Notes on: Deleuze G and Guattari, F (2004) A Thousand Plateaus, London: Continuum. Chapter 1 Introduction: Rhizome

Dave Harris


Any system of classification (trees or tracings), academic approach (psychoanalysis) or conventional narrative (in normal books) that rigidly applies fixed categories misses out on a lot of complexity and only takes snapshots of processes


They wrote this collectively to show how individuals are (a) pretty irrelevant and (b) merge into each other. They used ‘clever pseudonyms’ as well (3). I wonder who ‘Professor Challenger’ might be in ch. 3? [see below] A clue: it is someone who rambles on using examples from many different disciplines even though he admits he is not an expert in any of them, and whose audience tend to walk out in despair after his rambling asides). [Actually, he is a character in A Conan Doyle's stories -- an boringly aggressive adventurer]  They only keep their own names because want to be imperceptible (see ch.10 ) because it is nice to be an ordinary person sometimes [luvvie!]. The book is actually an assemblage or multiplicity so it is unattributable [the old death of the author stuff, p. 4 --must get a plagiarising student to try that one]. This book is a rhizome not a tree structure, a little machine, and it is about ‘multiplicities, lines, strata and segmentarities, lines of flight and intensities, machinic assemblages and their various types, bodies without organs and their constructions and selection, the plane of consistency and in each case the units of measure’ (5).

In writing a book like this, 'The multiple must be made, not by always adding a higher dimension but ...with the number of dimensions one already has available -- always n-1 (the only way the one belongs to the multiple: always subtracted). Subtract the unique from the multiplicity to be constructed: write at n-1 dimensions...A system of this kind would be called a rhizome' (7). [I added this bit later after reading a reference to n-1 in an educational commentary -- the author -- Gale I think -- saw this as trying to remove oneself from one's writings. I see it, after reading DeLanda, as a commentary on a multiplicity as a complex set of singularites and vectors, where all this is hidden in actual events. Writing at n-1 then means getting behind the apparent self-sufficiency of the actual event or object, stepping back from the concrete detail].

‘Becoming’ – probably just a long winded and hyped up way of saying the above, that complexity and potential should be allowed for, as a kind of familiar anti-positivsm [aimed at Hegel's separation of thesis and antitheses, I have since read] . Hence we need to work on understanding people on the plateaus (not at some mythical beginning or imposed end). Example of becoming in Ch 1 – Little Hans, who was oppressed by Freud in the familiar way (reduced to Oedipus etc) and not allowed to have heard his attempt to build a rhizome (expressed as lines connecting locations like his house and the street etc – because D&G were in the middle of some blurb about the difference between maps and tracings). Then this: ...’how the only escape route left to the child is [surely as] a becoming-animal [is] perceived as shameful and guilty (the becoming-horse of Little Hans, truly a political option)’ (16). I find this just baffling. Freud did impose his usual stuff on the kid, and Hans was tormented by wanting to find out about sex while being threatened by his father if he explored his mother (and by his mother who threatened to chop off his widdler if he masturbated). But horses only came into it all by accident really [maybe this contingency is common]-- they had big widdlers and they did odd protosexual things like falling over, drumming with their feet (an infantile image from a primal scene, although his Dad denied it) and hauling large boxes/wombs. Hans wanted to play with the street urchins in the yard opposite his house – discouraged no doubt on class grounds. But nowhere does Hans really want to be like a horse – like his father certainly, as his final fantasy showed. How on earth was he on his way to ‘becoming’ a horse in any sense? Why would any schizoid politics of becoming have helped the poor little sod who was phobic (of horses,  as it happened)? Hans is being forced to play a part for Deleuze and Guattari here – to get more schizy to help make their point. It is just as repressive as making the poor little lad act out the Oedipal drama. It is an example of the dubious kind of subjective liberation that awaits if we plunge into the world of D&G --free yourself from all constraint is a great idea for professional intellectuals but for normal people it would lead --to anomie?

