Where structuralism comes from

With Chomsky seemingly off the stage – exit left, the script reads, brooding on the sins of American foreign policy – it is now or never for Ferdinand de Saussure to take his place. One theorist of language at a time is probably all the popular awareness has room for, and over the past twenty years Chomsky has been it, investing grammar with a new, deep-seated charm, and bringing us to see language as a mysterious acquisition which does our species, and our brains, much credit. But as a linguist at least Chomsky has gone quiet, and even before that so much of his work had turned into a kind of algebra it could no longer make sense to the world at large. This is a fine moment, therefore, to exchange Chomsky for a theorist who saw language whole and in terms accessible to all: for Saussure, dead these seventy years, yet still, how wrongly, under a cloud in the English-speaking world.

He had to wait 43 years to get into English in the first place: it was 1959 before an English translation was provided of the epochal Cours de Linguistique Générale, which had itself been published three years after Saussure’s death in 1913. The view of language which this remarkable book puts forward was new and radical then, and will strike many coming to it as new and radical today. A work so disruptive of orthodox ideas about language and its workings was bound to travel slowly and erratically. By 1939 it had been translated only three times: into Japanese, in which language it flourished, into German, and into Russian – an edition of 1933 which once made was not heard of again, having, one can suppose, run foul of that opinionated, jealous and vindictive theorist of language, Joseph Stalin.

In this country, whatever future lay open for Saussure early on was blighted by the remarks made about him by C.K. Ogden and I.A. Richards in The Meaning of Meaning, where he is dead and buried by page six, charged with having ‘concocted’ the chief object of his inquiry, la langue, and of being at one with A.N. Whitehead in naively falling for the ‘Method of Intensive Distraction’. Saussure’s definition of language (langue) as an abstract system of which actual languages were the living evidence was too platonic to be stomached by the behaviouristic Ogden and Richards. And throughout the Twenties and Thirties, Saussure’s forthright ‘mentalism’ put him quite out of court for those linguistic thinkers, the influential Leonard Bloomfield foremost among them, who had set a taboo on Distraction, intensive or otherwise, and saw language as the sum total of real or potential verbal utterances. Where Saussure worked by abstraction, to the hypothesis of linguistic structure, the Bloomfieldians worked by addition, towards an inventory of everything that could be said or written.

By the time that the Cours was finally translated into English, at the end of the Fifties, Behaviourism, now in its even more fanatically empiricist Skinnerian guise, was on the run before the brilliant new mentalism of Chomsky. Saussurian ideas began to get more of a look in, though there were few British or American linguists to call themselves Saussurians. Chomsky referred to Saussure, generously at first, a little carpingly later, as a precursor. In fact, Chomsky’s own cardinal, much seized-on distinction between our ‘competence’ as users of language and our ‘performance’ is the selfsame distinction as that made much earlier by Saussure between langue and parole, or between the abstraction which he believed was the proper object of a linguist’s study and actual manifestations of language in speech or writing. Like Chomsky, Saussure wanted to get away from what he took to be the superficiality of mere databanks of linguistic facts: instead he wished to fix the rules, valid for all languages everywhere, without which there could be no language. His structuralism is psychological, as Chomsky’s is, and oppositional, or differential. It is from the Cours de Linguistique Générale that structuralism comes, in all its various applications (and from Jacques Derrida’s relentlessly acute interrogation of certain inconsistencies and biases in Saussure that post-structuralism comes).

Given the strong measure of overlap between Saussure and Chomsky, and that both of them look on language as our richest, most profound and most influential institution, there is good reason to hope that the one might now step into the place of the other, as the theorist of language from whom mere practitioners can best get their ideas about language. The appearance of a second, conspicuously more expert translation of the Cours will help a great deal. The arguments of the book aren’t easy, but neither are they obscure or brazenly technical. Saussure writes about language as we all know and use it, and no one could read him without coming to think more rationally and profitably on these basic matters. The Cours, moreover, is the more easily learnt from because it is not exactly a book: it is a reconstruction from the notes taken by his pupils on the lectures on general linguistics which Saussure gave at the University of Geneva between 1907 and 1911. The work could legitimately have been titled in English ‘Lectures on ... ’ instead of Course in General Linguistics, which sounds all too drably Teach-Yourself.

At Geneva, his native city, to which he came back in 1891 after teaching for ten years in Paris, Saussure was first professor of Sanskrit and Indo-European Languages, then professor of German Language and Literature, and finally, after 1907, professor of General Linguistics. His translator, Roy Harris, is also a professor of General Linguistics, at Oxford, having earlier been professor of Romance Languages there. He is Oxford’s first professor of General Linguistics, not because he has occupied that chair for an indecently long time but because it was an indecently long time before Oxford set it up: in 1978, to be precise, 71 years after Geneva had seen the light and realised that linguistics is the study of language, not of languages.

