Tracing Saussure’s intellectual roots is not easy. In his writing, including the notes, he rarely referred to ideas or authors outside technical linguistic studies. Saussure’s personal library, eventually bequeathed to the University of Geneva, consisted exclusively of books on linguistics (it is possible, however, that what his family decided to give away was only part of his library: Stancati 2004). Whenever Saussure delves into general issues of methodology, his references typically become generic, even sweeping; he fights with “prejudices” and “stupidities,” and often qualifies and differentiates without specifying the target of his argument. Many of those who knew Saussure marvelled at his erudition in matters of philosophy, literature, art, and music,1 yet in his writings one can find only occasional clues hinting at this background. For instance, we would never have known that he was interested in music at all, if not for some cursory yet sharp remarks in the notes about Wagner’s leitmotifs and Bach’s sound symbolism.2 Saussure’s ability to raise fundamental epistemological questions concerning the subject and method of linguistic studies show him to have been abreast of contemporary epistemological ideas, without giving any tangible evidence about which authors and works he may have been aware of or familiar with. For instance, his treatment of the evolution of language reads at times like a linguistic incarnation of Bergsonian duration—without any sign of Bergson’s presence on his intellectual horizon. To use Saussurean terms, we can say that what surfaces as the outward representation of his ideas is only a set of “differences” by which his thought seeks to determine its place among a variety of concepts and approaches concerning theory of cognition, philosophy of language, sociology, anthropology, psychology, studies of myth, and modernist poetics. As to the “substance” of those differentiations, it remains almost entirely tacitly implied.
Much has been said about the Saussure family’s hallowed tradition of excelling in the natural sciences. This background, resounding with Saussure’s own supreme aptitude for pattern building (of which his book on Indo-European vowels is a prime example), could suggest that Saussure’s approach to “general linguistics” was inspired by the natural sciences and mathematics, an assumption reinforced by his own occasional declarations of the need to present language as a chain of “theorems.” This assumption reinforced the perception of timeless structural orderliness in which Saussure’s linguistics was cast by subsequent generations. What this perception missed was the irreconcilable contradiction inherent in la langue that set it apart from natural sciences—a contradiction stemming from the heterogeneous yet symbiotic duality of the linguistic sign. The Leipzig school’s straightforward conviction that the laws of language are equivalent to natural laws and can be described accordingly evoked Saussure’s scorn.
Saussure was one of the world’s foremost experts on Sanskrit. Sanskrit stood at the epicenter of his teaching curriculum in Geneva and (together with the classical languages) of his Indo-European studies. It is natural to presume that the famous Panini grammar of Sanskrit—a formidable compendium, written in the fourth century BC, of about four thousand rules or sutras3—must have played an important role in shaping Saussure’s approach to language. Panini’s striking ability to focus on linguistic forms and their mutual distinctions, a feature quite a few modern scholars have perceived as protostructuralist or even protogenerativist (Kiparsky and Staal 1969; Misra 1964), invited analogies with Saussure’s Course.
One of trademarks of Panini grammar was its precise and detailed description of speech sounds and their alternations at the boundaries of words and morphemes. This topic clearly fascinated Saussure. Even in the second and third lecture courses on general linguistics, with their heavy slant toward philosophical problems of language, substantial space was allotted to descriptive phonetics; consequently, a chapter on the subject appears in the Course. Yet this elaborate excursion into empirical matters stands isolated in a book whose central theoretical nerve asserts that the physical quality of speech sounds is irrelevant. Saussure’s occasional profession that he “had no taste for phonology” (ELG, 244/173), while consistent with his theoretical convictions, was hardly sincere. As happened more than once in situations when Saussure became aware of contradicting himself, he tried to explain it away on didactic grounds: one had to take on phonological matters in order to correct errors in existing views (ELG, 244/173).
Panini’s influence on Saussure’s dissertation on the genitive absolute construction (1881) was unmistakable. Saussure’s double classification of the data—according to both its formal features and its meaning—reflected the way Panini described the Sanskrit case system. Yet one can see Saussure’s doubts about the viability of his own description (which he was careful enough to voice only in the dissertation’s “supplement”), later coming to the surface in his statements to the effect that all grammatical notions (“as, for example, genitive”) are no more than conventions of descriptive metalanguage, valid only insofar as one is willing to play along with them.
During the last twenty years of Saussure’s life, when he was striving to define his philosophical approach to language, we see no evidence of his involvement with Panini as a theoretical work, despite the fact that for all those years Sanskrit remained the centerpiece of his teaching load. At that time there was a vivid polemic among scholars concerning the provenance of Panini’s compendium and its possible predecessors. Saussure remained uninvolved, even though some scholars, to whom he felt intellectually or personally close, such as Whitney and Wackernagel, played a prominent role in the debates. Saussure’s name does not appear in the extensive modern bibliography of works about Panini (Cardona 1976). It could be said that Saussure tried, but eventually rejected, the pattern-building approach to language for which Panini presented a protomodel.
Saussure’s repeated praise of Whitney and, less specifically, of the French philosophes, for their emphasis on the social conventionality of language is one of the rare instances when we can see him referring to general philosophical ideas about language in a positive mode. These positive judgments are often pronounced in conjunction with ferocious diatribes against the “German” approach to language as an organism (an attitude exemplified to Saussure in particular by Schleicher and to a lesser extent by Humboldt).4 As Saussure turns to defining his own approach to linguistic conventionality, however, he does so by distancing himself from his predecessors. What one observes in language is empty or “pure” convention, the arbitrary confluence of accidents of usage. This exempts language from social “institutions” grounded in substantial social practices. Saussure’s emphatic assertion that language knows neither “beginning” nor teleological ends sets him apart from the French Enlightenment philosophy of language and its modern echoes, exemplified in utopian plans for an artificial “universal language.” On this issue, Saussure stands on the side of Herder and Hamann and against Rousseau and Condillac, though he never mentioned either pair directly.
