Illuminating the Unseen in Language: Saussure and the Reconstruction of Linguistic Objects
“It is only with the heart that one can see rightly; what is essential is invisible to the eye.”
—Antoine de Saint-Exupéry, Le Petit Prince
Abstract
This article examines a methodological continuity in Saussure's linguistics: the problem of establishing linguistic objects that cannot be identified simply with their observable manifestations. From the reconstruction of Proto-Indo-European vowels in the Mémoire to the analysis of la langue in the Cours, Saussure repeatedly approaches linguistic entities through the relations that make them intelligible. The move from historical reconstruction to synchronic analysis does not abandon this problem but transforms it. In la langue, the linguistic system is itself differential, and this differential organization permeates the identification of its units: phonological, lexical and grammatical identities emerge through contrasts and oppositions rather than through independently possessed positive properties. The article therefore argues for a methodological continuity in Saussure's theory: the linguistic object is established through the relations within which it acquires its identity.
1. Introduction — The Problem of an Object That Cannot Be Seen
How does a science study an object that cannot simply be placed before the observer? The question is particularly acute in linguistics. A spoken utterance can be heard and recorded; a written form can be seen and compared. Yet neither necessarily coincides with the linguistic entity that the linguist seeks to understand. The object must instead be approached through the observable manifestations and relations from which its organization can be established.
This problem appears in different forms throughout Saussure's work. In the Mémoire (1879), he reconstructs an earlier Indo-European system whose sounds cannot be directly heard. What is available are forms preserved in later languages and the correspondences among them. In his mature reflections on langue, the problem takes a synchronic form. Speech provides observable material through which the linguist approaches a linguistic system, but the system cannot be identified with any individual utterance. At the same time, langue is conceived as a differential system: its elements have linguistic identity only through their relations to other elements.
This differential principle is not introduced only when Saussure turns to phonology or grammar. Rather, his discussions of phonological, grammatical, and lexical units show how the relational character of langue is manifested at different levels of linguistic organization. The specific relations are not identical at every level, but the broader principle remains: a linguistic unit does not possess its linguistic value independently of the system in which it functions.
The continuity between these stages should therefore not be understood as a single theory already present in the young Saussure. The Mémoire belongs to comparative-historical linguistics, while the Cours de linguistique générale develops a different conception of language. What persists is a methodological problem and, with it, a recurring procedure: when the linguistic object cannot be directly grasped, Saussure seeks to establish it through the relations among observable manifestations.
This article traces that procedure from the Mémoire to the Cours, following its movement from historical reconstruction to the synchronic analysis of langue, and then to the differential organization of phonological, lexical and grammatical units. The aim is not to collapse these different moments into one theory, but to show how a persistent problem of access contributes to Saussure's changing conception of the linguistic object.
2. The Mémoire: Reconstructing What Cannot Be Heard
The problem appears at an early stage of Saussure's career, before the theoretical vocabulary associated with structural linguistics had taken shape. The Mémoire sur le système primitif des voyelles dans les langues indo-européennes, published in 1879, belongs to the comparative-historical tradition. Its concern is the reconstruction of the vowel system of Proto-Indo-European and, more specifically, of elements whose effects can be observed in the historical forms of the daughter languages. Yet the work already raises a question that will remain important throughout Saussure's linguistic thought: how can a linguistic entity be established when it is not itself available to direct observation?
A historical linguist cannot listen to Proto-Indo-European. No recording, speaker, or directly observable pronunciation can provide access to the reconstructed language. What is available instead is a body of linguistic evidence: forms preserved in different languages, recurring correspondences, alternations, and systematic patterns. The investigator begins with what can be observed and asks what organization is required if those observations are to be explained coherently.
The distinction between evidence and object is therefore essential. The forms found in the daughter languages are not themselves Proto-Indo-European forms. They are manifestations of historical developments whose relationships can provide information about an earlier linguistic system. Saussure's reconstruction depends upon these relationships. An element that cannot be heard directly may nevertheless be required by the structure of the evidence. Its status does not rest on the existence of an immediately available acoustic specimen but on the explanatory role it performs within the reconstructed system.
