ANNE HÉNAULT
Saussure (1857–1913) is the scientist, who, educated in sciences (mathematics and physics) as well as in philology and in the linguistics of his time, thought it necessary to transform this discipline into a hard science, by giving it a really rational basis. Linguistics, which had started to come together with the works of von Humboldt, Rémusat, de Chézy, Bopp, Steinthal amongst other authorities, could not, according to Saussure, constitute a rigorous and specific knowledge about language, expression and circulation of codified meanings unless an axiomatic approach was settled:
From the outset, we differ from the theoreticians who think that the point is to give an idea of the phenomenon of language or from those, already less numerous who endeavour to fix the operations of the linguist within these phenomena. Indeed, our point of view is that the knowledge of a phenomenon, or an operation of the mind, requires the previous definition of a term, whatever; not a definition at random, that can always be given of a relative term with regard to other relative terms, eternally running round in a vicious circle, but a consistent definition, starting at a given point from a base which I would not consider absolute, but explicitly chosen as an irreducible basis for us and central for the system. To imagine that in linguistics one can do without such an healthy mathematical logic, under the pretext that language is something concrete that ‘becomes’ and not something ‘abstract’ that ‘is’, represents to my mind a deep error.1
1 Long quotations are offered in the original French in the footnotes, as here. All English renderings of extracts from the Cours de linguistique générale are from Harris’ 1983 translation, hereafter cited as Saussure (1983), with the original French coming from Cours de linguistique générale (CLG 1916, 1922, 1972). Quotes are also taken from the critical edition of the Cours de linguistique générale (CLG/E 1967– 1974), rendered in English by the author. Likewise, English renderings from Saussure’s Ecrits de linguistique générale (ELG 2002) are those of the author.
Short quotations appear in English and, in brackets, in French in the main text.
Nous différons, depuis le principe, des théoriciens qui pensent qu’il s’agit de donner une idée des phénomènes du langage, ou de ceux, déjà plus rares, qui cherchent à fixer les opérations du linguiste au milieu de ces phénomènes. Notre point de vue est en effet que la connaissance d’un phénomène ou d’une opération de l’esprit suppose préalablement la définition d’un terme quelconque; non pas la définition de hasard qu’on peut toujours donner d’un terme relatif par rapport à d’autres termes relatifs, en tournant éternellement dans un cercle vicieux, mais la définition conséquente qui part à un endroit quelconque d’une base, je ne dis pas absolue, mais choisie expressément comme base irréductible pour nous, et centrale de tout le système. S’imaginer qu’on pourra se passer en linguistique de cette saine logique mathématique, sous prétexte que la langue est une chose concrète qui ‘devient’ et non une chose abstraite qui ‘est’, est, à ce que je crois, une erreur profonde.
(ELG, p. 34)
In those days, when David Hilbert was developing an axiomatic method, Saussure had the intuition to systematize or even to ‘mathematicize’ the relational interplays, which characterize ‘la langue’ and subtend the speech: ‘The day will come, when it will be recognised that [the values and]quantities of language and their relationship are regularly expressible in their fundamental nature by mathematical forms’;2 or, again: ‘Each sort of linguistic unit represents a relation and a phenomenon is also a relation. Therefore everything is relation. The units are not phonic, they are created by the mind … All phenomena are relations between relations.’3
He could envision the moment when it would be possible to transcribe the calculus of meanings into formalized languages that could be completely detached from verbal languages and that would also escape their slipperiness: ‘Some day, there will be a special and very interesting book to be written about the role of the word as the main disturber of the science of words.’4
Here again, such an approach is perhaps in a random accordance (the so-called ‘air du temps’) with epistemological propositions that are its contemporary, for instance Peano’s work on formalized languages. In the same way, for the next generation, the works of Hans Reichenbach were to provide A. J. Greimas with examples of a rigorous grasp of meaning outside of verbal language and were to encourage him towards what he called ‘symbolic language’ (hence, for instance, Greimas’ famous algebraic schematizations of narratives).
Such a theory of knowledge was to overturn the whole of linguistics and to throw it into a frenetic search for structures (that was going to end up, for a limited period, with the excesses of structuralist mechanicism). The debates surrounding this new orientation and its key terms were the most eagerly disseminated linguistic research between the years 1960 and 1980 and the story behind these debates is well known. Therefore, here we will concentrate on the strictly semiologic aspect of Saussure’s research and will also explain why it is no longer possible, today, to split the personality of Ferdinand de Saussure into several contradictory fragments as was frequently the practice during the heyday of structuralism. The very progress in the publishing of Saussure’s manuscripts and in the corresponding semiotic research allows us today to acknowledge the strong coherence of the totality of his cutting-edge thinking and shows that the same theoretical project is at work in what came to us from the real Cours de linguistique générale (CLG) and from his studies on anagrams and on Germanic legends.
2 ‘Il arrivera un jour (…) où l’on reconnaîtra que les [valeurs et] quantités du langage et leurs rapports sont régulièrement exprimables dans leur nature fondamentale par des formes mathématiques’(CLG/E, I, n. 10, p. 22).