This last point brings us to an issue that further obfuscates the discussion of Saussure’s intellectual roots—his peculiar relationship with Germany and her intellectual world. Saussure’s entire adult life took place under the shadow of the catastrophic consequences of the Franco-Prussian War. When in his lectures he cites a seemingly random example of a certain animal species being called either boeuf or Ochse on “this and that side of the border,” one can sense that the issue of that contentious “border” (with its personal dimension, due to his own Lotharingian roots) is never far from his mind. Lingering animosity toward “Germans” can be seen in the especially shrill vituperations in Saussure’s notes against the “idiocies” (bêtisses) of certain German ideas concerning language and linguistics. The French-German frictions that surrounded Saussure’s career erupted after his death in a controversy about who was to publish his papers in Indo-European linguistics. In a way, this pattern persisted in later (predominantly Swiss and French) Saussure scholarship, which rarely even glances at “that side of the border” in search of Saussure’s intellectual roots.5
Nonetheless, Saussure’s deep involvement with German language and culture was evident throughout his life. Not only did he receive his linguistic education there, but, together with Sanskrit, Germanistics comprised the core of his profile as a teacher. The full depth of Saussure’s involvement in the field is attested by his extensive studies of the Nibelung epos in the 1890s and 1900s.
It should be noted that Saussure’s ferocious critique of the hyperpositivism of the Leipzig school was done from methodological positions that bore the unmistakable imprint of the German philosophical tradition—from Kant to early Romantic philosophy of language and history (from Herder to Novalis) to the contemporary neo-Kantian, logical (Frege) and phenomenological critique of positivism. Explicating this tacit yet palpable background might have a considerable impact on the way Saussure’s pronouncements—often elliptical, aphoristically terse, and riddled with seeming contradictions—can be interpreted. Perhaps most importantly, it could help to resolve many of the apparent contradictions between the Course and Saussure’s notes by offering an intellectual framework in which they can both be read.
The question is, of course, how these analogies could be substantiated. For instance, the parallels between Saussure and Novalis in their approach to some fundamental properties of the sign are striking. Saussure seems to share with Novalis the idea of the inalienable duality of the two sides of the sign; the idea in itself is so original, so removed from conventional perceptions of the sign’s form and meaning, that no convincing analogy to it can be found anywhere else. Yet however much or little Saussure may have been aware of the Athenaeum metaphysics of sign and language, it left no explicit trace among his scarce references, although it is hard to imagine that he was not familiar with at least some of Novalis’s writing, various collections of which had appeared throughout the nineteenth century (including a very substantial two-volume edition in 1901).
Saussure’s reserved, even somewhat dour public persona presents a stark contrast to what we conventionally think of as a “Romantic personality.” Not only the generic image of the dishevelled “romantic,” but even its more dignified versions—Novalis’s passionate quest for the absolute or the young Schlegel’s irreverent wit—seem remote from Saussure’s personality or at least from its outward appearance. Thanks to Saussure’s private papers, we know about the passions, anxieties, and outbursts of anger and sarcasm that were hidden behind that reticent facade. Yet judging by the Course’s rhetorical attire, as well as the memories of his disciples, Saussure always projected an outward image of total intellectual composure, even though upon scrutiny it might turn out that what he was saying in his imperturbably lucid manner didn’t quite hang together.
Saussure’s declared goal was to lay out the foundations of linguistics with shining rationality, as a sequence of “theorems”—an achievement that would gain it entry into the compendium of modern sciences. Yet he was never able to come to terms with the intellectual compromises necessary to give “the science of language” an orderly rational shape. One could say that the wrenching contradiction between the clearly envisioned ultimate goal and the equally clear awareness of its unattainability revealed a state of mind akin to that of the early Romantics in the 1790s, before Romanticism succumbed to sweeping utopianism and flamboyant rhetorical postures. The predominant tradition of Saussure’s reception in the twentieth century highlighted his rationalist side while ignoring (or simply being ignorant of) the intellectual drama behind his seemingly unwavering if sometimes contradictory logic.
Exploring the philosophical ground from which Saussure’s ideas emerged could lead to a considerable reconfiguration of his entire intellectual landscape. In particular, it could help wrest Saussure from the utopian constructionism of the high modernism of the 1920s–1950s, while reaffirming, and casting a new light on, his connections with the epistemological revolution of early modernism (1890s–1900s) and its early Romantic antecedents.
Hidden in Plain View
It is time to explore clues that link Saussure’s idea of the unique epistemological predicament involving language with early Romantic philosophy of cognition and meaning. In purely personal terms, those clues look neither vague nor remote. In all probability, it was a tangible human continuity with roots in his family that underlay Saussure’s “elective affinities” to the world of the Athenaeum. That this line of the family tradition has remained hidden in plain sight for so long can be explained only by the blinding glare of the spirit of Enlightenment rationalism and scientific exploration embodied in generations of the Saussure family, which seemed to highlight Saussure’s scholarship and personality—as it was seen in the age of structuralism—in the most flattering way. By the same token, the early Romantic lineage of Saussure’s intellectual genealogy was doomed to remain in the background. I mean, of course, Saussure’s great-aunt, Albertine Adrienne Necker de Saussure (1766–1841), and her renowned—albeit not within the domain of Saussure studies—concept of “progressive education.”
A cousin (on her husband’s side) and a close friend of Germaine de Staël, Necker de Saussure had close relationships with such key figures of French and German early Romanticism as Benjamin Constant, August Wilhelm Schlegel, and Friedrich Schleiermacher. She was a frequent guest at Chateau Coppet, the Swiss residence of Mme. de Staël and a frequent meeting place of the early Romantics, which emerged in 1790s as an important center of the movement together with Jena.
At that time, following the birth of her four children, Necker de Saussure began noting down her observations of their behavior, beginning from the earliest stage. She was convinced early on of the necessity of turning her reflections into a book. However, it took her a long time to realize this project. In the course of the more than three decades that separated Necker de Saussure’s initial design from its completion, she translated and published A. W. Schlegel’s Course of Dramatic Literature and wrote a (rather hagiographic) book about de Staël’s life and work after her friend’s death in 1819.6
Upon its appearance, L’éducation progressive (vols. 1–2, 1828–1832, supplemented by Étude de la vie de femmes, 1837) enjoyed broad recognition. Several subsequent editions followed promptly in France (vol. 1, 1836 and 1843; a revised new edition of vols. 1–2, 1847) and in Germany (vols. 1–2 in 1838, followed by yet another translation in 1842; vol. 3 in 1839), where the book elicited a particularly wide response, perhaps because of its close intellectual kinship with German early Romantic philosophy (Maurer 1938:90).