This is one reason why the Mémoire is better understood as a work of systematic reconstruction than as an attempt simply to identify unknown sounds. Saussure's concern is not exhausted by the question of what a particular vowel might have sounded like. The larger problem is how the correspondences among forms can be explained by a coherent system. A reconstructed element derives its significance from the relations in which it participates. It is posited because the organization of the evidence calls for it.
The point becomes especially interesting in relation to the elements Saussure proposed as coefficients sonantiques. Their theoretical importance does not depend upon their being recoverable as familiar, directly observable vowel sounds. They occupy a place in an explanatory system. Their existence is inferred from the way they account for patterns otherwise left unexplained. The linguist thus moves from observable forms toward an element that cannot itself be observed in the same manner.
This does not mean that Saussure's early work had already reached the mature conception of linguistic value. The Mémoire remains firmly situated within nineteenth-century comparative linguistics, and its questions are historical rather than synchronic. It would therefore be misleading to read the later structuralist theory retrospectively into the young Saussure. What can be identified, however, is a methodological continuity. The object under investigation is not simply identified with any directly observable sound. Its status emerges from the relations that make the observed forms intelligible.
The significance of the reconstruction lies precisely in this relation between evidence and hypothesis. Saussure's reconstructed elements are not arbitrary inventions, but theoretically constrained entities whose justification depends upon the correspondences they help explain. Their status therefore does not depend upon direct observation. They are established through the systematic relations among the forms that remain accessible.
Reconstruction, then, does not mean inventing an invisible object to replace the visible one. The reconstructed element remains constrained by what can be observed. Its status depends upon the relations among attested forms and upon its capacity to account for them. The inaccessible is approached not by abandoning empirical evidence but by organizing that evidence in a way that makes another level of linguistic structure intelligible.
The Mémoire therefore presents an early form of a problem that will take a different shape in Saussure's mature linguistics. The young comparative linguist has to establish a linguistic system whose relevant elements cannot themselves be observed directly. The mature linguist will confront a related problem within a synchronic framework. The objects will change, and so will the analytical procedures, but the difficulty of access remains.
3. From Historical Reconstruction to Langue
The movement from the Mémoire to Saussure's mature reflections on language changes the object of investigation, but not the difficulty of gaining access to it. Comparative linguistics reconstructs an earlier linguistic system from forms preserved in later languages. General linguistics, by contrast, seeks to describe the organization of a language at a given moment. Yet langue is no more directly available to observation than Proto-Indo-European was. The linguist encounters individual utterances, but the system cannot simply be identified with any one of them.
Saussure's mature notes pose the problem directly: what are the concrete entities that constitute langue? The question might seem to call for identifiable objects, but Saussure's formulation complicates precisely this assumption. Linguistic entities cannot simply be isolated and inspected as independent things. They must be approached through the observable manifestations in which the organization of the system becomes accessible.
This explains the methodological importance of words and speech. Saussure notes that words can serve as specimens through which linguistic units become accessible to investigation. As he puts it, “Since we cannot have direct access to concrete entities and linguistic units,” words can provide “a rough idea” of the signs constituting a synchronic system (CGL, p. 158). The qualification matters. Words are not identical with the linguistic units they exemplify; they are observable manifestations through which the organization of the system can be approached.
The same distinction becomes especially clear in the Constantine notebooks, where Saussure writes: La parole ne figure ici que comme un document de langue—“Speech comes in here only as evidence bearing on the language.” The choice of document is revealing. A document is neither the reality it concerns nor an arbitrary substitute for it. It provides evidence from which something about that reality can be established. Speech therefore occupies an intermediate position: it is observable, but its significance for the linguist lies in what it reveals about the organization of langue.
This does not mean that langue should be understood as a hidden substance lying behind speech. If parole were merely a substitute for an already constituted object, the linguist's task would amount to recovering something that exists independently and fully formed behind individual utterances. Saussure's formulation suggests something more demanding. The system has to be approached through its manifestations and through the regularities that emerge among them. Speech provides indispensable material, but the system is not reducible to any particular act of speaking.