3 ‘Toute espèce d’unité linguistique représente un rapport, et un phénomène aussi est un rapport. Donc tout est rapport. Les unités ne sont pas phoniques, elles sont créées par la pensée … Tous les phénomènes sont des rapports entre des rapports’ (CLG 1908–1909).
4 ‘Il y aura un jour un livre spécial et très intéressant à écrire sur le rôle du mot comme principal perturbateur de la science des mots’ (CLG/E, IV, 3281, p.5).
Chronologically speaking, the Saussurean heritage first was exploited with the development of linguistics, then, second, by the debut of semiotics: the first results of Saussurean linguistics, as is well known, brought researchers like Barthes and Greimas to the intuition of a broader scientific project, encompassing linguistics, and which would be explicitly subtended by the really rational bases for which Saussure had called. Such researchers envisioned the semiological project as a consequence of Saussure’s global theoretical approach, even if the then published Cours had very few references to semiotics, in contrast with the unpublished manuscripts of Saussure, which do develop this new concept.
The onus today is to show how the semiological (or semiotic) approach imposed itself in order to conceive the phenomenon ‘Langue’ as a whole, seizable and analysable as such, but also, at the same time, as a component of a wider whole towards whose definition it would contribute. And to explain, at the same time, how the pyramid of descriptive languages to which this semiotic hypothesis gave birth embodied precisely the type of cognitive operations which build and guarantee, authentically, the scientific nature of a large number of human sciences.
We will then discuss how, from this largely implicit – and at least unpublished – theorization, Saussure’s direct heirs (Hjelmslev and Uldall in Denmark) and their followers (Greimas and Barthes in France) have been able to begin discovering certain constants of signification. Finally, we will consider how the publications that are now representative of Saussure’s work on textuality (anagrams or old legends) provide greater theoretical depth and significant breadth to what we hitherto knew about his views.
There has been much confusion attendant on the interpretation of the ‘cohesion’ of Saussure’s thought; from the outset we should eliminate some of these confusions including the very common ones about semiotics or semiology.
For Saussure, the evolution of language is merely change and alteration on the backdrop of an abstract and logical continuity. Under the pressure of the intense research to which they have been subjected, the meanings of certain terms in his manuscripts have evolved and have even split into antagonistic or complementary notions. Such is the case of ‘semiology’ exclusively used in the Cours and in all Saussure’s handwritten notes, though the word ‘semiotics’ had existed already in French since the sixteenth century. Semiotics indicated the set of corporeal signs that allow a physician to diagnose a state of health or illness. Nowadays, after several mutations by which even the terminological choices of Benveniste have been abolished (cf. the two articles – ‘La forme et le sens du langage’, 1967, and ‘Sémiologie de la langue’, 1974 – in Benveniste 1967, 1974), ‘semiology’, in the theorized definition given by Saussure, has turned into ‘semiotics’ whilst ‘semiology’ indicates a less formal analytical practice, more composite, more intuitive, linked to considerations concerning extra-linguistic references. Hence, in our contemporary studies, the substitution of the term ‘semiotics’ for that of ‘semiology’ is common.
One has only to sketch a comparison of Saussure’s work with that of his American contemporary, C. S. Peirce (1839–1914), also presented as one of the founding fathers of semiotics, to observe that the works of these two pioneers who never met are in opposition as much as ‘Theory of language and signification’ and ‘Philosophy of inference and deduction’ are. Peirce elaborates a philosophical system comparable to the great classical systems of Kant, Hegel and Schopenhauer. The type of abstraction he aims for, and which he attains, belongs to the realm of philosophy. All the notions he posits are thought out with the same realistic and substantial investment as the categories of Aristotle’s Metaphysics. These notions receive an ‘ontological’ definition: they exist, and they are. In contrast, Saussure’s problematics display a totally rational schematism with no ontology, a formal thinking that is neither nominalist nor realist. This attitude constitutes a radical break with the millennial philosophy of language subjacent to the various traditional grammars. This break, well ahead of its time, from the point of view of the aesthetics of knowledge, can only be compared to the ‘transmental’ experiments undertaken, in the same period, by Russian painting (Kandinski, Rozanova, Popova or Malevitch). We will return to this precise idea of ‘theory’ when the time comes to specify some of the characteristics of semiotics considered as a science.
For Saussure, this term concentrates the most intense reflexive activity and, as can be foreseen, its significations split according to the points of view under consideration. We will limit ourselves here in distinguishing three major meanings of this term and therefore three successive aspects under which we will examine this notion:
• On the one hand, semiology is the universal competency with which the living being is endowed to elaborate expression. It does look as if Saussure only attributes this competency to the human species, given that he conceives language as ‘what happens when mankind attempts to signify its thinking through a necessary convention’.5 For generations of researchers who have been exposed to phenomenology (from R. Ruyer to our times, including A. J. Greimas and R. Chambon), alternatively, the semiologic competency (alias semiotics) is not exclusive to human beings. They share
5 ‘Ce qui se produit lorsque l’homme essaie de signifier sa pensée au moyen d’une convention nécessaire’ (CLG/E, IV, 3342).
it, to various degrees, with the entire living realm. Greimas liked to say that the dog is a great semiotician.
