The Deleuzo-Guattarian Rhizome “between” philosophy, science, history and anthropology

1 In 1976, Deleuze and Guattari published a small book entitled Rhizome with Éditions de Minuit. This book is a point of passage between the two volumes of Capitalism and Schizophrenia: on the one hand, it records the two-person writing of Anti-Oedipus, published in 1972, and on the other hand, it announces the program of A Thousand Plateaus, to which it will serve as an introduction in 1980. [1] A decisive moment in their collaboration, Deleuze and Guattari theorized for the first time the new image of thought they were mobilizing—a nomadic image of thought (nomadology)—and the new practice of philosophy they were implementing—a philosophy of multiplicities. It is likely that this moment was particularly decisive for Deleuze, as the encounter with Guattari finally made possible the renewal he had been repeatedly calling for since the 1960s. In 1977, in Dialogues, Deleuze confided that:


In my earlier books, I tried to describe a certain exercise of thought; but describing it was not yet exercising thought in that way. (Similarly, proclaiming ‘Long live the multiple’ is not yet doing it, one must do the multiple. And neither is it enough to say, ‘Down with genres’; one must effectively write in such a way that there are no more ‘genres’, etc.) With Félix, all that became possible, even if we failed. [2]

3 However, why has it fallen to the rhizome to embody this renewal? The question arises all the more because Deleuze and Guattari give the rhizome a general, or rather transversal, scope. It is indeed both metaphysical and political. Metaphysical, since the rhizome testifies to a practical reformulation of immanence, which it is necessary to do at the same time as to think, so as to thwart all effects of transcendence. It is also political, since this practice is one with the process of nomadization, which Deleuze and Guattari illustrate by the exteriority of the war machine vis-à-vis the State as a form of interiority and transcendence. Why did Deleuze and Guattari choose to express this philosophical renewal through a plant system that had previously been deprived of any conceptual dignity? What relationship can there be between the rhizome—the yam, the cassava, the sweet potato, etc.—and nomadology?

4 Deleuze and Guattari’s interest in the rhizome stems first and foremost from its opposition to the root, to the tree. Indeed, whether it is a taproot or a fasciculated root, the root is a plant system that develops along a vertical and hierarchical axis (cutting a plant at the root most often amounts to killing it). On the other hand, the rhizome is a plant system that proliferates horizontally, most often underground, and that has no center or—what amounts to the same thing—that has several centers. In this sense, the root constitutes an image of the foundation or the hierarchical principle (arkhè), while, conversely, the rhizome is presented as an image of the becoming or of the network, of all multiplicity rebelling against centralization and hierarchization.

5 It is true that, historically, the root and the tree have been a powerful model for the unfolding of philosophical thought. In the history of philosophy, there is no better example than the importance of the Tree of Porphyry, which models the logic of gender and specific difference that Aristotle had elaborated in the Organon. [3] Indeed, we know that this logic will irrigate all of medieval scholasticism (the quarrel of universals, for example, is a long discussion with Aristotle), will animate the classificatory thought of the classical age (one may think of Linnaeus’s taxonomies), and will then again innervate the evolutionary schemes of the modern episteme (evolution understood as an arborescent process of differentiation). We will not be surprised, therefore, by the lack of philosophical dignity of the rhizome, by its minority situation. For Deleuze and Guattari, this is precisely the counterpart of the exaggerated importance that the West has given to the tree.


It is odd how the tree has dominated Western reality and all of Western thought, from botany to biology and anatomy, but also gnoseology, theology, ontology, all of philosophy…: the root-foundation, Grund, racine, fondement. [4]

7 But the initial question remains: in what way is the opposition of the tree and the rhizome something other than a metaphor for the image of thought, or even for political organization? How can we give it the value of a conceptual opposition? How can we give it the generality or transversality of a guiding scheme of thought and action? In order to answer this question, it is first necessary to restore the role played by Deleuze and Guattari’s references to the mathematical work of Petitot and Rosenstiehl and to the anthropological hypotheses of Haudricourt.