A vivid interest in Necker de Saussure’s work and personality persisted well into the twentieth century.7 This recognition, however, never spread across the compartmentalizing barrier separating the field of child psychology and education from theoretical linguistics. We find no mention of her grand-nephew in books dedicated to Necker de Saussure, while she receives only a few vague, sometimes condescending remarks in the literature about him.8
Saussure’s great-aunt died sixteen years before he was born. True, her library, containing a large collection of works of Romantic literature and philosophy, remained at Vufflens; true, in his youth Saussure’s teacher Pictet was a close friend of hers. But while Saussure’s familiarity with Necker de Saussure’s intellectual heritage seems highly probable, it cannot be stated for a fact.
Still, upon reading L’éducation progressive one cannot fail to discern numerous threads leading to Saussure’s thought and writing—from the general philosophical perspective to some key concepts and even certain favorite turns of speech. Moreover, as I will try to show, some aspects of Necker de Saussure’s work can be seen as occupying an intermediary position between early Romantic epistemology and Saussure’s philosophy of language. This makes Necker de Saussure an important albeit hypothetical link between Saussure and early Romantic metaphysics.
The World of a Small Child: Pure Differentiation
Necker de Saussure’s book, particularly the first volume, was focused on the child’s development beginning from the first few weeks of life. At this stage, the book asserts, the child does not yet possess the faculty of reflection, at least not as he begins to after the age of five (1:2–3). For the first eight days of his life in particular, a child is unable to differentiate between his inner sensations and outward impressions. Since he cannot externalize his impressions as if they were something that lay outside his self, he is unable to distinguish among them, since the act of distinguishing one impression from another is possible only through their externalization (1:142). All the child perceives is perceived as his own sensations, as in a dream. The newborn child is a “direct creation of God” (oeuvre immédiate de Dieu) (1:9): his state is that of an absolute wholeness which is, by the same token, a blank state of total indistinction.
When, from that very first moment on, the child moves toward cognizing the world around him, he does so by making distinctions. It is not a process based on preset categories, which in any case the newborn child is as yet unable to employ. Instead, he proceeds by groping for cognitive experiences in an arbitrary, random succession.
Concentrating on the reactions of small children allows Necker de Saussure to make a discovery whose significance goes beyond the domain of child psychology: namely, that the process of signification evolves as pure differentiation, i.e., as a sequence of acts of distinguishing one signified entity from another rather than as a “substantive” grasping of the content of particular phenomena. The child perceives a difference and seeks to express it: he calls all fruits “apricots” (thus distinguishing them from other samples of food), before arriving at differentiating between, and acquiring names for, pears, prunes, grapes, and cherries (1:212–213). The child does not proceed logically, from a general idea to its species; he does not think of “fruit” as a general concept. The fact that the first act of differentiation involved the name apricot and not, say, apple was a pure accident. The sign apricot, once distinguished from some adjacent semiotic domains (say, candy or ball), suits the child perfectly well as the designator for a number of things, until the child arrives at a further differentiation. We can recall Saussure’s argument that if one encountered a language whose vocabulary consisted of just two words, one would discover that for the speakers of that language the whole world was divided into two classes of phenomena corresponding to these two words. As a matter of fact, Necker de Saussure was aware that the principle of comprehending phenomena through differentiation does not apply only to infants. As a curious parallel to children’s strategy of signification, the book cites a “prince Lee Boo of the island of Pelew” who, upon arriving at Macao, saw a horse and called it dog.
Another variation of the idea of cognition as a chain of differentiations can be seen in Necker de Saussure’s observations, in her diary, of the evolving characters of her children. From a very early stage, she remarked, their behavior pointed in different directions, showing no uniform pattern of development. In this situation, it seemed more feasible to her to describe her children by pointing to differences between them rather than by trying to characterize each child separately (Mestral Combremont 1946:71).
The fact that one concept is cognized by being differentiated from another by means of representation makes a concept and its representation inseparable. Our soul is “condemned not to employ its activity otherwise than through impressions” (i.e., outward representations), the book argues (1:356). As if in anticipation of Saussure’s semiotic terminology, the book refers to the small child’s psychology as characterized by a “double nature.” Since the outward stimuli the child receives and his responses to them are unmediated by learned patterns of reason, Necker de Saussure claims, they stand in a kind of symbiotic relationship in which they simply cannot be considered separately. In the context of a small child’s world, what a certain phenomenon “means,” that is, what response it elicits in the child, has no direct connection with that phenomenon’s established values as they are known to us. It is impossible to determine with certainty what impression a certain event has left in a small child, nor upon what inner impulse he acted in a certain way. All one can observe is a symbiotic connection between a phenomenon and the meaning it elicits.
For Saussure, the duality of the linguistic sign marked the uniqueness of language among all material and spiritual phenomena. Much in the same way, Necker de Saussure argued for the uniqueness of the human infant among all living creatures. The world of an infant finds no analogy either in the spirit of a more developed human being, which has been imprinted by an increasing recognition of uniform patterns of reason, or in that of an animal, which is guided by natural instincts. From the moment of birth, the attention of an animal is directed by instinct toward phenomena related to the necessities of its survival; it seeks signs of food and shies away from signs of danger in its environment. In a maturing human being, a similar predictability of reactions to external phenomena is brought about by experience and the guidelines of education. But the attention of an infant is directed neither by reflection nor by any “natural” interest. The peculiar characteristic of the child’s behavior is its lack of any predictable sequentiality, grounded in the child’s ability to be engaged by “alien objects” (1:362)—“to act, not to obtain a result from his action” (1:365).
The idiosyncratic inseparability of external stimuli and internal responses in the small child’s “double consciousness” has two fundamental consequences. The first finds a close parallel in Saussure’s principle of the “mutability” of language. The lack of any constant ground or pattern for connecting what is going on “inside and outside us” (en dedans et en dehors de nous-mêmes) (1:113) plunges the world of a child into incessant motion. Every new event, internal or external, throws it off balance, making it move in an unpredictable direction to an unforeseeable new state. This is the principle on which the strategy of “progressive education” is founded.
