Are All Signs Shifters? Saussure, Jakobson, and the Question of Identity

Abstract

Shifters such as I, here, and now have often been regarded as a special challenge to structural linguistics because their interpretation depends upon the circumstances of utterance. Roman Jakobson and Émile Benveniste used these expressions to draw attention to dimensions of language that cannot be explained solely through the internal relations of the linguistic system. This article revisits the problem from a Saussurean perspective. Returning to Saussure’s reflections on identity, value, and difference, it argues that the fundamental question is not why certain signs shift, but how any linguistic sign remains identifiable across multiple acts of speech. From this perspective, shifters do not contradict the Saussurean conception of language. Rather, they reveal, in a particularly visible form, a condition that affects all linguistic signs: identity is not secured by material continuity or fixed reference but emerges from a system of relations.

Shifters and Linguistic Identity: A Saussurean Reconsideration

When two different people say the word I, the linguistic form remains identical while the person to whom it refers changes completely. The same phenomenon occurs with words such as here, there, now, yesterday, or tomorrow. Their interpretation seems inseparable from the situation in which they are uttered. If I say “I am tired,” the word I refers to me. If you utter the same sentence a moment later, the identical word refers to someone else. The sign remains the same; its reference shifts.

This peculiar class of expressions attracted considerable attention during the twentieth century. Roman Jakobson famously called them shifters because their reference changes according to the circumstances of speech. Linguists generally classify them as deictics, signs whose interpretation depends upon the context of utterance. Émile Benveniste would later make them central to his theory of énonciation, arguing that they reveal dimensions of language that cannot be understood solely by examining the linguistic system in abstraction from its use.

At first sight, such expressions appear exceptional. Most words seem to designate relatively stable categories. A chair remains a chair regardless of who pronounces the word. A tree remains a tree whether it is mentioned in New York, Chicago, or Las Vegas. But words such as I, here, and now appear to behave differently. Their interpretation changes from one speech situation to another. They seem to point beyond language itself toward the concrete circumstances in which language is used.

For this reason, shifters have often been regarded as a special problem for linguistic theory. They appear to occupy a threshold between language and the world, between the system of signs and the situations in which those signs are employed. Their meaning seems to require something beyond the internal relations of the linguistic system: the identity of the speaker, the position of the listener, the location of the utterance, and the moment in which the utterance occurs.

This is why they became so important for Benveniste. In his analysis, terms such as I and you cannot be understood in the same way as ordinary lexical items. They do not designate permanent objects or fixed concepts. Rather, they acquire their reference through the act of discourse itself. The word I means, in effect, “the speaker who is currently saying I.” Its identity cannot be established independently of the event of enunciation. Each new utterance renews its reference.

The consequences of this approach are significant. Attention shifts away from la langue, understood as a structured system of signs, toward énonciation, the act through which language is put into operation by a speaker in a particular situation. Language is approached not merely as a structure but also as an activity. The study of signs must therefore consider not only their relations within the system but also the circumstances in which they function.

From this perspective, Jakobson and Benveniste reveal something important: certain signs cannot be understood without reference to the conditions of utterance. Shifters expose a dimension of language that disappears when signs are considered only as abstract elements within a system.

Yet this interpretation raises a further question. The instability of shifters is undeniable. Their reference changes from one utterance to another. But are they truly exceptional? Or do they reveal a more general problem concerning linguistic identity itself?

Put differently, is the strange behaviour of I a unique feature of a particular class of expressions, or does it make visible a condition that affects all linguistic signs?

To approach this question, it is useful to return to a thinker who rarely appears in discussions of shifters but whose work provides a different perspective on the problem: Ferdinand de Saussure. For Saussure, the central question of linguistics was not primarily how words refer to objects or how speakers use language in particular situations. His concern was more fundamental. He wanted to understand what allows linguistic entities to be recognized as the same across multiple acts of speech.

This question emerges with particular clarity in Saussure’s discussion of what he calls synchronic identity in the Course in General Linguistics. Saussure begins with a deceptively simple example. During a lecture, a speaker may repeat the word Messieurs! several times. Each occurrence differs from the others. The tone changes. The volume changes. The rhythm changes. The emotional intention may change as well. Even the acoustic realization is never exactly identical from one utterance to the next. Yet listeners have no difficulty recognizing that the same linguistic unit has been repeated:

"For example, we may hear in the course of a lecture several repetitions of the word Messieurs! (‘Gentlemen!’). We feel that in each case it is the same expression: and yet there are variations of delivery and intonation which give rise in the several instances to very noticeable phonic differences – differences as marked as those which in other cases serve to differentiate one word from another. Furthermore, this feeling of identity persists in spite of the fact that from a semantic point of view too there is no absolute reduplication from one Messieurs! to the next."

This observation leads Saussure to an unexpected conclusion. The identity of a linguistic sign cannot depend upon its material substance. If it did, every new pronunciation would constitute a different entity. Nor can identity depend upon perfect coincidence of meaning, since the same word may be used in different circumstances and with different nuances without losing its identity.

The problem, therefore, is not why certain signs shift. Variation is everywhere. The more fundamental question is how linguistic identity survives variation.

To clarify this point, Saussure introduces a series of comparisons. Consider the 8:45 train from Geneva to Paris. The train departs every day, yet the locomotive may be different, the carriages may be different, and the personnel may be different. Nevertheless, we continue to speak of the same train. Or consider a street that has been demolished and rebuilt. Physically, little or nothing may remain of the original structure, but we still regard it as the same street. In both cases, identity does not reside in material continuity. What remains constant is a position within a system of relations.

