… is used in a way to express an opinion which we reject. The opinion e.g. that all other people except A (or myself) are shamming when they say that they have toothache. It seems that I want to || the solipsist wants to say of them that they haven't got toothache i.e. that he wants to say of them that which the sentence ‘they haven't got toothache’ says in the ordinary language. Whereas what he really wishes to do is to change the notation. Analogous when the idealist says that the tree doesn't exist when nobody looks at it. He is prompted to say this but the || his sentence in ordinary language expresses a view which he isn't at all keen on holding. He doesn't wish to say that the tree in the sense in which it might vanish while we look at it actually always vanishes as soon as everybody turns their backs on it. But he is not able to overcome the difficulty of expression which consists in his expressing himself in the phraseology in which views about things are uttered in ordinary language instead of the phraseology of grammar; i.e. giving || laying down rules. Why is he unable to emerge from that phraseology? Note the usage of the word ‘possibility’. The confusion we talk about is all here || prepared in the terminology which talks about physical & logical possibility. Now I am tempted to saythat || : whenever anything is really seen it is I who see. [Dieses Notizbuch] “For I was taught to use the word ‘to see’ for that which only I do.” But that isn't true. Rather I am tempted to use this word in a way different from that I was taught (drawing the boundary of a county). (Imagine someone who said: the word ‘red’ ought only to be used when something is said to be red not where a thing's said not to be red. One could perfectly well have such a notation & there is in fact quite a strong urge for one sometimes.) Now when I say: α) “Whenever anything is really seen it is I who see” I am pleading for a notation in which in certain cases [behaviourism] for me the use of the word ‘I’ is abolished. And that is what I want for I want to say that nothing in the visual experience necessarily points to my (or in fact a person) having it. I.e. a || A person may be seen (or may be) round the geometric eye but no person need be there. It is a || then a of experience whether a person sees, when something || anything is seen. When I said α it seemed as though I wished to be partial but as a matter of fact I wished to be impartial. Let us see what experiences determine who sees what is seen. What I wished was to give a description of what is seen without mentioning a person. “Whenever anything is really seen it is I who see.” 1) Why then use the ‘I’ at all? What you plead for is a notation in which you say “so & so is seen” instead of ‘I see so & so’. But further: whom did you mean by ‘I’? Wittgenstein? Now he is distinguished from other people primarily by his body. But there are many criteria for identifying the person & if facts were a bit || we imagine facts a bit different from what they are we should || shall see that the personal name is used in a much more complicated way than it || use of the personal name is not as simple & straightforward || simple as it appears. Examples: 1) Imagine all men looking alike but there existing definite sets of characteristics (like the four temperaments) inhabiting them & sometimes changing their habitation. One might in this case be inclined to call a particular character Jones. ¤
     Think¹ of Jekyll & Hyde.
     In which cases would we talk of two persons inhabiting the same body? Memories. Imagine my memory today to leave out yesterday the day before the day before yesterday etc. but tomorrow remembering yesterday etc. We could then say that two persons inhabited my body. But we needn't say this.
     Imagine everybody getting two names for the odd & even days of his life because of memory leaving out every other day.