Gilles Deleuze & Félix Guattari

Axiom I. The war machine is exterior to the State apparatus. Proposition I. This exteriority is first attested to in mythology, epic, drama, and games.

Georges Dumézil, in his definitive analyses of Indo-European mythology, has shown that political sovereignty, or domination, has two heads: the magician-king and the jurist-priest. Rex and flamen, raj and Brahman, Romulus and Numa, Varuna and Mitra, the despot and the legislator, the binder and the organiser. Undoubtedly, these two poles stand in opposition term by term, as the obscure and the clear, the violent and the calm, the quick and the weighty, the fearsome and the regulated, the “bond” and the “pact,” etc. But their opposition is only relative; they function as a pair, in alternation, as though they expressed a division of the One or constituted in themselves a sovereign unity. “At once antithetical and complementary, necessary to one another and consequently without hostility, lacking a mythology of conflict: a specification on any one level automatically calls forth a homologous specification on another. The two together exhaust the field of the function.” They are the principal elements of a State apparatus that proceeds by a One-Two, distributes binary distinctions, and forms a milieu of interiority. It is a double articulation that makes the State apparatus into a stratum.

It will be noted that war is not contained within this apparatus. Either the State has at its disposal a violence that is not channeled through war— either it uses police officers and jailers in place of warriors, has no arms and no need of them, operates by immediate, magical capture, “seizes” and “binds,” preventing all combat—or, the State acquires an army, but in a way that presupposes a juridical integration of war and the organisation of a military function. As for the war machine in itself, it seems to be irreducible to the State apparatus, to be outside its sovereignty and prior to its law: it comes from elsewhere. Indra, the warrior god, is in opposition to Varuna no less than to Mitral He can no more be reduced to one or the other than he can constitute a third of their kind. Rather, he is like a pure and immeasurable multiplicity, the pack, an irruption of the ephemeral and the power of metamorphosis. He unties the bond just as he betrays the pact. He brings a furor to bear against sovereignty, a celerity against gravity, secrecy against the public, a power (puissance) against sovereignty, a machine against the apparatus. He bears witness to another kind of justice, one of incomprehensible cruelty at times, but at others of unequaled pity as well (because he unties bonds…). He bears witness, above all, to other relations with women, with animals, because he sees all things in relations of becoming, rather than implementing binary distributions between “states”: a veritable becoming-animal of the warrior, a becoming-woman, which lies outside dualities of terms as well as correspondences between relations. In every respect, the war machine is of another species, another nature, another origin than the State apparatus.

Let us take a limited example and compare the war machine and the State apparatus in the context of the theory of games. Let us take chess and Go, from the standpoint of the game pieces, the relations between the pieces and the space involved. Chess is a game of State, or of the court: the emperor of China played it. Chess pieces are coded; they have an internal nature and intrinsic properties from which their movements, situations, and confrontations derive. They have qualities; a knight remains a knight, a pawn a pawn, a bishop a bishop. Each is like a subject of the statement endowed with a relative power, and these relative powers combine in a subject of enunciation, that is, the chess player or the game’s form of interiority. Go pieces, in contrast, are pellets, disks, simple arithmetic units, and have only an anonymous, collective, or third-person function. “It” makes a move. “It” could be a man, a woman, a louse, an elephant. Go pieces are elements of a nonsubjectified machine assemblage with no intrinsic properties, only situational ones. Thus the relations are very different in the two cases. Within their milieu of interiority, chess pieces entertain biunivocal relations with one another, and with the adversary’s pieces: their functioning is structural. On the other hand, a Go piece has only a milieu of exteriority, or extrinsic relations with nebulas or constellations, according to which it fulfils functions of insertion or situation, such as bordering, encircling, shattering. All by itself, a Go piece can destroy an entire constellation synchronically; a chess piece cannot (or can do so diachronically only). Chess is indeed a war, but an institutionalised, regulated, coded war, with a front, a rear, battles. But what is proper to Go is war without battle lines, with neither confrontation nor retreat, without battles even: pure strategy, whereas chess is a semiology. Finally, the space is not at all the same: in chess, it is a question of arranging a closed space for oneself, thus of going from one point to another, of occupying the maximum number of squares with the minimum number of pieces. In Go, it is a question of arraying oneself in an open space, of holding space, of maintaining the possibility of springing up at any point: the movement is not from one point to another, but becomes perpetual, without aim or destination, without departure or arrival. The “smooth” space of Go, as against the “striated” space of chess. The nomas of Go against the State of chess, nomas against polis. The difference is that chess codes and decodes space, whereas Go proceeds altogether differently, territorialising or deterritorialising it (make the outside a territory in space; consolidate that territory by the construction of a second, adjacent territory; deterritorialise the enemy by shattering his territory from within; deterritorialise oneself by renouncing, by going elsewhere…). Another justice, another movement, another space-time.

“They come like fate, without reason, consideration, or pretext…” “In some way that is incomprehensible they have pushed right into the capital. At any rate, here they are; it seems that every morning there are more of them.” Luc de Heusch analyses a Bantu myth that leads us to the same schema: Nkongolo, an indigenous emperor and administrator of public works, a man of the public and a man of the police, gives his half-sisters to the hunter Mbidi, who assists him and then leaves. Mbidi’s son, a man of secrecy, joins up with his father, only to return from the outside with that inconceivable thing, an army. He kills Nkongolo and proceeds to build a new State. “Between” the magical-despotic State and the juridical State containing a military institution, we see the flash of the war machine, arriving from without.