The other consequence of the unpredictability of the ongoing differentiations is that its “progression” can take an infinite number of roads. The sequence of concrete steps taken on the road of cognition is unique for every child; each is as good as any other. Yet, for each individual child, its particular road is “most important”; we see events happening to us as “true [veritable] events” (1:11). This thought finds a parallel in what Saussure calls the “immutability” of sign systems: for all the fortuitousness of the interconnections between signifieds and signifiers to be found in any particular language, for a speaker of that language they present themselves as a unique and inexorable reality.
To appreciate these fundamental features of the world of infants—its uncontrollable “mutability” and inexorable “immutability,” to use Saussure’s terms—one has to abandon all preconceptions concerning child psychology and development and look at things from the perspective of the child “himself” (lui-même) (1:19–20). The opening of Saussure’s Course, in which he strives to formulate the subject of linguistic studies, shows an almost ineffable resemblance to the way Necker de Saussure addresses what she sees as the principal task of child psychology in the opening pages of her book. The common mistake of works on education so far, Necker de Saussure argues, has been that they proceeded from the point of view of the educator. If observed from an outside perspective, child development can be seen as a succession of universal stages that follow each other in a prescribed order. Yet from the perspective of the child lui-même, his movement from one universal checkpoint to another reveals itself as a unique chain of steps that are neither sequential nor teleological. Once one adopts this approach, what one sees in a small child is, first and foremost, the unstoppable “progressive” movement of his spirit.
While Necker de Saussure applied her principle of “dual consciousness” specifically to the world of small children, she was aware of its universal validity as the fundamental condition of the human spirit. To be sure, the very nature of “progressive education” makes it a continual lifetime process: “Everything in human life is education” (1:12). As this process evolves, the spirit retains its mutability, albeit, as Necker de Saussure always emphasized, to a reduced degree, in the face of learned orderly patterns of thought and accumulated knowledge of the world. Here is where the gender difference comes to the foreground, since, according to Necker de Saussure, women are more able to retain spontaneity. In any event, preserving, at least to a degree, the “dual” nature of the spirit, which enables it to continue its free-flowing evolution, is crucial for every person. Necker de Saussure’s adherence to the principle of “duality” showed itself in her admonition to her (by now grownup) children “to grasp the meaning” of the Gospel “without ever stripping it from its form. Penetrate into its spirit, while returning unceasingly to its letter” (2:408–409).9 “Separated from its envelope, the spirit evaporates or becomes altered,” she concludes. The metaphor of the meaning “evaporating” when stripped of its “envelope” vividly recalls an image Saussure used to elucidate the interdependence of the form and the meaning of a sign, which he compared with a balloon filled with hydrogen (ELG, 115/78).
It seems that the similarities and coincidences between Necker de Saussure’s insights into the world of infants and Saussure’s revolutionary concept of the arbitrary duality of linguistic signs are simply too numerous and too weighty to be accidental. What separated Saussure from the conventionalist approach of les philosophes and Whitney—namely, the emphasis on the arbitrariness and, as its principal consequence, volatility of linguistic conventions—was precisely the point his theory shared with the idea of “progressive education.” In a remarkable coincidence of terminology, Saussure and his great-aunt called the phenomenon of idiosyncratic duality in language and in (children’s) thought “pure convention” and “pure intelligence,” respectively.
Necker de Saussure’s “infant” and Saussure’s “speaker” can be seen as transmutations of the same fundamental principle—that of cognition as a chain of differentiating steps whose progression is devoid of any guiding principle but differentiation itself. What Saussure discovered was that this phenomenon of ultimate negativity—and by the same token, ultimate freedom—was not limited to a prereflective age, as L’éducation progressive had claimed, but was the universal principle of signification by means of signs.
Progressive Education as a Romantic Concept
The main theme of Necker de Saussure’s book, suggested by its title, had little to do with “progressiveness” in the conventional sense. Rather, it referred to “progressivity,” a philosophical concept that constituted one of the cornerstones of early Romantic metaphysics.10 The notion of “progressive education” could be viewed as an extension of early Romantic epistemology into the domain of child development, particularly at its earliest stages. The “progressive” approach, in this Romantic sense, rejected the Enlightenment idea of a priori given predispositions, be they transcendental categories of cognition or universal parameters of child development.
The early Romantic response to Kant put at the core of cognition the free movement of the spirit catalyzed by its dialogical interaction with the world, a movement that neither follows any prescribed pattern nor evolves in a predetermined direction. Necker de Saussure’s book carried this principle into the domain of child development. She calls “Stoic” the Enlightenment’s ideal of submitting one’s self, through reasoning and experience, to the universal dictates of reason, in contradistinction to the Romantic principle of the uniqueness of every individual soul, which she calls “Christian” (1:60). The book takes as the fundamental premise of child psychology and education the ceaseless spontaneous movement of the child’s inner world. In this sense, a newborn child shows some of the fundamental features of the early-Romantic cognizing subject, which in fact make it possible to take a close look at those features at the crucial moment of their emergence. According to Necker de Saussure, one can never understand the human soul without acknowledging its mobile nature (1:43);11 “the character of the spirit is constantly modified,” it finds itself “in incessant activity” (1:13). From the moment of his birth, the human being lives in a torrent of ever-changing states of the spirit, whose commotion is “impetuous, blind” (1:103). (These expressions sound very close to how Saussure later conveyed the sense of the unstoppable commotion of language in his notes.)
An educator may have goals that are perfectly reasonable by themselves, but he must be aware that his input falls on the mobile spiritual terrain of a child, whose constant ferment defies any prefigured pattern. There is, and indeed should be, an “accidental” element in education (1:13). It is the fundamental freedom of the human being—rather than deviation from the “natural” ideal—that makes perfection unattainable and our desire for it unquenchable. The road to perfection suggests an ultimate goal toward which we strive again and again, while again and again it “slips away” (s’éleve) from our efforts (1:57); “to desire more than one can obtain is our destiny” (1:45). Yet without freedom, “if our will were chained,” the very striving for perfection would be impossible (1:83).
