One-dimensional thinking: The problem with Saussurean linguistics
One of the foundations of the “Theory” that still dominates the current academic scene is the work of the pioneer of semiotics, Ferdinand de Saussure (1857-1913).
It has long seemed to me that the interpretation of Saussure that guides Theory is founded upon the denial of both nature and history. I am not an expert in linguistics, however, so I invite criticism of this claim.
Saussure looked at language as a closed system, or collection of signs determined by their relations to each other, and specifically by their differences from each other. According to what seems to be the most common interpretation of his work, he saw the relationship between the elements of a sign (the signifier and the signified) as being arbitrary. The spoken word “cat” (the signifier) has nothing to do with the sound or shape of a cat. Furthermore, the signified in this case is not a physical cat, but a concept of a cat, defined in relation to other concepts.
Saussure’s idea of language is an abstraction, used for purposes of analysis. Saussure separated langue (language as a self-contained system) from parole (language as used by a specific person in a specific situation). He also focused on the synchronic (non-temporal or relational) aspects of language, and largely ignored the diachronic (temporal or historical) aspects of language.
There are a number of problems that Saussure’s ideas, and those of his followers, raise. I make no claim to originality here. Others have pointed to these problems, which tend to be dismissed or left unsolved among those who accept Saussure’s ideas as foundational.
1. The idea of the arbitrariness of the sign obviously has limited applicability to the study of the language of images such as paintings (iconic signs) or photographs (indexical signs) (to use terms borrowed from Charles Peirce). Further, it is not clear how a Saussurean should treat the elements of iconicity and indexicality that enter into the construction of letters, words, sentences, and other linguistic structures. Should these dimensions of discourse be treated as conventional as well? Some classic examples come from poetry: Ezra Pound explored the iconic dimension of language by using (semi-iconic) Chinese ideograms in his Cantos, and the Italian Futurist poets and performers made bold and extensive use of onomatopoeia. How can the sign be truly arbitrary if it contains iconic or indexical dimensions, which to all but the most committed perceptual relativists would display at least traces of non-arbitrary reference or influence? Why should we limit ourselves to understanding a language as merely a complex and self-referential, but ultimately arbitrary, “code”?
2. The signifier and signified form a sign that is supposed to be studied only within the closed system of langue. According to Raymond Tallis, if I understand him correctly, post-Saussurean theorists tend to ignore or blur the distinction between the signifier and the sign, and between the signified and external reality. They thus blur the distinction between langue and parole. (To clarify: they imagine that we live in, and are constituted by, langue, when in fact we are always outside it, even if we use it in particular speech acts. Derrida’s critique of the ‘transcendental signified’ is a reaction against Saussure, but he launched this critique after already blurring the distinction between langue and parole. He had to start with the assumption that the subject is constituted through and within language before he could begin his deconstructive project.) This blurring, as I see it, is emphasized in specific circumstances, in order to either undermine the idea of the stability or rootedness of the self (something that Tallis addresses in the work of Jacques Derrida) or simply to portray particular objects of analysis as mere sign-systems, and thus to show that these objects of analysis are arbitrary constructions that can claim no authenticity or connection to a stable ‘reality.’
3. Saussure seems to assume that systems of language have always existed. But how does a language system come into being? Which came first, langue or parole? My intuitive answer is parole. Grunts and gesticulations became primitive words, which eventually formed systems of language that can seem autonomous when seen synchronically as langue. Parole is the use of langue, but it is also its creator and developer. It would seem that langue cannot be detached from its diachronic roots in parole and lived experience in the physical world. Language is, on some level, derived from nature and from human history.
What was an abstraction for the purpose of analysis in Saussure became an absolute truth in Theory, where radically new understandings of identity and culture were developed based upon the idea of the fundamental arbitrariness of the sign. The linguistic connection to nature and history was denied so that identity could be seen as a complex semiotic arrangement, capable of dissolution and reconstruction.
For Tallis’s critique of post-Saussurean thought:
Tallis, Raymond. “The Linguistic Unconscious: Saussure and the Post-Saussureans.” Theory’s Empire: An Anthology of Dissent. Eds. Daphne Patai and Will H. Corral. New York: Columbia University Press, 2005. 126-46. First published in Tallis, Raymond. Enemies of Hope: A Critique of Contemporary Pessimism. New York: St. Martin’s Press, 1997. 272-89.