• On the other hand, ‘semiotics’ designates the various products of this competency, i.e. the different languages that circulate amongst living beings. With mankind, this competency takes on extremely varied forms. This is liable to generate numerous distinct semiologies, and, in this case, a ‘semiology’ (une sémiotique) is a particular language, as in ‘The particular semiology called “language”’.6 However, Saussure also mentions ‘general semiology’ (D 222, ibid.) and thus indicates the vast ensemble of which each semiology (each language, alias each semiotic) is but a particular implementation. As we are about to see, this notion of ‘general semiology’ plays a key role for Saussure’s theory: it is essential for the constitution of semiotics as a science, as, mutatis mutandis, the invention of the ‘zero’ for mathematics.
• Finally, ‘semiotics’ (for contemporary researchers) must also be understood as the science of all the various particular semiologies and/or that of general semiotics as well as the study of the ability for language, which generates these various semiotics.
To sum things up, in Saussure’s thinking a sentence such as ‘Man is a born semiotician’ could be understood as:
• ‘Man is an animal which could not survive without expressing and exchanging significations.’ Or
• ‘Man has the capacity to create various means of expression; he elaborates significations through all sorts of semiotics which have it in common to be articulate, codified and systematized (to various degrees, however). Amongst these semiotics, verbal language is the most articulate.’ Or, finally, as
• ‘Man intends to construct a rational knowledge, called “semiotics”, which is a science of the different languages at his disposal.’
In a certain sense, Saussure takes verbal language for granted. For him there is a kind of self-evidence of language as an institution that he never challenges in its genesis, its acquisition, or in its concrete aspect and in its expressive function. The question of the faculty of language only occurs twice in the CLG. Its clearest occurrence is noted by Constantin (Komatsu, Le cahier de Constantin, as cited in Saussure 1993: 276): ‘For us, language will be the social product the existence of which allows the individuals to exercise the faculty of language.’7
6 ‘La sémiologie particulière dite langage’ (N 10 in Godel 1957: 275).
7 ‘La langue sera pour nous le produit social dont l’existence permet à l’individu d’exercer la faculté de langage.’
Through his family background as well as his own inclinations Saussure was born into the learned linguistics of his day, to which many of his youthful ‘memories’ bear witness. On the other hand, he had a totally different experience of significations, through the work he dedicated to anagrams and old German legends. Obviously, when he devoted lengthy studies to a corpus, Saussure was extending what were common practices in his century: Auguste Comte, for instance, had shown great interest in anagrams, and the Romanist philologists (Gaston Pâris) and many Germanists were analysing the ancient legends spread around by the oral tradition of the entire Indo-European world. But, in revisiting this field, according to those radically formal and relational orientations that are his own, Saussure laid himself open to a semiological experience of a new sort. He discovered that, with man, the production of meaning does not depend only on the prison of words but is constantly on the point of springing up from nowhere, from the gathering of two straws or three feathers, as from some non-existent beings, evoked by a storyteller or shaped in clay (statues, figurines, any modelling). He, thus, had to take into account the fact that the transmission of meaning is not necessarily attached to the verbal expression as such, an experience, which led to the broadening of semiotic enquiry.
As he was working on the Cours, Saussure’s thinking reached a bedrock of ‘delimitations’ which began to bring forth the specificity of the domain of language while establishing some unarguable ‘truths’. What is the status of such operations of thought? They deal with the initial statements from which one must start in order to ‘lay out the foundations of the edifice’ and find ‘the foundations of language’.8 Thus Saussure had a clear perception of the two kinds of competence he needed to muster in order to endow linguistics with the hypothetico-deductive method which guarantees its scientificity. He had to continue to be the expert linguist, heir to the various European schools he had followed. At the same time he had to adopt the behaviour of an epistemologist familiar with ‘hard’ science. In his own eyes, only a true epistemological commitment was liable to establish a system of language out of the mass of facts already observed and collected.
Here are some of the initial statements, starting points for an acceptable theoretical path:
1 ‘In a language there are only differences, and no positive terms’ (Saussure 1983: 118) (‘Dans la langue, il n’y a que des différences sans termes positifs’, CLG, 166).
2 ‘A language constitutes a system’ (Saussure 1983: 73) (‘La langue est un système’, CLG, 107).
3 ‘The linguistic sign is arbitrary’ (Saussure 1983: 67) (‘Le signe linguistique est arbitraire’, CLG, 100).
4 ‘Linguistic phenomena always present two complementary facets, each depending on the other’ (Saussure 1983: 8) (‘Le phénomène linguistique présente perpétuellement deux faces qui se correspondent et dont l’une ne vaut que par l’autre’, CLG, 23).
It may be observed that these statements do not apply only to natural languages but, in fact, to all systems of signification. These various statements specify the field of language in general and thus place linguistics in a mental whole of which it is but a particular component. With the result that one can embrace linguistics in its totality (just as when one looks, from a distant and external point of view, at a geographical zone it becomes possible to embrace this zone as a whole, be it a given country or a continent). It is only once the phenomenon ‘language’ is thus apprehended as a whole that its ‘internal order’ of ‘la langue’ can be discerned as a unitary system.
