The Table Behind Classification: Borges, Linnaeus, and Foucault on the Order of Things
1. Borges: A Classification That Makes Us Laugh
Why does Borges's strange encyclopedia make us laugh?
In the preface of The Order of Things, Michel Foucault quotes a supposed Chinese encyclopedia in which 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 list is funny because the categories seem to belong to entirely different worlds. Some describe physical states, others kinds of animals, ownership, appearance, or even relations to representation. Yet they are presented as if they were equivalent divisions within a single taxonomy.
Our first reaction is therefore to call the list absurd. But what, exactly, is absurd about it?
The individual categories are not necessarily meaningless. We know what it would mean for an animal to be tame, embalmed, or fabulous. We can understand what a stray dog is, and we can imagine an animal belonging to an emperor. Even “innumerable” can be understood as a description of quantity. The difficulty begins when these intelligible expressions are placed side by side as categories of one classificatory system.
This distinction matters because a classification can contain unfamiliar or even imaginary objects without ceasing to be a classification. A zoological system can include extinct species; a mythology can distinguish among fabulous creatures; a legal system can classify actions that never actually occurred. Unusual content, by itself, does not destroy an order. What matters is whether the elements can be related according to some recognizable principle.
Borges's list withholds that principle. We are given categories, but not an ordering operation that would tell us why they belong together, why one stands beside another, or what makes them divisions of the same whole. The alphabetical sequence gives the list a visible succession, but it does not explain the relation among its contents. We can follow the order of the letters without discovering an order of thought.
The laughter that Foucault describes therefore gives way to uneasiness. The list deprives us, however briefly, of something that ordinarily remains unnoticed: the background against which things can be compared, separated, grouped, and named. Familiar classifications usually conceal this background because we encounter them only after their organizing principles have become familiar.
Borges makes that familiarity unstable. That is the disturbance Foucault takes seriously. Borges does not merely invent an eccentric taxonomy. He creates a situation in which the normally unnoticed conditions of classification become perceptible precisely because they seem to have disappeared.
2. Not Strange Categories, but a Missing Relation
Consider two of Borges's categories: “fabulous” and “stray dogs.” There is nothing particularly difficult about either expression. We understand what each might designate. The same is true of “tame,” “embalmed,” or “belonging to the Emperor.” If we encountered any of these descriptions separately, we would have little trouble understanding them.
The difficulty appears when they are presented as neighboring divisions of the same classification. What connects them? What operation places them on the same level? If “fabulous” names one kind of animal, what kind of difference separates it from “stray dogs”? If “tame” is a category, is it being used in the same classificatory sense as “sucking pigs”? And if “belonging to the Emperor” is another division, what makes ownership relevant at the same level as biological or physical characteristics?
To classify is not merely to collect things under names. It is to establish relations among the named things according to distinctions that are treated as relevant. A classificatory system therefore needs some common locus: a space in which its elements can occupy positions relative to one another.
Foucault develops this problem through the image of a table. The word is deliberately ambiguous. It refers both to the physical table on which objects can be placed together and to a tabula on which thought can arrange entities according to their identities, differences, and relations. A physical surface can bring almost anything into proximity. An umbrella and a sewing machine can occupy the same table. Their physical coexistence, however, does not by itself establish a meaningful relation between them.
This distinction is crucial to Borges. His animals can be brought together linguistically. They can appear one after another on a page, just as objects can be placed next to one another on a physical surface. What is missing is a convincing conceptual locus in which their differences and similarities could acquire a common classificatory significance.
Foucault puts the point in unusually strong terms: what is impossible is not the proximity of the things listed but “the very site on which their propinquity would be possible.” The absurdity lies, in other words, not in the objects themselves but in the disappearance of the ground that would allow them to coexist as parts of an intelligible order.
The alphabetical sequence makes the problem particularly clear. It provides a principle of succession—A, B, C, D, and so forth—but alphabetical order does not establish a relation among the things named. It tells us where each item occurs in the list, not why these items should constitute divisions of one whole. Borges thus supplies an order of presentation while withholding the kind of order that would make the presentation intelligible as a taxonomy.
The distinction is subtle but fundamental. A list can have sequence without having classification. It can contain names without establishing the relations that make those names parts of a common structure.
Once this becomes visible, the problem can be turned around. If Borges gives us intelligible categories without an intelligible classificatory relation, what makes another arrangement appear orderly to us? Why do some groupings seem obvious while others appear absurd?