An important aspect of early Romantic critique of Kantian pure reason consisted in pointing to its distinctly “masculine” character. Friedrich Schlegel personified the rationalist predilection for abstract universal patterns and the isolated, nondialogical nature of the rationalist subject as features of the stereotypical “male” character, which, according to Schlegel, need to be counterbalanced by such stereotypically “female” features as a heightened aesthetic sensitivity and a predisposition for dialogical interaction.12 Schlegel argued that Kant’s over-rationalizing approach was rooted in his lack of any deep understanding of art and his treatment of women as a “deviation” (Abart).13 In a similar vein, he suggested that the Christian idea of God could not be whole unless it were expanded to include a feminine component (the Virgin Mary).14
Necker de Saussure’s idea of the “education of women” was grounded in this attitude.15 The early Romantic dialogical duality of the masculine and feminine elements as the embodiment of the dualism of reason and “genius” (i.e., aesthetic and religious sensibility) acquires an additional dimension in Necker de Saussure’s work by bringing the world of small children into the picture. In Necker de Saussure’s vision, the symbolic figures of a woman and a child stand together in opposition to the relative uniformity and predictability of the “masculine” element, which is more inclined to conform to universal cognitive and behavioral patterns. This makes women and children the first and foremost carriers of the principle of the “progressive,” in the Romantic sense of the term. While men of mature age tend to a higher extent to adhere to a uniform mold, the book argues, women never cease evolving (1:60). It is because “women listen when they are spoken to” that their thought, as a consequence, knows “little rest” (1:15). The essentially dialogical nature of the consciousness of the woman and the child allows them to wrest themselves free of the presets of pure reason, with its “Stoic” acceptance of its own limitations for the sake of coherence and universality, and to reach for the mobile, “progressive” Romantic world.
By projecting the central idea of early Romanticism—the incessant motion of the cognizing spirit, whose efforts, due to the representational character of the process, are doomed to remain ever fragmentary and inconclusive—on the world of an infant, whose cognizing efforts are inseparable from acquiring and engaging signs of language, Necker de Saussure’s work posited itself as a crucial link, or transition point, between early Romantic semiotics of cognition and Saussure’s semiotics of language.
What underlay this intellectual succession was the vision of a “progressive” free-flowing succession of fragmented acts of oppositive signification, arbitrarily related to each other. According to Schlegel and Novalis, this was the only way Kantian pure reason could reckon with that peculiar endowment of humanity—its free will or, to put it in Necker de Saussure’s words, its ability to be interested in “alien objects”—to act for action’s sake. Like Necker de Saussure’s toddler, Saussure’s la langue could find its cognitive home neither in rationalism nor in empiricism, because adhering to either would mean giving up its most fundamental feature—its unconditional freedom.
Although Necker de Saussure began her close observation of her children’s behavior and characters in the 1790s and came up with the principal ideas of her book during the next decade, it took her another twenty years to overcome her perpetual uncertainties and procrastination and bring her work to completion. She struggled with her notes, more than once putting the project aside for years, despite the support and encouragement her friends gave to her project (de Staël suggested that it be entitled, “The Education of the Heart by Life Experience” (L’éducation du coeur par la vie); Schlegel exhorted her to move on with it, praising her “intellectual vigor” and “subtlety of observation”).16 The situation bears a curious “family resemblance” to the encouraging efforts of Saussure’s friends and disciples concerned with the public fate of his lectures and private notes. When Necker de Saussure spoke—in a transparent allusion to herself—of those burdened with “seeing a thousand faces in every object, which deprives them of “the energetic impulse that would follow some single motive out of many” (2:87), it sounds as if she were addressing her great-nephew directly. Still, she succeeded in pulling her work together despite the debilitating depressions from which she suffered.17 But her remarkable book, which was in essence a major intellectual document of early Romanticism, appeared in a later historical epoch. Although some theses and observations in L’éducation progressive proved to be influential in its immediate field, its philosophical background was lost on its readers. That may be one of the reasons why the book’s impact on Saussure’s ideas has remained unnoticed in Saussure studies.
Despite the book’s great success, its author, late in her life, having experienced many disappointments in her relationship with her own children,18 expressed doubts about her educational ideas, which she had developed and tried to practice in a more dynamic epoch. As she acknowledged in a letter to her daughter, “a great deficiency of the education that I had given and received consisted in its constant demand for new impressions, distraction, diversity, and a strong fear of boredom. The ideas of developing oneself, of activating one’s faculties, of giving one’s soul constant nourishment—those were the ideas I was preoccupied with.”19 Whether these words of regret, addressed to a daughter who had turned outright hostile to her, truly reflected her later convictions, is hard to say. In any event, in her book Necker de Saussure did not waver from the idea of progressivity that stemmed from her years of immersion in the spirit of the early Romantic era.
Cognition as Semiotic Process: The Interdependence of the “Defined” and the “Defining”
A century prior to Saussure’s mid-1890s notes, in which he put at the foundation of theoretical linguistics the incongruous yet indissoluble duality of the sign—whose “spirit” and “matter,” on the one hand, are totally independent from each other, yet, on the other, and precisely because of the lack of any external common ground, incapable of existing except in a mutual correlation—Novalis, in an extensive series of fragmentary notes now known as Fichte-Studien (1795–96), addressed the epistemological dilemma exemplified by the sign in a strikingly similar fashion. The Fichte-Studien offered a powerful critique of the subject-centered epistemological strategy, whose tenor showed intrinsic links to Saussure’s critique of the monolithic vision of language.
Novalis’s study was triggered by Fichte’s repsonse to Kant, which, by carrying out the principle of consciousness-oriented cognition to the utmost, exposed the limitation of that approach. Fichte countered Kant’s thesis of the categorically conditioned nature of cognition by asserting the absolute, unconditional character of the subject’s awareness of his own “I.” Fichte’s famous formula “Ich = Ich; Ich bin Ich” made the absoluteness of the subject’s identity with himself as the firm ground upon which all cognitive appropriation of the world by the subject can be built.20
Novalis’s principle objection pointed to the inescapably representational character of all the appropriations of the world by consciousness, including its awareness of itself. According to Novalis, the Fichtean absolute “I,” defined in terms of self-equivalence (“Ich bin Ich”), would simply remain inaccessible to itself. Nothing can be perceived from inside without being externalized as an object of perception; to perceive himself, to become aware of his own existence, the subject has to gain an outside perspective from which he can contemplate his own being.
