The [ ] walked down the street

Michael Silverstein

  • Saussure by John Joseph
    Oxford, 780 pp, £30.00, March 2012, ISBN 978 0 19 969565 2

Ferdinand de Saussure, who died in 1913 at the age of 55, sowed the seeds of structuralist thought that first took root in linguistics, then effloresced throughout the 20th century in fields as seemingly distinct as literary criticism, architecture, social anthropology and psychoanalysis. Yet, as John Joseph’s biography shows, Saussure struggled for his entire career to systematise as general theory what he had implicitly understood and put to stunningly successful use at the age of 19: that human language is the prime abstract ‘semiological’ (we now say semiotic) structure on which the possibility of interpersonal communication depends. He died without ever getting close to writing his magnum opus – which nevertheless exists, an ‘as if’ work imaginatively simulated in his name shortly after his death.

Saussure’s influence has thus been almost entirely posthumous: the influence, we might say, of an ancestor who has become a secular saint, to be propitiated by ritual pieties of citation and by curatorial devotion to his relics. He was born in Geneva in 1857 to a family in the uppermost echelons of local society, long prominent in science and letters. At university in Leipzig, he demonstrated precocious, controversial and prickly brilliance in comparative Indo-European linguistics. He felt himself very much to be a rival of his teachers, who, for the most part, did not appreciate the originality of his systematic reconstruction of the prehistoric ancestor language. He then settled in Paris, but his academic ambition was thwarted, and in the end he gave up hope of a professorship. He went back to Geneva and settled into a dull professional life in the provinces, teaching Sanskrit and comparative philology for twenty years to relatively few students. At the same time, however, he maintained a private, intense, seemingly anguished intellectual life devoted to his quest for the essence of language and an almost kabbalistic study of ancient poetic texts.

By the time of his death Saussure had published very little, having spent twenty frustrating years making fragmentary efforts at organising his ideas about how to study the ‘double essence’ of human language as a ‘semiological system’. First, there is a systematic character at any given moment to the way the members of a language community form and use words and lengthier expressions to communicate: he called this the ‘synchronic dimension’. Second, over the timespan of that community’s history we can sometimes document, sometimes only inferentially reconstruct, the continually changing configuration of parts in such a system: this is the ‘diachronic dimension’. Saussure’s great Mémoire, which he self-published in 1878 and which inferred the system of word structure in the reconstructed proto-Indo-European language, had in large part already demonstrated the ‘structural’ approach to linguistic form and its historical transformation that he never explicitly theorised to his satisfaction. The Mémoire’s influence, too, would be for the most part posthumous, its reconstructive inferences indirectly vindicated only in 1927.

In an extraordinary act of devotion, two of Saussure’s former students at the University of Geneva, Charles Bally and Albert Sechehaye, undertook to collate the course notes of students who were enrolled in one or another of Saussure’s three pedagogical attempts at a ‘course of general linguistics’ in 1907-8, 1908-9 and 1910-11, together with the few scribblings they could find in Saussure’s own hand. Bally and Sechehaye cut and pasted and – it has since emerged from parallel-column publication of the student notes alongside passages of the Cours – wrote much of the continuous prose themselves from phrases taken or derived from Saussure. The text they created, the Cours de linguistique générale (1916), was an orderly exposition of the field of linguistics of a then revolutionary sort. In the 1950s, scholars began a deeper archaeology of the Cours that continues still; the original students’ notes have been published and copiously annotated, and collections of further fugitive notes published. But the original fabricated book, slightly revised and repaginated by Bally and Sechehaye in 1922, is the source of the body of ideas we have ascribed to ‘Saussure’ and the basis for his pervasive influence in the human sciences.

Joseph’s massive, meticulous book is a heroic biography, tightly focused on its protagonist and with a melodramatic, if not tragic, tone of knowing retrospection. Joseph lays out the many crushing psychological, familial, social, institutional and professional forces arrayed against his brilliant, headstrong subject. Saussure was a chip off the old Calvinist block, a Genevan aristocrat of apparently impeccable, seemingly distant and modest, perhaps even condescending manners, who, unsurprisingly, functioned poorly in the intensely professionalised and competitive academic milieus of Leipzig and Paris. There is a growing sense of inevitability as Joseph’s story unfolds: Saussure dies just as the European Belle Epoque is about to give way to the cataclysms of the 20th century, which would leave the US as the dominant power in the scholarly and scientific fields.