8 In a 1974 article entitled “Automate asocial et systèmes acentrés,” Pierre Rosenstiehl and Jean Petitot, two specialists in mathematics applied to the human sciences, sought to conceptualize, in the tradition of René Thom’s work, the idea of an acentric system, that is to say, the idea of a coherent organization proper to the multiple. Of course, they admit, “the custom is to consider the notion of centralization as a kind of obligatory correlation of that of system or organization,” [5] but this necessary correlation is in reality unfounded. Thus they show, for example, that the mathematical and military problem of the Firing Squad (can a firing squad manage to fire simultaneously without having a general as an external decision center?) can receive an acentric solution. The solution to such a problem, which concerns the coordination between cells deprived of a coordinator, suggests more generally that,


Contrary to the belief that numbers breed either disorder or uniformity, effective coordination can take place in an acentric community of individuals of limited memory, yet arbitrarily numerous. [6]

10 Such a solution to the problem of the Firing Squad, which constitutes a precious model for thinking about the reticular arrangement proper to the war machine (that of a guerrilla logic refusing any central command), thus makes it possible to demonstrate the possibility of an organization proper to the multiple, that is to say the validity of a coordination between elements that is independent of any central and hierarchical authority of command.

11 However, Rosenstiehl and Petitot do not only intend to establish the real opposition between hierarchical and centralized structures and acentric reticular systems. They claim above all to denounce the prevalence granted to the first over the second, and at the same time the privilege granted to arborescent structures, especially as this primacy tends to “exaggerate the social scope” [7] of centralized organization, and in particular to use it as an argument to naturalize it. Not content with contesting this primacy, Rosenstiehl and Petitot argue, on the contrary, for the primacy of acentric systems, according to the idea that any organization can be reduced to that of an acentric system, and that such a system tends, moreover, to reject the effects of centralization (so much so that the question becomes rather of knowing under what conditions a hierarchical system is able to form). For the authors, it is then necessary to draw all the consequences: mathematical, with the need to give full scope to graph theory; technological, with the possibility of designing acentric computer systems; biological and socio-political, with the aim of rethinking both “animal societies” and the famous evolutionary typology of human groups—the family, the horde, the tribe, the State—including bands, gangs, and the whole variety of social groups.


It would be simplistic to think that the hierarchical concepts imposed by those who exercise power really correspond to the nature of things. Biological organisms, animal societies, as we say human ‘hordes’ of all kinds, reveal, on closer inspection, centres almost everywhere, or even an absence of center. [8]

13 One understands the interest that Deleuze and Guattari could then find in such a hypothesis from the point of view of the theorization of machinic arrangements opposed to any principle of structural or genetic unification.

14 In another register, that of anthropology, no less importance will be given to the work of André-Georges Haudricourt in the development of the concept of rhizome. A relatively unknown anthropologist, specialist in ethnobotany, ethnozoology and ethnolinguistics, Haudricourt was endowed with a prodigious erudition which did not prevent him from formulating hypotheses as audacious as they were dazzling. Among these hypotheses is the one formulated in an article published in 1962 in the famous anthropological journal L’Homme. In “Domestication des animaux, culture des plantes et traitement d’autrui,” Haudricourt makes the hypothesis that there is a correspondence between the treatment of nature and the treatment of others, between the way of cultivating plants and of raising domestic animals on the one hand and interhuman political social relations on the other. In the treatment of nature and the treatment of others, Haudricourt distinguishes two types of action, two practical schemes: negative indirect action and positive direct action. Negative indirect action, of which he finds the archetype in the cultivation of yams among the Melanesians or rice cultivation in Asia, consists in favoring the conditions for the development of the domesticated being rather than acting directly on them: thus, the environment of each plant is arranged so that it can deploy its growth possibilities to the best of its ability. Conversely, positive direct action, of which sheep farming in the Mediterranean basin constitutes the archetype, consists in acting directly on the individual, in permanent contact with them, and in treating them collectively: thus, the herd depends for its nourishment and protection on the shepherd who constantly accompanies and directs it. However, as Haudricourt points out, such an opposition of behavior does not derive from the difference between cultivated plants and domestic animals, since this opposition is found elsewhere. Thus, cereal agriculture involves a positive direct action analogous to that involved in sheep farming, namely a series of coercive operations on plants considered collectively. Conversely, the breeding of buffaloes in the Indochinese countryside (or reindeer in the arctic zones), implies a negative indirect action, because the buffaloes sometimes protect the child, who is supposed to “guard” them, from the attacks of the tiger. Finally, according to Haudricourt, we also find such opposition at work in the political treatment of humans: in Europe and the Near East, the image of the sovereign who commands as the pastor directs his flock (or the pilot his ship) has indeed developed, as testified by the Bible or Aristotelian thought. On the other hand, in the Chinese Confucian tradition as in certain Melanesian chiefdoms, a non-interventionist attitude prevails which allows things to happen. Haudricourt can then close his reflection on the following question, a question that is both rhetorical and open to further research:


