Saussure and the Problem of the Sign

1. The Problem of Conceptual Vocabulary

Words such as word, meaning, concept, and sign seem familiar enough in ordinary use. We can usually understand one another without deciding exactly what each denotes or how the terms relate to one another. In everyday language, speakers intuitively recognize the units that matter in discourse, even though they may not be able to explain systematically how those units are distinguished or related. The difficulty begins when these expressions enter a theoretical vocabulary, where distinctions that ordinary language leaves implicit may need to be made explicit.

This raises a more specific question: What happens when the vocabulary available to a theory does not provide the distinctions that the theory needs to make?

At that point, the task may involve more than determining what a familiar concept means. It may require refining an inherited term, assigning it a more precise function, or introducing new terms whose relations make a theoretical distinction explicit. The purpose is not to replace ordinary language as such, but to develop the vocabulary required for a particular intellectual task.

Rudolf Carnap later gave this kind of activity a systematic methodological formulation in his theory of explication. Ferdinand de Saussure provides an earlier and more concrete instance of the problem. Carnap approaches the problem from above; Saussure encounters it from within.

2. Carnap and the Method of Explication

Carnap's method of explication begins with a distinction between an existing concept and a newly constructed one. The original, relatively vague or inexact concept is the explicandum; the more precise concept constructed for a particular theoretical purpose is the explicatum.

The relation between the two is important. An explicatum should remain sufficiently connected to the original concept to preserve relevant uses, but it need not reproduce everything that ordinary usage associates with it. Carnap is therefore not asking what the original concept “really” means and then uncovering its definitive form. He is asking how a more exact version can be constructed for a specified theoretical purpose.

Carnap identifies four main desiderata for an adequate explicatum: similarity, exactness, fruitfulness, and simplicity. These do different kinds of work. Similarity preserves a connection with the earlier concept; exactness gives the new concept clearer rules of use; fruitfulness makes it useful within a wider theoretical system; and simplicity favors a less complicated construction when the other requirements are adequately met. The four criteria may pull in different directions, and their relative importance depends on the purpose of the explication.

This makes explication less a search for the one true definition than a form of deliberate conceptual construction. An explicatum can therefore be better suited to one theoretical task than another without becoming the uniquely correct version of the original concept.

Adequacy, then, is a matter of balance rather than maximization. The aim is not to eliminate every imperfection inherited from ordinary usage, but to construct a concept that preserves enough continuity with its starting point while gaining the precision and usefulness required for a particular purpose.

3. Saussure as a Case Study

Saussure's discussion of the linguistic sign gives us a concrete instance of the kind of problem that Carnap later formulates in methodological terms. He begins by describing the sign as an association of a concept and an image acoustique. The formulation seems clear enough until he notices that signe already has a different range of use in ordinary language. It can refer to the sensory element alone, as when arbor is called a sign. Saussure's analysis, however, requires signe to designate the association of the two sides.

The difficulty is therefore terminological in a precise sense. Saussure needs terms that will distinguish the two elements from one another and from the totality they form. He states the problem explicitly:

« Cette définition pose une importante question de terminologie. Nous appelons signe la combinaison du concept et de l’image acoustique : mais dans l’usage courant ce terme désigne généralement l’image acoustique seule, par exemple un mot (arbor, etc.)»

If signe is retained for the whole, the two elements within it need names that correspond to their respective roles in the analysis.

His solution is to reorganize the terminology. He proposes retaining signe for the whole and replacing concept and image acoustique respectively with signifié and signifiant. The resulting vocabulary is correlated: signifiant and signifié designate the two sides, while signe designates the totality formed by their association:

« L’ambiguïté disparaîtrait si l’on désignait les trois notions ici en présence par des noms qui s'appellent les uns les autres tout en s'opposant. Nous proposons de conserver le mot signe pour désigner le total, et de remplacer concept et image acoustique respectivement par signifié et signifiant »

The new terms thus make it possible to refer to the components of the analysis without confusing either one with the whole.

