Selections from Problèmes de linguistique générale by E. Benveniste

Émile Benveniste

Problèmes de linguistique générale

Vol. II. Paris: Gallimard 1974

Table des matières

Note that all my comments are from the point of view of language as interaction (in a discourse analytic and ethnomethodological fashion) and are thus anachronistic given that Benveniste wrote before these traditions became public and in explicit acknowledgement of Saussure. I have to check whether these traditions ever discussed Benveniste. I am testing the argument here that much of what he says about the processes of "saying" in the present when something has to be said is actually quite compatible, and in fact necessary, to establish the theoretical foundations of these traditions.

On syntagmatic relationships and syntagms

Definitions, sort of

... si notre langue nous donne le moyen de construire des phrases, c'est que nous conjoignons des mots qui valent à la fois par les syntagmes* et leurs opposition.

* (footnote) Groupe de mots formant une unité à l'intérieur d'une phrase. (p. 32)

 

 

"syntagme" ... s'applique à n'importe quel groupement, même occasionel, opéré, par des fins syntaxiques (p. 172)

 

both of these quasi definitions appear in passing (the first one in a footnote) and in the context of a discussion about something else. The second is actually the most enlightening as it appears in a discussion of compound words that have themselves become words even when the parts are recognizable (e.g., in English, "jetplane," "rock-n-roll," "bloodless"). The issue is that syntagms are about saying something through selection and combination. In other words the two sentences "I like rock-n-roll" contrasts to "I like country and western" through the paradigmatic choice of two words for types of music that, by themselves, do not say (that is, do) anything. A word by itself is a potentiality ("un transfer au virtuel" p. 161), a syntagm is an act.

As B. states it:

la sémantique résulte d'une activité du locuteur qui met en fonction la langue... la phrase ... n'est que particulière... avec la phrase, on est relié aux choses hors de la langue ... le sens de la phrase implique référence à la situation du discours, et l'attitude du locuteur.

... le sens s'établit dans et par une forme spécifique, celle du syntagme, à la différence du sémiotique qui se définit par une relation de paradigme. (p. 225)

 

This is also another way of saying what Merleau-Ponty began writing in The Prose of the World

On the syntagmatic

Il y a donc deux propriétés inhérentes à la langue... Il y a la propiété qui est constitutive de sa nature d'être fromée d'unités signifiantes, et il y a la propriété qui est constitutive de son emploi de pouvoir agencer ces signes d'une manière signifiante... Entre ces deux propriétés le lien est établit par un troisième propriété ... la propriété syntagmatique, celle de les combiner dans certaines règles de consécution et seulement de cette manière. (p. 97)

 

Here the syntagmatic becomes a THIRD property (I think that B. is thinking of morphology and syntax as the first two properties) but remains that aspect of language that actually produces something that is said about something by someone. On this page B. uses the word "interpretant" (but does not mention Pierce) and makes a statement that prefigures much post-modernism (and may have to do with why B. is a favorite linguist among many)

C'est grace à ce pouvoir de transmutation de l'expérience en signes ... que la langue peut prendre pour object n'importe quel ordre de données et jusqu'à sa propre nature. Il y a une métalangue, il n'y a pas de métasociété. (p. 97)  

An instance of syntagmatic relationships

il frappe ~ il a frappé
il frappe ~ il est frappé
il frappe ~ il peut frappé
(p. 179ff)

 

This is followed by a long discussion of the ways through which various possibilities can be specified through syntactic properties. Here again, "meaning" as saying something with consequences is built up by the sequencing of various parts of speech. In syntax the framing is extremely strict and strongly constrains "what can easily be said" (Sapir/Whorf). But much can be said--and other means are available to say what cannot easily be said (e.g. the evidence on which any of these statements are made that can be marked syntactically in certain American Indian languages).


July 20,, 2007