Other point: have we ditched the whole array of theoretical stuff developed in Anti-Oedipus? Machinic breaks/flows have now become rhizomes? Where are the conjunctions and disjunctions?  The molar and the molecular, subjugated and subject groups? Desiring machines are mentioned and there is a whole chapter on bwo which I will get to, but it all seems far less obsessively detailed than AntiOedipus [boy, was I wrong about that!]. Has Deleuze recovered from his delirium? I think any reader should start with this volume and leave the more obsessive one for later, and then only if you are particularly interested in postructuralist revolts in French academic life and how they overthrew the dominance of Marx, Freud and  Levi-Strauss: Foucault explains it all tersely and well in the Preface (and in his essay in  Power/Knowledge)

OK, so I have revisited, after reading a lot more of this turgid bloody book,  and tried a more detailed set of notes. I am still converting bullshit into Portsmouth, however:

Books are composite without an object or subject, influenced by the exterior relations of its subject matters, with both lines of articulation, but also lines of flight etc., in an assemblage.  The book is a multiplicity [as well?].  One side of the assemblage faces the strata, but the other side faces a body without organs, which preserves open access, not limited by being attributed to a subject, 'causing asignifying particles or pure intensities to pass'.  Several things are found on this body without organs according to which lines are followed and how or whether they converge on or are selected by a plane of consistency.  The aim is to produce a book which talks about something and simultaneously about how it is made.  There is no object.  As an assemblage, the book relates only to other assemblages and other BWOs.  There is therefore nothing to understand (from an external reference system like a theory) , no signification [no external reference], although there are relations with the outside [which shape some of the relations inside?].  The book is therefore a machine, related to other machines like war machines, love machines, and ultimately to an abstract machine.  This book is clearly plugged into a literary machine which has produced other texts, hence the quality for which they have been criticized - 'over quoting literary authors' [which helps us see the self defensive nature of this whole explanation].  Literature itself is an assemblage, not an ideology - 'There is no ideology and never has been'(5) [that is, no general theory of ideology, like the rather reductionist version of Marxism they operate with]. 

They want to talk about 'multiplicities, lines, strata and segmentarities, lines of flight and intensities, machinic assemblages and their various types, bodies without organs...the plane of consistency, and in each case the units of measure' (5) [intensive units that is of course].  So their writing is quantified [no doubt a response to another criticism?], and sets out to measure 'something else'.  It is not about signifying, but rather offers 'surveying, mapping, even realms that are yet to come'[that is, mapping the whole of reality including the virtual].

We often find a tree structure in classical books, lending it to some natural or organic qualities, but also linking it to [interiorizing or internalizing] strata.  These books often demonstrate a binary structure - 'the One that becomes two'.  Of course, this is not a natural structure at all, and it already separates out nature and art.  Binary structures are not improved even by dialectic.  However, we do not always find binary divisions in nature, as with ramifying systems of roots.  Thought would do well to examine these natural structures.  However, even advanced disciplines like linguistics retain this tree structure [as with Chomsky and his grammatical trees].  The sort of discussion of multiplicity that we find involves ramifications of an initial unity, a series of objects.  A similar model involves a circle of objects around the unity, each offering 'biunivocal' forms of communication [sequential monologues].  Even if we change the model slightly to consider ramifying taproots, therefore we will still not be able to understand multiplicity.  Tree structures also dominate psychoanalysis, structural linguistics, and information science.

We might consider a slightly different kind of root -'the radicle system or fascicular root'(6), which offers a multiplicity of secondary roots.  We find this kind of model in our culture, apparently offering a more extensive totality, but still with  'comprehensive secret unity'.  An example here is the W. Burrough's 'cut-up' approach to texts, folding one text into another, adding dimensions to text by folding.  Apparently fragmented work can still be seen as offering some total insight.  Modern conceptions of offering series to produce multiplicities reproduce a linear direction for one series with a unified total collection of series in a circular or cyclic dimension.  This still restricts the notion of a multiplicity by specifying a reduced set of laws of combination: we can see this in the work of Joyce which although rejecting linear unity, still preserves cyclic unity; Nietzsche's aphorisms, similarly, still invoke 'the cyclic unity of the eternal return'[I thought other work denied this was a cyclic unity].  We still have dualism, notions of a subject and object, a natural reality and a spiritual reality, where a unity is dissolved in the object, only to triumph in the subject [that is, imposing some sort of subjective unity in thought or spirit].  This is a bad development, because the subject then finds itself in some sort of relation of ambivalence or over determination, something that always exists outside the object, while the world itself 'has become chaos'.  We find this in books which also offered this image, somehow copied from the world, both fragmented yet held together by some higher unity.

