ludwig Wittgenstein

My rating: 5 of 5 stars

If you read first Wittgenstein’s Tractatus, and then follow it with his Philosophical Investigations, you will treat yourself to perhaps the most fascinating intellectual development in the history of philosophy. Wittgenstein has the distinct merit of producing, not one, but two enormously influential systems of philosophy—systems, moreover, that are at loggerheads with one another.

In fact, I wouldn’t recommend attempting to tackle this work without first reading the Tractatus, as the Investigations is essentially one long refutation and critique of his earlier, somewhat more conventional, views. But because I wish to give a short summary of some of Wittgenstein’s later views here, I will first give a little précise of the earlier work.

In the Tractatus, Wittgenstein argues that language has one primary function: to state facts. Language is a logical picture of the world. A given proposition mirrors a given state of affairs. This leads Wittgenstein to regard a great many types of utterances as strictly nonsense. For example, since ethics is not any given state of affairs, language couldn’t possible picture it; therefore, all propositions in the form of “action X is morally good” are nonsense.

Wittgenstein honestly believed that this solved all the problems of philosophy. Long-standing problems about causation, truth, the mind, goodness, beauty, etc., were all attempts to use language to picture something which it could not—because beauty, truth, etc., are not states of affairs. Philosophers only need stop the attempt to transcend the limits of language, and the problems would disappear. In his words: “The solution of the problem of life is seen in the vanishing of this problem.”

After publishing this work and taking leave of professional philosophy (as he thought it had been dealt with) Wittgenstein began to have some doubts. Certain everyday uses of language seemed hard to account for if you regarded language as purely a truth-stating tool. These doubts eventually culminated in a return to Cambridge, and to philosophy. His posthumously published Investigations represents the fullest expression of his later views.

So what are these views? Well, first let us compare the styles of the two works. The writing in both the Tractatus and the Investigations is extraordinary. Wittgenstein is one of the very finest writers of philosophy, in a league with Nietzsche and Plato. He uses almost no technical terms, and very simple sentence-structures; yet his phrases can stick in the mind for months, years, after first reading them. Just the other day, I was having a conversation with my German tutor about learning a foreign language. To something I said, she responded, “Die Grenzen meiner Spracher bedeuten die Grenzen meiner Welt.” (“The limits of my language are the limits of my world”—a quote from the Tractatus.)

Although the the writing in both works is equally compelling, the structures are quite different. In the Tractatus, Wittgenstein’s argument is unified, complete; he even numbers his sentences as primary, secondary, and tertiary in terms of their importance to the argument. In that work, we can clearly see the influence of Bertrand Russell’s logicism: language is reduced to logical propositions, and the argument is organized along logical grounds.

The reader of the Investigations will encounter something quite different. Wittgenstein writes in similarly terse aphorisms; he even retains a numbering-system for his points—each individual point getting its own numbered paragraph. The numbering of these paragraphs, however, is cumulative, and does not express anything about their significance to his larger design. It is almost as if Wittgenstein wrote down his thoughts on numbered flash cards, and simply constructed the book by moving the flash cards around. Unlike the Tractatus, which resolves itself into a unified whole, the Investigations is fragmentary.

I begin with style because the contrast in writing is a clue to the differences in thought between the earlier and later works. Unlike the Tractatus, the Investigations is rather a collection of observations and ideas. The spirit of Wittgenstein’s later enterprise is anti-systematic, rather than systematic. Wittgenstein aims not at erecting a whole edifice of thought, but at destroying other edifices. Thus, the text jumps from topic to topic, without any explicit connections or transitions, now attacking one common philosophical idea, now another. The experience can often be exasperating, since Wittgenstein is being intentionally oblique rather than direct. In the words of John Searle, reading the Investigations is “like getting a kit for a model airplane without any explanation for how to put it together.”

Let me attempt to put some of these pieces together—at least the pieces that were especially useful to me.

Wittgenstein replaces his old picture metaphor with a new tool metaphor. Instead of a word being meaningful because it pictures a fact, the meaning of a word is—at least most of the time—synonymous with the social use of that word. For example, the word “pizza” does not mean pizza because it names the food; rather, it means pizza because you can use the word to order the food at a restaurant. So instead of the reference to a type of object being primary, the social use is primary.

This example reveals a general quality of Wittgenstein’s later thought: the replacement of the objective/subjective dichotomy with the notion of public, social behavior.

Philosophers have traditionally posited theories of meaning that are either internal or external. For example, pizza can mean the particular food either because the word points to the food, or because the word points to our idea, or sensation, of the food. Either language is reporting objective states of affairs, or subjective internal experiences.

Wittgenstein destroys the external argument with a very simple observation. Take the word “game.” If the external theory of meaning is correct, the word game must mean what it does because it points to something essential about games. But what is the essential quality that makes games games? Is there any? Some games are not social (think of solitaire), some games are not trivial (think of the Olympic Games), some games are not consequence-free (think of compulsive gambling), and some games are social, trivial, and consequence-free. Is a game something that you play? But you also play records and trombones. So what is the essential, single quality of “game” that our word refers to?

