General Grammar vs. Universal Grammar: an unbridgeable chasm between the Saussureans and Chomsky - Archive ouverte HAL Access content directly
Conference Papers Year : 2018

General Grammar vs. Universal Grammar: an unbridgeable chasm between the Saussureans and Chomsky


As Harris (2003: 152-170) has illustrated -especially by his concentration on the notion of ‘creativity’- ‘Chomsky the Saussurean’ is nothing but “an academic fable”. This fable is a result of misreading –by Chomsky himself (1964) and also by others-, assimilating Saussure’s la langue (in the singular form) with the generativist concept of ‘competence’ and, therefore, its grammar with the Universal Grammar (UG). Chomsky’s approach to deviant utterances, from his standpoint of individual psychology, never brings him to a concept of ‘grammar’ which function would also be to explain poems, puns and any kind of wordplay. The contradiction here is that, on one hand, he claims just to speak about the ‘individual faculty of language’ (which can lead to an infinitive number of individual grammars), and on the other hand, his aim is to discover the UG which means “a framework of principles and elements common to attainable human languages” (Chomsky, 1986: 3) (which for him would be a concrete unique absolute one). This situation leads him to assume a completely transcendental postulate which claims that all human beings share an innate, genetically determined language faculty that contains/knows the rules of UG. As a result, Chomsky and others in the huge generativist camp concentrate their attempts on the search for a vouchsafed ‘universal rule-and-concept system’ which is a reproduction of an old traditional dream of a ‘universal language of thought’. In a Saussurean perspective, this assumption, aside from its failure to observe the diversity and specificity of languages, is a metaphysical and, therefore, an incoherent basis for linguistic theory. Basing linguistic theory on language acquisition or biological facts is not at all acceptable for Saussure because any understanding of the faits de parole and ‘substantive facts’ presupposes an understanding or an implicit definition of language which is, in Chomsky’s case, the common modular understanding of language. The generativist modular conception of language, therefore, turns a deaf ear to the fundamental problems propounded by Saussure concerning the very essence of language, especially the arbitrariness of linguistic sign. How does Saussurean linguistics define its own grammar as General Grammar, and how does it deal with the common or universal linguistic facts which are the main goals for the Chomskyans? In the present study, this is a main concern, which I intend to consider as a matter of ‘algebra of language’ and also as a question of typology by investigating the few indications by Saussure in the CLG and ELG and the explications given by Hjelmslev. From this perspective, we will find only the universal arbitrary structural rules in our search for a general framework/calculus susceptible to describe all possible languages and language types. Finally, in agreement with Harris and De Mauro, we will claim that the Chomskyians and the Saussureans are in two fundamentally different paths in dealing with grammar, where the latter –and notably Hjelmslev- provides a broader possibility for theorizing language.
Fichier principal
Vignette du fichier
CLG2016-Shakeri-SE103.pdf (1.08 Mo) Télécharger le fichier
Origin : Publisher files allowed on an open archive

Dates and versions

hal-01773237 , version 1 (27-04-2018)


  • HAL Id : hal-01773237 , version 1


Mohammad Amin Shakeri. General Grammar vs. Universal Grammar: an unbridgeable chasm between the Saussureans and Chomsky. Le Cours de Linguistique Générale, 1916-2016. l'émergence, Jan 2017, Genève, Switzerland. pp.3-10. ⟨hal-01773237⟩
319 View
541 Download


Gmail Facebook Twitter LinkedIn More