We have to do something quite different if we want to really get to a proper conception of multiplicity: 'The multiple must be made' (7).  Strangely enough, we do this not by adding dimensions to reality, but subtracting them: 'always N-1 (the only way the one belongs to the multiple: always subtracted).  Subtract the unique from the multiplicity to be constituted'[in other words, do not try to focus excessively on what is unique, but try to look behind it to see how unique objects are actually produced by multiplicities, together with other unique objects, even if they look quite separate from the point of view of empirical reality].

'A system of this kind could be called a rhizome', a subterranean root, not like the other roots we have considered [taking the multiplicity here to exist 'beneath' empirical reality].  Using this definition, we can move away from the normal notion of bulbs and tubers: all plants might seem to have these underground connections, so might animals: 'rats are rhizomes'.  Rhizomes can proliferate on the surface in all directions, or be concentrated into things like bulbs and tubers.  Both nice and nasty things can be seen as rhizomes, 'potato and couch grass, or the weed'.

Let us try to convince people further [evidently, people were pretty unconvinced before].  If we consider this abstract notion of a rhizomes, we can see that any point on it can be connected to any other point or thing 'and must be'.  We are moving away from points which turn into series through dichotomies on single dimensions.  Not all of the points on an abstract rhizome corresponds to linguistic features or conventional signs.  We are considering other kinds of semiotic chain and other kinds of coding as well: 'biological, political, economic, etc.'These use different 'regimes of signs'[of which more later], but also connect things with different statuses [for example gases of different temperatures?].  Together, these semiotics systems produce 'collective assemblages of enunciation', functioning within machinic assemblages (8).  It is not just a matter of detaching regimes of signs from objects [to study them in the abstract], since even explicit communication refers to something implicit like types of social power, or 'particular modes of assemblage'.  We see this by examining Chomsky and his argument for well developed grammar as an abstract quality: we want to insist a power relation is involved [discussed much further in the chapter on sign regimes].  [As we will argue] Chomsky's approach is not abstract enough, because it does not discuss the abstract machine that connects language to assemblages of enunciation and to micro political struggles over language.

By contrast, 'a rhizome ceaselessly establishes connections between semiotic chains, organizations of power, and circumstances relative to the arts, sciences, and social struggles', connecting diverse acts, not just linguistic ones but gestural and cognitive ones.  'There is no language in itself'[that is, no abstract model existing independently of pragmatic uses of language, enunciations].  Trying to find internal structural elements is like searching for conventional roots, avoiding what people actually do with language.  The rhizome method fully includes these other dimensions and registers, and language is therefore 'never closed upon itself'.

We must consider all complex substances as a multiplicity, breaking any relation to a single origin, subject or object, fundamental image or picture of the world.  'Multiplicities are rhizomatic', and we can use this notion to criticize 'arborescent pseudo-multiplicities'.  There is no fundamental unity, no subject or object, only 'determinations, magnitudes, and dimensions' (9).  Changes in those determinations and dimensions only occur when the multiplicity itself grows.  There is no underlying control by an artist or puppeteer, but rather 'a multiplicity of nerve fibers, which form another puppet in other dimensions connected to the first'.  If we think of some actor as pulling strings, we must complicate this picture by remembering that 'the actors nerve fibers in turn form a weave' [there is no coherent actor or subject fully separated from outside influences].  As multiplicities extend [and condense?] their dimensions into an assemblage, they change their nature, but there are no defining points or structures of the kind you find in trees in the rhizome, 'only lines'.

This sort of conception makes us see music quite differently as a connection of lines between musical points.  Even the normal conception of numbers changes if we see them as multiplicities, not just stages in a single dimension, and this helps us see that there are many units of measurement.  A power struggle can lead to the imposition of a unity, insisting on the importance of the signifier, or constructing particular privileged subjects [again discussed more in the chapter on different regimes of signs], and this often leads to the notion of some fundamental One, or for a 'pivot-unity' [privileged sets of biunivocal relationships].  This unity has to posit itself outside of the system however, and to engage in 'overcoding', from outside.  This never takes place with rhizomes or multiplicities because there is no extra dimension outside the elements that constitute it.  In this sense 'all multiplicities are flat' [never controlled by something outside them in a hierarchy].  This provides us with the concept of 'the plane of consistency of multiplicities', even though the plane itself can increase as more and more connections are made on it. There is one type of outside to a multiplicity, 'the abstract line, the line of flight or deterritorialization' (9-10) [otherwise multiplicities would never change]. The plane of consistency is also outside particular multiplicities.  Multiplicities happily fill a finite number of dimensions with no supplementary dimensions until a line of flight arises, connecting multiplicities on a plane of consistency.  Multiplicities are 'asignifying and asubjective' as argued above.