From the standpoint of the State, the originality of the man of war, his eccentricity, necessarily appears in a negative form: stupidity, deformity, madness, illegitimacy, usurpation, sin. Dumézil analyses the three “sins” of the warrior in the Indo-European tradition: against the king, against the priest, against the laws originating in the State (for example, a sexual transgression that compromises the distribution of men and women, or even a betrayal of the laws of war as instituted by the State). The warrior is in the position of betraying everything, including the function of the military, or of understanding nothing. It happens that historians, both bourgeois and Soviet, will follow this negative tradition and explain how Genghis Khan understood nothing: he “didn’t understand” the phenomenon of the city. An easy thing to say. The problem is that the exteriority of the war machine in relation to the State apparatus is everywhere apparent but remains difficult to conceptualise. It is not enough to affirm that the war machine is external to the apparatus. It is necessary to reach the point of conceiving the war machine as itself a pure form of exteriority, whereas the State apparatus constitutes the form of interiority we habitually take as a model, or according to which we are in the habit of thinking. What complicates everything is that this extrinsic power of the war machine tends, under certain circumstances, to become confused with one of the two heads of the State apparatus. Sometimes it is confused with the magic violence of the State, at other times with the State’s military institution. For instance, the war machine invents speed and secrecy; but there is all the same a certain speed and a certain secrecy that pertain to the State, relatively, secondarily. So there is a great danger of identifying the structural relation between the two poles of political sovereignty, and the dynamic interrelation of these two poles, with the power of war. Dumézil cites the lineage of the Roman kings: there is a Romulus-Numa relation that recurs throughout a series, with variants and an alternation between these two types of equally legitimate rulers; but there is also a relation with an “evil king,” Tullus Hostilius, Tarquinius Superbus, an upsurge of the warrior as a disquieting and illegitimate character. Shakespeare’s kings could also be invoked: even violence, murders, and perversion do not prevent the State lineage from producing “good” kings; but a disturbing character like Richard III slips in, announcing from the outset his intention to reinvent a war machine and impose its line (deformed, treacherous and traitorous, he claims a “secret close intent” totally different from the conquest of State power, and another —an other—relation with women). In short, whenever the irruption of war power is confused with the line of State domination, everything gets muddled; the war machine can then be understood only through the categories of the negative, since nothing is left that remains outside the State. But, returned to its milieu of exteriority, the war machine is seen to be of another species, of another nature, of another origin. One would have to say that it is located between the two heads of the State, between the two articulations, and that it is necessary in order to pass from one to the other. But “between” the two, in that instant, even ephemeral, if only a flash, it proclaims its own irreducibility. The State has no war machine of its own; it can only appropriate one in the form of a military institution, one that will continually cause it problems. This explains the mistrust States have toward their military institutions, in that the military institution inherits an extrinsic war machine. Karl von Clausewitz has a general sense of this situation when he treats the flow of absolute war as an Idea that States partially appropriate according to their political needs, and in relation to which they are more or less good “conductors.”

Trapped between the two poles of political sovereignty, the man of war seems outmoded, condemned, without a future, reduced to his own fury, which he turns against himself. The descendants of Hercules, Achilles, then Ajax, have enough strength left to proclaim their independence from Agamemnon, a man of the old State. But they are powerless when it comes to Ulysses, a man of the nascent modern State, the first man of the modern State. And it is Ulysses who inherits Achilles’ arms, only to convert them to other uses, submitting them to the laws of the State— not Ajax, who is condemned by the goddess he defied and against whom he sinned. No one has portrayed the situation of the man of war, at once eccentric and condemned, better than Kleist. In Penthesilea, Achilles is already separated from his power: the war machine has passed over to the Amazons, a Stateless woman-people whose justice, religion, and loves are organised uniquely in a war mode. Descendants of the Scythians, the Amazons spring forth like lightning, “between” the two States, the Greek and the Trojan. They sweep away everything in their path. Achilles is brought before his double, Penthesilea. And in his ambiguous struggle, Achilles is unable to prevent himself from marrying the war machine, or from loving Penthesilea, and thus from betraying Agamemnon and Ulysses at the same time. Nevertheless, he already belongs enough to the Greek State that Penthesilea, for her part, cannot enter the passional relation of war with him without herself betraying the collective law of her people, the law of the pack that prohibits “choosing” the enemy and entering into one-to-one relationships or binary distinctions.

Throughout his work, Kleist celebrates the war machine, setting it against the State apparatus in a struggle that is lost from the start. Doubtless Arminius heralds a Germanic war machine that breaks with the imperial order of alliances and armies, and stands forever opposed to the Roman State. But the Prince of Homburg lives only in a dream and stands condemned for having reached victory in disobedience of the law of the State. As for Kohlhaas, his war machine can no longer be anything more than banditry. Is it the destiny of the war machine, when the State triumphs to be caught in this alternative: either to be nothing more than the disciplined, military organ of the State apparatus, or to turn against itself to become a double suicide machine for a solitary man and a solitary woman? Goethe and Hegel, State thinkers both, see Kleist as a monster, and Kleist has lost from the start. Why is it, then, that the most uncanny modernity lies with him? It is because the elements of his work are secrecy, speed and affect.” And in Kleist the secret is no longer a content held within a form of interiority; rather, it becomes a form, identified with the form of exteriority that is always external to itself. Similarly, feelings become uprooted from the interiority of a “subject,” to be projected violently outward into a milieu of pure exteriority that lends them an incredible velocity, a catapulting force: love or hate, they are no longer feelings but affects and these affects are so many instances of the becoming-woman, the becoming-animal of the warrior (the bear, she-dogs). Affects transpierce the body like arrows, they are weapons of war. The deterritorialisation velocity of affect. Even dreams (Homburg’s, Pentheselea’s) are externalised, by a system of relays and plug-ins, extrinsic linkages belonging to the war machine. Broken rings. This element of exteriority—which dominates everything, which Kleist invents in literature, which he is the first to invent—will give time a new rhythm: an endless succession of catatonic episodes or fainting spells, and flashes or rushes. Catatonia is- “This affect is too strong for me,” and a flash is: “The power of this affect sweeps me away,” so that the Self (Moi) is now nothing more than a character whose actions and emotions are desubjectified, perhaps even to the point of death. Such is Kleist’s personal formula: a succession of nights of madness and catatonic freezes in which no subjective interiority remains There is much of the East in Kleist: the Japanese fighter, interminably still who then makes a move too quick to see. The Go player. Many things in modern art come from Kleist. Goethe and Hegel are old men next to Kleist. Could it be that it is at the moment the war machine ceases to exist, conquered by the State, that it displays to the utmost its irreducibility, that it scatters into thinking, loving, dying, or creating machines that have at their disposal vital or revolutionary powers capable of challenging the conquering State? Is the war machine already overtaken, condemned, appropriated as part of the same process whereby it takes on new forms, undergoes a metamorphosis, affirms its irreducibility and exteriority, and deploys that milieu of pure exteriority that the occidental man of the State, or the occidental thinker, continually reduces to something other than itself?