Is it absurd to wonder whether the gods that command, the morals that order, the philosophies that transcend might not have something to do with the sheep, through a predilection for slave and capitalist modes of production, and whether the morals that explain and the philosophies of immanence might not have something to do with yams, taro, and rice, through the modes of production of Asian antiquity and bureaucratic feudalism? [9]

16 However schematic it may be, such an opposition between two types of action has been confirmed by the comparative work of the philosopher and sinologist François Jullien. In Traité de l’efficacité, he distinguishes two paradigms: one, dominant in Greek and European philosophy, consists in planning the effect “by thinking of efficiency from the abstraction of ideal forms, built as models, which one would project onto the world and which the will would set as the goal to be achieved”; the other, which irrigates Chinese thought, supposes, on the contrary, that one should collect the effect rather than seek it, that one should “let it happen” rather than aiming at it directly. [10] The importance of such a distinction of paradigms lies in its degree of generality, or transversality. For these schemas of action are valid for all fields of practice, both in the treatment of nature and in the treatment of others. This characteristic of Haudricourt’s hypothesis will also play a decisive role in the genesis of Philippe Descola’s anthropological project. In search of these structures of experience that are the “integrating schemas” of practice, which are recognized by the fact that they “are activated in the greatest number of situations, both in the treatment of humans and in that of non-humans,” Descola suggests that Haudricourt was showing such schemas by distinguishing between the two forms of treatment of nature and of others, which are positive direct action and negative indirect action. [11] Following Descola, the interest of Haudricourt’s typological hypothesis is therefore twofold. On the one hand, it offers the possibility of integrating under the same schema the treatment of nature and the treatment of others, that is to say, behaviors involving relations to entities ordinarily perceived as belonging to separate ontological domains, thus allowing the neutralization of this dualism of nature and culture which is anthropologically relevant only from the point of view of the modern West. On the other hand, by simply invoking “correspondences” between the treatment of nature and the treatment of others in the East and in the West, such a hypothesis makes it possible to exhibit homologies without reducing them either to the projection of interhuman relations onto relations to nonhumans or to the extension to humans of relations to nonhumans, and thus to exhibit the integrative character of these schemas.

17 When Deleuze and Guattari invoke Haudricourt’s hypothesis in Rhizome, it is clear that they see in it the same double benefit that Descola would later point out. On the one hand, the opposition of the rhizome and the tree immediately takes on a transversal validity for them, which goes beyond the sole framework of botany, and at the same time removes the use of these concepts from any metaphor. Thus, one must read their diagnosis literally, according to which Europe has “lost the rhizome, or the grass,” and the counterpart formulated by Henry Miller: “China is the weed in the human cabbage patch.” [12] The other lesson Deleuze and Guattari take from Haudricourt’s hypothesis is his refusal to reduce these correspondences or homologies between the treatment of nature and the treatment of others to a material or symbolic causality. In accordance with the theory of agency, the rhizome is no more to be understood according to a mechanism of material determination—where the conditions of production, that is to say the infrastructure or the economic base, would be the cause of an ideological conception of the world—than according to an imaginary or symbolic determination—according to which the mythical or mental structures would be actualized in real agropastoral practices. It is therefore not by default that Haudricourt writes that the philosophies of immanence have “something to do with” the cultivation of tubers, and that the philosophies of transcendence have “something to do with” sheep. This is, on the contrary, a quite positive formula, which revokes unilateral determinations in favor of a reciprocal presupposition of the treatment of nature and the treatment of others.