An ideal book would lay everything out on a plane of consistency -'lived events, historical determinations, concepts, individuals, groups, social formations' (10).  Kleist is admired for developing writing of this type.

Multiplicities are not internally divided into structures.  Rhizomes [of the garden type this time I assume] show us the possibilities of reforming, starting up again even if cut.  So do ants , 'an animal rhizome'.  Rhizomes do show 'lines of segmentarity' [concrete, actual lines] which can be stratified and territorialized, but they also have 'lines of deterritorialization', and 'the line of flight is part of the rhizome'.  Artificial attempts to polarize them, say into dualities will never work.  At the same time, lines of flight can be restratified, restrained by power, which might assign some authorized signifier, or understood as the attribute of a subject.  This danger is always present, as 'microfascisms just waiting to crystallise', so it is naive to divide the characteristics of the rhizomes into good and bad, unless we recognise we are making a selection, and we need to constantly revisit it.

As an example of the relative interrelationships of deterritorialization and reterritorialization, let us consider the orchid and the wasp [sigh].  Both swap processes of de and reterritorialization as they interlink [the orchid deterritorializes by forming its image, and the wasp reterritorializes on it , and lots more tedious interweavings take place, 11].  Wasp and orchid are heterogeneous but they form a rhizome. Their actions look as if they take place in one dimension, but several dimensions are in fact are involved, and not just simple imitation or capture of code - the 'becoming-wasp of the orchid and a becoming-orchid of the wasp'.  As one becoming develops, patterns of deterritorialization and reterritorialization interweave and increase deterritorialization.  It is not just a matter of resemblance, rather 'an exploding of two heterogeneous series on the line of flights composed by a common rhizome'.  No signifying is involved.  The process has been described as '"aparallel evolution" of two beings that have absolutely nothing to do with each other'.  We might have to consider a number of evolutionary schemes in this light, and abandon the old tree metaphor.  Once we have adjusted our perspective, for example, we can see aparallel evolution connecting the baboon and the cat [apparently], again making the point the two animals do not have to copy each other.  Another example might be the transfer of genetic material through viruses offering 'transversal communications between different lines' (12).  This reminds us of the importance of molecular or submolecular particles - flu is a rhizome not a classic evolving disease.  'The rhizome is an antigenealogy'.

Consider the book as in aparallel evolution with the world: it deterritorializes it, but the world reterritorializes.  Books do not mimic the image of the world, but form a rhizome with it.  Crocodiles do not imitate or reproduce their surroundings, nor do chameleons.  [And then a really silly section about the pink panther, which paints the world pink and thus becomes-world, aiming at imperceptibility, refusing to signify].   [Even non-rhizomatic in the normal sense] plants form rhizomes with something else, like the wind or animals or human beings.  We should always follow rhizomes until we get to the most abstract and tortuous connections.  The lines of flight will eventually lead us to the abstract machine operating on the plane of consistency [homely quote from Castenada about the way seeds are dispersed].  Music is another example, capable of overturning its own codes, so 'musical form…is comparable to a weed, a rhizome'.

Rhizomes have no 'genetic axis or deep structure'(13), no central pivotal point, no elements to be united by subjectivity.  At best, these features produce 'principles of tracing' and reproduction of the structure.  In classic psychoanalysis, the pivotal object is the unconscious that becomes crystallised into various complexes, developing along a genetic axis, and 'distributed within a syntagmatic structure'.  The point is to maintain some sort of 'balance in intersubjective relationships', relying on an unconscious that is already there to be uncovered, or traced. 

