[T]he relation between information and ideology in the Encyclopedie raises some general issues about the connection between knowledge and power. Consider, for example, a totally different kind of learned book, the Chinese encyclopedia imagined by Jorge Luis Borges and discussed by Michel Foucault in The Order of Things. It divided animals into: “(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.” This classification system is significant, Foucault argues, because of the sheer impossibility of thinking it. By bringing us up short against an inconceivable set of categories, it exposes the arbitrariness of the way we sort things out. We order the world according to categories that we take for granted simply because they are given. They occupy an epistemological space that is prior to thought, and so they have extraordinary staying power. When confronted with an alien way of organizing experience, however, we sense the frailty of our own categories, and everything threatens to come undone. Things hold together only because they can be slotted into a classificatory scheme that remains unquestioned. We classify a Pekinese and a Great Dane together as dogs without hesitating, even though the Pekinese might seem to have more in common with a cat and the Great Dane with a pony. If we stopped to reflect on definitions of “dogness” or on the other categories for sorting out life, we could never get on with the business of living.
Pigeon-holing is therefore an exercise in power. A subject relegated to the trivium rather than the quadrivium, or to the “soft” rather than the “hard” sciences, may wither on the vine. A misshelved book may disappear forever. An enemy defined as less than human may be annihilated. All social action flows through boundaries determined by classification schemes, whether or not they are elaborated as explicitly as library catalogues, organization charts, arid university departments. All animal life fits into the grid of an unconscious ontology. Monsters like the “elephant man” and the “wolf boy” horrify and fascinate us because they violate our conceptual boundaries, and certain creatures make our skin crawl because they slip in between categories: “slimy” reptiles that swim in the sea and creep on the land, “nasty” rodents that live in houses yet remain outside the bounds of domestication. We insult someone by calling him a rat rather than a squirrel. “Squirrel” can be a term of endearment, as in Helmet’s epithet for Nora in A Doll’s House. Yet squirrels are rodents, as dangerous and disease-ridden as rats. They seem less threatening because they belong unambiguously to the out-of-doors. It is the in-between animals, the neither fish-nor-fowl, that have special powers and therefore ritual value: thus the cassowaries in the mystery cults of New Guinea and the tomcats in the witches’ brews of the West. Hair, fingernail parings, and feces also go into magic potions because they represent the ambiguous border areas of the body, where the organism spills over into the surrounding material world. All borders are dangerous. If left unguarded, they could break down, our categories could collapse, and our world dissolve in chaos.
Setting up categories and policing them is therefore a serious business. A philosopher who attempted to redraw the boundaries of the world of knowledge would be tampering with the taboo. Even if he steered clear of sacred subjects, he could not avoid danger; for knowledge is inherently ambiguous. Like reptiles and rats, it can slip from one category to another. It has bite. Thus Diderot and d’Alembert took enormous risks when they undid the old order of knowledge and drew new lines between the known and the unknown.
from The Great Cat Massacre and Other Episodes in French Cultural History by Robert Darnton, rev. ed. 1984, Chapter 5.