This is also why the distinction between langue and parole should not be understood simply as a division between an abstract realm and a concrete one. The distinction is methodological as well as theoretical. Parole provides access to linguistic phenomena, while langue names the organized system whose relations must be established through them. The individual utterance is therefore indispensable without being sufficient.
At this point, however, another difficulty appears. If langue is a system, what makes its elements linguistic elements in the first place? The answer cannot be that each element possesses a self-sufficient identity and is subsequently inserted into a system. Saussure's conception of the system points in the opposite direction. The elements are what they are within the relations that distinguish them from one another.
The differential character of langue therefore should not be treated as a principle added after the system has been identified. It is part of what it means for langue to be a system. The units that compose it do not first possess linguistic identities independently and then enter into relations. Their linguistic identities emerge within those relations.
This is where the continuity with the Mémoire becomes visible, while the difference between the two projects remains intact. In historical reconstruction, attested forms provide evidence from which a non-attested system can be inferred. In synchronic linguistics, speech provides evidence through which the organization of langue can be approached. The procedures are not identical: one operates across historical correspondences, the other within a contemporary linguistic system. What they share is a refusal to equate the linguistic object with whatever happens to be immediately observable.
The difference is that the mature Saussure makes the relational character of the object itself increasingly explicit. The linguistic system is not merely something inferred from evidence; it is a system in which the identity of its elements depends upon their relations to one another. The problem of access thus leads directly to the problem of linguistic identity.
4. The Phoneme: Difference at the Level of Sound
Saussure's discussion of phonological units makes visible what the differential conception of langue means at the level of sound. The issue is not that difference suddenly becomes relevant when Saussure turns to phonology. Rather, phonology provides a particularly clear case in which the linguistic identity of an element cannot be derived from its positive physical properties alone.
Saussure states the point directly:
“What characterises those units is not, as might be thought, the specific positive properties of each; but simply the fact that they cannot be mistaken for one another. Speech sounds are first and foremost entities which are contrastive, relative and negative” (CGL, pp. 164–165).
The formulation requires some care. Saussure is not claiming that speech sounds lack physical properties. A sound can be described acoustically and articulatorily; it has measurable characteristics and a material realization. His argument concerns the source of its linguistic identity. Physical properties alone do not determine whether a difference matters within a particular language.
His example of French r makes the point clearly. The sound can be realized as a uvular consonant or as an apical trill without thereby becoming a different linguistic unit. French does not require an identical acoustic event every time r is pronounced. What matters is that its realizations remain distinguishable from the other consonants of the system. The same physical or articulatory realization can have a different linguistic status elsewhere: Saussure notes that a French r could even be pronounced like the German ch of Bach, whereas German maintains a distinction between r and ch.
The example separates physical realization from linguistic function. A particular acoustic event does not carry its linguistic identity with it independently of the language in which it occurs. The same or similar sound may participate in different linguistic oppositions in different systems, while different physical realizations may count as instances of the same linguistic unit within a single system.
This is why Saussure describes phonological units as negative. “Negative” does not mean unreal or nonexistent. It means that their linguistic identity is determined through distinctions from other units rather than through an autonomous positive essence. To identify a phonological unit is therefore to determine its place within a network of contrasts.
Variation makes the point especially clear. Speakers do not reproduce precisely the same acoustic event every time they pronounce what linguists identify as the same phonological unit. Individual realizations vary according to speaker, context, and other conditions. Yet such variation does not necessarily produce a new linguistic unit. What remains stable is not a single physical substance but the distinction maintained by the language.
The converse is equally important. A small physical difference can become linguistically decisive when a language organizes it as a contrast. Two sounds that are close phonetically may belong to distinct phonological categories if replacing one with the other affects the identity of a form. The linguistic relevance of a difference therefore depends not simply on its physical magnitude but on the distinctions maintained within the system.
This is where Saussure's formulation that there are “only differences without positive terms” becomes particularly useful. Applied to phonology, the statement does not deny material sound. It denies that material substance, considered in isolation, is sufficient to establish linguistic value. A sound acquires its linguistic place through the differences that distinguish it from other sounds.
