The Order Behind Language: Borges, Saussure, and the Problem of Classification

“Language (le langage) in its entirety has many different and disparate aspects. It lies astride the boundaries separating various domains. It is at the same time physical, physiological and psychological. It belongs both to the individual and to society. No classification of human phenomena provides any single place for it, because language (le langage) as such has no discernible unity.

A language (la langue) as a structured system, on the contrary, is both a self-contained whole and a principle of classification. As soon as we give linguistic structure (la langue) pride of place among the facts of language, we introduce a natural order into an aggregate which lends itself to no other classification”.

– Ferdinand de Saussure, Course in general linguistics

1. Borges and the Classification That Makes Us Laugh

Why does Borges’s strange encyclopedia make us laugh?

In Borges’s supposed Chinese encyclopedia, animals are divided into categories such as:

(a) belonging to the Emperor, (b) embalmed, (c) tame, (d) sucking pigs, (e) sirens, (f) fabulous, (g) stray dogs, (h) included in the present classification, (i) frenzied, (j) innumerable, (k) drawn with a very fine camelhair brush, (l) et cetera, (m) having just broken the water pitcher, (n) that from a long way off look like flies.

The immediate temptation is to call the classification absurd, yet that does not quite explain the joke. Most of the categories are perfectly intelligible on their own. We understand what it means for an animal to be tame or embalmed, and we can make sense of a stray dog, a fabulous creature, or an animal belonging to an emperor. Even “innumerable,” though an unusual zoological category, has a comprehensible meaning. Nothing becomes meaningless simply because it appears in Borges’s list.

The difficulty emerges when these expressions are treated as divisions within a single classificatory system. A taxonomy normally gives us a reason to regard its categories as comparable: its objects are being considered according to some common basis that explains their place within the same order. Borges removes that basis. His categories seem to belong to different conceptual worlds. Some describe properties, others identify kinds of creatures, while others concern ownership, behaviour, representation, or a particular circumstance.

The list therefore produces a peculiar effect. We can understand every entry, yet we cannot easily understand what makes the entries members of one classification. The alphabetical sequence gives them a visible succession without providing an order of thought. The letters tell us which category comes next, but they tell us nothing about why these categories belong together.

This is what makes the list funny: it frustrates an expectation so basic that we normally remain unaware of it. When we encounter a classification, we assume that its elements have been selected and arranged according to some underlying principle. Borges makes that principle impossible to recover. We recognize the categories while losing sight of the order that would make them parts of a whole.

2. A Collection Is Not Yet an Order

Borges’s encyclopedia thus raises a more fundamental question: what turns a collection of items into a classification?

Almost anything can be placed beside almost anything else. A table can hold a book, a glass, a key, a telescope, a pipe, and a typewriter. Their physical coexistence tells us nothing about the relations in which they should be considered. They are together in space without belonging to a common order of knowledge.

A classification requires a basis for bringing things together as members of the same field. The relevant similarities must be identified, their differences made significant, and their positions determined in relation to some shared criterion. Without such a basis, there may be accumulation or juxtaposition, yet nothing tells us why the elements should constitute one system rather than an arbitrary collection.

This is why Borges’s list reveals more than an eccentric failure of taxonomy. It exposes something ordinary classifications tend to hide. Once an organizing principle has become familiar, we usually attend to the objects and categories rather than to the framework that makes them intelligible as parts of a whole. Borges removes that background condition while retaining the appearance of a classification. The result is the strange experience of recognizing the elements without being able to recover the order that relates them.

Now consider what would happen if the problem arose in a science. We might ordinarily assume that the objects to be classified are already there, and that the task is simply to identify them and determine where they belong. But what happens when the objects of inquiry are themselves not given in a form that establishes what belongs together?

That is where Ferdinand de Saussure’s problem begins. The phenomenon he calls langage does not present the linguist with a single, self-evident object. It extends across several domains, so the difficulty is prior to sorting its manifestations into categories. One must first determine what kind of unity could make them belong to one specifically linguistic field.

3. Saussure’s Problem: What Is the Object?

Saussure begins his discussion of linguistic method with a deceptively simple question: what exactly is the object of linguistics? His difficulty is that language does not provide the sort of object that other sciences seem to receive in advance. As he puts it, “The object is not given in advance of the viewpoint.”

His example is the French word nu, “naked.” It may appear to be an obvious linguistic object, yet different perspectives disclose different phenomena: a sound, an articulation, the expression of an idea, a form derived from Latin nūdum, and so on. The same occurrence can therefore be approached phonetically, physiologically, psychologically, or historically, with each perspective yielding something legitimate.

Saussure then broadens the problem. Linguistic phenomena involve complementary dimensions that cannot simply be detached from one another. What the ear perceives depends on vocal articulation, while the movements of the vocal organs matter linguistically only in relation to their auditory effect. Speech sounds are connected with thought; language has both an individual and a social aspect; a language exists as an established system at a given moment while also carrying the effects of its history. Each perspective reveals something real, yet none of them, taken by itself, provides the unified object that linguistics requires.

