Major Authors

The Linguistic Sign The linguistic sign was first defined by the Swiss linguist Ferdinand de Saussure in his Course in General Linguistics in the early 20th century. Saussure addressed the seemingly obvious but also somewhat revolutionary insight that words are not the things they attempt to represent (contrast, for example, the beliefs of the Symbolist Andrei Belyi, who is a contemporary of Saussure). Language does not function through some sort of special connection between words and things; rather, language, for Saussure, is a relatively self-contained system with its own rules. Language is made up of signs, which have little to do with the referent, the actual objects in the world. The signs themselves are composed of two parts: the signifier (the letters on the page or the sound that bounces off our eardrum) and the signified (the concept that appears on our brain when we read or hear the signifier). Be careful not to confuse the signified and the referent--the signified is the product of our mind produced when we process a signifier; the referent is the thing out in the world (whatever that might be) that this signifier is trying to evoke. What Saussure goes on to demonstrate is that the linguistic sign is arbitrary, conventional, and differential. The sign is arbitrary because there is no reason that the letters "c-a-t" (or the sound of those phonemes) produce a four-legged domesticated feline on our brains. There is nothing special or inherent in those orthographic symbols or auditory impressions that is somehow mystically connected to the nature of cats. Most of us imagine a cat when we read "cat" as a result of convention: as users of the same language, we have agreed (or been told) that those letters or sounds mean certain things. If I write "lepidopterist," it is likely that nothing (or at least nothing specifically relevant) pops into your brain because there is nothing about that series of letters that would suggest someone who studies butterflies. Moreover, the linguistic sign is differential (and this is an insight that will be explored with much more radical implications by Derrida and the Deconstructionists). The way in which we process signifiers has nothing to do with their special relationship with signifieds, much less referents, but is a result of their difference from other signifiers. When you say "cat," I produce a certain signified, as opposed to another one, because you didn't say "pat." Because of the incredible marvel that is the human brain, I am able to process in a nanosecond the possibilities that those phonemes could produce with variation (pat, can, cad, etc.) and then situate the correct signified in the system. Perhaps the best way to get a handle on Saussure's insight here is to show difference at work: why is it that a parent seems to understand a toddler's non-standard language use when others hear only gibberish? Aside from some of the non-linguistic elements involved, the primary reason is that the parent understands the child's developing differential language use. The child may always substitute a certain letter/sound for the proper one; but because it is regular, and different from other possible substitutions, the parent can comprehend it, while those not familiar with the child's linguistic system are befuddled and marvel at the parent's interpretive capabilities. This same phenomenon can be seen with non-native speakers of a language. Why is it that we can understand people with so-called accents? Even though they may speak in "non-standard" pronunciation, they do so within the limits allowed by the language (otherwise we would not be able to sense the necessary possible differentiation). Non-native speakers, not having fully internalized the full system of differences and possible substitutions of the language, can have tremendous difficulty with non-standard language, even though they might have a tremendous vocabulary and grasp of the grammar of the language.

Major Authors

Revised: May 21, 2003

Contact: Prof. Christine Roth or Cary Henson