This also clarifies why calling the Saussurean phoneme simply “abstract” can be misleading. If abstraction suggests an entity detached from speech, the term obscures the relation between linguistic structure and material realization. Saussure does not remove physical sound from linguistics. He distinguishes the physical realization from the linguistic function that a distinction assumes within the system.
The phoneme therefore cannot be understood as a small physical object hidden inside speech, waiting to be discovered. Nor is it merely an arbitrary mental category imposed upon sound. Its linguistic identity lies in the system of distinctions through which it is differentiated from other units. What persists across variable realizations is the opposition maintained by the language.
The significance of phonology for the larger argument is therefore methodological. The linguistic object is not made inaccessible by the fact that speech is material. Rather, material speech becomes linguistic evidence only when it is considered in relation to the distinctions organized by the system. What the linguist seeks is not a physical substance concealed within the sound but the network of differences that determines which differences matter.
Phonology thus makes explicit a principle already implicit in the conception of langue: linguistic identity is relational. The differential character of the system does not remain at the level of an abstract definition of langue; it is operative in the identification of particular units.
But the principle does not stop at sound. If the differential organization of langue is genuinely general, it should also become visible wherever linguistic units acquire their value through relations with other units. Saussure's discussion of grammatical facts provides a particularly clear example.
5. Difference Beyond Phonology
The phoneme provides one instance of a broader principle. The differential organization of langue is not confined to sound. It is manifested, in different forms, wherever linguistic units acquire their identity or value through relations within the system.
Saussure's example of the German Nacht : Nächte is particularly revealing:
“What is usually called a ‘grammatical fact’ corresponds in the final analysis to our definition of a linguistic unit. For there is always an opposition of terms involved.”
The distinction between Nacht and Nächte is not simply a matter of two independently existing forms, one singular and the other plural. The grammatical opposition is realized through several formal differences: the singular lacks the umlaut and final -e, whereas the plural has both. Yet these features acquire their grammatical significance through the opposition in which they participate.
Saussure's conclusion is deliberately strong: “In isolation, Nacht and Nächte are nothing: the opposition between them is everything” (CGL, p. 168). He does not mean that the two forms cease to exist phonetically or graphically outside their grammatical relation. Rather, their grammatical value cannot be determined by considering either form independently. What matters is the function of the distinction between them.
The example is important because it prevents us from restricting the differential principle to phonology. The opposition between singular and plural is not simply a difference between two positive forms that happen to coexist. The grammatical identity of each form depends upon the system of oppositions in which it participates. The relation is therefore constitutive of the linguistic value of the forms.
The same principle extends to the lexical level. A word's value is not exhausted by its association with a particular concept; it is also shaped by the differences separating it from other terms in the same language. A lexical term occupies a position within a field of alternatives, and its value depends partly upon what it is not.
The mechanisms involved at these different levels are not identical. A phonological opposition is not the same thing as a grammatical opposition, and neither is reducible to a lexical relation. The point is not that all linguistic differences work in exactly the same way. It is that the linguistic identity of units at each level cannot be understood independently of the relations that organize that level.
This distinction is essential. If we say that the differential principle "trickles down" from langue into phonology, grammar, and the lexicon, we risk imagining langue as a higher-order entity that first exists independently and then imposes its structure upon subordinate levels. Saussure's conception is more relational than that. Langue is a system precisely because its elements stand in organized relations, and those relations are what become visible when we analyze its different levels.
The movement from phonology to grammar therefore does not represent a movement from one theory to another. It demonstrates the same general problem in a different domain. In both cases, the linguist must determine what makes an element linguistically what it is. In both cases, the answer lies not in isolating a positive essence but in identifying the relations that distinguish the element within the system.
The linguistic object is consequently not established by isolating a positive substance. Its identity becomes intelligible through the relations in which it functions. What changes from one level to another is the specific form those relations take, not the fact that linguistic identity is relational.