18 Nevertheless, there is an essential difference between the philosophical reading that Deleuze and Guattari make of Haudricourt’s hypothesis and the comparative and structuralist reading that Descola makes of it. For the latter, the typology of the schemas of practice partly inspired by Haudricourt forms a classification of the structures of experience that does not imply any particular privilege of one type over another. Thus, in Par-delà nature et culture, naturalism, animism, totemism and analogism constitute different modes of identification distributed in a contrastive way in the space of possibilities, without one of them enjoying any primacy over the others. However, when they take up Haudricourt’s hypothesis, Deleuze and Guattari consider for their part that the two models of action do not have the same status. It is not only that, historically, the West has lost the rhizome or the grass. It is also and above all that the rhizome does not form a model in the same sense as the tree, so that their dualism is not really one. Strictly speaking, only the tree-root acts as a transcendent model, even if it means generating its own rhizomatic growths, while the rhizome acts “as an immanent process that overturns the model […], even if it constitutes its own hierarchies”. [13] In other words, just like Rosenstiehl and Petitot’s opposition between hierarchical system and acentric system, the apparent dualism between processual rhizomatic system and arborescent system is in reality resolved in a primacy of the first in relation to the second, which is only a centering and a hierarchization that is always second and provisional.

19 In addressing the historical-geographical dimension of Haudricourt’s hypothesis, it is not without reason that Deleuze and Guattari emphasize the impossibility of sticking to a static opposition between the philosophies and morals of transcendence, dear to the West since Greek antiquity, and the philosophies and morals of immanence in the East, a sort of eternal lesson of China. For them, it is not only a question of arguing that the West has lost immanence and the rhizome, but also that it can go in search of them and find them again. That it has lost them, and that this loss has been replayed at different moments of its history, it is up to a critical analysis of the “Neolithic Revolution” and its aftermath to show this. That it can find them again, it is up to the analysis of this great deterritorialization of modern history that is the “discovery of America” to show it, because America embodies precisely the extra-European territory on which, in modern history, the Europeans went to lead this search for immanence. For Deleuze and Guattari, despite the fact that America is not exempt from its own search for roots, especially in the organic relationship of the East Coast to the old European world, “everything important that has happened or is happening takes the route of the American rhizome: the beatniks, the underground, bands and gangs, successive lateral offshoots in immediate connection with an outside.” [14] Kerouac’s entire œuvre bears witness to this ambivalence: a movement of absolute deterritorialization in On the Road, Lonesome Traveler, and The Dharma Bums, the invention of a new relationship to the urban people and territory with the beatniks of The Subterraneans, the schizoparanoid ambiguity of deterritorialization in Big Sur, the failed fantasy of a return to Breton-Celtic origins in Satori in Paris, etc. Invoking the work of Leslie Fiedler, who distinguished different types of American literature according to the mythical power of the different cardinal points, Deleuze and Guattari could note how the opposition of East and West was transformed in America, since the East is impregnated with the genealogical relation to old Europe, while the West embodies the shifting and always displaced relation to the Frontier.


America reversed the directions: it put its Orient in the West, as if it were precisely in America that the earth came full circle; its West is the edge of the East. (India is not the intermediary between the Occident and the Orient, as Haudricourt believed: America is the pivot point and mechanism of reversal. [15]