Saussure himself showed signs of impatience when he was still at school to pass from being a student of languages to a student of language. At 15, having by then got English, German, Greek and Latin to go with his French, he submitted a paper to one of his senior teachers called, bravely, ‘Sur les Langues’, in which he maintained that certain phonetic combinations were to be found in words bearing similar meanings in all languages. This presumed universal link between the phonetic and semantic aspects of words is one which he denies absolutely in the Cours, whose central principle it is that the linguistic sign is arbitrary, related only conventionally to the thing in the world it is the sign of, its form being therefore determined by culture, not by nature. On the other hand, he did not go back on his ambition to derive general rules for language as such, and then test them against known languages. The mentor to whom he had shown his essay deplored such callow universalism, which went entirely against the prejudice of the later 19th century for the historical and positivistic in language questions. (That prejudice long outlived the 19th century. Thirty years ago, as a student of French at Professor Harris’s own university, the nearest I came to general linguistics was a lugubrious initiation into the history of French. What language itself might be was something better left unasked; and Saussure’s was a name never then heard.)

It was an act of considerable philanthropy on Saussure’s part to work to undo this stifling and exclusive concern with evolution in language. Linguistics of that sort is merely erudite: it will not engage with the deeper, more philosophical issue of how language works. Saussure paints it almost as facile:

Generally speaking, static linguistics is much more difficult than historical linguistics. Facts of evolution are more concrete, and stir the imagination more readily; the connections link series of terms which can be easily grasped. It is simple, and often entertaining even, to follow through a series of linguistic changes. But a linguistics concerned with values and co-existing terms is much harder going.

He knew what he was up against in thus trying to wrench linguistics away from the diachronic to the synchronic view, to study language as a going concern, at a given moment in time, and regardless of how it came to be in its present state. Linguists who were scholars and not thinkers were incapable of making the change, and the institutions where they were strong resisted Saussurianism – there are surely pockets of retarded philologists who do so yet. More interestingly, Saussure also had to contend against a certain romanticism – ‘imagination’ he calls it – for which language is the bearer of all manner of heartening messages from the past and therefore to be evaluated retrospectively.

It is still change that stirs, if not the imagination of lay commentators on language, then their bile; most of them, though they may not know it, are diehard diachronicists, for whom etymology is also explanation. They are oblivious to what Saussure has to say about ‘values and co-existing terms’. The values of words are not the same as their meanings, although it is not always easy to stop the two from merging. Meaning is reference, a word’s relation to what is outside language: value is one word’s relation to other words, its place within the language. If meanings are stable – a horse is always a horse, so to say – values are not, they are determined by other words: a horse is not, semantically, quite the same as a charger or a steed, nor, phonetically, as a house or a hearse. Any change in a language affects more than the item which changes, it affects the value of other, adjacent items. So to relate it backwards, to its superseded form, rather than sideways, to co-existing items, is to miss the real effect of linguistic change. Contemporary changes in vocabulary or usage frequently enrich the possibilities of the language by providing us with alternative forms to use, and no two alternatives can have exactly the same value. The person who says ‘cheers’ and not ‘thank you’ achieves whatever value we assign to ‘cheers’-sayers.

This preoccupation with linguistic values lends a powerfully social side to Saussure, and to structuralism, a social side which for Professor Harris is just about the most attractive thing there is in the Cours. He makes two dramatically large claims at the start of his quite short introduction: that the book ‘occupies a place of unique importance in the history of Western thinking about man in society’; and that it is ‘without doubt one of the most far-reaching works concerning the study of human cultural activities to have been published at any time since the Renaissance’. These are not claims that could ever sensibly have been made for Chomsky, let us say, the psychological implications of whose work are not so simply extended into the social and the cultural. In the case of Saussure, Harris’s claims may be polemical, since he is set on rescuing this transcendent thinker from relative eclipse, but they are also defensible, because of the vast applications which can be made of Saussure’s insights into the nature and operations of language.

Saussure himself did not live to apply them, but he foresaw, famously, the hugely enlarged scope of his new linguistics as part of semiology:

It is therefore possible to conceive of a science which studies the role of signs as part of social life. It would form part of social psychology, and hence of general psychology. We shall call it semiology (from the Greek semeion, ‘sign’). It would investigate the nature of signs and the laws governing them. Since it does not yet exist, one cannot say for certain that it will exist. But it has a right to exist, a place ready for it in advance.

Three-quarters of a century later, semiology, the study of signs, exists, but tenuously, if the test of its existence is the number of places where it is taught and the number of those practising it. The world contains an infinite number of signs, and only a handful of semiologists. If there are not more of them, this is because semiology is an oddly centrifugal study, one for poachers or parasites, who may set up shop with their specific techniques and axioms in any academic discipline at all and throw much light on how it is conducted without being able to claim any centrality or superiority for their own discipline in the process. Roland Barthes, who died as professor of Literary Semiology at the Collège de France, likened the semiologist’s role in his inaugural lecture to that of a wild card in the university pack. Semiology is everywhere and nowhere: there can be a semiology of music, of art, of anthropology, of mathematics, of all the multiple sign-bound activities which add up to culture. Its inspiration is the knowledge that everything cultural is conventional and thereby interpretable.