As Joseph notes, various gossipy biographical accounts have tried to explain Saussure’s failure as a practising academic: stories of secret dissoluteness, drunkenness, schizophrenic tendencies and so on have all filled the factual void. Joseph isn’t interested in these but gives us all the particulars that can be had from the archive, more than one imagined could be so carefully documented. Seekers of detail will not be disappointed: Saussure begins in the ancient family seat in Lorraine, Saulxures-lès-Nancy, sometime before the year 1440; Ferdinand-Mongin is born, on 26 November 1857, a hundred pages in. Sixteen chapters intervene, each of them covering three or so years of his life, before his lonely death on 22 February 1913 at the imposing Château de Vufflens, his wife’s family house outside Lausanne.

Saussure was onto something big, something comparable in aspiration to what another tortured and little published genius, Charles Sanders Peirce (1839-1914), had been onto slightly earlier. Peirce created his ‘semiotic’ as an architecture of experiences-as-signs – a universe populated by properties (like loudness), occurrences (events) and regularities (seeming recurrences). He called the preponderant type of human linguistic signs ‘symbols’ because, as conventional signs, they stand not for specific things or states of affairs but for conceptual categories determining things in classes (the verbal sign ‘table’ stands for the concept ‘table’, and can therefore be used to refer to tables in general, or any table in particular) and possibly descriptive states of affairs (‘walking’ as a kind of event, whenever and wherever such an event may happen). In order that they might represent the worlds of experience and imagination, such symbols have to be put together with ‘indexical’ signs, as Peirce termed them, such as articles (the, some), demonstratives (this, those), tense-inflections (walk-s, walk-ed), moods (may/might walk, shall/should walk) etc.

Saussure’s ideas match Peirce’s in many respects. Thus Peirce differentiates between the fact of linguistic regularities-as-signs – Saussure’s langue – and the discursive activity of realising them on any given occasion – Saussure’s parole. And in his description of the special character of symbols, Peirce differentiates between the ‘arbitrariness’ or community-relative conventionality of the relation between an abstract linguistic signifier (Saussure’s signifiant) and a correlative abstract conceptual category (Saussure’s signifié). These basic axes of Saussure’s schema were neither new nor very original, all things considered. Not only Peirce but also the linguist William Dwight Whitney (whom both Peirce and Saussure knew personally) had already made such distinctions clear in trying to write about ‘general linguistics’. And, as argued by Hans Aarsleff among others, parts of this general framework had already been systematised by widely influential figures in Paris before and during Saussure’s years teaching there, for example the linguist Michel Bréal, the historian and critic Hippolyte Taine and others.

But if the proposals commonly associated with Saussure are not, in fact, so innovative, what is? For this, we must look to Saussure’s work as a comparative-historical linguist. By 1877, when he matriculated at the University of Leipzig, the linguists there were fairly confident about the power of their inferential methods. The idea was to compare the phonetic shapes of words in various languages and, on discovering cross-language correspondences in their sounds, to infer a common ancestor. Thus, in words that derive from Indo-European, the English or German f- occurs where in Greek or Sanskrit p- occurs, both sounds pointing to a hypothetical ancestral p-. The various later forms derive, over time, from the earlier form by means of distinct processes (Lautgesetze, or ‘sound-laws’). The champions of this comparative method, the Neogrammarians, considered the pronunciation of the forms of language a physical phenomenon independent of the symbolic relation about which Peirce and Saussure worried, and they thought that historical transformations of pronunciation were governed by essentially inviolable ‘laws’.

