ABSTRACT
THE point just established being important, I want to reinforce the argument by comparing the sentences ‘I am asleep’ and ‘I am in pain’. The reason for this particular comparison is that there is an inclination to think that these two sentences must have the same sort of use, namely, to describe states of oneself. One does not find out that one is in pain by employing a criterion. Indeed it makes no sense to speak of finding out that one is in pain, when this would imply that one was previously in pain but not aware of it. There is, however, a criterion for determining whether someone uses the sentence ‘I am in pain’ correctly—and this makes it an intelligible sentence. The behaviour and circumstances of an infant are the original criterion of his being in pain. As he grows and begins to talk it will normally come about that often when his behaviour and circumstances are those of a person in pain he will say the words ‘It hurts’, or some synonymous ones; and hardly ever will he say them when either his behaviour or circumstances do not satisfy the original criterion of pain. This development fulfills our criterion of his understanding those words. Now his saying them serves as a new criterion of his being in pain. We shall 16conclude sometimes that he is in pain from his mere say-so, even though his behaviour and circumstances are not a paradigm of pain. He can tell us that he is in pain, and we can know that he is because he has told us. This is possible because his verbal expression has been conjoined with certain behaviour (the natural, primitive, behaviour of a person in pain) occurring in certain circumstances (e.g. he has been hit or cut or burned). (Wittgenstein, § 244). His saying ‘I am in pain’, either to others or to himself, can be a use of language only because a connection has already been established between those words and the outward phenomena that are the original criterion of pain.