Yet Saussure does not present this reorganization as a perfectly satisfactory terminological solution. He explicitly acknowledges a limitation:

« Quant à signe, si nous nous en contentons, c’est que nous ne savons par quoi le remplacer, la langue usuelle n’en suggérant aucun autre »

He keeps signe because the ordinary language at his disposal offers no other term. Saussure therefore improves the vocabulary where a sharper distinction requires it, while accepting an inherited term where no satisfactory substitute is available. His solution remains constrained by the linguistic resources in which the theory is being articulated.

This is where the connection with Carnap becomes especially revealing. Carnap's explication provides a general framework for constructing concepts whose theoretical usefulness may require departures from inherited usage. Saussure's passage shows what such reconstruction can look like in practice: a theoretical vocabulary can be adjusted where necessary without becoming a perfectly systematic replacement for ordinary language. The aim is an adequate vocabulary for the work the theory has to perform.

The distinction also depends on what Saussure means by image acoustique. It is not the physical sound produced by the vocal apparatus. The relevant element is a psychological representation of sound rather than the material event of speech. The term signifiant gives this element a stable place within the theoretical vocabulary while preserving its distinction from physical articulation.

Saussure's intervention therefore does more than provide new labels. It establishes a set of terms whose relations allow the distinctions required by his analysis to be maintained. Terminology can be part of the work by which theoretical distinctions become precise and usable.

The comparison with Carnap should nevertheless be kept in perspective. Saussure was not applying a method of explication that Carnap would formulate only later. Rather, Carnap provides a general methodological account of conceptual reconstruction, while Saussure encounters a specific terminological difficulty in the course of developing his theory of the linguistic sign. His discussion therefore offers a historical case in which a more abstract methodological problem becomes visible in practice.

4. When Terminology Becomes Part of the Theory

The Saussurean example points to a broader consequence. A technical vocabulary becomes part of theoretical reasoning when the relations among its terms determine what the theory can distinguish, identify, and discuss. At that point, the terms do more than name ideas that were already fully formed: their articulation helps give the conceptual structure its usable form.

This does not mean that concepts are reducible to words, or that translation makes conceptual communication impossible. A distinction can be carried from one language into another, but its expression may require different lexical relations or a reorganization of the terminology. What matters is therefore not whether a concept has one permanently correct verbal form, but whether the vocabulary in which it is expressed can sustain the distinctions the theory needs.

The question with which the article began can now be answered more precisely. What happens when inherited vocabulary does not provide the distinctions a theory needs? The theory may have to reorganize that vocabulary, refining some terms, introducing others, and retaining familiar expressions where no better alternative is available. The result need not be a perfect or self-contained language. It needs to provide enough conceptual precision for the work the theory has to perform. A theory sometimes needs its own vocabulary, then, not because ordinary language is inadequate in general, but because a particular theoretical problem requires distinctions that ordinary usage does not yet make sufficiently visible.

References

Brun, G. (2016). Explication as a method of conceptual re-engineering. Erkenntnis, 81, 1211–1241.

Brown, H. D., & Abeywickrama, P. (2019). Language assessment: Principles and classroom practices (3rd ed.). Pearson.

Carus, A. W., & Friedman, M. (2025). Rudolf Carnap. In E. N. Zalta & U. Nodelman (Eds.), The Stanford Encyclopedia of Philosophy (Fall 2025 ed.). Metaphysics Research Lab, Stanford University.

Carnap, R. (1947). Meaning and necessity: A study in semantics and modal logic. University of Chicago Press.

Carnap, R. (1950). Logical foundations of probability. University of Chicago Press.

Derrida, J. (2002). There is no one narcissism. In P. Cheah & S. Guerlac (Eds.), Derrida and the time of the political (pp. 198–210). Duke University Press.

Dutilh Novaes, C. (2020). Carnapian explication and ameliorative analysis: A systematic comparison. Synthese, 197(3), 1011–1034.

Friedman, M., & Creath, R. (Eds.). (2007). The Cambridge companion to Carnap. Cambridge University Press.

Saussure, F. de. (1983). Course in general linguistics (C. Bally, A. Sechehaye, & A. Riedlinger, Eds.; R. Harris, Trans.). Duckworth.