The train is identified by its route, its timetable, and its place within a larger network. The street is identified by its location and its relations to surrounding streets. Their identity is not based on the persistence of a substance but on their position within a structure.

Saussure argues that linguistic identity functions in a similar way. A sign is not what it is because of some material element that remains unchanged through time. It is what it is because it occupies a particular position within a system. Identity is therefore relational rather than substantial.

The same logic appears in one of Saussure’s most famous analogies: the game of chess. A knight possesses value only through its place within the system of the game and the rules governing its movements. If the piece is lost, it can be replaced by an object of entirely different shape. A button, a coin, or a pebble can function as a knight provided that it assumes the same role within the structure of play. The physical object is secondary. What matters is the position it occupies within the system of relations that defines the game.

Seen from this perspective, the problem of shifters begins to appear in a different light. Jakobson and Benveniste drew attention to signs whose reference changes according to the situation of utterance. Their analyses remain essential because they reveal that certain expressions cannot be understood independently of the circumstances in which they are used. Yet Saussure’s reflections suggest that another question lies beneath this apparent exception. Before asking why certain signs shift, we might ask how any sign remains identifiable at all.

The word I appears unstable because its reference changes from one speaker to another. But Messieurs! is not materially identical across its various occurrences either. Nor is the word tree, nor any other linguistic sign. Every act of speech introduces new sounds, new contexts, new intentions, and new circumstances. The apparent stability of linguistic units is itself something that requires explanation.

The problem is therefore not simply why certain signs shift. The deeper problem is how any sign remains the same across different utterances, speakers, contexts, and acoustic realizations.

Once the question is reformulated in terms of identity rather than reference, shifters may no longer appear as an exception to the general functioning of language. They reveal, in a particularly visible form, a condition that affects all linguistic entities. What Jakobson and Benveniste discovered through expressions such as I, here, and now, Saussure approached from another direction through his reflections on identity, value, and difference.

This does not mean that there is no difference between a shifter and an ordinary lexical sign. The word tree and the word I do not function in exactly the same way. The reference of tree does not depend upon the identity of the speaker in the same direct manner as I. A shifter has a distinctive relation to the situation of enunciation built into its functioning.

The point, rather, is that this difference is not a difference between relational signs and non-relational signs. Both are relational. Both acquire their identity through a system of relations. The difference lies in the type of relation through which their value is produced. The value of tree emerges primarily through its position within the lexical network of the language: tree is what it is because it differs from bush, plant, forest, and countless other terms. The value of I, however, includes a relation to the structure of discourse itself: it is the position occupied by whoever assumes the role of speaker in a particular act of utterance.

From a Saussurean perspective, then, shifters do not escape the linguistic system by introducing something external to language. Rather, they reveal that the system itself is based on relations rather than fixed substances. They make visible a principle that is already present in every sign.

This becomes clearer when we consider words that appear more stable than shifters. Take a word such as tree. It may seem fundamentally different from I. Unlike the personal pronoun, its reference does not change each time it is uttered. Yet the identity of the sign itself does not rest upon a permanent essence. It depends upon its relations within the linguistic system. It is recognizable because it occupies a particular position among other signs.

The same observation could be extended to more abstract terms such as democracy. Different speakers employ the word in different historical contexts, political traditions, and ideological frameworks. Its uses may vary considerably. Yet we continue to recognize the same sign across these variations. Its identity does not arise from a single fixed content preserved unchanged through time. It emerges from a network of differences and oppositions that allows the sign to maintain a recognizable position.

This is perhaps where Saussure’s thought reaches its highest level of abstraction. The examples of the train, the street, the chess knight, and the linguistic sign all point toward the same conclusion. Identity is relational. What endures is not a thing but a position. What gives coherence to an entity is not a positive substance but the system of relations that makes it identifiable.

Everything returns to difference.

If one were searching for the central intuition around which Saussure’s thought revolves, this would be a strong candidate. Whether discussing phonemes, signs, values, units, or identities, Saussure repeatedly returns to the same insight: linguistic entities are not defined by what they positively contain but by the differences that separate them from other elements within a system.

The article began with a simple observation. When two different people say I, the word remains the same while its reference changes. This appeared to reveal a peculiar instability within language. The phenomenon fascinated linguists because it seemed to challenge the idea that meaning could be explained solely through the internal relations of a linguistic system.

After the detour through Saussure, however, the example appears in a different light. The mystery is no longer only who I refers to in a particular act of speech. A more fundamental question emerges: what allows us to recognize the same sign through endlessly changing acts of speech? How does linguistic identity persist despite variations in sound, context, intention, and use?

Jakobson and Benveniste revealed the instability of certain signs by showing that their interpretation depends upon the circumstances of utterance. Saussure reveals a deeper instability at the heart of linguistic identity itself: no sign possesses identity because of material continuity or an unchanging essence. Signs remain identifiable because they occupy positions within systems of relations.

Perhaps, then, shifters are fascinating not because they depart from the logic of language, but because they make visible, in an especially striking form, what is true of every sign. They do not expose a failure of the linguistic system. They reveal one of its deepest principles: identity is not something signs possess before entering into relations. It is something they acquire through those relations.

References

Benveniste, É. (1971). Problems in General Linguistics (M. E. Meek, Trans.). University of Miami Press. (Original work published 1966)

Jakobson, R. (1957). Shifters, Verbal Categories, and the Russian Verb. In R. Jakobson (Ed.), Selected Writings: Volume II: Word and Language (pp. 130–147). Mouton.

Saussure, F. de. (1983). Course in General Linguistics (R. Harris, Trans.). Duckworth. (Original work published 1916)