Rhizomes offer 'a map and not a tracing'[this reminds me of the debates between advocates of behavioural objectives and knowledge structures - the former highlight the particular favoured route through the territory, while the latter offers the whole map].  [NB they also talk of 'decalcomania' here -- an art form where things are stuck on to the surfaces eg of pots]. It does not just reproduce the real though [unlike real maps then], and links fields, undoes blockages on bodies without organs, opens onto a plane of consistency.  Their kind of map is 'open and collectable in all of its dimensions', and you can move along it in various ways, constantly modifying your route if you wish.  Individuals, groups or social formations can establish routes.  Maps can appear as works of arts, political actions or meditations.  One of its most important aspects is its 'multiple entryways' (14), and burrows inside it [like those of the pack rat] can be lines of flight or living spaces on strata.  Maps stress performing rather than just competence.  Thus schizoanalysis refuses to develop any predestined tracing.  Conventional psychoanalysts like Klein operated with ready made tracings stemming from Oedipus [in the case of Little Richard], and the child's performance was misconstrued.  Although she identified particular stages like attachment to part objects, these are really 'political options for problems', constraining children and presenting them with impasses.

Is this just not another dualism between map and tracing, one being good and the other bad?  Obviously, maps can produce tracings, just as rhizomes intersect with roots.  All multiplicities have strata on which conventional social processes like mimesis, conventional subjectivity, or power takeovers can develop.  Even lines of flight can reproduce the formations which they intend to dismantle.  But the reverse process is the key to method - put the tracing back on the map.  This does not reproduce a map.  Tracings are selections from maps, images of maps that domesticate rhizomes 'into roots and radicles', stabilizing multiplicities by signifiance [NB] and subjectification, and ending by just reproducing itself.  If you collected a series of tracings, you would only get all the problems reproduced and imposed on the map - 'impasses, blockages, incipient tap roots, or points of structuration'(15).  Conventional psychoanalysis and linguistics has produced a number of tracings, but we see the consequences, say in the case of Little Hans [I am glad he's back, I have missed him], whose map was broken and blocked, who had shame and guilt 'rooted' in him, producing his phobias.  Freud charted some of the map, but then imposed his views of the family on it [he manage to 'project it back on to the family photo'].  Klein did the same for Little Richard.  Once your rhizome has been channeled, 'no desire stirs; for it is always by rhizomes that desire moves and external, productive outgrowths'.

What we should have done is help Little Hans build a rhizome linking the family house, the line of flight into the street and so on.  Professor Freud imposed a signifier on his desires, producing only 'a subjectification of affects'(16).  Hans's option was a 'becoming-horse', 'a truly political option', but this escape route was associated with shame and guilt.  We need to reconstruct the whole map, with Freud's tracing placed on it.

We can do this on a group scale as well, placing 'massification, bureaucracy, leadership, fascisation etc.' on a group map.  This will also help us preserve lines which continue 'to make rhizome in the shadows'.  Apparently, Deligny did this with combining the maps of several autistic children.  It might be possible to enter such a map through particular tracings, 'assuming the necessary precautions are taken', like avoiding 'any manichean dualism'.  We often do have to pursue tracings to dead ends to uncover the maps of the unconscious, exploring 'rigidified territorialities that open the way for other transformational operations' [as in Chaosmosis] .  Alternatively, we can follow a line of flight from the start to demolish strata and roots, and make new connections.  Of course there are root structures in rhizomes and conversely, and there is no theoretical analysis to distinguish them, only pragmatics to compose multiplicities or aggregate intensities.  Even accounting and bureaucracy can turn into rhizomes, 'as in a Kafka novel'.  Intensive traits, like hallucinations or play with images can challenge 'the hegemony of the signifier'.  Childish gestures and play can extricate kids from tracings, like that 'dominant competence of the teacher's language - a microscopic event upsets the local balance of power'.  Even Chomsky's trees can be turned into a rhizome.  The trick is to take stems that seem to be roots, and 'put them to strange new uses' (17).  We should celebrate the rhizomes and not the tree.  Amsterdam is 'a rhizome city' connected to 'a commercial war machine'.

Brains are not ramified into dendrites, and messages can leap across the structures, making the brain itself a multiplicity, a probabilistic system, grass more than tree.  We can see this with studies of memory, and reinterpret the differences between long-term and short term memory as being the difference between the tree and the rhizome respectively.  This helps us produce and validate 'the splendour of the short term Idea'[ as in spontaneous or delirious writing?].  The traces of long-term memory are continually affected by the shorter term actions.