Problem I. Is there a war of warding off the formation of a State apparatus (or its equivalents in a group)? Proposition II. The exteriority of the war machine is also attested to by ethnology (a tribute to the memory of Pierre Clastres).

Primitive, segmentary societies have often been defined as societies without a State, in other words, societies in which distinct organs of power do not appear. But the conclusion has been that these societies did not reach the degree of economic development, or the level of political differentiation, that would make the formation of the State apparatus both possible and inevitable: the implication is that primitive people “don’t understand” so complex an apparatus. The prime interest in Pierre Clastres’s theories is that they break with this evolutionist postulate. Not only does he doubt that the State is the product of an ascribable economic development, but he asks if it is not a potential concern of primitive societies to ward off or avert that monster they supposedly do not understand.

Warding off the formation of a State apparatus, making such a formation impossible, would be the objective of a certain number of primitive social mechanisms, even if they are not consciously understood as such. To be sure, primitive societies have chiefs. But the State is not defined by the existence of chiefs; it is defined by the perpetuation or conservation of organs of power. The concern of the State is to conserve. Special institutions are thus necessary to enable a chief to become a man of State, but diffuse, collective mechanisms are just as necessary to prevent a chief from becoming one. Mechanisms for warding off, preventive mechanisms, are a part of chieftainship and keep an apparatus distinct from the social body from crystallising. Clastres describes the situation of the chief, who has no instituted weapon other than his prestige, no other means of persuasion, no other rule than his sense of the group’s desires. The chief is more like a leader or a star than a man of power and is always in danger of being disavowed, abandoned by his people. But Clastres goes further, identifying war in primitive societies as the surest mechanism directed against the formation of the State: war maintains the dispersal and segmentarity of groups, and the warrior himself is caught in a process of accumulating exploits leading him to solitude and a prestigious but powerless death. Clastres can thus invoke natural Law while reversing its principal proposition: just as Hobbes saw clearly that the State was against war, so war is against the State, and makes it impossible. It should not be concluded that war is a state of nature, but rather that it is the mode of a social state that wards off and prevents the State. Primitive war does not produce the State any more than it derives from it. And it is no better explained by exchange than by the State: far from deriving from exchange, even as a sanction for its failure, war is what limits exchanges, maintains them in the framework of “alliances”; it is what prevents them from becoming a State factor, from fusing groups.

The importance of this thesis is first of all to draw attention to collective mechanisms of inhibition. These mechanisms may be subtle, and function as micromechanisms. This is easily seen in certain band or pack phenomena. For example, in the case of gangs of street children in Bogota, Jacques Meunier cites three ways in which the leader is prevented from acquiring stable power: the members of the band meet and undertake their theft activity in common, with collective sharing of the loot, but they disperse to eat or sleep separately; also, and especially, each member of the band is paired off with one, two, or three other members, so if he has a disagreement with the leader, he will not leave alone but will take along his allies, whose combined departure will threaten to break up the entire gang; finally, there is a diffuse age limit, and at about age fifteen a member is inevitably induced to quit the gang.These mechanisms cannot be understood without renouncing the evolutionist vision that sees bands or packs as a rudimentary, less organised, social form. Even in bands of animals, leadership is a complex mechanism that does not act to promote the strongest but rather inhibits the installation of stable powers, in favour of a fabric of immanent relations. One could just as easily compare the form “high-society life” to the form “sociability” among the most highly evolved men and women: high-society groups are similar to gangs and operate by the diffusion of prestige rather than by reference to centres of power, as in social groupings (Proust clearly showed this noncorrespondence of high-society values and social values). Eugene Sue, a man of high-society and a dandy, whom legitimists reproached for frequenting the Orleans family used to say: I’m not on the side of the family, I side with the pack.” Packs, bands, are groups of the rhizome type, as opposed to the arborescent type that centres around organs of power. That is why bands in general, even those engaged in banditry or high-society life, are metamorphoses of a war machine formally distinct from all State apparatuses or their equivalents, which are instead what structure centralised societies. We certainly would not say that discipline is what defines a war machine: discipline is the characteristic required of armies after the State has appropriated them. The war machine answers to other rules. We are not saying that they are better, of course, only that they animate a fundamental indiscipline of the warrior! A questioning of hierarchy, perpetual blackmail by abandonment or betrayal, and a very volatile sense of honour, all of which, once again, it impedes the formation of the State.

But why does this argument fail to convince us entirely? We follow Clastres when he demonstrates that the State is explained neither by a development of productive forces nor by a differentiation of political forces. It is the State, on the contrary, that makes possible the undertaking of large-scale projects, the constitution of surpluses, and the organisation of the corresponding public functions. The State is what makes the distinction between governors and governed possible. We do not see how the State can be explained by what it presupposes, even with recourse to dialectics. The State seems to rise up in a single stroke, in an imperial form, and does not depend on progressive factors. Its on-the-spot emergence is like a stroke of genius, the birth of Athena. We also follow Clastres when he shows that the war machine is directed against the State, either against potential States whose formation it wards off in advance, or against actual States whose destruction it purposes. No doubt the war machine is realised more completely in the “barbaric” assemblages of nomadic warriors than in the “savage” assemblages of primitive societies. In any case, it is out of the ‘ question that the State could be the result of a war in which the conquerors imposed, by the very fact of their victory, a new law on the vanquished, because the organisation of the war machine is directed against the State-form, actual or virtual. The State is no better accounted for as a result of war than by a progression of economic or political forces. This is where Clastres locates the break: between “primitive” counter-State societies and “monstrous” State societies whose formation it is no longer possible to explain. Clastres is fascinated by the problem of “voluntary servitude,” in the manner of La Boetie: In what way did people want or desire servitude, which most certainly did not come to them as the outcome of an involuntary and unfortunate war? They did, after all, have counter-State mechanisms at their disposal: So how and why the State? Why did the State triumph? The more deeply Clastres delved into the problem, the more he seemed to deprive himself of the means of resolving it. He tended to make primitive societies hypostases, self-sufficient entities (he insisted heavily on this point). He made their formal exteriority into a real independence. Thus he remained an evolutionist, and posited a state of nature. Only this state of nature was, according to him, a fully social reality instead of a pure concept, and the evolution was a sudden mutation instead of a development. For on the one hand, the State rises up in a single stroke, fully formed; on the other, the counter-State societies use very specific mechanisms to ward it off, to prevent it from arising. We believe that these two propositions are valid but that their interlinkage is flawed. There is an old scenario: “from clans to empires,” or “from bands to kingdoms.” But nothing says that this constitutes an evo- lution, since bands and clans are no less organized than empire-kingdoms. We will never leave the evolution hypothesis behind by creating a break between the two terms, that is, by endowing bands with self-sufficiency and the State with an emergence all the more miraculous and monstrous.