3. What Makes a Classification Coherent?
Imagine a simple biological classification in which organisms are grouped into increasingly specific kinds. Even without knowing much about taxonomy, we recognize that such an arrangement differs from Borges's encyclopedia. The categories appear to belong together, and their positions within the system can be related to one another. If we encounter an unfamiliar organism, we can ask where it belongs by examining characteristics that the system has made relevant.
But why does such an arrangement seem coherent?
The answer cannot simply be that the categories correspond to things that exist in nature. Existence alone does not tell us how things should be grouped. The world contains an enormous number of possible similarities and differences. Two animals may share a color, habitat, size, behavior, anatomical feature, or evolutionary history. Whether any of these resemblances matters for a particular classification depends upon the purpose and structure of the system in which they are considered.
Classification therefore requires criteria. We need some way of determining which characteristics establish a meaningful similarity, which differences warrant separation, and how those distinctions can be related. A criterion does not merely follow observation; it helps determine what counts as relevant within observation.
This is close to the point Foucault makes when he describes even the simplest order as requiring a “system of elements”: a way of defining the segments through which similarities and differences can be displayed, the possible variations affecting those segments, and the threshold at which a difference becomes significant. What appears to be an immediate perception of resemblance already involves an operation of comparison.
This does not mean that classification is arbitrary. Nor does it mean that the objects being classified are inventions of the classifier. The point is more precise: perception alone does not explain why one relation becomes significant rather than another. The same object can enter different systems of comparison, depending on what the classificatory framework allows us to recognize as relevant.
A biological classification, for example, may treat anatomical structure as decisive, while another system might organize animals according to habitat or behavior. Neither possibility follows automatically from perception. Each establishes a particular field within which some features become consequential and others recede.
The question is therefore no longer whether classifications have criteria, the more difficult question is why particular criteria come to seem self-evident. Why does one way of distinguishing things acquire the appearance of necessity while other possible arrangements seem arbitrary, eccentric, or simply wrong?
That question takes us from the logic of classification toward its history. To see what is at stake, we need an example of an ordering system that has become so familiar within Western knowledge that its organizing principles can easily disappear from view.
4. Linnaeus: An Order of Things We Recognize
The taxonomy associated with Carl Linnaeus provides a useful contrast to Borges because its organizing relations are historically recognizable to us. Linnaeus did not simply compile a list of living things. He developed a systematic method for naming and grouping organisms, arranging them according to specified characteristics and placing them within a structured hierarchy.
The important point here is not to decide whether Linnaeus gave us the final or correct picture of nature. His importance for the present argument lies elsewhere: his work allows us to see what it looks like when a classificatory framework supplies recognizable principles for relating its elements.
In Linnaean natural history, organisms were grouped according to observable characteristics, and these groupings were organized into a hierarchy of ranks. A particular organism could therefore be considered not simply as an isolated object but as a member of a succession of larger and smaller groupings. The system established distinctions that allowed some organisms to be placed together and others to be separated.
What makes such an arrangement intelligible is not merely the presence of categories. It is the network of relations connecting them. A group has a position within a larger organization; a distinction separates one group from another; characteristics become relevant because the system has assigned them a classificatory function. The arrangement can therefore be inspected, extended, and applied to organisms that were not previously included in the immediate act of observation.
This is quite different from Borges. We may disagree with particular Linnaean groupings, and later biology may reorganize them, but we can reconstruct the logic by which the system operates. We can ask which characteristics are being used, how groups are differentiated, and how the resulting distinctions are arranged. The system gives us a recognizable procedure for moving from observed characteristics to classificatory placement.
That recognizability is precisely what Borges withholds. In Borges's encyclopedia, “fabulous,” “stray dogs,” and “drawn with a very fine camelhair brush” may each be intelligible descriptions, but we cannot determine what makes their differences comparable as divisions of a single taxonomic field. Linnaeus, by contrast, gives us an order whose distinctions can be followed even when we no longer accept all of its conclusions. This difference is important because it prevents us from confusing coherence with correctness.
Linnaeus therefore gives us a more useful contrast with Borges than a simple opposition between rationality and absurdity. His taxonomy shows what it means for a classificatory order to have recognizable rules. But once those rules become familiar, another problem emerges.
Why these characteristics? Why this arrangement of similarities and differences? Why should certain visible features provide the basis for ordering living beings while others remain irrelevant?