By assuming an external perspective, “I” becomes a representation of itself, a “thing” perceivable from a particular point of view. Paraphrasing Fichte’s thesis that the existence of “I” is “unconditional” (unbedingt), Novalis declares, with a tinge of irony: “Wir suchen überall das Unbedingte, und finden immer nur Dinge” (We seek the unconditional everywhere, and always find merely things) (BS, no. 1). To translate Novalis’s aphorism into the terms of Saussurean semiology, it could be said that we are always seeking meaning and are always left with signs. Prior to Saussure’s Course de linguistique générale, perhaps no other author of modern times showed such an appreciation of the key importance of the fact that the substance of cognition can be reached only through semiotic externalization.
The crucial point that connects Saussure’s “general linguistics” with Jena epistemology lies in the essentially representational nature of both the cognizing and the linguistic consciousness. According to Novalis, cognition exists only in and through external projections, as does language as a semiotic phenomenon. It is the semiotic duality of the sign—the fact that it exists only as a reified externalization of itself—that sets it apart from all other phenomena, both material and spiritual. This is what makes language so intractable as the subject of linguistics. Both Novalis and Saussure were able to appreciate the fatal epistemological consequences of the symbiosis of the signifier and signified: namely, that all their duality can offer is an indirect and partial comprehension, made from a contingent point of view and bound to be challenged by the infinite plurality of alternative acts of comprehension made from other points of view.
Novalis was as explicit as Saussure in asserting the inseparability of meaning from its external representation: “We know something only insofar as we express it” (VF, no. 267); “Something becomes comprehensible only through representation” (AB, 246, no. 49). The signifier breaks the intractable wholeness of the signified phenomenon (its Beharrlichkeit) by exposing its heterogeneous constitution (heterogene Bestandtheilen) (FS, no. 464). By doing so, it becomes a “differential” of the idea. In a statement presaging Saussure’s semiotics of differences, Novalis proclaims: “Phenomenal appearances are differentials of ideas” (Novalis 1901:93). By “carving out” a phenomenon, through its representation, from the continuum of the world (the Weltall), a sign makes it contingent on its stance vis-à-vis other signs: “An object emerges out of a counterpart, a counterpart out of the object, so that the identity of the object for itself must be determined contrastively. Coexistence and co-determination are one and the same” (FS, no. 284).
As if anticipating Saussurean terminology, Novalis avoided the traditional terms form and meaning, which presented the two poles of the sign as separate. Instead, both authors adopted terms whose close etymological kinship highlighted the inseparable bond between the sign’s “spiritual” and “bodily” aspects. The close parallel between Novalis’s German terms, das Bestimmende (“defining, distinguishing”) and das Bestimmte (“defined, distinguished”) (FS, no. 1), and their Saussurean French counterparts, signifiant and signifié, reflected how similarly they understood the double nature of the sign. The way Novalis emphasizes the interdependence of the “defined” and the “defining” sounds almost like a quotation from Saussure’s Course: “The subject matter is defined not otherwise than by its defining, its counterpart defines not otherwise than by its defined” (FS, no. 291).
Cognition through externalization is a process whose results are of necessity fragmentary and partial. It is impossible to access the world directly; all that can be accessed is an infinite variety of partial visions created by various representations. The fragmented character of signification causes it to evolve “parabolically,” by leaps from one instance of cognition to another (TF, no. 349). The process essentially lacks any consistent pattern—it is an “inner matter” of the sign, not to be subjugated to or directed by any outward principle.
Novalis’s famous Monologue is an emphatic declaration of the total freedom of language, grounded in the immanency of the double structure of the sign:
It is a ridiculous delusion, which causes one to marvel, to think that when one is speaking, one does so for the sake of things. Nobody is aware of the proper nature of language, namely, that it is concerned with nothing but itself.… If only one could make people understand that language is like a mathematical formula. Both constitute a world of their own—they play with nothing but themselves, express nothing but their own wonderful nature, and it is precisely because of this that they are so expressive.… They become a part of nature solely as a result of their freedom, and it is solely through their free movement that the universal soul expresses itself, making them a gentle scale and foundation of things. (Novalis 1981:672)
Novalis’s free-flowing discourse (itself a vivid representation of what he strives to convey) could easily be interpreted as mere “Romantic” effusions about the wonders of glossolalia-like irrationalism. Yet seen in the context of Novalis’s critique of contemporary theories of the sign and cognition, the Monologue highlights points of crucial importance about the most fundamental nature of language, which foreshadow Saussure’s philosophy of language. That seeing language as a tool through which one reaches the phenomena of the world is “a ridiculous delusion” (here, Novalis’s lächerlicher Irrtum anticipates Saussure’s favorite ridicule); that it is in fact a free interplay, responsible to and expressing nothing but itself; and that it is precisely this total freedom of language that connects it to nature, making it a “gentle scale and foundation” of things rather than a label attached to them—these points of the Monologue contain, in embryonic form, the quintessence of Saussure’s teaching about language: its immanent “emptiness” resulting in its unbounded freedom.
Asymmetry of the Sign: “Linearity” as a Philosophical Concept
One of the points in Saussure’s Course that provoked much controversy even at the heights of its influence concerned the notion of “linearity.” Saussure presented the “linearity of the signifier” as one of the two fundamental (a priori) properties of the sign, on a par with the principle of arbitrariness. Once again, he resorts to a grave admonition to potential addressees who might fail to recognize this “obvious” issue, whose consequences, meanwhile, are “incalculable,” without further elaborating on what he actually means by this (CLG, 103/70). And once again, as in the case of arbitrariness, this typically Saussurean emphatic yet elliptical assertion gives rise to all kinds of critique based on numerous examples, indeed quite “obvious,” that seem to contradict his thesis.