Wittgenstein says there isn’t any. Rather, the word “game” takes on different meanings in different social contexts, or modes of discourse. Wittgenstein calls these different modes of discourse “language-games.” Some examples of language games are that of mimicking, of joking, of mourning, of philosophizing, of religious discourse. Every language game has its own rules; therefore, any proposed all-encompassing theory of language (like Wittgenstein’s own Tractatus) will fail, because it attempts to reduce the irreducible. You cannot reduce chess, soccer, solitaire, black-jack, and tag to one set of rules; the same is true (says Wittgenstein) of language.

Another popular theory of meaning is the internal theory. This theory holds that propositions mean things by referring to thoughts or sensations. When I refer to pain, I am referring to an internal object; when I refer to a bunny, I am referring to a set of visual sensations that I have learned to call ‘bunny’.

Wittgenstein makes short work of this argument too. Let’s start with the argument about sensations. Wittgenstein points out that our ‘sensations’ of an object—say, a bunny—are not something that we experience, as it were, purely. Rather, our interpretations alter the sensations themselves. To illustrate this, Wittgenstein uses perhaps the funiest example in all of philosophy, the duck-rabbit:

As you can see, whether you interpret this conglomeration of shapes, lines, and spaces as a rabbit or a duck depends on your interpretation; and, if you had never seen a duck or a rabbit in your life, the picture would look rather strange. Ernst Gombrich summed up this point quite nicely in his Story of Art: “If we look out of the window we can see the view in a thousand different ways. Which of them is our sense impression?”

The point of all this is that trying to make propositions about sense-impressions is like trying to hit a moving target—since you only see something a certain way because of certain beliefs or experiences you already hold.

The argument about inner feelings is equally weak. For example, when we learned the word pain, did someone somehow point to the feeling and name it? Clearly, that’s impossible. What actually happens is that we (or someone else) exhibited normal behavioral manifestations of pain—crying, moaning, tearing, clutching the afflicted area. The word pain then is used (at least originally) to refer to pain-behavior, and we later use the word ‘pain’ as a replacement for our infantile pain-behavior—instead of moaning and clutching our arm, we tell someone we have a pain, and that it’s in our arm. This shows that the internal referent of the word ‘pain’ is not fundamental to its meaning, but is derivative of its more fundamental, public use.

This may seem trivial, but this line of argument is a powerful attack on the entire Cartesian tradition. Let me give you an example.

René Descartes famously sat in his room, and then tried to doubt the whole world. He then got down to his own ego, and tried to build the work back up from there. This line of thought places the individual at the center of the epistemological question, and makes all other phenomena derivative of the fundamental, subjective experience of certainty.

But let us, as Wittgenstein advises, examine the normal use of the word “to know.” You say, “I know Tom,” or “I know American history.” If someone asked you, “What makes you say you know Tom and American history?” you might say something like “I can pick Tom’s face out of a crowd,” or “I could pass a history test.” Already, you are giving social criteria for what it means to know. In fact, the word “to know” presupposes the ability to verify something with something that is not yourself. You would never verify something you remember by pointing to another thing you remember—that would be absurd, since your memory is the thing being tested. Instead, you indicate an independent criterion for determining whether or not you know something. (The social test of knowledge is also explicit in science, since experiments must be repeatable and communicable; if a scientist said “I know this but I my can’t prove it once more,” that would not be science.)

So because knowing anything apparently requires some kind of social confirmation, the Cartesian project of founding knowledge on subjective experience is doomed from the start. Knowing anything requires at least two people—since you couldn’t know if you were right or wrong without some kind of social confirmation.

Wittgenstein brings this home with his discussion of private language. Let’s say you had a feeling that nobody has told you how to name. As a result, you suspect that this feeling is unique to yourself, and so you create your own name for it. Every time you have the feeling, you apply this made-up name to it. But how do you know if you’re using the name correctly? How do you know that every time you use your private name you are referring to the same feeling? You can’t check it against your memory, since your memory is the very thing being doubted. You can’t ask somebody else, because nobody else knows this name or has this sensation. Therefore, merely thinking you’re using the name consistently and actually using the name consistently would be indistinguishable experiences. You could never really know.

Although Wittgenstein’s views changed dramatically from the early to the late phase of his career, you can see some intriguing similarities. One main current of Wittgenstein’s thought is that all philosophical problems result from the misuse of language. Compare this statement from the Tractatus, “All philosophy is ‘Critique of language’,” with this, from the Investigations: “Philosophy is a battle against the bewitchment of our intelligence by means of language.” In both works, Wittgenstein is convinced that philosophical problems only arise because of the misuses of language; that philosophers either attempt to say the unsayable, or confuse the rules of one language-game with another—producing nonsense.

I cannot say I’ve thought-through Wittgenstein’s points fully enough to say whether I agree or disagree with them. But, whether wrong or right, Wittgenstein already has the ultimate merit of any philosopher—provoking thought about fundamental questions. And even if he was wrong about everything, his books would be worth reading for the writing alone. Reading Wittgenstein can be very much like taking straight shots of vodka—it burns on the way down, it addles your brain, it is forceful and overwhelming; but after all the pain and toil, the end-result is pleasant elation.

View all my reviews