Thus the statements we have just listed function as ‘a good generalization’, rigorously delimited and, for that very reason, discriminating for the field of language. The next operations consist in exploring the specific properties of the various languages inasmuch as they relate to each other, hence establishing operators of individualization for each one of them. For example, one of the very obvious specificities of the semiotic ‘verbal language’ is its linear dimension which allows for a temporal deployment. In contrast, visual language is necessarily planar or n-dimensional. A verbal semiotic is also by necessity arbitrary, whereas a pictographic semiotic may appear partially motivated, because it appears to resemble. Thus, progressively, by alternate operations of generalizations and differentiations, operating concepts build up, leading to a formal method for the analysis of significations.
An extremely detailed presentation of the way these sequences of statements guarantee the soundness of semiotic work is to be found in Pariente (1973). The rationality of this work proceeds from the establishment of a rigorous hierarchy of inferences and implications, and this first step leads to the formulation of fundamental concepts and to the constitution of a metalanguage as reliable as the language of mathematics or logic, as well as rigorously specific and adequate to the object ‘language’.
Let us give the floor to Saussure himself who sums up, as follows, his practice, his ‘faire’, in 1894: ‘The two things, a good generalization on language which can interest anybody or a sound method assigned to comparative grammar for precise everyday operations, are in fact one and the same.’9
9 ‘Les deux choses, une bonne généralisation sur le langage, qui peut intéresser qui que ce soit, ou une saine méthode à proposer à la grammaire comparée pour les opérations précises de chaque jour sont en réalité la même chose’ (CLG/E, IV, 3297).
The phrase ‘sémiologie, partie de la psychologie sociale’ (Godel 1957: 275) is frequently to be found in Saussure’s manuscripts, and this, in the CLG, becomes the ‘prediction’ so often quoted:
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 séméion, ‘sign’).10
This formulation is disturbing because all Saussure’s declarations and what has been possible to infer from his work is that there is nothing psychological about his linguistics. How is one supposed to understand this idea of ‘psychologie sociale’ which is nonetheless at the heart of the definition of his scientific endeavour? We have developed elsewhere an answer to this enigma (Hénault 1996: 9–54). In short, in his thinking, on the ‘carte forcée’, on the constraints that the social dimension of the ‘langue’ places upon individuals and, especially, on the factors of linguistic change that no individual can change, Saussure represents the social mass made of all the speaking individuals as a ‘weightiness’, active through its mass alone, and subject to another ‘weightiness’ – the time factor. None of these actors is endowed with an individualized psychology and, if there is continuous creation in language, it acts more like a series of glacial deposits: ‘Of these vast moraines to be seen on the edge of our glaciers, picture a prodigious accumulation of things borne down over the centuries.’11 However, these relations of uncertainty, linked to the disorder inflicted by the passage of time and by the speaking mass, are recaptured at a less superficial level by a sort of logical fibrillation of the language: ‘All the logical features of the language depend or can depend on immutable data that the accidents of time or the geographical place do not explain.’12
This ‘logic face of langue’ is made up of the deep constants which are a sort of particular reason, specific to the linguistic field, a linguistic rationalism (in a meaning akin to Bachelard’s ‘electric rationalism’, designating the sum of the regular occurrences and laws of the physics of electricity). This linguistic
10 ‘On peut concevoir une science qui étudie la vie des signes au sein de la vie sociale; elle formerait une partie de la psychologie sociale et par conséquent de la psychologie générale; nous la nommerons sémiologie (du Grec séméion, “signe”)’(CLG, 33).
11 ‘De ces grandes moraines qu’on voit au bord de nos glaciers, tableau d’un prodigieux amas de choses charriées à travers les siècles’ (Première conférence à l’Université de Genève, 1891, CLG/E, IV, 3281, p.5).
12 ‘Toute la face logique de la langue dépend ou peut dépendre de données immuables que les accidents du temps ou du lieu géographique n’ atteignment pas’ (Notes pour le CoursIII, 1910–1911; ELG, 306).
rationalism, also described by Saussure as describing ‘the forces operating permanently and universally in all languages’ (Saussure 1983: 6) (‘ces forces qui sont en jeu d’une manière permanente et universelle dans toutes les langues’ (CLG, 20), can be considered as an expression of the ‘esprit collectif’. It is therefore in this area that we might seek what Saussure means by ‘psychologie sociale’. The task assigned to the social psychology called semiotics would be to bring to light the constant and universal relations that subtend significations and allow them to take shape.