We are compelled to say that there has always been a State, quite perfect quite complete. The more discoveries archaeologists make, the more empires they uncover. The hypothesis of the Urstaat seems to be verified- The State clearly dates back to the most remote ages of humanity.” It is hard to imagine primitive societies that would not have been in contact with imperial States, at the periphery or in poorly controlled areas. But of greater importance is the inverse hypothesis: that the State itself has always been a relation with an outside and is inconceivable independent of that relationship. The law of the State is not the law of all or Nothing (State societies or counter-State societies) but that of interior and exterior. The State is sovereignty. But sovereignty only reigns over what it is capable of internalizing, of appropriating locally. Not only is there no universal State but the outside of States cannot be reduced to “foreign policy,” that is to a set of relations among States. The outside appears simultaneously in two directions: huge worldwide machines branched out over the entire ecumenon at a given moment, which enjoy a large measure of autonomy in relation to the States (for example, com- mercial organization of the “multinational” type, or industrial complexes, or even religious formations like Christianity, Islam, certain prophetic or messianic movements, etc ) but also the local mechanisms of bands, margins, minorities, which continue to affirm the rights of segmentary societies in opposition to the organs of State power. The modern world can provide us today with particularly well developed images of these two directions: worldwide ecumenical machines, but also a neoprimitivism, a new tribal society as described by Marshall McLuhan. These directions are equally present in all social fields, in all periods. It even happens that they partially merge. For example, a commercial organisation is also a band of pillage, or piracy for part of its course and in many of its activities; or it is in bands that a religious formation begins to operate. What becomes clear is that bands, no less than worldwide organisations, imply a form irreducible to the State and that this form of exteriority necessarily presents itself as a diffuse and polymorphous war machine. It is a nomas very different from the “law” The State-form, as a form of inferiority, has a tendency to reproduce itself, remaining identical to itself across its variations and easily recognisable within the limits of its poles, always seeking public recognition (there is no masked State). But the war machine’s form of exteriority is such that it exists only in its own metamorphoses; it exists in an industrial innovation as well as in a technological invention, in a commercial circuit as well as in a religious creation, in all flows and currents that only secondarily allow themselves to be appropriated by the State. It is in terms not of independence, but of coexistence and competition in a perpetual Held of interaction, that we must conceive of exteriority and inferiority, war machines of metamorphosis and State apparatuses of identity, bands and kingdoms, megamachines and empires. The same field circumscribes its interiority in States, but describes its exteriority in what escapes States or stands against States.

Proposition III. The exteriority of the war machine is also attested to by epistemology which intimates the existence and perpetuation of a “nomad” or “minor science.”

There is a kind of science, or treatment of science, that seems very difficult to classify, whose history is even difficult to follow. What we are referring to are not “technologies” in the usual sense of the term. But neither are they “sciences” in the royal or legal sense established by history. According to a recent book by Michel Serres, both the atomic physics of Democritus and Lucretius and the geometry of Archimedes are marked by it. The characteristics of this kind of eccentric science would seem to be the following;
1. First of all, it uses a hydraulic model, rather than being a theory of solids treating fluids as a special case; ancient atomism is inseparable from flows, and flux is reality itself, or consistency.
2. The model in question is one of becoming and heterogeneity, as opposed to the stable, the eternal, the identical, the constant. It is a “paradox” to make becoming itself a model, and no longer a secondary characteristic, a copy; in the Timaeus, Plato raises this possibility, but only in order to exclude it and conjure it away in the name of royal science. By contrast, in atomism, just such a model of heterogeneity, and of passage or becoming in the heterogeneous, is furnished by the famed declination of the atom. The clinamen, as the minimum angle, has meaning only between a straight line and a curve, the curve and its tangent, and constitutes the original curvature of the movement of the atom. The clinamen is the smallest angle by which an atom deviates from a straight path. It is a passage to the limit, an exhaustion, a paradoxical “exhaustive” model. The same applies to Archimedean geometry, in which the straight line, defined as “the shortest path between two points,” is just a way of defining the length of a curve in a predifferential calculus.
3. One no longer goes from the straight line to its parallels, in a lamellar or laminar flow, but from a curvilinear declination to the formation of spirals and vortices on an inclined plane: the greatest slope for the smallest angle. From turha to turbo: in other words, from bands or packs of atoms to the great vortical organisations. The model is a vortical one; it operates in an open space throughout which things-flows are distributed, rather than plotting out a closed space for linear and solid things. It is the difference between a smooth (vectorial, projective, or topological) space and a striated (metric) space: in the first case “space is occupied without being counted,” and in the second case “space is counted in order to be occupied.”
4. Finally, the model is problematic, rather than theorematic: figures are considered only from the viewpoint of the affections that befall them-sections, ablations, adjunctions, projections. One does not go by specific differences from a genus to its species, or by deduction from a stable essence to the properties deriving from it, but rather from a problem to the accidents that condition and resolve it. This involves all kinds of deformations, transmutations, passages to the limit, operations in which each figure designates an “event” much more than an essence; the square no longer exists independently of a quadrature, the cube of a cubature the straight line of a rectification. Whereas the theorem belongs to the rational order, the problem is affective and is inseparable from the metamorphoses, generations, and creations within science itself. Despite what Gabriel Marcel may say, the problem is not an “obstacle”; it is the surpassing of the obstacle, a projection, in other words, a war machine. All of this movement is what royal science is striving to limit when it reduces as much as possible the range of the “problem-element” and subordinates it to the “theorem-element.”