At this point, we reach the limit of a purely structural account of classification. We can describe how a system works without yet explaining why its particular criteria have acquired authority. The question is no longer simply how organisms are arranged. It is why this way of arranging them can appear natural, intelligible, and worth preserving.
That question leads beyond Linnaeus himself. It concerns the historical field in which a classificatory system becomes possible and within which its criteria can come to count as knowledge.
5. But Why Do These Criteria Count?
Once we recognize that classification depends upon criteria, a further question becomes unavoidable: why do these criteria count?
It is tempting to answer that they count because they correspond to the way things really are. On this view, a successful classification discovers distinctions already present in nature. The history of knowledge can then be imagined as a gradual improvement in our ability to perceive those distinctions correctly: earlier systems contain errors, later ones correct them, and scientific progress brings us progressively closer to an objective order that was there from the beginning.
Foucault's archaeology begins by putting this picture into question.
The issue is not that classifications have nothing to do with the things they classify. Nor is it that different systems are simply arbitrary constructions. The point is that the same beings can become objects of knowledge through different arrangements of relations. What changes is not necessarily the existence of the objects, but the manner in which they can be observed, compared, described, and related. A characteristic that is decisive within one system may have little significance within another.
The history of the life sciences illustrates the point. Classical natural history organized living beings through a particular configuration of visible characteristics, resemblances, differences, and taxonomic ranks. Later biology increasingly approached organisms through concepts such as organization, function, development, and historical descent. The transformation cannot be adequately described as the simple accumulation of more information about the same objects. What counted as an intelligible object, and what counted as a significant relation between objects, had changed.
This is why Foucault's references to Linnaeus, Cuvier, and Darwin should not be read as stages in a straightforward story of scientific improvement. Linnaeus did not simply know less than Darwin, nor did Darwin merely add missing facts to a pre-existing taxonomic picture. Their work belongs to different configurations of knowledge. The relations through which living beings became intelligible were themselves transformed.
The question “Why these criteria?” therefore has a historical dimension. A characteristic can appear obvious to people who have learned to perceive the world through a particular organization of knowledge. Its obviousness does not establish its universality. It may instead reveal how thoroughly a certain arrangement has become established.
Consider how easily we can look at a Linnaean classification today and assume that its distinctions simply reflect the structure of nature. That reaction is understandable. A successful classificatory system becomes most effective when its operations cease to feel like operations. Once the criteria have become familiar, we no longer experience ourselves as applying them. We experience ourselves as simply seeing what is there.
This is one reason historical change can be difficult to perceive. When a classificatory framework is replaced, the transformation may initially appear to concern terminology, technical refinements, or newly discovered facts. From Foucault's perspective, something deeper may have shifted: the relations through which things could be made intelligible.
The point is not to replace scientific objectivity with subjectivity. Foucault is not saying that Linnaeus invented organisms or that biological classifications are nothing more than arbitrary decisions. He is asking a different question. What historical conditions allow particular distinctions to function as knowledge? What makes certain ways of comparing things available, productive, and convincing at a given moment?
Once we ask that question, classification becomes an entry point into a larger history. We are no longer concerned only with the arrangement of objects. We are asking about the field within which such an arrangement can acquire meaning and authority.
This is where Foucault's table becomes more than an image for classification. Behind every visible arrangement lies a less visible space in which the relations composing that arrangement can take shape.
6. The Table Behind the Table
Foucault's image of the table allows us to distinguish between an arrangement that we can see and the conditions that make that arrangement intelligible. The distinction changes what we are looking for when we examine a classificatory system.
Imagine a physical table covered with objects. We can arrange them according to color, size, material, function, ownership, or any number of other characteristics. The surface itself does not tell us which arrangement to choose. Yet without some surface—or some equivalent space—the objects could not occupy positions relative to one another at all.
Foucault's tabula is not simply an empty surface waiting for someone to impose an order upon it. Nor is it merely a mental picture of the world. It is the space of relations within which identities, differences, similarities, and distinctions can become available for thought. It is therefore not another object within the classification. It is what permits objects to enter into classificatory relations in the first place.
This helps clarify Foucault's apparently paradoxical claim that order is both “in things” and in the grid through which things are encountered. We are accustomed to separating the world from our representations of it: first there are objects, and then we describe or classify them. Foucault directs attention to what this distinction leaves out. Before an object can function as an object of knowledge, there must be ways of distinguishing it, relating it to others, identifying similarities, and establishing differences that matter.