The first Saussurean readers – convinced by the examples in the editions of the Cours by Bally and Sechehaye (1916) and Tullio de Mauro (1972) and, in particular, by the famous oval-enclosed binaries (p. 99) with the superposition of a tree and the word ‘arbour’ or the French word ‘arbre’ and the Latin word ‘arbor’ – have portrayed Saussure as the ‘Sign guy’, the man who advocates analysis of isolated signs. A. J. Greimas (1917–1992) used to explain frequently that the first of his own contributions to the theorizations of signification had been to break the straitjacket of the dimension-sign, supposedly imposed by Saussure, to consider all dimensions of systems of significations and of the two planes of the signifying (or expression) and the signified (or content), and to describe them in their full extent. In reality, Greimas himself, very early on, had suspected that the first publishers of the Cours had insisted on the idea of the sign taken in isolation, because they maintained a conception of language as nomenclature (i.e. like a list of signs, each sign associated with its referent and fundamentally separated from the other signs) which is precisely what Saussure had aimed to overturn. In 1984, in still unpublished conversations with the present author, A. J. Greimas indicated, however, that he thought he had perceived in the metaphors of the ‘Royaume flottant’ and of the sheet of paper an explicitly planar conception of the approach to ‘sounds’ and ‘ideas’: ‘So, we can envisage the linguistic phenomenon in its entirety – the language, that is – as a series of adjoining subdivisions simultaneously imprinted both on the plane of vague, amorphous thought (A), and on the equally featureless plane of sound (B)’ (Saussure 1983: 110). (Here a schema shows two undulating surfaces: the more or less parallel waves are segmented by parallel dotted lines.)
The characteristic role of a language in relation to thought is not to supply the material phonetic means by which ideas may be expressed. It is to act as intermediary between thought and sound, in such a way that the combination of both necessarily produces a mutually complementary delimitation of units.
(Saussure 1983: 110)
And some lines further on, the junction of these two ‘indefinite planes’ is described as ‘the coupling of the mind with phonic matter’.13
None of this has prevented the proliferation of so-called Saussurean theories of the sign taken in isolation, approaches which, from the point of view of theoretical semiotics, are so many deadends. Umberto Eco acknowledges this fact in A Theory of Semiotics (1976: 12): the semiotic universe is not made up of signs but of ‘semiotic functions’, a term that Saussure would not have taken exception to.
Chapter III of the CLG opens with the enumeration of some of these dualities constantly called upon by Saussure to demonstrate the character of language, so difficult to apprehend. For example, this formulation that has already been quoted: ‘linguistic phenomena always present two complementary facets, each depending on the other’14 (Saussure 1983: 8).
With a very literary mind, Jakobson named Saussure as the great revealer of linguistic antinomies. He saw this as a sort of character trait and felt free to reproach Saussure for his perpetual dual thinking, calling him ‘the great doubter who always saw the two sides of the problem’ (1939: 237).
The earliest commentators on these dualities, of this perpetually dual thinking, insisted on its antithetic aspect and radicalized the contrasts to the point of turning them into contradictions. The dichotomies became unmanageable antinomies and semiological thought was, in an act of flagrant abuse, made rigid. The publication of larger proportions of the handwritten notes left by Saussure, on the other hand, has enabled the demonstration of how, for instance, far from being totally hampered by the discontinuist binarism of structuralist thought, his theory is conceived of evolution and gradualism in many fields. This can be constantly witnessed in the recently published Ecrits.
What is the exact meaning that Saussure gave to the word ‘theory’? A short note in his own hand leaves no room for doubt:
Baudouin de Courtenay and Kruszewski have got closer than anyone to a theoretical view of language, without departing from purely linguistic
13 ‘Nous pouvons nous représenter le fait linguistique dans son ensemble, c’est-à-dire la langue, comme une série de subdivisions contiguës, dessinées à la fois sur le plan indéfini des idées confuses et sur celui non moins indéterminé des sons’; ‘Le rôle caractéristique de la langue vis-à-vis de la pensée n’est pas de créer un moyen phonique matériel pour l’expression des idées, mais de servir d’intermédiaire entre la pensée et le son, dans des conditions telles que leur union aboutit nécessairement à des délimitations réciproques d’unités’; ‘l’accouplement de la pensée avec la matière phonique’ (CLG, 155–156).
14 ‘Le phénomène linguistique présente perpétuellement deux faces qui se correspondent et dont l’une ne vaut que par l’autre’ (CLG, 23).
considerations. Most western scholars anyway ignore them. The American Whitney, whom I revere, has never said a word on these same subjects that was not right, but like for all the others, it does not occur to him that language needs a systematic.15
(cited by Godel 1957: 51)
On the one hand, his praise of these two Russian researchers impugns any proposition that does not stem from the very technicity of linguistics; thus understood, scientific thinking leads to the apprehension of regularities that have been discovered. There is no longer a case for formulating arguments that have merely been invented. On the other hand, his reservations about the erudite work of Whitney show that, for Saussure, a theory cannot be construed without a ‘system’ which allows the results that have been obtained to contribute to the elucidation of similar cases.