When examined against the empirical data of language, the thesis that linearity is universal seems to run up against a number of instances in which the relations between linguistic units supersede a linear order. Even on a purely phonetic level, the Saussurean pronouncement “c’est une ligne” (CLG, 103/70) seems to be an oversimplification. As Jakobson convincingly argued, relying on this principle in accessing the flow of sounds in speech would create a vicious circle: one considers speech to be a succession of sounds because one segments speech into successive sounds (Jakobson and Waugh 1987:22). The phonic texture of speech at large could easily refute the idea of its linearity: in continuous speech, the intonational curve, word stresses, and changes of speech tempo evolve according to rhythms of their own, which overlap with each other and frustrate the linearity of the sound progression (Harris 1987:70). When all the sound dimensions of speech are taken into account, its phonic texture appears to be, at the very least, “multi-linear.”21
Even more obviously, the linear progression is often broken at the syntactic level, where structural relations between words often supersede their linear order in a phrase. Lucien Tesnière (1959), one of the founders of structural syntax, suggested a distinction between what he called the “structural” and the “linear” syntactic order (l’ordre structurale versus l’ordre linéaire)—a concept clearly aimed at amending Saussure’s thesis. Generative “syntactic rules” departed even further from the idea of a sentence as a sequence of words, overriding not only the linear but even the structural order of “surface” syntactic structures.
Finally, elements of the rhetorical, or “poetic,” fabric of a text feature manifold patterns of repetitions and parallelisms that defy the linearity on the textual level (Jakobson 1987). Jakobson (1960) famously defined the “poetic” aspect of language as that in which “the axis of selection” (i.e., paradigmatic copresence) is superimposed upon “the axis of combination.” According to Jakobson, nonlinear associative relations between various points in a text rearrange it in multiple ways, complicating, or even frustrating, its linear progression. Moreover, the discovery of Saussure’s anagram studies revealed his own awareness of the nonlinear aspect of discourse (Holdcroft 1991:58–60).22
Both Saussurean fundamental principles—linearity and arbitrariness—received a very similar treatment in the critical literature. In both cases what Saussure declared to be a transcendental property of language was understood—due in no small part to his elliptical presentation of the issues—as a statement about a certain empirical feature directly observable in linguistic data. Once placed in the empirical domain, however, the concept runs up against copious evidence that contradicts it—evidence so obvious that it seems strange that Saussure could overlook it.23 The question should be asked, however: when Saussure referred to linearity as to one of the two fundamental properties of the sign, did he really mean by that the rather trivial—and not entirely correct—fact that when we produce speech it emerges piece by piece in a “linear” temporal succession? Here as elsewhere, Saussure’s thought should be examined in the context of his general philosophical postulations about language—which he expressed at length and rather eloquently—rather than on the grounds of the cursory definitions and impromptu examples with which he often approached more specific issues in his lectures.
In assessing the issue, one should take heed of the fact that Saussure discussed linearity as a property specifically belonging to the signifier—in contradistinction to arbitrariness that characterized the bipolarity of the sign as a whole.24 In a letter to Gaston Paris (December 30, 1890) Saussure elaborated on this distinction, stating that linearity, which represents the point of view of “phonetics,” is “primordially incompatible” with the point of view of “morphology”; the former has to do with successivity, while the latter has to do with (nonsuccessive) “sense, valeur.”25
Once again, “phonetics” and “morphology” seem rather poorly chosen terms, capable of obfuscating even further the “primordial” nature of the issue. Here is a point at which Saussure’s putative link to Novalis can offer a way out of all the confusion. Thanks to his concentration on the metaphysical rather than specifically linguistic aspects of the sign, Novalis comes to the rescue of Saussure’s concept by giving it an unequivocal philosophical dimension that Saussure himself could not or would not articulate.
For Novalis, the indissoluble symbiosis of das Bestimmende and das Bestimmte is as important as their asymmetry. The signified belongs to the spiritual realm. It is not a material “object,” which means that it does not occupy a particular physical space. The semantic space allotted to a signified when it is “captured” in a signifier is not its own; to remember Saussure’s metaphor, the signified “fills” the signifier the way hydrogen fills a balloon—without the signifier, it would not have any distinct semantic space at all. Contrary to this, das Bestimmende belongs to the world of physical objects. It is a “thing”—a thing of a peculiar kind, to be sure, since it comes into existence solely by being connected to something it “defines”; still, with or without this connection, it remains a piece of matter. Like any matter, it is fragmentary: it is a piece, a particle that exists alongside many other particles of various sizes and shapes. Spiritual phenomena are not fragmentary: one cannot say that a portion of meaning captured in a sign is a “particle” of spiritual matter. The etherlike spiritual domain knows neither finitude nor any shape of its own. The spiritual realm is ultimately whole, although its ultimate unity is doomed to remain nameless (AB, 290, no. 285). It appears to us apportioned into signs, yet any phenomenon of meaning thus apportioned can dissolve into or fuse with any other at any moment. The world of signifiers, on the other hand, is a world of fragments. It is not whole: it exists only as a conglomeration of fragmentary entities; no matter how many signifiers are at hand, they never cease to be fragments. They never “fill up” the whole space of signifying possibilities because, by their very nature as “pieces,” they can never attain wholeness.
The credit for recognizing the metaphysical dimensions of Saussure’s linearity, which go beyond syntagmatic successivity, belongs to Roy Harris. According to Harris, Saussure’s linearity means that the signifier “occupies a certain temporal space (une étendue)” (Harris 1987:71); what linearity means in this more abstract sense is that “all human activities in the external world take place in time” (76). Indeed, linearity belongs to the “external world,” as opposed to the internal spiritual world of semiotic values. What Harris’s explication does not highlight, however, is the constitutional fragmentariness of semiotic activity, brought about by its attachment to the material world with its spatial/temporal dimensions.
The fragmentariness of signifiers constitutes a “primordial” contrast with the world of signified values, which always, under whatever configuration, constitutes a hermetically closed system. It creates a fundamental paradox in la langue: on the one hand, it is hermetic, always perfectly self-sufficient, but, on the other, it is perpetually fragmentary and, as a result, ever in a struggle with and under tension over whatever its current state is. Being a “thing” (das Ding)—a piece of the matter with which both Novalis’s subject and Saussure’s speaker try to capture the “unconditional” (das Unbedingte)—a signifier always turns out to be a fragment among a multitude of other fragments, in a field that could expand through all eternity without ceasing to be an agglomeration of piecemeal entities. It makes the drive for changes in the realm of signs (what Saussure calls “mutability” and Novalis describes with metaphors of “hunger” or erotic desire) unquenchable and unceasing.