Tree models produce their own models of the multiple as 'a centred or segmented higher unity' (18).  They feature different sorts of links, 'dipoles', sometimes running from bottom to top, sometimes taking the form of radiating spokes.  No matter how prolific these links are, we never get beyond the old binary system and 'fake multiplicities'.  Hierarchy remains, as do 'centres of signifiance and subjectification'.  This still limits modern computer science with its central databases and command trees [referring to recent commentaries].  One example is 'the famous friendship theorem'[we all know that!  The argument apparently is that if two individuals in a society have one mutual friend then there must be one individual who is the friend of all the others.  These two commentators suggest that this leads to an argument for the philosopher as the universal friend of humanity, a kind of benign dictator].  The proposal is to develop acentred systems instead, with flows being driven by differences in intensity [this sounds like a modern conception of embryology as well, discussed in DeLanda], forming a graph not a tree, a map in their terms.  Or take collective actions like the coordination of soldiers in a war machine: is a general really necessary, or could we operate with a 'war rhizome', 'guerrilla logic' (19) [texts advocating this like Debray's Revolution in the Revolution were very popular at the time].  We could model collectives as 'an acentred multiplicity possessing a finite number of states with signals to indicate corresponding speeds'[on the technical notion of speed, see notes on other chapters], and this might even be able to resist centralisation.  We could consider anything in this way [or as they put it 'Under these conditions, N is in fact always N-1'.  Prats], as a calculation, trying to induce a change in state.  We can even recast psychoanalysis away from its authoritarian tracings and develop schizoanalysis, 'the unconscious as an acentred system, in other words, as a machinic network of finite automata (a rhizome)'.  The same goes for linguistics.  In both cases, we have to produce the new conception, and with its new statements, different desires: the rhizome is precisely this production of the unconscious' (20).

The tree has been a powerful model of reality in the west, but then 'The West has a special relation to the forest'[SIC, 20], and to deforestation.  In the East, the steppe and the garden, or the desert and the oasis have produced 'a different figure', the 'cultivation of tubers by fragmentation of the individual', and a rejection of sedentary animal raising.  [Let's hear it for the nomads in other words, as in ch 12].  This gets close to claims that the East has developed something that might be a rhizomatic model and leads to some fanciful work explaining all sorts of differences between western and eastern morality and philosophy, including a preference for transcendence rather than immanence, or different images of god.  And music and sexuality ['of the earth' that is].  In particular, 'the rhizome...  is a liberation of sexuality not only from reproduction but also from genitality' [great news for 60s permissives].  Henry Miller says China is the weed threatening the orderly cabbage patch of the West [Oh good.  The authors remind us that this might be an imaginary China.  The politics of actual China was of course a major issue for French radicals].

What about America?  It has been dominated by trees, such as the interest in European ancestry, but anything that's important 'takes the route of the American rhizome: the beatniks, the underground, bands and gangs'(21).  'The conception of the book is different'.  There is a difference between the arborescent East, with its concern for ancestry, and the 'rhizomatic West, with its Indians without ancestry [not after Europeans destroyed their societies anyway], its ever receding limits, its shifting and displaced frontiers' [these blokes have watched too many Westerns].  This reverses the European tensions between west and east.  [And, as a clincher], 'The American singer Patti Smith sings the Bible of the American dentist: Don't go for the root, follow the canal'.

Western bureaucracy may also have divided into different types, whether based on agrarian societies with trees, more modern societies like feudalism, inventing property and developing the state, and expansion through warfare.  'The Kings of France chose the lily because it is a plant with deep roots that clings to slopes'[more totally convincing argument].  It might be different in the orient, where the state is not so arborescent, and bureaucracy works with the hydraulic model, where the state channels and distributes classes [apparently based on Wittfogel on Asiatic modes of production].  We see metaphors of rivers to describe rulership.  [One obvious symbol can be easily dismissed by assertion] 'Buddha's tree itself becomes a rhizome' (22).  America as an intermediary state, both liquidating people [geddit?] and having flows of immigration and capital, so rootnd rhizome come together.  This means that 'There is no universal capitalism, there is no capitalism in itself; capitalism is at the crossroads of all kinds of formations'.  [Dear god!  Their stand depends on this sort of 'evidence'?]