This Archimedean science, or this conception of science, is bound up in an essential way with the war machine: iheproblemataaie the war machine itself and are inseparable from inclined planes, passages to the limit, vortices, and projections. It would seem that the war machine is projected into an abstract knowledge formally different from the one that doubles the State apparatus. It would seem that a whole nomad science develops eccentrically, one that is very different from the royal or imperial sciences. Furthermore, this nomad science is continually “barred,” inhibited or banned by the demands and conditions of State science. Archimedes, vanquished by the Roman State, becomes a symbol.The fact is that the two kinds of science have different modes of formalisation, and State science continually imposes its form of sovereignty on the inventions of nomad science. State science retains of nomad science only what it can appropriate; it turns the rest into a set of strictly limited formulas without any real scientific status, or else simply represses and bans it. It is as if the “savants” of nomad science were caught between a rock and a hard place, between the war machine that nourishes and inspires them and the State that imposes upon them an order of reasons. The figure of the engineer (in particular the military engineer), with all its ambivalence, is illustrative of this situation. Most significant are perhaps borderline phenomena in which nomad science exerts pressure on State science, and, conversely, State science appropriates and transforms the elements of nomad science. This is true of the art of encampments, “castrametation,” which has always mobilised projections and inclined planes: the State does not appropriate this dimension of the war machine without submitting it to civil and metric rules that strictly limit, control, localise nomad science, and without keeping it from having repercussions throughout the social field (in this respect, Vauban is like a repeat of Archimedes, and suffers an analogous defeat). It is true of descriptive and projective geometry, which royal science would like to turn into a mere practical dependency of analytic, or so-called higher, geometry (thus the ambiguous situation of Monge and Poncelet as “savants”). It is also true of differential calculus. For a long time, it had only parascientific status and was labeled a “Gothic hypothesis”; royal science only accorded it the value of a convenient convention or a well-founded fiction. The great State mathematicians did their best to improve its status, but precisely on the condition that all the dynamic, nomadic notions—such as becoming, ‘ heterogeneity, infinitesimal, passage to the limit, continuous variation —be eliminated and civil, static, and ordinal rules be imposed upon it (Carnot’s ambiguous position in this respect). Finally, it is true of the hydraulic model, for it is certain that the State itself needs a hydraulic science (there is no going back on Wittfogel’s theses on the importance of large-scale waterworks for an empire). But it needs it in a very different form, because the State needs to subordinate hydraulic force to conduits, pipes, embankments, which prevent turbulence, which constrain movement to go from one point to another, and space itself to be striated and measured, which makes the fluid depend on the solid, and flows proceed by parallel, laminar layers. The hydraulic model of nomad science and the war machine, on the other hand, consists in being distributed by turbulence across a smooth space, in producing a movement that holds space and simultaneously affects all of its points, instead of being held by space in a local movement from one specified point to another. Democritus, Menaechmus, Archimedes, Vauban, Desargues, Bernoulli, Monge, Carnot, Poncelet, Perronet, etc.: in each case a monograph would be necessary to take into account the special situation of these savants whom State science used only after restraining or disciplining them, after repressing their social or political conceptions.

The sea as a smooth space is a specific problem of the war machine. As Virilio shows, it is at sea that the problem of the fleet in being is posed, in other words, the task of occupying an open space with a vortical movement that can rise up at any point. In this respect, the recent studies on rhythm, on the origin of that notion, do not seem entirely convincing. For we are told that rhythm has nothing to do with the movement of waves but rather that it designates “form” in general, and more specifically the form of a “measured, cadenced” movement. However, rhythm is never the same as measure. And though the atomist Democritus is one of the authors who speak of rhythm in the sense of form, it should be borne in mind that he does so under very precise conditions of fluctuation and that the forms made by atoms are primarily large, nonmetric aggregates, smooth spaces such as the air, the sea, or even the earth (magnae res). There is indeed such a thing as measured, cadenced rhythm, relating to the coursing of a river between its banks or to the form of a striated space; but there is also a rhythm without measure, which relates to the upswell of a flow, in other words, to the manner in which a fluid occupies a smooth space.

This opposition, or rather this tension-limit between the two kinds of science—nomad, war machine science and royal, State science—reappears at different moments, on different levels. The work of Anne Querrien enables us to identify two of these moments; one is the construction of Gothic cathedrals in the twelfth century, the other the construction of bridges in the eighteenth and nineteenth centuries. Gothic architecture is indeed inseparable from a will to build churches longer and taller than the Romanesque churches. Ever farther, ever higher… But this difference is not simply quantitative; it marks a qualitative change: the static relation, form-matter, tends to fade into the background in favor of a dynamic relation, material-forces. It is the cutting of the stone that turns it into material capable of holding and coordinating forces of thrust, and of constructing ever higher and longer vaults. The vault is no longer a form but the line of continuous variation of the stones. It is as if Gothic conquered a smooth space, while Romanesque remained partially within a striated space (in which the vault depends on the juxtaposition of parallel pillars). But stone cutting is inseparable from, on the one hand, a plane of projection at ground level, which functions as a plane limit, and, on the other hand, a series of successive approximations (squaring), or placings-in-variation of voluminous stones. Of course, one appealed to the theorematic science of Euclid in order to find a foundation for the enterprise: mathematical figures and equations were thought to be the intelligible form capable of organising surfaces and volumes. But according to the legend, Bernard de Clairvaux quickly abandoned the effort as too “difficult,” appealing to the specificity of an operative, Archimedean geometry, a projective and descriptive geometry defined as a minor science, more a mathegraphy than a matheology. His journeyman, the monk-mason Garin de Troyes, speaks of an operative logic of movement enabling the “initiate” to draw, then hew the volumes “in penetration in space,” to make it so that “the cutting line propels the equation” (le trait pousse le chiffre). One does not represent, one engenders and traverses. This science is characterised less by the absence of equations than by the very different role they play: instead of being good forms absolutely that organise matter, they are “generated” as “forces of thrust” (poussees) by the material, in a qualitative calculus of the optimum. This whole current of Archimedean geometry was taken to its highest expression, but was also brought to a temporary standstill, by the remarkable seventeenth-century mathematician Desargues. Like most of his kind, Desargues wrote little; he nevertheless exerted a great influence through his actions and left outlines, rough drafts, and projects, all centred on problem-events: “Lamentations,” “draft project for the cutting of stones,” “draft project for grappling with the events of the encounters of a cone and a plane,… Desargues, however, was condemned by the parlement of Paris, opposed by the king’s secretary; his practices of perspective were banned. Royal, or State, science only tolerates and appropriates stone cutting by means of templates (the opposite of squaring), under conditions that restore the primacy of the fixed model of form, mathematical figures, and measurement. Royal science only tolerates and appropriates perspective if it is static, subjected to a central black hole divesting it of its heuristic and ambulatory capacities. But the adventure, or event, of Desargues is the same one that had already occurred among the Gothic “journeymen” on a collective level. For not only did the Church, in its imperial form, feel the need to strictly control the movement of this nomad science (it entrusted the Templars with the responsibility of determining its locations and objects, governing the work sites, and regulating construction), but the secular State, in its royal form, turned against the Templars themselves, banning the guilds for a number of reasons, at least one of which was the prohibition of this operative or minor geometry.