Saussure’s comparative and reconstructive study in the Mémoire, by contrast, presumed that even the ‘sounds’ of language are really an integral part of the grammatical structure of both the original Indo-European parent language and of each of its descendants. Such phonemes – as we now call them – are not sounds per se, but cognitive categories of sound: abstract, pronunciation-determining discriminations that people rely on each time they articulate or hear the words of a language. Phonemes are not merely physical entities: they are, as Peirce would say, ‘symbolic’ counters, the abstract contrasts which differentiate words and expressions in a language. For example, swap the initial k- unit of the word we write as ‘cat’ for a p- unit, and we get ‘pat’. The -k v. -p distinction recurs at the end of a syllable in ‘tack’ v. ‘tap’, but these two sounds are never found adjacent to each other in the same syllable in English, as in the hypothetical kpat, pkat, takp or tapk. The distribution of phonemes and their sequences is governed by rules, and Saussure explicitly relies on this combinatorial – rather than any phonetic – property to infer the existence of a whole otherwise unattested class of phonemes in ancestral Indo-European, with what the Neogrammarians thought to be a shocking disregard for their phonetic character. For Saussure, though, the systematic correspondences between the distribution of phonemes in the descendant languages pointed unerringly to the prior existence of a category of ‘coefficients sonantiques’ that others, decades later, would come to identify. The task of elaborating a general linguistics from Saussure’s structural view of language, which he proved unable to manage, was left to the great early generations of linguists in the 20th century, led by Edward Sapir and Leonard Bloomfield in America, Daniel Jones in Britain, and Roman Jakobson and Prince Nikolai Troubetzkoy on the Continent.

Saussure’s most important breakthrough, the one that later allowed a vigorous structuralism to flourish, was the notion that all semiotic systems – human languages above all – organise the relation of sign-form (signifier) to sign-meaning (signified) as a system of ‘values’. The term misleads because of the economic metaphor: Saussurean value is best thought of as ‘valence’ in the chemical sense. Mendeleev organised the chemical elements into classes according to their potential to combine with one another. Saussure, too, realised that every linguistic sign must be defined by two combinatory properties: first, its privileges of ordered combination with other linguistic signs (what he called the syntagmatic relations); second, its paradigmatic relations with all other signs with similar privileges of occurrence in syntagmatic combinations. Thus, in the position indicated by brackets in the sentence ‘The [ ] walked down the street,’ any one of an essentially infinite paradigmatic class of signs can occur, whether simple or complex: for example, ‘man’, ‘large shaggy dog’, ‘individual whose name I cannot now recall’. Paradigmatic classes are categories of potential units that the grammarian must assemble based on the rules of syntagmatic distribution.

In a stroke, with this structural-functional analysis of ‘value’, Saussure rendered any natural human language a formal system, langue. From the paradigmatic contrasts of linguistic forms, contrasts of meaning can be projected. Users of language may not know it as they communicate (Saussure’s realm of parole) but this structure of sense-determining contrasts underlies their intuition that such and such a word or expression refers to something in the world of experience. By the mid-20th century, students of other ‘meaning’ and ‘function’-laden areas of human experience created a structuralist moment – and movement – in the human sciences. Paradigmatic relations in one area of social life, they realised, have analogous relations in another, orienting the way in which we identify ourselves – and are identified – with cultural values. Think of political affiliation. In an earlier period of American politics, one could be accused of being ‘a red’. But notwithstanding its origins in the colour of a historical revolutionary banner, this made sense to Americans only as opposed to being ‘[true] blue’, where red is to blue as communist ideology is to capitalist ideology as was the Soviet Union to the United States. (Socialists, when people differentiated, were then thought of as ‘pink’.) In the post-Soviet era the differential structure of political coloration has been renewed, but now America’s ‘red states’ vote Republican while its ‘blue states’ vote Democrat.

Claude Lévi-Strauss, Edmund Leach, Mary Douglas and Marshall Sahlins have pointed out that our normative ‘totemic’ systems of colour – urban gang colours, sports team colours, school colours, gender colours etc – are precisely analogous to those of the indigenous groups traditionally studied by anthropologists: we no less than they inhabit a social world that makes sense on the basis of a world of metaphors and metonymies that we happen to live in as well as to conceptualise. For people the world over, the myths we tell ourselves, our children and our psychiatrists, and the rituals we perform and attend – social texts in words and action – actualise the never-ceasing dialectic of social life, where the langue-like semiotic ‘code’ is always being combined, and recombined and potentially changed, by the parole-like creations we dream up.