|
|
|
|
|
|
To name A by k is a case of synecdoche (the veil for the ship, pars pro toto). Since k could even be the seme «crown» characterizing the sememe «king», to name A by k can also be a case of metonymy (in traditional terms; in the terms of our present approach, such a difference tends to disappear). But k happens to be also a seme of another sememe, namely, D. Therefore, by an amalgamation through k, one can substitute A with /D/. This is a case of metaphor. A long white neck being a property both of a beautiful woman and of a swan, the woman can be metaphorically substituted for by the swan. Apparently, one entity is in the place of the other by virtue of a mutual resemblance. But the resemblance is due to the fact that in the code there exist already fixed relations of substitution which, in some way or other, link the substitute entities to those substituted for. |
|
|
|
|
|
|
|
|
Now suppose that there exists a practice of language in which A is habitually substituted for k. In this case k becomes, by convention, one of the possible connotations of A. The metaphor, once it has become usual, enters as part of the code and in the long run can fix itself in a catachresis ('the neck of the bottle', 'the leg of the table'). The fact remains, however, that the substitution took place because of the existence, in the code, of connections and therefore contiguities. This would lead us to state that the metaphor rests on a metonymy. If Model Q is based upon unlimited semiosis, every sign, sooner or later, must depend upon a connection anticipated by the code. Obviously there can be produced connections of which no one had ever thought. We then have an ambiguous message. The aesthetic function of language seeks to create connections which as of yet do not exist; thus it also seeks to enrich the code's possibilities. Even in this case metaphoric substitution can rely upon metonymic practice. |
|
|
|
|
|
|
|
|
2.7. The Crown and the White Collar |
|
|
|
|
|
|
|
|
At this point we need to elucidate the terms 'metaphor' and 'metonymy', since it seems that up to now the latter term has been used in a metaphoric sense. |
|
|
|
|
|
|
|
|
Every theory of metaphor defines this figure as the substitution of one element of language for another (the operation is completely internal to the semiotic circle), but by virtue of a resemblance between their referents. It is this necessary recourse to referents (and to their presumed relations of analogy or resemblance) that has prodded us to criticize the notion of metaphor as something that cannot be founded on autonomously semiotic terms. The risk is now that (reconducting every metaphor to a chain of metonymies) even these will demand to be founded on a recourse to referents. In reality, rhetoricin having recourse to |
|
|
|
|
|