Is Anne Querrien right to find yet another echo of the same story in the case of bridges in the eighteenth century? Doubtless, the conditions were very different, for the division of labor according to State norms was by then an accomplished fact. But the fact remains that in the government agency in charge of bridges and roadways, roadways were under a well-centralised administration while bridges were still the object of active, dynamic, and collective experimentation. Trudaine organised unusual, open “general assemblies” in his home. Perronet took as his inspiration a supple model originating in the Orient: The bridge should not choke or obstruct the river. To the heaviness of the bridge, to the striated space of thick and regular piles, he opposed a thinning and discontinuity of the piles, surbase, and vault, a lightness and continuous variation of the whole. But his attempt soon ran up against principled opposition; the State, in naming Perronet director of the school, followed a frequently used procedure that inhibited experimentation more than crowning its achievements. The whole history of the Ecole des Ponts et Chaussées (School of Bridges and Roadways) illustrates how this old, plebeian “corps” was subordinated to the Ecole des Mines, the Ecole des Travaux Publics, and the Ecole Polytechnique, at the same time as its activities were increasingly normalised. We thus come to the question, What is a collective body Undoubtedly, the great collective bodies of a State are differentiated and hierarchical organisms that on the one hand enjoy a monopoly over a power or function and on the other hand send out local representatives. They have a special relation to families, because they link the family model to the State model at both ends and regard themselves as “great families” of functionaries, clerks, intendants, or farmers. Yet it seems that in many of these collective bodies there is something else at work that does not fit into this schema. It is not just their obstinate defence of their privileges. It is also their aptitude—even caricatural or seriously deformed—to constitute themselves as a war machine, following other models, another dynamism, a nomadic ambition, over against the State. As an example, there is the very old problem of the lobby, a group with fluid contours, whose position is very ambiguous in relation to the State it wishes to “influence” and the war machine it wishes to promote, to whatever ends.

A body (corps) is not reducible to an organism, any more than esprit de corps is reducible to the soul of an organism. Spirit is not better, but it is volatile, whereas the soul is weighted, a center of gravity. Must we invoke a military origin of the collective body and esprit de corps? “Military” is not the part that counts, but rather the distant nomadic origin. Ian Khaldun defines the nomad war machine by: families or lineages plus esprit de corps. The war machine entertains a relation to families that is very different from its relation to the State. In the war machine, the family is a band vector instead of a fundamental cell; a genealogy is transferred from one family to another according to the aptitude of a given family at a given time to realise the maximum of “agnatic solidarity.” Here, it is not the public eminence of a family that determines its place in a State organism but the reverse; it is the secret power (puissance), or strength of solidarity, and the corresponding genealogical mobility that determine its eminence in a war body. This has to do neither with the monopoly of an organic power (pouvoir) nor with local representation, but is related to the potential (puissance) of a vortical body in a nomad space. Of course, the great bodies of a modern State can hardly be thought of as Arab tribes. What we wish to say, rather, is that collective bodies always have fringes or minorities that reconstitute equivalents of the war machine—in sometimes quite unforeseen forms—in specific assemblages such as building bridges or cathedrals or rendering judgments or making music or instituting a science, a technology… A collective body of captains asserts its demands through the organisation of the officers and the organism of the superior officers. There are always periods when the State as organism has problems with its own collective bodies, when these bodies, claiming certain privileges, are forced in spite of themselves to open onto something that exceeds them, a short revolutionary instant, an experimental surge. A confused situation: each time it occurs, it is necessary to analyse tendencies and poles, the nature of the movements. All of a sudden, it is as if the collective body of the notary publics were advancing like Arabs or Indians, then regrouping and reorganising: a comic opera where you never know what is going to happen next (even the cry “The police are with us!” is sometimes heard).