However 'we are on the wrong track with all these geographical distributions.  An impasse.  So much the better'[nothing can stop them].  The rhizomes also have their own despotism and hierarchy, and this only goes to show that there is no dualism either ontological or axiological, no simple good and bad, but rather arborescence in rhizomes and vice versa.  Rhizomes can generate their own peculiar 'despotic formations of immanence and channelization', and root systems can have 'anarchic deformations'.  [After all this] the two systems are not opposed, although one is transcendent and the other immanent.  We should also choose either a model that is forever constructing or collapsing, or one that prolongs itself.  If we seem to have sneaked in another dualism, this is because of 'the problem of writing'.  If we want to explain things, 'anexact expressions are utterly unavoidable'[and are indeed celebrated in later chapters].  We also tactically use one dualism to challenge another, moving through dualist models to 'arrive at a process that challenges all models'.  We have to constantly correct the dualisms through which we pass until we arrive that 'the magic formula we all seek—PLURALISM = MONISM' (23) [an underlying level of being that explains all empirical variety—reductive for Badiou].

To summarize, rhizomes connect any point to any other points, and display traits of different natures, covering different regimes of signs 'and even non-signed states'. They are not reducible to the One or the multiple [certainly not the one that leads to a binary series, and certainly not to the falsely multiple].  It [the rhizome] is composed of dimensions or 'directions in motion'.  It has no beginning or end, only 'a middle (milieu) from which it grows and over spills' [I often wonder if translating milieu as middle rather than context is helpful here].  It constructs linear multiplicities with N dimensions.  It has no subject or object.  It moves on a plane of consistency 'from which the One is always subtracted (N-1)' [that is, multiplicities turn into singularities when they 'cool down', or move to a state with fewer dimensions, like moving from N dimensions to three dimensions].  Changes of this kind are 'changes in nature' as well [changes in state would be better].  There is no underlying structure with points, positions and binary and biunivocal relationships between them.  The rhizome 'is an anti genealogy'[that is, it does not simply involve from earlier to later states].  It is a short term memory [as explained above].  It 'operates by variation, expansion, conquest, capture, offshoots'.  It 'pertains to a map that must be produced, constructed' with all the qualities listed above, and the trick is to locate tracings on the map not the opposite.  The rhizome is acentred and non hierarchical.  It is not a signifying system.  The questions it poses about sexuality, but also relations with animals, vegetables, the world, politics, the book, the difference between the artificial and the natural, are not answered in the usual arborescent way, but by positing 'all manner of "becomings"' (24).

A rhizome is made of plateaus, middles, borrowing from Bateson to refer to the region of intensities, developed in Balinese culture.  Plateaus evidently are found along planes of consistency.  Western thought tends to think in terms of beginnings and ends instead: 'for example, a book composed of chapters has culmination and termination points' instead of a series of plateaus communicating with each other.  A plateau is 'any multiplicity connected to other multiplicities by superficial underground stems in such a way as to form or extend a rhizome'.  'We are writing this book as a rhizome', in a circular form 'but only for laughs'.  Each writer decided to work on a particular plateau.  Hallucinatory experiences helped trace the lines between the plateaus.  Circles of convergence were constructed, but each plateau can be read starting anywhere and related to any other plateau.

Constructing the multiple requires a method, and 'no typographical cleverness, no lexical agility, no blending or creation of words, no syntactical boldness, can substitute for it'.  Indeed, these techniques must be broken out from their original use which was to express some hidden unity.  Only a few have managed to do this [and a note refers us to de la Casiniere Absolument necessaire: the emergency book, Paris 1973, and also 'research in progress at the Montfaucon Research Centre'].  'We ourselves were unable to do it', and used words 'that in turn function for us as plateaus 'RHIZOMATICS = SCHIZOANALYSIS= STRATOANALYSIS= PRAGMATICS=MICROPOLITICS'.  We consider these words to indicate concepts, and, in turn, lines, 'number systems attached to a particular dimension of the multiplicities'(25) including 'strata, molecular chains, lines of flight or rupture, circles of convergence etc.'.  They are not offering a science, which is as dubious a general concept as is ideology [ referring to the great science/ideology debate in marxism then raging?]  Instead 'all we know are machinic assemblages of desire and collective assemblages of enunciation'.  We are not offering significance [SIC - is this a typo?] or subjectification.  We are 'writing to the nth power' to avoid individuated enunciation which is trapped by significations and subjects [so these nasty developments are wished away by collective writing].