This brings us back to the problem with which the article began. If linguistic objects cannot simply be identified with observable manifestations, what allows us to establish them? Saussure's answer, across different stages of his work, is increasingly clear: they become intelligible through relations. Historical correspondences allow a non-attested system to be reconstructed; synchronic manifestations provide access to langue; phonological contrasts establish units of sound; grammatical oppositions establish values among forms.
The relation between the levels should therefore not be understood as a hierarchy in which one principle is successively applied from above. Rather, the differential character of langue is encountered at each level in the relations through which its units become linguistically identifiable.
6. Conclusion — A Method of Access, Not a Finished Theory
Saussure's work does not present a single theory fully formed from the beginning. The Mémoire of 1879 and the Cours belong to different intellectual contexts and address different linguistic problems. The young comparative linguist should therefore not be treated as an undeveloped structuralist whose later vocabulary was already implicitly present.
There is nevertheless a significant methodological continuity between these moments. In the Mémoire, Saussure reconstructs an unattested linguistic system from forms preserved in later languages and from the relations among them. In his mature linguistics, langue is approached through parole, while phonological and grammatical units are identified through the differences and oppositions that organize them. In each case, the linguistic object is not simply given in an immediately observable form.
What persists, then, is not a finished doctrine but a method of access. Saussure begins with manifestations that can be observed and asks what relations are necessary to establish their linguistic organization. Historical correspondence becomes synchronic relation; reconstruction becomes the analysis of a system; and the question of what an entity is increasingly becomes a question of what makes it distinct within that system.
The continuity should not be exaggerated. Saussure's reconstruction of Indo-European vowels cannot simply be equated with his later theory of linguistic value, and the concepts of the Cours should not be projected unchanged backward onto the Mémoire. What can be traced is a persistent intellectual difficulty: the linguistic object cannot be isolated from the relations through which it becomes intelligible.
This difficulty also explains why some of Saussure's methodological distinctions resist being kept completely separate. Langue and parole, system and element, material realization and linguistic unit, identity and difference: each distinction is necessary for analysis, yet neither side is fully intelligible independently of the other. The distinction between them is therefore not the discovery of two self-sufficient objects. It is a way of making visible a relation that constitutes the object of inquiry.
The differential character of langue should be understood in the same way. Difference is not a principle added to an already constituted system when Saussure turns to phonology or grammar. It is operative in the organization of the system itself, and the analysis of different linguistic levels reveals different forms of its operation. Phonological contrast and grammatical opposition are not identical, but both demonstrate that linguistic identity depends upon relations rather than upon independently possessed substance.
Saussure's achievement, then, may be understood not simply as the discovery that language is a system of differences. More fundamentally, he develops increasingly precise ways of studying a linguistic system whose objects are never simply given in any one of their manifestations. What cannot simply be seen or heard does not therefore remain inaccessible. It can be reconstructed, differentiated, and analyzed through the relations in which it becomes intelligible.
The linguistic object is thus neither a hidden physical thing waiting to be discovered nor an arbitrary construct imposed upon observable material. It emerges at the point where manifestations, differences, and relations become sufficiently organized for something to count as a linguistic unit. To study the unseen in language is therefore not to look behind its observable forms for another substance, but to reconstruct the relations through which those forms acquire linguistic identity.
Related Post
The Principle of Ascent: Illuminating the Unseen in Language
https://derridaforlinguists.blogspot.com/2024/04/blog-post_06.htmlReferences
Joseph, J. E. (2012). Saussure. Oxford University Press.
Saussure, F. de. (1959). Course in general linguistics (W. Baskin, Trans.). Philosophical Library. (Original work published 1916)
Saussure, F. de. (1995). Writings in general linguistics (C. Bally & A. Sechehaye, Eds.; S. Bouquet & R. Engler, Eds.). Oxford University Press.
Saussure, F. de. (2013). Mémoire sur le système primitif des voyelles dans les langues indo-européennes. In Recueil des publications scientifiques de Ferdinand de Saussure (pp. 1–268). De Gruyter. (Original work published 1879)
Comments
Post a Comment