Rastier, François: "On Signs and Texts"


3.3. Interpretation cannot be given a single model, because it is the work of situated subjects (Rialle 1996), and because __ despite Grice, or Sperber and Wilson __ the interpretive situation cannot be qualified in transcendental terms (Rastier, forthcoming).

IV. The issue of the text and of semiotic performances

Even if it is agreed that texts are the empirical objects and at the same time the objects of knowledge of linguistics, to what extent are textual issues relevant to the whole of semiotics, and not only to linguistics as defined as a semiotics of all natural languages.
The text is merely a kind of semiotic performance.1 In the next section, we specify what the primacy of text and semiotic performances owes to the primacy of practice in knowledge.

1. The semantic approaches to texts

Linguistic theories fall into five categories.2
a) Theories resulting from formal semantics, among which Kamp's is the best known. Their technical complexity is remarkable. On the other hand, they do not lend themselves to a comparative description of texts. For example, the concept of genre could not be reformulated within this framework. Descriptions, in fact, do not extend beyond units larger than the paragraph.
b) Pragmatic-enunciative theories resulting from discourse analysis. They have attempted to identify enunciative marks, such as pointers, to classify speech acts, to study the argumentative structures of texts, when it can be done. Those theories lend themselves to the analysis of micro-sociological interactions, in particular to the study of conversations. They are related to certain oral genres, but are not pliable enough to be the basis for a typology of texts, especially because they assert a transcendental definition of communication (Grice refers to Kantism; Sperber and Wilson petition the a priori principle of communicative relevance).
These the first two types of theories are compatible with one another, at least because pragmatics and formal semantics are germane within logical positivism.
c) The semantic theories resulting from the Saussurian current (Coseriu, the first Greimas). Their favourite haunting grounds are lexical semantics, isotopy theory and the analysis of narrative. Within this general framework, the program of interpretive semantics aims at integrating the three planes of linguistic description (word, sentence, text) as different stages of complexity, but nonetheless describing them in terms of the same basic concepts, that of "seme"3 among them.
d) The material hermeneutics sketched by Schleiermacher in his time, whose program was resumed by authors such as Szondi or Bollack: it is an ambitious form of philological hermeneutics that has remained incomplete.
e) The "rhetorical" theories resulting from the study of specialized languages (Swales, Bhatia). Most of them are tied to the anthropological tradition (Firth, then Halliday), they describe in detail the diversity of texts in terms of socialized practices, and have accumulated invaluable observations on the language of the law.


1. We would like to highlight some of the semes in the word: "expertise", "completion", "public nature", and thereby designate the semiotic level of a social practice. Performances are accomplished, because they correspond to practical sessions (e.g. a sermon, a thesis assessment), whereas practices (like religious or academic liturgies) have unspecified durations. Nonetheless, by virtue of this completedness, semiotic performances do not become objects we can interpret in their own terms, without resorting to the conditions of their production and interpretation. [RETURN]

2. In this section I am using elements from Rastier 1996b. [RETURN]

3. The reason why it is also qualified as differential and unified, which characterize its method. Interpretive refers to its epistemological prospect. [RETURN]

Page - 1     Page + 1

AS/SA Nº5, Article 4 : Page 17 / 27

© 1998, AS/SA

E-mail to the editors
Pour écrire à la rédaction