Take the seemingly simple act of comparing two animals. We can say that they resemble one another in color, anatomy, habitat, behavior, or evolutionary history. But none of these dimensions is simply “the” relation between them. Each belongs to a different way of organizing what can be observed. Comparison presupposes a field in which a particular property can become the basis for a meaningful distinction.
The tabula is that field.
This is why it would be misleading to imagine Foucault's grid as a screen placed between the observer and an otherwise transparent reality. A grid does not merely hide some things and reveal others. It helps determine what can appear as a relevant distinction at all. It permits objects to be compared in particular ways, allows recurring characteristics to be identified, and establishes relations through which things can be grouped and named.
The historical dimension is crucial. If the table were a neutral surface that remained unchanged while successive thinkers arranged the same objects upon it, the history of knowledge could be described largely as a sequence of increasingly accurate classifications. Foucault's archaeological inquiry proposes something more radical: the space of possible relations can itself be transformed.
When that happens, the change is not simply a matter of replacing one category with another. The principles according to which categories can be formed may shift. Relations that once seemed fundamental may lose their significance; distinctions that were previously invisible may become decisive. What changes is not only what is placed on the table but what the table allows us to do.
This is why Foucault can discover connections among domains that appear, from a conventional history of ideas, to have little to do with one another. The classification of living beings, the analysis of language, and the study of wealth are not necessarily connected because thinkers consciously borrowed concepts from one another. They can be connected because they participate in a common configuration of possible relations.
The tabula behind a classification is therefore not a concealed catalogue of principles. It is the historical space in which principles of ordering can function at all.
And this raises a further question. If such a space determines, at least in part, what can count as an intelligible relation, how should we describe its historical unity? What is the name for a configuration that extends beyond an individual classification or even beyond a single science?
This is the problem to which Foucault gives the name episteme.
7. From Classification to Episteme
The word episteme can easily be misunderstood if it is treated as a synonym for “worldview,” “belief system,” or “the dominant ideas of an age.” Foucault's use of the term is more specific. He is concerned with the historical configuration within which knowledge can be organized: what can appear as an object, what relations can be established among objects, what distinctions can count as significant, and what forms of explanation can become possible.
The movement from classification to episteme can now be seen as a series of questions.
A classification requires criteria. Criteria determine which similarities and differences matter. But those criteria do not exist in isolation. They belong to a broader field in which particular ways of seeing, naming, comparing, and ordering things become possible. That field is historical. It can change, and when it changes, objects can enter into new relations without necessarily changing their physical existence.
The episteme therefore operates at a level deeper than any individual taxonomy. It is not itself a classification, and it is not a theory consciously held by everyone in a period. A zoologist does not normally begin an investigation by formulating the historical conditions that make zoological investigation possible. A grammarian does not usually step outside language in order to identify the structure of knowledge that permits grammar to function as a science. People think, investigate, classify, and formulate statements from within a field whose limits they do not necessarily perceive.
This is why Foucault places the analysis of the episteme between the empirical orders of a culture and explicit theories about order. At one level are the practical and perceptual arrangements through which people encounter things: ways of speaking, observing, classifying, exchanging, and acting. At another are scientific theories and philosophical interpretations that attempt to explain why things have the order they do.
Between these lies the historical space in which those empirical arrangements can take shape and within which formal theories can emerge.
Foucault describes this as a middle region because it is neither the immediately given world nor a conscious theory imposed upon it. It is the level at which a culture can, under certain historical conditions, begin to encounter its own arrangements as arrangements. What previously seemed transparent can become visible as one possible organization among others. Once that happens, the existing order can be questioned, reorganized, or replaced.
This distinction prevents us from reducing historical change to a succession of beliefs. Foucault is not primarily asking why people stopped believing one proposition and began believing another. He wants to identify transformations in the relations that made particular propositions, objects, classifications, and forms of knowledge possible in the first place.
This is the force of the discontinuities that structure The Order of Things. Foucault identifies major transformations between the Renaissance, the Classical age, and modernity. These are not simply changes in the content of thought. They concern the configuration within which thought operates.
In the Classical episteme, representation occupies a central position. Language, natural history, and the analysis of wealth can be understood through a common organization of identities, differences, and relations. The world can be ordered through systems of representation, and language itself functions as a privileged means of linking words and things.
The modern configuration is organized differently. History penetrates more deeply into the objects of knowledge. Living beings come to be understood through organization, function, and development; the study of wealth gives way to political economy and the analysis of production; language acquires a historical depth and becomes an object of investigation in its own right. At the same time, “man” emerges as a distinctive object of knowledge within the human sciences.