21 The fact remains that, if the history and geography of modern and contemporary America have embodied a movement of deterritorialization and reconquest of the lost rhizome, it is in reality the entire history and geography of the West that can be read in terms of this vector, and of all the possible combinations between the transcendent and despotic model of the State and the immanent and nomadic process that exceeds it, resists it, and sometimes submits to it, even if it means working on it from within. It is a question of “all history and geography”, since it is since the end of prehistory, and more precisely in the Neolithic period, that the first forms of sedentary agropastoralism appear, with its share of domesticated plants and animals, and that the first States are born—and, in all likelihood, that social hierarchies and inequalities are instituted, or at least the problem of voluntary servitude, of which Deleuze and Guattari will make the fundamental problem of political philosophy. It is thus not a coincidence if Haudricourt himself placed his hypothesis on the concomitant change of the relations between man and nature and of the interhuman relations within the framework of the effects of the Neolithic Revolution. [16] Moreover, the problem of the effects of Neolithization on the structures of experience today finds a particularly rich extension in the archaeological and anthropological works and debates on the Neolithic, in connection in particular with the critique of State power and the ecological crisis. [17] The anthropologist James C. Scott—who belongs to a family of anarchist anthropologists, because he studies forms of resistance to the State (Sahlins, Clastres, Graeber)—shows that in societies currently living on the fringes of the States of Southeast Asia (China, Burma, Laos, Vietnam), resistance is achieved, among other things, through the cultivation of tubers, which constitutes a culture of avoidance of the State and the cereal monoculture that it favors in order to facilitate tax collection. Developing this idea on the question of the conditions of the appearance of the State during the Neolithic period, Scott hypothesized that the development of cereal cultivation at the time of the agricultural revolution was a condition of possibility for the formation of the first States, insofar as such a culture lends itself particularly well to the concentration of production, to tax levies, to appropriation, to storage and to calculation, much more than tuber crops—a hypothesis that is supported by the observation that, conversely, history has not recorded any trace of the existence of cassava, sago, yam, taro, plantain, breadfruit or sweet potato States… [18]

22 And although there are few traces of this, we have good reason to suppose that these archaic states, which rested on fragile foundations, always gave rise to various forms of resistance. Unless we fall victim to an optical illusion linked to the fact that history has been written almost exclusively from the point of view of sedentary and grain-producing States, it cannot be limited to their vicissitudes. Better still, given the fact that only a few centuries ago a minority of the world’s population lived in real States, it must be admitted that history was for a long time, until recently, the golden age of the opponents of the State, of the “savages,” the “barbarians,” and the “nomads.” There is nothing to prevent a gigantic re-evaluation of the importance given to the State-form and the significance of all the structures of experience that accompany it—including the very project of writing history. That is to say that the absence of philosophical dignity of the rhizome, as of the nomads, is the counterpart of the joint development of the State and of a certain image of thought, following two homologous organic and sovereign logics. The organon as State-form developed in thought.


History is always written from the sedentary point of view and in the name of a unitary State apparatus, at least a possible one, even when the topic is nomads. What is lacking is a Nomadology, the opposite of a history. [19]

24 It will be up to the “Treatise on Nomadology—The War Machine”, the twelfth of the Thousand Plateaus, to fulfill the program opened up by Rhizome, to proceed with this relativization of State-form by showing the real existence and the logical consistency of a nomadic organization capable of conjuring up the formation of a State apparatus or of its equivalents in a group or in thought. From the metaphysical-political point of view that we are dealing with, State-form is defined by its form of interiority and its form of transcendence—two typical features of idealism—the first one resulting from an operation of capture by which elements in relation of exteriority are gathered and homogenized within a whole that totalizes them, the second one proceeding from the autonomization of an element of the whole that starts to be worth as the presupposition of all that is. Quite different is the nomadic war machine that aims to maintain and actively create an immediate relation to the Outside, to preserve and construct the milieu of exteriority and immanence that all organization presupposes and that State conjures up with all its might (which explains in passing that the war machine can present itself at the same time as a dimension of all organization and an organization among others). Since Deleuze and Guattari’s rhizomatic demand is not a simple descriptive issue but also testifies to a concrete practice of the multiple in philosophy, it was logical that they take as an example, from the beginning of Rhizome, the opposition between two ways of writing books. Against the idea of the book as an image of the world, as an internalization of the world and a substitute for it, Deleuze and Guattari intend the book to be in direct contact with the Outside, in immediate relation to the socio-historical field and the real struggles that run through it, to territories and populations, to continents and races, in short to everything that desire immediately invests, as the analysis of schizophrenic delirium in Anti-Oedipus shows. Such a book commands at the same time a philosophical and non-philosophical, conceptual, and affective reading, that is to say, it thwarts the effects of interiority and transcendence, following the example of Nietzsche’s aphoristic writing described by Deleuze in “Nomadic thought.” [20] With the conception of rhizomatic systems and the idea of nomadology, thought and politics have now become indistinguishable.

