The reason is that "meaning exists only where there is a distinction between Intentional content and the form of its externalization and to ask for the meaning is to ask for an Intentional content that goes with the form of externalization" (ibid.
If meaning of a word were a mental content, then it would appear reasonable to try to discover it by taking the word in isolation and searching out the links leading from it into the mind.
Of course not (unless by meaning something by a word we understand furnishing the word with a mental content).
MeaningW of an expression amounts to some causal or "intentional" link between the expression and an extralinguistic thing (a real thing, a 'content of consciousness' or something like that).
The opposition between meaningL and meaningW is sometimes, especially in the context of Saussurian linguistics, reflected by such distinctions as ‘meaning’ vs. ‘content’, ‘Bedeutung’ vs. ‘Bezeichnung.
Providing a description elucidates the meaning to the extent to which it is the description of the meaning, or of that to which we hold the meaning to be reducible, e.g. the use of the expression, or a 'cognitive content' for which the expression is supposed to stand.
Restricting ourselves to the two most prominent reducienda of the meaning of an expression, namely the use of the expression and the mental entity ('cognitive content') 'behind' the expression, the following main possibilities seem to emerge as to what a diagram associated with a sentence, or, more generally, with an expression, can amount to: (i) a description of the meaning of the expression (ii) a description of the way the expression is used 6 (iii) a description of a mental entity associated with the expression (iv) a translation of the sentence into another language The first alternative seems to offer the most promising route: what could be a more direct realisation of the task of semantics than displaying expressions alongside with their meanings?8 However, this proposal is rather tricky; for what could count as a description of meaning, which, as we have concluded in the preceding section, is best seen not as a 'real' object, but rather as a value?
It might seem that in this case we may be able to pick up some relevant 'content of consciousness' independently of any linguistic articulation; however, it is hard to see how we could identify contentful mental entities save by way of language; we cannot describe the mental entity 'beyond' the sentence 'Every farmer owns a donkey' save by saying that it is the thought (or idea, or whatever) that every farmer owns a donkey, or the thought that for every x, if x is a farmer, then x owns a donkey etc. What is worse, even if we could give an independent 8 9 See Chomsky (1967).
The enterprise of semantic analysis, it is then claimed, consists in revealing the "semantic structure" of an expression (or of the mental content of an expression).
The structure of the expression, the story goes, is the structure of a mental entity behind the expression - be it called an idea, an intention, a cognitive content, or whatever.