One might object to this argument that all these interpretations pre-suppose another word-language. And this objection is significant if by “interpretation” we only mean “Translation into a word-language”. ‒ ‒ ‒ Let me give some hints which might make this clearer. Let us ask ourselves what is our criterion when we say that somseone has interpreted the ostensive definition in a particular way. Suppose I give to an Englishman the ostensive definition “this is what the Germans call ‘Buch’”. Then, in the great majority of cases, at any rate, the English word “book” will come into the Englishman's mind. We may say he has interpreted “Buch” to mean “book”. The case will be different if e.g., we point to a thing which he has never seen before and say: “This is a banjo”. Possibly the word “guitar” will then come into his mind, possibly no word at all but the image of a similar instrument, possibly nothing at all. Supposing then I give him the order “now pick a banjo from amongst those things”. If he picks what we call a “banjo” we might say “he has given the word ‘banjo’ the correct interpretation”; if he picks some other instrument:– “he has interpreted ‘banjo’ to mean ‘string instrument’”.