I am off work sick, my head feels as if it is stuffed with cotton wool, I have barely moved from bed all day and am too tired to stay awake and too awake to go to sleep, and to cap it all, yesterday when I started writing this we had a (minor) earthquake. Obviously, I am in the perfect frame of mind to tackle …
1: Ferdinand de Saussure, The Object of Study/Nature of the Linguistic Sign
The first thing to note is that we are not actually reading Saussure. The book, Course in General Linguistics (1915) upon which his reputation rests, was put together after his death by his students working from his lecture notes. And, moreover, these extracts are from a translation, (by Roy Harris, 1983). What’s more, if there is one thing I know about Saussure is that the central idea in his work relates to the signifier and the signified, yet Harris has chosen to translate those terms as ‘signal’ and ‘significance’.
Somehow, this layering of baffles and interpretations between what Saussure said and what we hear seems entirely appropriate: his work seems to me to be intimately concerned with the distance between what is said and what is understood, and the processes by which we bridge that gap.
It is worth pausing to consider exactly what Saussure was doing. Whenever I have come across his name and his ideas it has been in connection with philosophy and literary studies, yet Saussure was neither a philosopher nor a literary theorist. He was a scientist concerned with language, at the time it was known as philology and based on historical principles; for Saussure, ‘linguistics’ was a new branch of language study he was opening up based on scientific principles. From what I can gather, outside the field of linguistics itself his work was first taken up in Russia, it didn’t become widely known in Western Europe until after the Second World War, and in Britain and America he didn’t become known until the 1970s, probably the late-70s. (When I was studying philosophy in the early 70s his name was never mentioned, even though linguistics was a constant point of reference in post-Wittgensteinian plain language philosophy.)
The breakthrough idea that set Saussure’s theories in motion was that language did not replicate a real-world object. The underlying theory, indeed more like the unquestioned assumption, behind all language study to this point was that words named objects. Hence the way speech works is that you look at an object, you say the name of that object, the name triggers in the mind of the hearer an image of the object, and that is how the speech is understood. (I note that Saussure does seem to refer to other forms of communication, writing, sign language, etc, but all the models presented within these two extracts are based on speech, and I am not familiar enough with the rest of Saussure’s work to know how well his models really do apply to such other forms of communication. I’m not even sure whether this is a relevant object of concern.)
Saussure argued that when a word is spoken several different things are going on at the same time. There is the physical sound, uttered and perceived; there is the mental action that identifies the sound with an idea, by both the speaker and the listener; there is the cultural tradition that assigns the word to the idea, and this social aspect in turn has both an historical evolutionary aspect and a current practice aspect. Given this complex view of language, Saussure concentrates on the structure of language, but even this is a multi-valent thing with much going on in the same action. He argues, for instance, that ‘it is not spoken language which is natural to man, but the faculty of constructing a language, i.e. a system of distinct signs corresponding to distinct ideas’ (4). It follows, therefore, that any individual language (one of the slight confusions I find in these extracts, or perhaps more accurately in this translation, is between language as a whole and a language, such as English, French, Latin) is a social construction independent of any particular speaker. In other words, any conversation between two speakers consists of three elements, the speaker, the listener and the language.
I find this notion completely unproblematic. It ties in, for instance, with Wittgenstein’s discussion of a private language in Philosophical Investigations, though I’m not sure Wittgenstein would have been aware of Saussure’s work. What bothers me more is that the ideas that emerged from Saussure’s work, both in philosophy and literary theory, are often presented as being antithetical to the anglo-saxon tradition of linguistic philosophy, and at this stage in the game I cannot see why that should be. Anyway, on with Saussure …
Having identified language as an independent social construct, the field of linguistic study now open to him is to examine the structure of that construct, and this he analyses as consisting of a system of signs, to the study of which he gives the name ‘semiology’. He is very careful at this point to emphasise that since language is not nomenclature, ‘a linguistic sign is not a link between a thing and a name, but between a concept and a sound pattern’ (10). Notice that by this stage common everyday ideas like ‘words’ are disappearing from the discussion, I think this is because Saussure wants to avoid any suggestion of a direct connection between word and object. (Somewhere I have, or had, a book by W.V.O. Quine called something like Word and Object, which I haven’t read because I never really got on with Quine. It occurs to me that herein might lie some suggestion of that antithesis I was musing about a paragraph ago.)
So a sign is made up of a concept and a sound pattern, or the idea we wish to express and the expression. This is where Saussure introduces two new terms: the concept he terms the ‘signified’ and the sound pattern the ‘signifier’ (Harris uses ‘signification’ and ‘signal’ respectively, but for the sake of my own sanity I am going to stick with the terms more commonly used by those who came after Saussure). My interpretation, then, is that the basic linguistic unit is a ‘sign’ which in turn consists of the thing pointed to, and the way we point at it.
For a while I had difficulty working out how ‘signifier’ and ‘signified’ differed from ‘word’ and ‘object’. It was only when I looked more closely at what he meant by ‘sign’, rather than its constituent parts, that I realised how fundamental the difference is. A word is both a mental idea and an utterance; an object is thing out there in the real world. So word/object is, if you like, a linguistic version of the old mind/body problem. But Saussure faces no such problem because the sign is a purely mental object; both the signified and the signifier take place in the mind. The sound pattern isn’t pointing to a thing out there in the real world but to a concept in the mind, a concept which may or may not have any relationship with any real world object.
This makes sense to me. But as a rationalist empiricist rather than an idealist, I like to think that there is a world of physical objects outside my body, including the physical object of another person with whom I engage in the social construct of language. In other words, to identify language as a social construct the way Saussure does is to imply the existence of a real objective world. At some point language must relate to that real world; what society constructs is actually the character of that relationship. So it would seem that in making the sign entirely mental, Saussure is only pushing the mind/body problem of language one step further away.
This isn’t really an issue for Saussure: his concern is to construct the idea of a science of language and to specify the object of such enquiry. Thus the nature and structure of the sign and how it operates in individual languages is something that can now be examined with scientific rigor. With two principles that the sign is arbitrary, i.e. there is no natural connection between signifier and signified in reality; and that the signifier is linear these two extracts come to a close.
The implications of Saussure’s work for literary theory appear to be in part a matter of analogy (that you can produce a scientific analysis of an underlying literary structure in the same way that he called for a scientific analysis of an underlying linguistic structure), and in part a direct consequence (that literature like language is a social construct). Since I can buy in to Saussure’s linguistic analysis, I think I can also buy in to the second of these. But I suspect I will still need some convincing of the value of the argument by analogy.
First published at LiveJournal, 4 March 2009.