When we think about a general problem of the meaning, we must take into account three following functions, all of which must operate at the same time as far as a word can stand as a word in language ; F_1 (Designation)-a function to designate a referent ; F_2 (Signification)-a function to signify something which is not necessarily equal to the referent ; F_3 (Syntactic Construction)-a function to construct a sentence. According to G. Blocker, among these functions there are very interesting phenomena. In one direction F_1 has a tendency to absorb F_2 into itself, and in another direction F_2 does F_1. But at the moment this absorption seems to be accomplished in either direction, the language itself can no more operate as a language. Because, in the first case, meaning, something signified, becomes a kind of real object which cannot be discriminated from a physical object ; in the second case, the referent seems to exist without any signification-nothing else but an illusion. This attraction and repulsion between F_1 and F_2 compells us to recorrect the formalistic view of musical meaning, which tries to drive both F_1 and F_2 out of consideration and to insist as if the musical meaning could exist simple as one kind of F_3-F_3 functions here as an activiation of F_2 through the intrusion or deviation of an different sound term in a direction-oriented context (L. B. Meyer). When we think it impossible to return to the simple referentialistic view which cannot explain the raison d'etre of music, we arrive at N. Goodman's concept, "exemplification", which can explain how a work which exemplifies something can posess not only the function of F_3 but also those of F_1 and F_2. And if we were right up to this point, we must also turn our concerns from semantics to pragmatics and beginn to make it clear what the implicit or underground referent is in music (Ch. Hubig).