Foucault's point is not that the modern arrangement is simply more rational than the Classical one. Indeed, one of the central claims of the book is that the apparent continuity between periods can conceal profound transformations in the conditions of knowledge. The change is not best described as reason gradually discovering more of the truth. The mode in which things can be ordered and known has itself changed.
This is why the concept of episteme matters for our original problem. When we ask why a classification appears coherent, we are eventually led beyond its visible categories and explicit criteria. We arrive at the historical configuration that makes certain criteria available and intelligible in the first place.
The episteme names the historically specific configuration of knowledge within which, at a given period, particular forms of knowledge become possible.
Seen from this perspective, the humble act of classification becomes surprisingly consequential. To classify is to do more than put things into groups. It is to participate in a historically specific field of relations that determines, among other things, what can count as a meaningful similarity, difference, object, or distinction.
Borges's encyclopedia brought us to this problem by making such relations impossible for us to reconstruct. We can now return to the list with a better sense of what its disturbance reveals.
8. What Borges Makes Visible
We can now return to the laughter with which the discussion began.
At first, Borges's encyclopedia appears to be a joke about a ridiculous way of classifying animals. Its categories are so heterogeneous that they seem to mock the very idea of taxonomy. But the joke acquires a different force once we notice what ordinary classifications require.
The strange thing is not simply that “fabulous” animals are placed beside stray dogs, or that creatures “drawn with a very fine camelhair brush” appear alongside animals belonging to the Emperor. The deeper disturbance comes from our inability to reconstruct the relations that would make these categories divisions of a common field.
Borges therefore performs a kind of conceptual experiment. He leaves the categories in place while removing—or making inaccessible—the relations that would normally connect them. The result is not pure meaninglessness. On the contrary, the individual terms remain intelligible. What collapses is our ability to determine how their meanings belong together.
That failure turns our attention back toward familiar classifications.
When we encounter a biological taxonomy, we tend to look at the organisms, the names assigned to them, and the groups into which they have been placed. We rarely attend to the historical conditions that allow the arrangement to appear orderly. Its principles have become so familiar that they no longer announce themselves as principles. They present themselves as an obvious way of seeing.
Borges interrupts that familiarity.
The effect is similar to what happens when a normally invisible structure is suddenly removed. We become aware of its presence through its absence. We search for a criterion, a hierarchy, a relevant similarity, a significant difference, or a relation of inclusion. We expect to find the conceptual surface on which the items can occupy intelligible positions, but the expected structure does not materialize.
What Borges exposes, then, is not simply the difference between a good classification and a bad one. He exposes our dependence upon an order that usually remains in the background.
This is also why Foucault's response should not be reduced to the claim that some systems are rational while others are irrational. The more unsettling possibility is that what appears to us as rational order has a history. The relations through which things become comparable and classifiable are not timeless features of thought. They belong to particular configurations of knowledge.
Seen this way, Borges's encyclopedia does not offer an alternative taxonomy that we are expected to accept. Its importance lies precisely in its failure. It prevents us from taking the ordinary conditions of intelligibility for granted.
The lesson extends beyond zoology. Whenever a category appears self-evident, we can ask what makes it so. Whenever two things seem naturally related, we can ask what allows that relation to count as significant. Whenever a classification appears inevitable, we can look for the historical arrangement within which its distinctions have acquired their force.
This does not mean that every classification is equally arbitrary or that reality imposes no constraints upon knowledge. It means that between the things we encounter and the theories we construct about them lies a historical field of relations that deserves examination in its own right.
That is what makes Borges such an effective beginning for Foucault's project. The absurdity of the list does not teach us how animals ought to be classified. It makes us notice the space that ordinary classifications conceal.
The most revealing question is therefore not only “What belongs in this category?” It is also “What makes this category, and its relation to other categories, possible at all?”
Borges makes the question visible. Foucault turns it into an archaeology of knowledge.
References
Borges, J. L. (1999). The analytical language of John Wilkins. In Selected non-fictions (E. Weinberger, Ed.; E. Allen et al., Trans., pp. 229–234). Viking. (Original work published 1942)
Foucault, M. (1994). The order of things: An archaeology of the human sciences. Vintage Books. (Original work published 1966)
Linnaeus, C. (1758). Systema naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis (10th ed., Vol. 1). Laurentii Salvii.
Comments
Post a Comment