There can even be a semiology of semiology, which is not exempt from investigation by its own methods. The meaning of the word ‘semiology’ is the study of signs, but that is not its value, as one can soon see from considering the title of Sandor Hervey’s book, Semiotic Perspectives. Semiotics, or sometimes, more archaically, semiotic, without the s, means the same as semiology: it, too, is the study of signs. Its value, however, is other. To call yourself a semiologist is to ally yourself with Saussure and with the mainly European tradition he has led to. To declare yourself a semiotician is to post an affiliation with the alternative and rival North American tradition of sign-study which descends from the logician and polymath C.S. Peirce. Semiology and semiotics have diverged to the point, it is said, of incompatibility: if this is so, it is deplorable, given how little headway they have managed to make in all this time.

If they are to do better in the future, it will be in spite of books like Semiotic Perspectives, whose grouchy scientism is a throwback to Ogden and Richards. Dr Hervey draws, as they did, a confidently sharp dividing-line between technical and emotive language, becoming as he does so emotively technical. Hervey is very much a Peirce, not a Saussure man. The latter’s Cours de Linguistique Générale has never struck me as a very emotive volume, but a bare-faced mistranslation can always make it appear so: where Saussure writes, neutrally enough, of semiology studying the life of signs ‘au sein de la vie sociale’ (or, in Harris’s version, ‘as part of social life’), Hervey pretends that the English equivalent of that innocuous au sein de is ‘nurtured in the womb of’, admits that that may be a trifle flowery but allows it to stand, as an imputation against Saussure’s scientific standing. Admittedly, Hervey has come to semiotics from the study of Chinese, but should we trust a man who does not discriminate between a bosom and a womb?

Semiotic Perspectives is a run-through of the doctrines and formulations of Saussure, Peirce and other semiologists or semioticians. It is not the plain and unprejudiced introduction to the subject we could do with. Hervey is unnecessarily anxious that semiotics should not only be a science but be seen to be a science, and his pages are full of diagrams and tables, many of which do nothing for his argument but are mere emblems of scientificity, like the dreadfully ugly new terms which he has found the need for in order himself to qualify as an innovator in this crowded domain. Not only should semiotics be able to do without barbarisms like ‘plerematic’ and ‘cenotactic subsystems’: it will have to do without them if it is to keep any appeal at all.

And appeal it can, in ways which Hervey would like to stamp out. His greatest scorn in Semiotic Perspectives is directed at the most appealing semiologist we have yet had, Roland Barthes, whose superb, provocative, less than scientific Mythologies Hervey finds no better than a ‘parlour-game’. It seems to me, on the contrary, that it is this very parlour-game aspect of semiology that will be the saving of it, and its lifeline to the intellectual world outside. Semiology must be an art, whether or not it one day proves to be a science. We may none of us practise it with the verve and intelligence which Barthes had, but there is insight and enjoyment to be got from trying. The point of the art is to read the products of one’s culture for their value, not just for their function, since all of them in greater or lesser degree bear messages hidden from the innocent, pre-semiological eye or ear.

Culture is all signs, and signs – this analysis is close to being the heart of the Cours – are twofold: they are at once sensible and intelligible, material and mental. When we use signs, as we do all the time, we signify, because signs are the socially accredited bearers of meaning and value; they are an institution with which we have to comply if we are to communicate at all. The lay semiologist needs little apparatus, beyond the ability to keep the two aspects of the sign separate from each other. Saussure’s own names for these two aspects were the signifiant and signifié, or, as they’ve customarily been known in English, ‘signifier’ and ‘signified’. Roy Harris translates them differently, as ‘signal’ and ‘signification’; he does not say why. As a pair ‘signifier’ and ‘signified’ have advantages which ‘signal’ and ‘signification’ don’t: they look suitably inseparable and they are exclusive to semiology. ‘Signal’ and ‘signification’ are less obviously related and they have uses outside the analysis of signs. ‘Signal’, also, is a notably more active term than ‘signifier’, with clear connotations of intentionality: but not all signs by any means seem to signal to us, and it is often the reticent ones that best repay investigation.

Harris’s departure from the standard translation in this instance is odd and unhappy: but his translation as a whole is the one to use from now on. It serves Saussure well, though there is not a lot to the annotation promised on the cover: Harris’s notes are few and abbreviated, the edition an austerity one. But in an age when professionalism threatens to be the death of linguistics, it is a supreme virtue of the Course in General Linguistics that it can be read without help.