The question is do we want a “more clear/communicative” version of Deleuze and Guattari? How about ourselves?

We wrote Anti-Oedipus together. Since each of us was already multiple, there was always a crowd. We drew on whatever came within range—what was nearest as well as what was farthest. We used pseudonyms to avoid recognition. Why, then, did we keep our own names? Out of habit, nothing more. To make ourselves unrecognizable in another way—not by hiding who we are, but by obscuring what makes us act, feel, and think. It is also pleasant to speak like everyone else, to say “the sun rises,” even though we know it is only a figure of speech. The goal is not to stop saying “I,” but to reach the point where it no longer matters whether one says “I” at all. We are no longer ourselves. Each will recognize their own. We have been aided, inspired, multiplied.

A book has neither object nor subject. It is composed of materials of different kinds, dates, and speeds. To attribute a book to a subject is to ignore the interplay of these materials and the external nature of their relations. It is like inventing a benevolent God to explain geological movements. In a book, as in everything else, there are lines of articulation or segmentarity—strata and territories—but also lines of flight, movements of deterritorialization and destratification. The varying speeds along these lines produce effects of slowness and density, or, conversely, of acceleration and rupture. All of this—these lines and measurable speeds—forms an assemblage.

A book is such an assemblage, and as such, it cannot be attributed to a subject. It is a multiplicity. Yet we do not yet know what the multiple entails when it is no longer tied to attribution, when it becomes a substantive in its own right. One side of a machinic assemblage faces the strata, giving it the appearance of an organism, a signifying totality, or a subject-based unity. Another side faces what we call the body without organs, which continually dismantles the organism, allowing intensities to circulate and leaving behind only names as traces.

What, then, is the body without organs of a book? There are many, depending on the kinds of lines involved, their density, and whether they converge on a “plane of consistency” that selects them. Here, as elsewhere, measurement is essential: writing must be quantified. There is no difference between what a book says and how it is made. A book has no object. As an assemblage, it exists only in relation to other assemblages and other bodies without organs. We will never ask what a book means, as signifier or signified. We will not look for something to interpret. Instead, we ask what it connects with, what flows it transmits, into which multiplicities it enters and transforms, and with which bodies without organs it converges.

A book exists only through the outside and on the outside. It is itself a small machine. The question is how this literary machine connects with other machines—a war machine, a love machine, a revolutionary machine—and with the abstract machine that carries them along. We have been criticized for excessive quotation. But writing always involves asking which other machine the literary machine must connect to in order to function. Kleist connects to a war machine, Kafka to a bureaucratic machine. Literature is an assemblage; it has nothing to do with ideology. There is no ideology, and there never has been. We speak only of multiplicities, lines, strata, segmentarities, lines of flight, intensities, machinic assemblages, bodies without organs, planes of consistency, and the units that measure them. Writing does not signify; it maps and surveys, even territories yet to come.

There is, first, the root-book. The tree is already the image of the world, just as the root is the image of the world-tree. This is the classical book: noble, signifying, centered on subjective interiority. The book imitates the world as art imitates nature, using its own procedures. Its law is reflection: the One becomes two. This formula, however it is expressed, represents the oldest and most tired form of thought. Nature does not operate this way. Its roots spread laterally and multiply; they do not simply divide. Thought lags behind nature.

Even when the book is taken as a natural object—with a spine and leaves—it still functions, as an image, according to this logic of unity dividing into multiplicity. Binary logic governs this model. Even advanced disciplines like linguistics retain this structure, as in grammatical trees that branch from a single point. Such systems fail to grasp multiplicity, since they always presuppose an underlying unity. Whether unity appears in the subject or in the object, the result is the same: binary logic persists.

A second model is the radicle or fascicular system. Here, the main root is aborted, and a multiplicity of secondary roots proliferates. Yet unity persists, either as something past, future, or possible. Modern methods often operate this way: they produce multiplicity in one dimension while reintroducing unity in another. Fragmented works may still present themselves as totalities. Joyce disrupts the unity of the word but restores unity at the level of the text. Nietzsche breaks linear knowledge but invokes the eternal return as a higher unity. In this model, unity is displaced but not abolished. It remains, especially in the subject, as a higher-order synthesis.

Thus, even in fragmentation, the book continues to reflect the world. The world may appear chaotic, but the book remains its image—a “chaosmos.” This is a mystification: a book all the more total for being fragmented. Simply proclaiming “long live the multiple” is not enough. The multiple must be constructed. It is not achieved by adding dimensions but by subtraction—by working with what is already there, at “n – 1” dimensions. The one belongs to the multiple only as something subtracted.

Such a system is called a rhizome. A rhizome is a subterranean stem, fundamentally different from roots and radicles. Bulbs and tubers are rhizomes. Even organisms with roots may exhibit rhizomatic features. Some animals, such as rats in their pack formations, are rhizomatic, as are burrows in their functions of movement and escape. A rhizome spreads in all directions, taking diverse forms—from surface networks to dense clusters. It includes both the best and the worst: cultivated plants and invasive weeds alike. It is both order and proliferation. To make this clear, we must now identify some of its defining characteristics.