How we intend a word might be significant to its meaning, but how we experience the word is not. This is not to say that there are not characteristic experiences associated with certain words, but these experiences do not fix the meanings of these words. We can often speak without any notable experience. Meaning and intention lack "experience-content": though we may have certain experiences while meaning or intending something, these experiences are not themselves the meaning or intending. The experiences accord with how we choose and value words, but how we use them to mean something is not determined by these accompanying experiences. Choosing words is not a mental experience. If "the word is on the tip of my tongue" were never followed by finding the w ord, we would not think to say this. The characteristic behavior that surrounds such expressions, and not a mental experience, give them their sense.
We do not "know" our inner experiences like pain or speaking to ourselves any more than we believe, suspect, or do not know them. "Knowledge" is not the correct term here because the general criteria for talking about knowledge are missing. This point is illustrated by comparing the assertions "a goose has no teeth" and "a rose has no teeth." We can verify that a goose has no teeth by looking in its mouth, but how do we determine whether a rose has no teeth? One place is as good to look as another, so we could say that "a rose has teeth in the mouth of a beast": the cow's teeth chew the grass that passes through its digestive tract and comes out as dung that feeds the rose. We are misusing language, not discovering a special kind of first-person kno wledge, when we talk about "knowing" our inner life.
When I confess what I was thinking, my truthfulness is not checked against inner criteria: truthfulness is not determined by whether or not I am correctly describing an inner process. The skeptical conclusion that other people's feelings are hidden from u s is usually irrelevant. We can still reach conclusions about people's inner states that are not necessarily incomplete: when someone is writhing in pain, I do not reflect that this person's inner feelings are hidden. We can know, even be certain, what so meone else is feeling, and our not feeling it ourselves is no detriment to this certainty.
We do not, as a rule, come to blows over the result of an equation, because we can agree on methods of determining these results. The difference between mathematical certainty and certainty regarding other people's feelings is not a matter of degree, but a matter of different language-games: there are not established and sure-fire ways I can make other people share my certainty regarding someone's feelings. And regarding other people's feelings, there are people who are better and worse at judging, a nd such judgment can be taught; the difference is that there are not fixed and clearly defined methods.
The final three sections are brief, and deal with the nature of our concepts as determined by our forms of life, with remembering and its lack of experiential content, and with psychology's conceptual confusions inhibiting the possibility of any scien tific progress.
Skepticism about other minds is made up of the many doubts we can raise about other people's inner experience. There is no outward difference between someone who is unhappy, and a good actor who is simply pretending. We can even raise the question of whet her other people are in fact automata. A masterfully programmed robot could conceivably simulate all our outward behavior, but without any kind of inner experience.
These sorts of skeptical questions are based on the perception that we have a different kind of epistemic access to other people's experience than we do to our own. In my own case, I know that my tears, smiles, speech, and gestures, are all only outward m anifestations of my inner life. This inner life is, as it were, "hidden" from everyone but myself. There are certain things that no one but myself can know about myself.
Wittgenstein sets about disintegrating this sort of skepticism in a number of ways. One of his more powerful observations is that I do not actually "know" my own inner life. The things we talk about knowing are the same things we talk about finding out, s uspecting, believing, or doubting. There is no process of "finding out" whether I am in pain. How would we set about determining whether or not I know I am in pain? This investigation would be confounded in the same way that an investigation as to wheth er a rose has teeth would be confounded: we do not even know how to look. The idea that "a rose has teeth in the mouth of a beast" is a peculiar, but sly, solution to the question of whether a rose has teeth. Because there is no obvious mouth on a rose to look into, we might as well look for these teeth anywhere. Our investigation had no clear direction from the outset, so we are as justified in claiming the rose's teeth are in the cow's mouth as anywhere else.
The push toward skepticism relies on the contrast between first-person and third-person knowledge, in pointing out that people have clearer knowledge of their inner lives than we do, and concluding that our knowledge is therefore lacking. If this contrast is false, there is no foundation for our claim that our own knowledge regarding other people's feelings is lacking. The obvious objection to saying that there is no distinction between first- and third-person knowledge is that we lack knowledge of other people's states. I can think someone is in pain who is just faking it: there is a fact of the matter here that I can be wrong about. Wittgenstein is not trying to assert the plainly false here, that we have clear access to other people's lives. Rather, he is showing us that there is no higher degree of certainty that we can aspire to. There is no fact, no item of knowledge, that exists only in the subject's mind, which would settle the matter for us if only we could have access to it. Wittgenstein careful ly shows us how we construct our language-games regarding other people's feelings. When we discuss things like knowledge, uncertainty, doubt, and conviction, our attention is directed exclusively toward outward behavior. All the criteria for judging on th ese matters are before our eyes. Because I cannot know someone's inner state (nor was there ever any question of knowing it), this inner state does not factor into my discussion of how I know what the person is feeling. This is not to say, in a behavioris t vein, that outward pain-behavior is the pain. Pain is pain, and not pain-behavior, but knowledge of pain is knowledge of pain-behavior, and not knowledge of inaccessible inner sensations.
Wittgenstein anticipates the further objection that, surely, our certainty regarding other people's inner states is less complete than our certainty regarding mathematical results. True enough, but this assertion just highlights the way language-games fun ction differently in different forms of life. If there were no fixed rules for solving mathematical equations, or if the ink and paper mathematicians used often morphed to shift the results that had been written down, our concept of mathematical certainty would no longer be the same. It is not so much that our knowledge regarding other people is less certain than our knowledge of mathematics; it is that certainty functions differently in this context. My certainty regarding someone's inner life is an expr ession of conviction. This expression can be questioned, disputed, and proven wrong in different ways than expressions of conviction regarding mathematical results, but in the language-games dealing with other people's feelings, there is no higher degree of certainty that I could aspire to that I am now somehow missing.