In the spiritual domain, everything can potentially be connected with everything else. There is nothing to prevent thought from running from one idea to any other, from conflating them, transforming them one into another, modifying them into infinite variations in which they brush up against an infinite number of other ideas. Yet the oppositive character of the sign system makes it locked, or “immutable,” in mutual oppositions in which the “values” exist literally by clinging to each other. Under these conditions, there seems to be no room to set the system in motion, to unlock its crystalline nodal junctions. Here, however, is where “linearity” comes into the picture. The noncompartmentalizable, ethereal, volatile nature of meaning evolves alongside its apportioning and reapportioning into enclosed semiotic entities. By exposing the inherent fragmentariness of every sign, linearity undermines the seemingly impervious intrasystemic balances on which the whole system rests.
Viewing the system in all its hermetic splendor, one may feel that nothing can be done to sign A because it is correlated with sign B, while nothing can be done to B because of its correlation with A. That would indeed be the case if the signs were devoid of “linearity.” However, the nature of signifiers as “things” whose number and configurations are absolutely fortuitous means that the manner in which they divide meaning into signifieds is always provisional and inconclusive. Two signs defined by the mutual difference “A-is-not-B” and “B-is-not-A” never present an airtight opposition. A “breach” will always remain between the two signifiers, leaving room for the signs’ proliferation or elastic extension. The semiotic space expands as if ever striving to fill the gap between existing signs; in Novalis’s words, “and thus philosophy grows into infinity, outward and inward, while it strives to fill the infinite space between its elementary components” (LF, no. 18).
Form and Substance Revisited
In his notes, Saussure repeatedly speaks about the impossibility of reaching beyond the network of purely oppositive and negative differentiations in which the perpetually elusive essence of language presents itself: “One will never ever penetrate the purely negative, purely differential essence of every element of language to which we precipitously accord reality”—albeit, Saussure “has to admit,” perhaps we need to adhere to the usage of conventional “fictions” as labels, without which those very differences would be impossible even to point out (ELG, 64–65). Saussure’s writing moves along with this elusiveness of language, never reaching beyond fragmentary adumbrations of his ideas. In this it recalls the Athenaeum strategy of fragmentary writing as the only way to convey the constant motion of the spirit as its cognizing condition. The “truth” about the matter never allows itself to be addressed directly; it is lurking, as it were, in the cracks between fragmentary representations, each manifestly inconclusive because of the very fact of its representational character: “We speak neither of axioms, nor of principles, nor of theses. These are, simply put and in the purely etymological sense, aphorisms, or delimitations … yet the ones [indicating] limits between which the truth can always be found, whatever the point of departure” (ELG, 123).
It was in 1906, more than a decade after his sketchy linguistic notes and just before he began his lectures, that Saussure made an attempt at reckoning with the “substance” of the problem in his notes on the Rig Veda. There, he spoke about the “ardent preoccupations directed at (making) the object of cognition unequivocally absolute” that drove the thought of the Indian Brahmans. Yet such an object is “absolutely indifferent insofar as it is absolute”; no one can declare that he has “grasped” the absolute substance (qu’on l’a saisie), since there is “nothing to cognize” in it (Saussure 1993a:218–219). The ideal condition for reaching the absolute would be dreamless sleep; in this case the subject could be said to have reached an ideal harmony between the world and his propre moi precisely by becoming totally oblivious of both (221, 223).
Quite characteristically, Saussure does not fail to plunge into an angry outburst against this unreasonable longing: “It is paradoxical in the highest degree, though, or childish if you prefer, this picture of so ardent a preoccupation directed toward an object that is doubtless absolute but also, insofar as it is absolute, absolutely indifferent. We, adherents of the spirit of the Occident, understand the dilemma: cognize, or, if it is inaccessible, don’t try to cognize. None of this is understood in India.”26 For a moment the “Western” spirit emerges in all the rationalist glory of the Kantian critique of cognition, scoffing at “Oriental” irrationality. And yet—“néanmoins, toutes notres distinctions, toute notre terminologie, tout nos façons de parler sont moulées sur cette supposition involontaire d’une substance” (nevertheless, all our distinctions, all our terminology, our whole manner of speaking are molded by that involuntary supposition of a substance).27
As it turns out, the principle of the indifference of the substantial absolute does not make “ardent preoccupation” with it futile. The only way to “reach the supreme substance” is to channel it into a plurality of “forms” that are “conditional, diverse, and multiple.” Which one of these relative “forms” one chooses is of no importance in itself. Their relation to each other is not hierarchical: they are a company of “neighbors,” each related to its absolute “base” in one way or another (Saussure 1993a:220). Whatever choice one makes among those possibilities, it is necessarily an arbitrary choice (198). One should abandon attempts to unite all these pluralities into a single rationally founded doctrine: “rien de plus découragement que de chercher une formule rationnelle” (194). The substantial object can be grasped from all the alternative possible systems of its differential representation, with no need to settle on any one of them (196).
We can now see the implications of Saussure’s thesis that language is a form and not a substance, that its nature is “oppositive and not positive, absolute,” and that there is nothing in language but differences. It would be a profound mistake to take these assertions of the Course as professions of an unperturbed relativism (or, otherwise, as a perversion of Saussure’s thought perpetrated by the publishers). The connection with early Romantic philosophy of language reveals what stands behind these terse formulations. By endowing it with freedom, the representational nature of language defies any universal order. You can “grasp” language as a whole only by giving up its representational utility—e.g., by assuming a deathlike state of dreamless sleep in which, ironically, you would not be aware of having grasped it. (Novalis presented a grimmer version of the same thought, with which he became increasingly captivated in the last years of his life, namely, that death is the only state in which one can hope to attain the absolute, while Necker de Sassure posited the absolute of nondifferentiation at the opposite pole of the life spectrum—in her vision of a newborn as “the direct creation of God”). As long as one is awake, or alive, one can “grasp” only tangible forms (Novalis’s Dinge). But, whenever one grasps at a palpable entity, all one ends up with is its relation to another entity. Saussure’s thesis, like that of his early Romantic predecessors, is anti-utopian in tenor. In this it stands in marked contrast to the totalizing visions of twentieth-century “Saussurean” linguistics.