Saussure is therefore caught between two unsatisfactory approaches. One can concentrate on a single aspect and risk excluding others that depend upon it, or attempt to consider the different dimensions together, in which case, he warns, “the object of study becomes a muddle of disparate, unconnected things.” Linguistics would then overlap with psychology, anthropology, philology, prescriptive grammar, and other disciplines without possessing a clearly defined domain of its own.

The significance of the passage lies partly in the fact that Saussure does not simply announce his solution. He considers the alternatives and arrives at his conclusion only after showing what each approach fails to accomplish. The formulation “One solution only, in our view, resolves all these difficulties” therefore reads as the endpoint of a methodological inquiry. The qualification “in our view” is worth retaining: the sentence proposes a solution to a difficulty rather than presenting an obvious fact that requires no argument.

Saussure’s answer is to make linguistic structure (langue) the primary concern and to relate the other manifestations of language to it. A language (langue) understood as a structured system is then described as “both a self-contained whole and a principle of classification.” The important point is not simply that Saussure selects one aspect of language and sets the others aside. He is looking for a form of organization that can give the diverse manifestations of langage a specifically linguistic relation to one another.

The question has therefore changed. Saussure is no longer asking how many different things language includes; the earlier discussion has already established its heterogeneity. He is asking what kind of order could allow those different manifestations to constitute a coherent object of linguistic inquiry. That is the problem to which langue will respond.

4. Why Langue? From Heterogeneity to a Principle of Classification

The phrase “principle of classification” tells us what Saussure’s solution is meant to accomplish. Classification is often imagined as an operation performed after the objects have been identified: one discovers the things, distinguishes them, and places them under appropriate headings. Saussure’s problem comes earlier. He needs a basis on which linguistic phenomena can count as members of the same field at all.

A structured language (langue) provides such a basis because it establishes relations among phenomena that would otherwise remain dispersed across different domains. A sound considered solely as a physical event belongs to phonetics; considered as part of a linguistic system, it acquires a linguistic function through its relations with other sounds. What matters to linguistics is therefore not merely that a phenomenon exists, but the position it occupies within the organization of a language.

The system also determines which differences are relevant to linguistic analysis. Two sounds may differ physically without constituting a meaningful distinction within a language, while a relatively small difference may matter greatly because of the opposition it establishes within the system. Their linguistic significance depends on the relations in which they participate.

This helps explain Saussure’s description of a language as a “self-contained whole.” He does not mean that langue is detached from physiology, psychology, history, or society. Those dimensions remain involved in language, and other disciplines can legitimately investigate them. A linguistic system is “self-contained” in the more specific sense that its internal relations form a coherent field that can be studied as such.

The classificatory principle therefore does more than arrange a collection of facts. It establishes the terms under which phenomena can be compared as linguistic phenomena. Once language is approached as a structured system, the linguist has a basis for deciding which relations belong to the linguistic domain and for treating apparently diverse manifestations as elements of the same organized whole.

This is why langue should not be understood simply as one component of langage that Saussure happens to prefer. Its significance lies in the order it provides. It identifies the level at which linguistic phenomena can be related systematically without requiring them to be reduced to one another.

The move toward langue is therefore methodological before it is doctrinal. Saussure is not first choosing an especially interesting part of language and then constructing a theory around that preference. He begins with a phenomenon that does not provide linguistics with a unified object and seeks the kind of organization that could give it one. The structured system of langue answers that requirement by providing a determinate domain in which linguistic relations can be studied.

Seen in this light, “principle of classification” has a stronger meaning than “method of arranging facts.” It names the order that makes linguistic phenomena comparable as members of the same field. Saussure’s problem was never simply where to put the various manifestations of language; it was what kind of order would allow them to belong to one object of inquiry in the first place.

5. Returning to Borges

Seen from this perspective, Borges gives us a useful way of grasping what Saussure’s formulation means. The categories in the encyclopedia remain intelligible, but the principle that would make them divisions of one whole is unavailable. The list exposes, through humour, a condition that ordinary classifications allow us to overlook: things become comparable as elements of a system only within an order that determines how they are to be related.

Saussure begins from a different problem. Langage encompasses a multiplicity of physical, physiological, psychological, social, individual, and historical phenomena, yet their coexistence does not by itself produce the object of linguistics. His move toward langue identifies the structured order in which those phenomena can be treated as belonging to a common linguistic domain.

What Borges makes perceptible is the dependence of classification upon an organizing order, while Saussure confronts that dependence as a methodological problem: before linguistics can classify its phenomena, it needs a principle that gives them a specifically linguistic unity.

This brings us back to the table with which the problem began. An umbrella, a telescope, and a pipe can share a surface without thereby forming a system. Their coexistence becomes intelligible as an ordered whole only when some principle establishes the relations under which they are to be considered together.

Why does Borges’s encyclopedia make us laugh? Because it momentarily removes that principle while leaving the categories themselves recognizable. Saussure’s langue can be understood as his attempt to identify the kind of order that language requires if its heterogeneous manifestations are to constitute a coherent object of linguistic knowledge.

References

Borges, J. L. (1964). Other inquisitions, 1937–1952 (R. L. C. Simms, Trans.). University of Texas Press. (Original work published 1952)

Saussure, F. de. (1983). Course in general linguistics (C. Bally & A. Sechehaye, Eds.; R. Harris, Trans.). Open Court. (Original work published 1916)