figure im1
figure im2


  • [1]
    There are differences between the two versions of this text, that of 1976 (which has become unavailable) and that of 1980 (reprinted in A thousand Plateaus). When quoting passages where this is the case, we will indicate it in a footnote.
  • [2]
    D, pp. 16-17.
  • [3]
    See Aristotle, Organon V, Topics; Porphyry, Isagoge. Deleuze comments on the Tree of Porphyry and the Aristotelian logic of gender and specific difference in DR, p. 21 and p. 47-48. On the opposition between the arborescent model inherited from Aristotle and Porphyry and the rhizomatic model of Deleuze and Guattari, see U. Eco, Sémiotique et philosophie du langage, Paris, PUF, 1988, rééd. 2013, coll. Quadriga, p. 112; Écrits sur la pensée du Moyen Âge, Paris, Grasset & Fasquelle, 2016, p. 560.
  • [4]
    TP, p. 18.
  • [5]
    P. Rosenstiehl & J. Petitot, “Automate asocial et systèmes acentrés”, Communications, Vol. 22, 1974, p. 48. For an analogous critique (of the function of the center in the idea of structure), see J. Derrida, “La structure, le signe et le jeu dans le discours des sciences humaines”, in L’Écriture et la différence (1967), Paris, Seuil, 1979, p. 409-428.
  • [6]
    P. Rosenstiehl & J. Petitot, op. cit., p. 61.
  • [7]
    Ibid., p. 48-49.
  • [8]
    Ibid., p. 45.
  • [9]
    A.-G. Haudricourt. “Domestication des animaux, culture des plantes et traitement d’autrui”. L’Homme. 1962, Vol. 2 (1), p. 50.
  • [10]
    F. Jullien. Traité de l’efficacité (1996), in La Pensée chinoise dans le miroir de la philosophie. Paris: Seuil, 2007, p. 1557.
  • [11]
    P. Descola. Par-delà nature et culture. Paris: Gallimard, 2015, coll. « Folio », p. 193. On the role of Haudricourt in the genesis of his project: see P. Descola. La Composition des mondes. Paris: Flammarion, 2017, coll « Champs », p. 201-204. On the posterity of Haudricourt’s typology in anthropology, see for example: C. Ferret. « Vers une anthropologie de l’action. André-Georges Haudricourt et l’efficacité technique ». L’Homme. 202/2012, p. 113-139.
  • [12]
    MP, p. 28. (It should be noted, however, that from a strictly botanical point of view, grass is not a rhizome, but a plant system with fasciculate roots.) The reference to Miller is an addition to the 1980 version. On literality in Deleuze, see F. Zourabichvili. La Littéralité et autres essais sur l’art. Paris: PUF, 2011.
  • [13]
    TP, p. 20.
  • [14]
    Ibid, p. 19.
  • [15]
    Ibid, p. 19. See L. Fiedler. The Return of the Vanishing American, New York: Stein and Day, 1968.
  • [16]
    The concept of “Neolithic Revolution” was coined in the 1930s by the Marxist archaeologist Vere Gordon Childe, to mark the depth, simultaneity and rapidity of the changes that occurred during the Neolithic period. Today, the term “neolithization” is preferred, to indicate the variety of phases, rhythms and independent geographical areas in which these transformations occurred. For a (relatively recent) synthesis of the question, see J.-P. Demoule (ed.). La Révolution néolithique dans le monde. Paris: CNRS Éditions, 2009.
  • [17]
    A very rich debate is currently taking place in the journal L’Homme on the anthropological interpretations of the Neolithic period (birth of inequalities and of the State, break with the savage and domestication of non-humans, birth of “modernity,” etc.): see the contributions of Emmanuel Guy, Charles Stépanoff, Rémi Hadad, Christophe Darmangeat…
  • [18]
    See J. C. Scott. The Art of Not Being Governed: An Anarchist History of Upland Southeast Asia. New Haven: Yale University Press, 2009; Against the Grain. New Haven: Yale University Press, 2017 (See the review of this book by Quentin Badaire in this issue).
  • [19]
    TP, p. 23 (the last sentence of the quotation is an addition to the 1980 version). In relation to anarchist anthropology and theory, it would be necessary to assess the relationship between nomadism and anarchism in Deleuze and Guattari.
  • [20]
    DI, p. 350-364.