Husserl speaks of a protogeometry that addresses vague, in other words, vagabond or nomadic, morphological essences. These essences are distinct from sensible things, as well as from ideal, royal, or imperial essences. Protogeometry, the science dealing with them, is itself vague, in the etymological sense of “vagabond”: it is neither inexact like sensible things nor exact like ideal essences, but anexactyet rigorous (“essentially and not accidentally inexact”). The circle is an organic, ideal, fixed essence, but roundness is a vague and fluent essence, distinct both from the circle and things that are round (a vase, a wheel, the sun). A theorematic figure is a fixed essence, but its transformations, distortions, ablations, and augmentations, all of its variations, form problematic figures that are vague yet rigorous, “lens-shaped,” “umbelliform,” or “indented.” It could be said that vague essences extract from things a determination that is more than thinghood (choseite), which is that of corporeality (corporate), and which perhaps even implies an esprit de corps. But why does Husserl see this as a protogeometry, a kind of halfway point and not a pure science? Why does he make pure essences dependent upon a passage to the limit, when any passage to the limit belongs as such to the vague? What we have, rather, are two formally different conceptions of science, and, ontologically, a single field of interaction in which royal science continually appropriates the contents of vague or nomad science while nomad science continually cuts the contents of royal science loose. At the limit, all that counts is the constantly shifting borderline. In Husserl (and also in Kant, though in the opposite direction: roundness as the “schema” of the circle), we find a very accurate appreciation of the irreducibility of nomad science, but simultaneously the concern of a man of the State, or one who sides with the State, to maintain a legislative and constituent primacy for royal science. Whenever this primacy is taken for granted, nomad science is portrayed as a prescientific or parascientific or subscientific agency. And most important, it becomes impossible to understand the relations between science and technology, science and practice, because nomad science is not a simple technology or practice, but a scientific field in which the problem of these relations is brought out and resolved in an entirely different way than from the point of view of royal science. The State is perpetually producing and reproducing ideal circles, but a war machine is necessary to make something round. Thus the specific characteristics of nomad science are what need to be determined in order to understand both the repression it encounters and the interaction “containing” it.

Nomad science does not have the same relation to work as royal science. Not that the division of labor in nomad science is any less thorough; it is different. We know of the problems States have always had with journeymen’s associations, or compagnonnages, the nomadic or itinerant bodies of the type formed by masons, carpenters, smiths, etc. Settling, sedentarising labor power, regulating the movement of the flow of labor, assigning it channels and conduits, forming corporations in the sense of organisms, and, for the rest, relying on forced manpower recruited on the spot (corvee) or among indigents (charity workshops)— this has always been one of the principal affairs of the State, which undertook to conquer both a band vagabondage and a body nomadism. Let us return to the example of Gothic architecture for a reminder of how extensively the journeymen traveled, building cathedrals near and far, scattering construction sites across the land, drawing on an active and passive power (mobility and the strike) that was far from convenient for the State. The State’s response was to take over management of the construction sites, merging all the divisions of labor in the supreme distinction between the intellectual and the manual, the theoretical and the practical, modelled upon the difference between “governors” and “governed.” In the nomad sciences, as in the royal sciences, we find the existence of a “plane,” but not at all in the same way. The ground-level plane of the Gothic journeyman is opposed to the metric plane of the architect, which is on paper and off site. The plane of consistency or composition is opposed to another plane, that of organisation or formation. Stone cutting by squaring is opposed to stone cutting using templates, which implies the erection of a model for reproduction. It can be said not only that there is no longer a need for skilled or qualified labor, but also that there is a need for unskilled or unqualified labor, for a dequalification of labor. The State does not give power (pouvoir) to the intellectuals or conceptual innovators; on the contrary, it makes them a strictly dependent organ with an autonomy that is only imagined yet is sufficient to divest those whose job it becomes simply to reproduce or implement of all of their power (puissance). This does not shield the State from more trouble, this time with the body of intellectuals it itself engendered, but which asserts new nomadic and political claims. In any case, if the State always finds it necessary to repress the nomad and minor sciences, if it opposes vague essences and the operative geometry of the trait, it does so not because the content of these sciences is inexact or imperfect, or because of their magic or initiatory character, but because they imply a division of labor opposed to the norms of the State. The difference is not extrinsic: the way in which a science, or a conception of science, participates in the organisation of the social field, and in particular induces a division of labor, is part of that science itself. Royal science is inseparable from a “hylomorphic” model implying both a form that organises matter and a matter prepared for the form; it has often been shown that this schema derives less from technology or life than from a society divided into governors and governed, and later, intellectuals and manual labourers. What characterises it is that all matter is assigned to content, while all form passes into expression. It seems that nomad science is more immediately in tune with the connection between content and expression in themselves, each of these two terms encompassing both form and matter. Thus matter, in nomad science, is never prepared and therefore homogenised matter, but is essentially laden with singularities (which constitute a form of content). And neither is expression formal; it is inseparable from pertinent traits (which constitute a matter of expression). This is an entirely different schema, as we shall see. We can get a preliminary idea of this situation by recalling the most general characteristic of nomad art, in which a dynamic connection between support and ornament replaces the matter-form dialectic. From the point of view of nomad science, which presents itself as an art as much as a technique, the division of labor fully exists, but it does not employ the form-matter duality (even in the case of biunivocal correspondences). Rather, it follows the connections between singularities of matter and traits of expression, and lodges on the level of these connections, whether they be natural or forced. This is another organisation of work and of the social field through work.

