Rastier, François: "On Signs and Texts"


1. The unifying strategy

It summarizes the semiotic level with only one type of signs, and the semantics associated to it. The prevalent strategy in classical cognitivism is a case in point. As the arbitrary and uncritical reduction to one type of signs is the means through which unification is achieved, this strategy can be considered to be dogmatic.

2. The "Ecumenical" Strategy

It creates several general categories of the sign and interpretation, under which it summarizes all the kinds of signs and interpretations. It was illustrated by Eco, and Enjalbert (1996). We will give a broad outline before specifying our point of view.

2.1. The general category of sign

Eco's great treatise (1975) borrows from "medieval" thought a general definition of the sign: aliquid stat pro aliquo.1 This definition appears unusable to us. On the one hand, it is static (as the stat indicates, resumed in Ogden and Richards' stands for) whereas, according to us, the sign results from the process of interpretation, for its signifier is not offered up to simplex apprehensio, but is identified within a practical process, and its signified is not intrinsic: in short, a sign can be identified only as a station on an interpretive path.
Moreover, the interpreter is a great absentee in this definition of the sign.2 Then again, it neglects the fact that the two "objects", aliquid and aliquo, are not at all unspecified. This school definition, which none of the great medieval semioticians adopts, smooths over the differences between relata (that can be two things, two signs, or a sign and a thing, a sensory or intelligible element). The only binary sign recognized by the Ancients was the semeion, an inferential sign defined as such by the rhetoric theory of forensic evidence. The canonical example is: If she is lactating, she gave birth. This type of sign forms the basis for the theory of natural signs in Augustine's thought.
The consequences of this generalization are tremendous. Thus Eco manages to fuse under the aristotelian model of signification (or semiotic triangle) all the major Western models of the sign (1973). The move is not immune to simplification: for example, that leads Eco to equate Saussure's concept and his signified, whereas Saussure's stroke of genius was precisely to make a distinction between them. Eco also juxtaposes the Saussurian signified and Peirce's interpreter (whereas these two notions have no common ground), etc.3
Lastly, this definition does not distinguish strictly between the three basic relationships (inference, difference, reference) and remains stuck with the single model of the semiotic triangle, which is certainly amenable for logical positivism (we stressed that the syntax / semantic / pragmatic division originated from it in Rastier 1990), but not very compatible with the semantics of natural languages, and at any rate not with the problems of the text, because it defines semiosis in relationship to the sign in isolation.

2.2. The General Category of Interpretation.

The difficulties it raises are no smaller than the ones raised by the general category of the sign. I will not expand here upon the interpretation conceptualized by the philosophical hermeneutics of Heideggerian descent (because it merges altogether with human life) and will focus on the semiotic problem itself.
Enjalbert resumes the general category of the sign to offer a model of interpretation reminiscent of the rewriting of aliquid into aliquo. The input (or source) as well as the output (or target) could consist in objects, percepts, in signs, or concepts.

1. Jakobson, in his inaugural conference of the Second International Congress of Semiotics, included the formula aliquid stat pro aliquo as a general description of the various types of signs. It can be found almost anywhere, as in Lalande's dictionary (1968, p.991), or Abbagnano's (1971, p.777). [RETURN]

2. Whereas it is present in Peirce, according to whom a sign "is something which stands to somebody for something in some respect or capacity" (Collected Papers, 2.228). [RETURN]

3. One extra difficulty is this: the referent (the Scholastics's res) is assimilated with the stoical tynchanon; however the Stoics had a propositional theory of signification, and the referent is a state of affairs or more precisely an action (the traditional example is Dion runs), not an object (contra Eco 1973, p.36). If Eco chooses to downplay these difficulties, or to mention them on the paraliptic mode, it is because he intends to unite the models of the sign around the semiotic triangle. His argumentation turns to the flimsy and the whimsy: "As we can see, on the basis of common sense __ isn't it the most fairly distributed of worldly goods? __ there is agreement as to how to call this tripartition, but not on the names to be given to each of the three poles thereof" (p.39). [RETURN]

Page - 1     Page + 1

AS/SA Nº5, Article 4 : Page 14 / 27

© 1998, AS/SA

E-mail to the editors
Pour écrire à la rédaction