243. A human being can encourage himself, give himself orders, obey, blame and punish himself; he can ask himself a question and answer it. We could even imagine human beings who spoke only in monologue; who accompanied their activities by talking to themselves. -An explorer who watched them and listened to their talk might succeed in translating their language into ours. (This would enable him to predict these people's actions correctly for he also hears them making resolutions and decisions.) But could we also imagine a language in which a person could write down or give expression to his inner experiences - his feelings, moods, and the rest - for his private use? - Well can't we do so in our ordinary language? - But that is not what I mean. The individual words of this language are to refer to what can only be known to the person speaking; to his immediate private sensations. So another person cannot understand the language.
This paragraph though not actually going so far as to bluntly state that private languages are impossible, does not fall far short. As a defender of the hypothesis that private languages are possible I feel bound to contest two points. The first is the assumption he makes that the words of language do not refer to what can only be known to the person speaking. There is some difficulty however arising through the use of the word 'know' here. It seems to be the case that very few if any of the words in ordinary language which might be thought to refer to 'inner experiences' such as 'pain' are commonly believed to refer to what can only be known to the person speaking. We often believe ourselves to know that someone else is in pain, and if we are right in so believing in these and similar cases then Wittgenstein's assumption is very tempting. It is quite essential to a proper statement of the empiricist position that if we claim that our experiences are private, and mean by private that we alone can have knowledge of them, then we do not accept that people are right in claiming for example, to "know" that we are in pain. In order to maintain this position, the empiricist must either assert that people are often wrong in their claims to knowledge or alternatively he may concede that he is using the word know in a special way. Neither of these steps would meet the approval of Wittgenstein. A case could be presented for either of these steps but I shall for the sake of brevity confine myself to the latter and leave open questions concerning the ordinary usage meaning of 'know'. The special sort of knowledge which is required to formulate the definition of a private language is something like `knowledge which logically excludes the possibility of error'. There is no doubt that this is a problematic definition of knowledge. It is problematic from my viewpoint because it is not obvious that we can ever be logically incapable of error in any respect at all. I shall assume for the present that this particular difficulty can be dealt with and that we can argue convincingly that our knowledge of such things as our sensations, feelings emotions and thoughts is of this special infallible sort. The definition is unacceptable to Wittgenstein for other reasons which we will come to later. The importance of this distinction lies in the relationship between the 'private language' problem and other areas of philosophy. If a broad notion of 'knowledge' is taken to be the relevant one for the definition of private language then the non existence or impossibility of such languages will be much less remarkable. If a broad notion is taken there may be nothing at all about which we may not all claim some knowledge, and in a world thus stripped of all privacy the demise of the private language would be of little concern. To have a real impact Wittgenstein must disprove the existence of private languages as defined using the narrower concept of 'Knowledge', and this would have the corollary that none of the things to which we may refer in ordinary language can be in the relevant sense 'private', the point I am presently contesting.
DB3. The second point I want to dispute is that in the last sentence of the paragraph. Wittgenstein infers from the fact that a language is used by someone to refer to the things of which he has private knowledge that no other person will be able to understand that language. Given only the additional premise that we do understand each other's talk about pains sensations and emotions we can infer that these things referred to in ordinary language are not things of which we have private knowledge, which is just the claim I have argued against in the previous paragraph. I maintain, in default of seeing any good case to the contrary, that a language private in the sense of being capable of referring to private objects (or what have you's), need not also be private in the sense of being understood by only one person, and ordinary language is private in the first sense, but not in the second.