Without such a systematic approach, the gathering of facts does not result in the construction of cumulative knowledge. Thus one can see that Saussure intends to constitute a theory which would provide foundations for a theory of language, and which would eventually be modelized or formalized in the same way as ‘hard’ sciences. For him, the word ‘theory’ does not carry the speculative meaning intended by a literary mind; it takes on the meaning to be found in expressions such as ‘Theory of Relativity’ or ‘Theory of Electricity’. For scientific theories, the establishment of observables that specify the area of knowledge which is in the process of formation does not proceed from that immediate common sense which is supposed to drive ordinary life. It requires lengthy preliminary work, identical to the founding movement of a new science about to emerge. There is no doubt that Saussure hoped that some of the findings of linguistics would attain the status of scientific laws in the sense understood by his friend, the epistemologist from Geneva, Naville. In fact Saussure considered that there were two ways of treating the facts of language, one scientific alias ‘Theorematic’, the linguistics of language and the other, which attained a weak theoretical level, dealing with speech. Sometimes known as ‘Stylistique’, the linguistics of speech were confined to a mere practise of observations and classifications:
Stylistics … seeks its object above all in the observation of what is spoken, in the living forms of language, recorded or not in a text … It is not a normative science edicting rules. It claims and is correct in claiming to be a science of pure observation, registering facts and classifying them.16
15 ‘Baudouin de Courtenay et Kruszewski ont été plus près que personne d’une vue théorique de la langue, cela sans sortir de considérations linguistiques pures; ils sont d’ailleurs ignorés de la généralité des savants occidentaux. L’Américain Whitney, que je révère, n’a jamais dit un seul mot sur les mêmes sujets qui ne fût juste, mais comme tous les autres, il ne songe pas que la langue ait besoin d’une systématique.’
16‘La stylistique … voit avant tout son objet dans l’observation de ce qui est parlé, dans les formes de langage vivantes, consignées ou non dans un texte … Elle n’est pas une science normative édictant des règles. Elle prétend et a droit de prétendre être une science de pure observation, consignant les faits et les classant’ (‘Rapport sur la création d’une chaire de linguistique’, in ELG, 2002: 209).
Whereas for linguistics of ‘langue’, short working notes show themselves to be far more exacting, as here, under the title ‘Laws’: ‘Laws: The universal laws of language [which are imperative] (theorematic)’.17
What is the purpose of this theory? ‘a) to describe all known languages and record their history …; b) to determine the forces operating permanently and universally in all languages, and to formulate general laws which account for all particular linguistic phenomena historically attested; c) to delimit and define linguistics itself’ (Saussure 1983: 6). Which, in Saussure’s authentic manuscripts, is formulated as follows:
Some recurrent truths … Let us not speak of axioms, principles or propositions. They are [merely, and] in the pure etymological sense, aphorisms, delimitations, [but] limits in between which, wherever we start from, the truth is to be found.18
Under the impulse of the Lithuanian-born French researcher A. J. Greimas, a largely international group of researchers which used to meet in Paris, and was thus known as the ‘Ecole de Paris’, elaborated, between 1956 and 1980, what appears as a theory of signification, rigorously inherited from Saussurean tenets. The most representative works of this semiotic current are Sémiotique structurale, Maupassant and Sémiotique, dictionnaire raisonné de la théorie du langage, a manifesto published in the form of a dictionary. These publications advertise the coherence and the analytical power of a systematic of which Greimas had an intuition in the 1950s and from which he never departed: that is, starting from the base of the indefinable and first strivings for definition, selected according to Saussure’s own criteria, making explicit the relational system that Saussure considered the only possible path leading to rigorous definitions in matter of language. The name of metalanguage applies to the entirety of biunivocal and interdefined concepts (avoiding the confusion of common language) thus brought to light. Amongst the models and concepts, let us take note of the following: semic analysis, narrative schema, constitutional model
171 ‘“Langue” tandis que pour la linguistique de la langue, de brèves notes de travail stipulent une tout autre exigence, comme ici, sous le titre “Lois” “Lois: 1 Les lois universelles de la langue [qui sont impératives] (théorématique)”’ (CLG/E, IV, 3310, 8).
18 ‘a) faire la description et l’histoire de toutes les langues … ; b) chercher les forces qui sont en jeu d’une manière permanente et universelle dans toutes les langues et dégager les lois générales auxquelles on peut ramener tous les phénomènes particuliers de l’histoire; c) se délimiter et se définir soi-même’ (CLG, 20). ‘Quelques vérités qui se retrouvent … Ne parlons ni d’axiome, ni de principes, ni de thèses. Ce sont [simplement et] au pur sens étymologique des aphorismes, des délimitations…. [mais] des limites entre lesquelles se retrouve constamment la vérité, d’où que l’on parte’ (CLG/E, N 19, IV, 42).
(alias semiotic square), modalities and generative trajectory, which have led to numerous demonstrative analyses in visual semiotics as well as in other semiotics (auditive and musical, tactile or gustatory) in many fields of social life.
The first works of the Ecole de Paris inspired by Saussurean tenets mainly concerned the analysis of tales, as narrativity appeared to be the very form of the verbal expression of the transformations that human action brings to (or inflicts on) the world. Binarism proved to be perfectly adequate to rationally segment narrative sequencesas well as the inversions of contents that that entailed. When it is the task to programme an action for a clearly formulated objective, one deals mainly with rather cerebral and generally ‘clear and distinct’ (Descartes) significations. Binarism was precisely this.