It is instructive to contrast two models of science, after the manner of Plato in the Timaeus. One could be called Compars and the other Dispars. The compars is the legal or legalist model employed by royal science. The search for laws consists in extracting constants, even if those constants are only relations between variables (equations). An invariable form for variables, a variable matter of the invariant: such is the foundation of the hylomorphic schema. But for the dispars as an element of nomad science the relevant distinction is material-forces rather than matter-form. Here, it is not exactly a question of extracting constants from variables but of placing the variables themselves in a state of continuous variation. If there are still equations, they are adequations, inequations, differential equations irreducible to the algebraic form and inseparable from a sensible intuition of variation. They seize or determine singularities in the matter, instead of constituting a general form. They effect individuations through events, not through the “object” as a compound of matter and form; vague essences are nothing other than haecceities. In all these respects, there is an opposition between the logos and the nomas, the law and the nomas, prompting the comment that the law still “savours of morality.” This does not mean, however, that the legal model knows nothing of forces, the play of forces. That it does is evident in the homogeneous space corresponding to the compars. Homogeneous space is in no way a smooth space; on the contrary, it is the form of striated space. The space of pillars. It is striated by the fall of bodies, the verticals of gravity, the distribution of matter into parallel layers, the lamellar and laminar movement of flows. These parallel verticals have formed an independent dimension capable of spreading everywhere, of formalising all the other dimensions, of striating all of space in all of its directions, so as to render it homogeneous. The vertical distance between two points provided the mode of comparison for the horizontal distance between two other points. Universal attraction became the law of all laws, in that it set the rule for the biunivocal correspondence be- tween two bodies; and each time science discovered a new field, it sought to formalise it in the same mode as the field of gravity. Even chemistry became a royal science only by virtue of a whole theoretical elaboration of the notion of weight. Euclidean space is founded on the famous parallel postulate, but the parallels in question are in the first place gravitational parallels, and correspond to the forces exerted by gravity on all the elements of a body presumed to fill that space. It is the point of application of the resultant of all of these parallel forces that remains invariable when their common direction is changed or the body is rotated (the center of gravity). In short, it seems that the force of gravity lies at the basis of a laminar, striated, homogeneous, and centred space; it forms the foundation for those multiplicities termed metric, or arborescent, whose dimensions are independent of the situation and are expressed with the aid of units and points (movements from one point to another). It was not some metaphysical concern, but an effectively scientific one, that frequently led scientists in the nineteenth century to ask if all forces were not reducible to gravity, or rather to the form of attraction that gives gravity a universal value (a constant relation for all variables) and biunivocal scope (two bodies at a time, and no more). It is the form of interiority of all science. The nomos, or the dispars, is altogether different. But this is not to say that the other forces refute gravity or contradict attraction. Although it is true that they do not go against them, they do not result from them either; they do not depend on them but testify to events that are always supplementary or of “variable affects” Each time a new field opened up in science—under conditions making this a far more important notion than that of form or object—it proved irreducible to the field of attraction and the model of the gravitational forces, although not contradictory to them. It affirmed a “more” or an excess, and lodged itself in that excess, that deviation. When chemistry took a decisive step forward, it was always by adding to the force of weight bonds of another type (for example, electric) that transformed the nature of chemical equations. But it will be noted that the simplest considerations of velocity immediately introduce the difference between vertical descent and curvilinear motion, or more generally between the straight line and the curve, in the differential form of the clinamen, or the smallest deviation, the minimum excess. Smooth space is precisely the space of the smallest deviation: therefore it has no homogeneity, except between infinitely proximate points, and the linking of proximities is effected independently of any determined path. It is a space of contact, of small tactile or manual actions of contact, rather than a visual space like Eu- clid’s striated space. Smooth space is a field without conduits or channels. A field, a heterogeneous smooth space, is wedded to a very particular type of multiplicity: nonmetric, acentered, rhizomatic multiplicities that occupy space without “counting” it and can “be explored only by legwork.” They do not meet the visual condition of being observable from a point in space external to them; an example of this is the system of sounds, or even of colours, as opposed to Euclidean space.

When we oppose speed and slowness, the quick and the weighty, Celeritas and Gravitas, this must not be seen as a quantitative opposition, or as a mythological structure (although Dumézil has established the mythological importance of this opposition, precisely in relation to the State apparatus and its natural “gravity”). The opposition is both qualitative and scientific, in that speed is not merely an abstract characteristic of movement in general but is incarnated in a moving body that deviates, however slightly, from its line of descent or gravity. Slow and rapid are not quantitative degrees of movement but rather two types of qualified movement whatever the speed of the former or the tardiness of the latter. Strictly speaking, it cannot be said that a body that is dropped has a speed, however fast it falls; rather it has an infinitely decreasing slowness in accordance with the law of falling bodies. Laminar movement that striates space, that goes from one point to another, is weighty; but rapidity, celerity, applies only to movement that deviates to the minimum extent and thereafter assumes a vortical motion, occupying a smooth space, actually drawing smooth space itself. In this space, matter-flow can no longer be cut into parallel layers, and movement no longer allows itself to be hemmed into biunivocal relations between points. In this sense, the role of the qualitative opposition gravity-celerity, heavy-light, slow-rapid is not that of a quantifiable scientific determination but of a condition that is coextensive to science and that regulates both the separation and the mixing of the two models, their possible interpenetration, the domination of one by the other, their alternative. And the best formulation, that of Michel Serres, is indeed couched in terms of an alternative, whatever mixes or compositions there may be: “Physics is reducible to two sciences, a general theory of routes and paths, and a global theory of waves.”

A distinction must be made between two types of science, or scientific-procedures: one consists in “reproducing,” the other in “following “ The first involves reproduction, iteration and reiteration; the other, involving itineration, is the sum of the itinerant, ambulant sciences. Itineration is too readily reduced to a modality of technology, or of the application and verification of science. But this is not the case: following is no, at all the same thing as reproducing, and one never follows in order to reproduce The ideal of reproduction, deduction, or induction is part of royal science at all times and in all places, and treats differences of time and place as so many variables, the constant form of which is extracted precisely by the law for the same phenomena to recur in a gravitational and striated space it is sufficient for the same conditions to obtain, or for the same constant relation to hold between the differing conditions and the variable phenomena Reproducing implies the permanence of a fixed point of view that is external to what is reproduced: watching the flow from the bank. But following is something different from the ideal of reproduction. Not better, just different. One is obliged to follow when one is in search of the “singularities” of a matter, or rather of a material, and not out to discover a form- when one escapes the force of gravity to enter a field of celerity; when one ceases to contemplate the course of a laminar flow in a determinate direction to be carried away by a vortical now; when one engages in a continuous variation of variables, instead of extracting constants from them, etc. And the meaning of Earth completely changes: with the legal model, one is constantly reterritorialising around a point of view, on a domain, according to a set of constant relations; but with the ambulant model, the process of deterritorialisation constitutes and extends the territory itself “Go first to your old plant and watch carefully the watercourse made by the rain By now the rain must have carried the seeds faraway. Watch the crevices made by the run-off, and from them determine the direction of the now. Then find the plant that is growing at the farthest point from our plant. All the devils weed plants that are growing in between are yours. Later you can extend the size of your territory.” There are itinerant, ambulant sciences that consist in flowing a flow in a vectorial field across which singularities are scattered like so many “accidents” (problems). For example why is primitive metallurgy necessarily an ambulant science that confers upon smiths a quasi-nomadic status? It could be objected that in these examples it is still a question of going from one point to another (even if they are singular points) through the intermediary of channels, and that it is still possible to cut the