We often take philosophy to be a matter of digging out what is hidden from view. When we think about it theoretically, language seems mysterious. It allows us to express thoughts and talk about the world, as if there were some direct correlation between the sentences we utter, the thoughts we have, and facts about the world. In logic, we find a world of pure, rigid relationships that we can apply to language, thought, and the world. Logic seems to contain the a priori order of things: it expresses all possibilities in the clearest manner, not clouded by any vagueness.
If language can be analyzed into perfect logical relationships, the sense of every sentence must be totally clear. Even sentences that may sound vague in ordinary speech must have a definite sense. This ideal of a logically perfect language eliminates all vagueness and uncertainty, but it also loses its connection to the vagueness and uncertainty we employ in everyday speech. Wittgenstein likens a world of ideal, logical forms to a smooth, frictionless surface as contrasted to the "rough ground" of ordinary language: "We have got on to slippery ice where there is no friction and so in a certain sense the conditions are ideal, but also, just because of that, we are unable to walk. We want to walk: so we need friction. Back to the rough ground!" (section 107). In order to make sense of how we actually use words, we must abandon this notion of ideal logical relationships and recognize there is no essence to language. "Language" is a series of more or less related family resemblances.
When we remove words like "knowledge," "proposition," and "being," from their ordinary use and then puzzle over what sorts of things they are independent of the contexts we find them in, we are bewitched by "pictures" that our language presses upon us. Wittgenstein's method is to describe how these words are used in their everyday contexts, and so show that the metaphysical questions we ask about them are not appropriate to the words as they are actually used. This method does not provide any great enlightenment, but only a sober sense of how little is achieved in metaphysical investigations. Philosophy does not tell us things we did not know; it reminds us of what we have always known by putting it clearly before our eyes. Philosophy should remind us of the ordinary use of the words that we puzzle about only when we extract them from their ordinary contexts. The methods he employs are thus like "therapies" that lead us out of metaphysical puzzlement.
We might think we have found a general definition of a proposition in saying that a proposition is the thing about which we can say it is true or false. But this simply describes how we talk about propositions: it does not explain anything about their deeper nature.
In criticizing the search for a logically ideal world of essences and rigid forms, Wittgenstein criticizes the position he himself espouses in the Tractatus. That text, which builds upon the earlier work of Frege and Russell, seeks to explain what language, thought, and the world must be like so that one can reflect the others. He develops what has been called the "picture theory of propositions," according to which propositions can depict the world by virtue of sharing a logical form with the facts they depict.
The remarks of sections 65–91 lead naturally into this criticism of his earlier view, as those sections attack the notion of essences and fixity of meaning. If words and sentences do not have to have a fixed sense, then how can we take them even to approximate rigid logical forms? The key to understanding language is not to identify the hidden structure of propositions but to appreciate how we actually use language to say what we need to say.
This approach caused something of a revolution in analytic philosophy, and spawned a movement known as "ordinary language philosophy" that came to prominence at Oxford in the decades following the Second World War. The debate between ordinary language philosophy and traditional philosophical methods has been rife with misunderstanding, largely because Wittgenstein's thought does not simply adopt a position contrary to tradition, but also introduces new methods and new approaches that have not been assimilated by the tradition.
One way of understanding Wittgenstein's new approach is to say that there is no special body of knowledge that belongs to philosophy. Physics, for instance, has its own field of investigation and its own special terminology—words like "mass," "energy," "electron," and so on. We might think analogously that philosophy can also investigate concepts such as knowledge, selfhood, and language, and adopt its own special terminology to deal with these theoretical issues. But unlike physics, philosophy employs a double standard. On one hand, in inquiring into the nature of knowledge, we treat "knowledge" like a technical term that refers to something whose nature is as yet unclear and that we must uncover. But on the other hand, this investigation draws on the word "knowledge" as we use it in everyday speech. A physicist who researches what an a electron is investigates something as yet undiscovered; in investigating what knowledge is, we are puzzling over a word with which we are already familiar.
We all know perfectly well how to use "knowledge" in ordinary contexts. But when we extract "knowledge" from these ordinary contexts and simply ask, "what is knowledge?" we are at a loss how to respond. Metaphysical speculation, according to Wittgenstein, arises when we take such words out of their ordinary contexts and ask about the nature of the thing itself. Wittgenstein points out that "knowledge" is a word, and words mean what they do by virtue of being used in the contexts that they are used. If we remove "knowledge" from all contexts and ask what the thing itself is we are at a loss precisely because, outside of all contexts, the word cannot mean anything at all.
Wittgenstein's method can be called "therapeutic." He is not making new discoveries or providing new explanations, but is employing a purely descriptive method to untie the knots in our thinking that are caused by metaphysical speculation. In section 127 he says, "The work of the philosopher consists in assembling reminders for a particular purpose." His method is to remind us of the various contexts in which words such as "knowledge" make sense, and to show that they have no meaning beyond what we find in those contexts. We end up not with an understanding of what knowledge is, but only with a reminder of what we have always known: we can form coherent sentences that include the word "knowledge."
If Wittgenstein's philosophy is a therapeutic untying of mental knots, his methods are as diverse as the various temptations that lead us into metaphysical thinking. He uses language-games, discusses rule following, and investigates supposed mental states and mechanisms. Most of his methods are entirely new because his project is entirely new. He is going in the opposite direction of traditional philosophy, and trying to lead us away from complex theories rather than into them. While this section of the text outlines the direction of his thought, we are best able to understand it by delving into the various "therapies" he applies in the rest of the text.