Then, from the 1980s onwards, this same Ecole de Paris moved towards a continuist epistemology, closer to the sensitive and/or affective apprehension of experienced meanings. The attempt was to handle ‘less cerebral’ significations in order to account for that portion of meaning which is experienced closest to the body.
Thus it seemed that semiotics was therefore under the obligation to square the circle: that is, to conceive procedures that would analyse a continuum of feelings which, taken, problematically, on their own, were unstable in the first place. Could not analysis simply be confined to the process of segmenting? Half a century of research has allowed for experimentation with a theory and a method that Saussure discovered without having the time to explicate its various articulations but whose orientations he managed nonetheless to impose enduringly. This happens to be particularly clear with his work on legends.
A certain number of works have recently thrown light on the analyses that Saussure himself devoted to the old legends taken from the Germanic tradition (cf. especially, Arrivé 2002, 2007; Kim 1993). Through them one can now follow, as if in a running commentary, a storm in Ferdinand de Saussure’s brain, an authentic problem of semiotic epistemology experienced under Saussure’s pen. Let us give a much-accelerated account of the discoveries made by Michel Arrivé from the short notes dealing with these Germanic legends (Arrivé 2007).
In the Cours de linguistique générale, as in all the manuscripts that deal with the linguistic sign, Saussure never explicitly refers to semiosis of legends, and the other sign systems that are quoted as semiologies alternative to language are always very close to verbal language. In contrast, in the work on legends, proven to have been undertaken at the same time as the study of language and linguistic sign, there are abundant references to the Cours de linguistique générale. Why?
The semiologic unit of the legend (either the character, the geographical location or the historical period) is not arbitrary like the linguistic sign. From the beginning, it is motivated by onomastics, history and geography. One can frequently retrace and prove that these units have a referential origin and, therefore, a precisely limited meaning. As such, the unit is not a semiological sign. This is how Saussure first describes the unit of legendary meaning.
Later on, Saussure manages to convince himself of the fact that there comes a moment in the evolution of the great legends when the mythical character is totally detached from his origin and from all initial referential charge. The character then becomes an empty form, an ‘inexistent being’ like a letter in the alphabet or like a word. Such a form is then liable to enter into the open combinatoire that bestows on language its infinite creativity. It attains the same degree of plasticity as the linguistic sign, it has turned into a real semiological sign, liable to be loaded with meaning and value by anybody searching for the specific expression conveyed by the ancient legends. Thus, a new area of meaning with a very particular texture and colouring is placed at the disposal of the ‘Corps Social’. This new semiotic reaches areas of meaning other than those of the linguistic sign, as is exactly the case for musical or visual semiotics.
The area of meaning specific to the semiosis of legends could be described as the mental space for dreams of love, heroism and glory, or even secret identifications with legendary characters. The object ‘mythical person’ is both an ‘inexistent being’ (Saussure’s expression) and a possible prop for our affections:
As one can see, the incapacity to maintain a certain identity should not be attributed to the effects of Time – such is the remarkable error committed by those who deal with signs – but is placed beforehand within the being that one cherishes and observes as an organism though it is no more than a fleeting combination of two or three ideas … The association – that we sometimes cherish – is but a soap bubble.19
Thus the legend generates an affective and very intimate sort of meaning, it is an autonomous semiology and, exactly as with verbal language, it goes through incalculable changes and modifications in the course of time (LEG, 31): ‘What philosophers and logicians have missed here is that, from the moment a system of symbols is independent from the designated objects, it is prone, for its part, to suffer displacements that the logician is unable to calculate.’20
19 ‘Comme on le voit, l’incapacité à maintenir une identité certaine ne doit pas être mise sur le compte des effets du Temps – c’est là l’erreur remarquable de ceux qui s’occupent des signes – mais est déposée d’avance dans l’être que l’on choye et observe comme un organisme, alors qu’il n’est que la combinaison fuyante de deux ou trois idées … L’association – que nous chérissons parfois – n’est qu’une bulle de savon’ (LEG, 192).
20 ‘Ce qui a échappé ici, aux philosophes et aux logiciens, c’est que, du moment qu’un système de symboles est indépendant des objets désignés, il était sujet à subir, pour sa part, par le fait du temps, des déplacements non calculables pour le logicien’ (ELG, 20).
Hence we witness, almost ‘live’, the birth of another idea of semiotics in the mind of Saussure. Not only, any more, is there the logical cum grammatical semiology that is the necessary foundation for proposing the limits between which is brought into play the entire system of relations empty of meaning that serves to describe the abstract system of ‘la langue’ (language), but also a sensitive and affective semiology that intends to take into account feelings as truer to life and the experience of meaning at the affective level. In this moment of improbable synthesis that is one of the most respected aspects of Saussure’s thinking, we witness the point that all semiotic schools have reached today: all have, at the same time, but following very different paths, set their agendas, for their research, on the question of instinct, emotions and passions.
