4.011 At first glance a sentence – as it exists printed on paper perhaps – seems not to be a picture of the reality with which it deals. But so too do written notes seem at first glance not to be a picture of music, nor our written signs for sounds (letters) to be a picture of our spoken language.

And yet these symbolisms prove to be pictures – even in the ordinary sense of the word – of what they represent.

Really? A written sentence is a picture, in the ordinary sense, of a spoken sentence? Notes picture music? They represent these things, I think we can say that. So perhaps “Bild” is better translated as representation than as picture. Then propositions represent reality as we conceive of it, or imagine it, or take it to be. This sounds fair enough, albeit perhaps not very interesting.

Black (p. 163) says that the bit about pictures even in the ordinary sense “can hardly be defended.” On the same page, above this, he quotes Moore Papers p. 263 to the effect that Wittgenstein admitted that when he wrote the TLP he had not noticed that the word ‘pictures’ was vague, but that, nevertheless, “he still … thought it “useful to say ‘A proposition is a picture or something like one’” although … he was willing to admit that to call a proposition a “picture” was misleading; that propositions are not pictures “in any ordinary sense”.” So something odd is going on here. According to Moore, Wittgenstein said that he merely wished to stress a similarity between the grammar or use of ‘proposition’ and that of ‘picture.’ Why, though?