Assemblages act on a number of flows, semiotic material and social.  These effects are often recapitulated by science or theory, but at the expense of the division between reality and representation, and the author [all seen as 'fields' here].  The assemblage connects multiplicities from each of these fields, so books do not just refer to reality as its object nor to authors as its subject, nor to other books ['a book has no sequel', a rather ironic comment in the circumstances].  They are not trying to represent some outside, because this would involve image, signification [again] and subjectivity.  The book as an assemblage rather than an image of the world, a  'rhizome book'.  We should never send down roots.  We should let things occur to us from the middle not from any roots.  This is not easy, but we should 'try it, you'll see that everything changes'.  This is like Nietzsche arguing that aphorisms had to be ruminated, hence 'never is a plateau separable from the cows that populate it, which are also the clouds in the sky'[I am so pissed off with this poetry].

History is written from a sedentary point of view, and from the point of view of the state, even when discussing nomads.  We need to develop instead a nomadology.  Rare examples of this include Schwob's book on the children's crusades, with multiple narratives and variable numbers of dimensions.  Another one they like is Andrzeweski's The Gates of Paradise, which apparently has a single uninterrupted sentence representing flows of children and their confessions they give to the monk, 'a flow of desire and sexuality' (26).  What is important is the collective assemblage of enunciation, the machinic assemblage of desire, both plugged into outside multiplicities.  Another example is Farachi on the fourth crusade, with unusually spaced sentences, and typography that begins 'to dance as the crusade grows more delirious'.

Writing needs a war machine and lines of flight, but still runs the risk of copying or modeling or imaging something outside, even the books above.  Maybe the crusades are only limited examples of nomadism.  The struggle will be to find writing about proper heterogeneous nomadism.  Most 'cultural books' are but tracings, however, including tracings of previous books, other books, established concepts and words.  Anti cultural books on the other hand pursue 'forgetting instead of remembering...underdevelopment instead of progress towards development' making maps.  [In this sense, 'RHIZOMATICS=POP ANALYSIS'- no doubt they had in mind Pop Art?].  However, these still might contain 'the blocks of academic culture or pseudo scientificity' [this book certainly does].  The best of modern science and mathematics are nomadic, rather than tracing concepts.  In most cases though, the state has long been the model for the book [see chapter 12].  War machines make 'thought itself nomadic', as in Kleist and Kafka.

So what we should do is 'write to the nth power, the N-1 power'[via a collective enunciations and links between different regimes of signs and so on].  Write with slogans, such as 'run lines, never plot a point!'.  Develop speed, follow lines of chance and lines of flight.  Don't be a general.  'Don't have ideas just have an idea (Godard)'[obscure here, but eventually explained in the books on cinema].  Make maps.  Be the pink panther'.  Take comfort from the old song about Old Man River [with a verse reproduced on 27].

The rhizome has no beginning and end.  It is a matter of alliance rather than filiation.  It proceeds by the conjunction 'and…and…and' [again obscure here but discussed elsewhere, probably first in Anti Oedipus.  It is really urging us to join together things that are heterogeneous].  'Uproot the verb "to be"'[in other words do not try to find out what a thing actually is].  Do not ask useless questions about origins or destinations.  Do not seek foundations or beginnings - 'all imply a false conception of voyage and movement'.  Follow and admire Kleist, Lenz and Buchner, who proceed from the middle.  American literature and some English literature shows this 'rhizomatic direction' (28), following a 'logic of the AND'[more of this in Deleuze's work on literature in the clinical project].  They practice pragmatics.  We should not see the middle as the average but something where speed increases [which I think means something where we can see the connections between things that are widely distributed].  The notion of between should be seen as 'a perpendicular direction, a transversal movement that sweeps one and the other away', just as we can see a stream without beginning or end picking up speed and undermining its banks.

back to menu page