The work on legends introduces Saussure’s reflection on some textual workings that he considers because they are not literary and because, for this reason, they present units of meaning which are not set (‘figées’), but, on the contrary, totally subject to the pressure of the ‘masse parlante’ that moves them around like it moves the meaning and the value of everything, in language.
The contradiction between the referential and motivated aspect of the mythical character and the necessarily arbitrary nature of the real sign disappears here in order to bring in, at a deeper and more theoretical level of understanding, a ‘linguistication’ of the mythical character which becomes available for just about any signification. Thanks to this exigency of thought, ‘semiotics’ is not only an all-embracing logical concept that enables the theory to function. It is also a process whose beginning and evolution Saussure is capable of experiencing and seeing with his own eyes: one cannot imagine the origin of language, but one can, however, witness at each moment the appearance of new semioses assumed by the Corps Social.
It is therefore thanks to this extremely honest work on his own systematicity that Saussure loses nothing of what he perceives in the universe of the living, in the living functioning of semiosis. In CLG/E, 2780 B, one finds the expression ‘the living units beneath the word’ (‘les unites vivantes au-dessous du mot’) and this is in relation to the infinite plasticity of true semiotics. He thus attains stronger and more anticipatory views than, say, Hjelmslev, who did not share this mental flexibility and open-mindedness which bring together the hazards of concrete observation and the most open experiences with ongoing theoretical commitments. The solution of the enigma which the legends present rests upon two major criteria, in constant demand in order to distinguish what is ‘semiotic’ and what is not:
• A real semiotic constantly evolves under the double and combined influence of time and of society.
• True semiotics should be able to take into account any signification.
One will have observed that Saussure (ELG, 20) dismisses unnamed ‘philosophers and logicians’ so as to claim his scientific autonomy and so render autonomous the field of knowledge that he calls into existence, the field of the rational analysis of emergence and creation of meaning in all its forms. Our last quotation of Saussure will not surprise by the gravity it takes on here:
The nobleness of legend as of language stems from the fact that both – set to make use of elements brought before them and bearing random meanings – bring them together and continuously draw from them a fresh meaning. A solemn law stands which one might be well advised to ponder before reaching the conclusion that this conception of legend is incorrect: nowhere do we witness the flowering of something that is not the combination of inert elements and nowhere do we see that matter be anything but the continuous food that thought digests, ordains, commands, unable however to do without.21
Saussure’s theory is difficult and its editing and translation has promoted misunderstandings of all sorts. Nonetheless the entire semiotic field is making progress, in an astonishing harmony with Saussure’s Ecrits. Thus, components like these works on the legends, which had practically seemed aberrant to his contemporaries, now appear liable to be the most fecund in that they reinvent the phenomenological experience of meaning and that they raise, with the utmost pointedness, some major questions such as: How come meaning is constantly renewed from any kind of material? How should we rationalize our perceptions and put together all heterogeneous units in analysis? On the one hand, we deal with the clear and distinct minimal units (syntactic items and phonological, morphological or semic components of words) of the linguistic level. On the other hand, we handle enigmatically (narratively or otherwise) modelized macro-units like texts and enormous ensembles of texts (corpus) considered as one single unit of meaning. We reduce them to some singular ‘whole of signification’ (‘un Tout de signification’) and, in between, mysterious middle-sized units such as emotional expressions or legendary characters, plus spaces (for instance, Pompeii, for Freud’s Gradiva) and remote or future times, which may become semiotic because they are permeated by some vague scents of the collective unconscious.
Universal negativity, which, for Saussure, is the founding operation of language – with his strictly formal, differential and oppositional definition of all the elements of language never credited with the slightest positive element – is not to be interpreted as a ‘pulsion de mort’ (a death drive) and destruction like mainly literary minds may be led to believe. Saussurean negativity is but the
21‘Ce qui fait la noblesse de la légende comme de la langue, c’est que condamnées, l’une comme l’autre, à ne se servir que d’éléments apportés devant elles et d’un sens quelconque, elles les réunissent et en tirent continuellement un sens nouveau. Une loi grave préside qu’on ferait bien de méditer avant de conclure à la fausseté de cette conception de la légende: nous ne voyons nulle par fleurir une chose qui ne soit la combinaison d’éléments inertes et nous ne voyons nulle part que la matière soit autre chose que l’aliment continuel que la pensée digère, ordonne, commande mais sans pouvoir s’en passer’ (LEG, 307).
abstract mental mechanism (at the same degree of abstraction as negative numbers in mathematics), which allows, on the contrary – as we have just seen with the metaphor taken from food (supra, LEG, 307) – the perpetual surge, the perpetual renewal of meaning.
Arrivé, M. (2007) A la recherche de Ferdinand Saussure, Paris: PUF.
Cahiers Ferdinand de Saussure, passim, Geneva: Droz.
Saussure, F. de (1967–1974) Cours de linguistique générale, edition critique, Vols 1 and 2, ed. R. Engler, Wiesbaden: Harrassowitz.
Saussure, F.de (1983) Course in General Linguistics, trans. R.Harris, London: Duckworth.
Saussure, F. de (1986) Le leggende germaniche, ed. A.Marinetti and M.Meli, Este: Zielo.