Where and when do networks start or end? Or how do they start or end?
We are deeply embedded in the “mesh” [Morton, 2011], or ecological relational systems which link everything to everything in a “web of life”. There are also all sorts of so-called networks around us – transportation, communication, sanitation networks, and so on. We negotiate them and they negotiate with us daily. Thus, we are networked all the time. Too much even. But are we really at net-work at all?
First, let's leave aside the idea of networks as structures or infrastructures, and instead think them as composed of infra-actions between actors, of movements tendings tensions towards-an-other. Net-work is a way of acting – of doing politics, economy and ethic, all at the same time. In last instance, net-work is a way of making the social and being in it. Net-work is a floating state of “radical inclusion” [Raunig, 2013], a multiplying multiplicity open to not-yet-multiplicity, to undifferentiated singularities.
This text is a beginning of a search for net-work as a way of becoming multitude.
“The multitude is a multiplicity, a plane of singularities, an open set of relations, which is not homogeneous or identical with itself and bears an indistinct, inclusive relation to those outside of it.” [Hardt and Negri, 2000: 103] But, despite what Hardt and Negri state, the multitude is not present already.
Today it is not sufficient to claim that we are already participating in the “mesh”, or that we are in the world wide web. We might be members of ecosystems or digital ecologies, but this does not mean that we are networking. “it is not enough to say, 'Long live the multiple',... The multiple must be made,...” [Deleuze and Guattari, 1987: 6] Made and remade.
Network, if we were to follow Deleuze and Guattari in their rhizomatic reading, “has neither beginning nor end, but always a middle (milieu) from which it grows and which it overspills”. But the middle is fragile and unstable, and the ends as well as beginnings of networks are everywhere.
If we start from the idea that network is essentially a middle ground or milieu created in a way to allow for “growth” and/or “overspilling”, that means there need to be open ends, invitations into and exits out of a network. Actually, the terms 'inside' and 'outside' do not correspond much to the case of networks. A net-work must be radically open, so what seems to lie 'outside' at a given moment is more precisely not-yet-network, but it could potentially be. It is not excluded or delimited.
If there is no biunivocal relationship between interior and exterior, this applies to the dynamics of the middle as well. There is no priority, there is no subject nor object, no one comes first and the other second.
Networking does not run along the lines AB, BC, CA between nodes A, B, and C. A network would be something like ACBACBA or CBABACBACBC, and so on. Networks are flat ontologically speaking, but they are not bidimensional spatially or temporally, they cannot be mapped or diagrammed.
Networks are not shapes nor forms, but they are real-time events or performances. Hence the difficulty or improbability of integrally representing or visualising a network. This is because the graph theory does not represent adequately the question of agency, it snapshots the network therewith cancelling the temporalities of its relations. [Galloway and Thacker, 2007: 33]
Another obstacle to understanding or visualising networks is that they defy mechanics of causality: action-reaction, answer-reply, input-output are not adequate categories or models to understand the networking's working. Causes and effects do happen within networks, but they are only the segments where the network loses its tension and eventually breaks down. Plus, networking does not obey to the laws of exchange either, be it communicational or economic. There is no give and take between distinct poles. And network does not live according to supply or demand. No one can win or lose in a network.
For all these reasons, networks cannot be merely seen, but I think that a network can be felt. If you are in the middle of one.
What happens in networks is the making of potentiality. Better said, networks make the potentiality happen. This is what Pierre Lévy calls the virtual, a higher ontological state than the 'possible' and diverse from the 'actual'. I would say that the virtual aggregates a hypercomplex intermeshing of heterogeneous energies, whereas the possible consists of a determinate series of homogeneous possibilities or outcomes.
“By the virtual we understand the set of powers to act (being, loving, transforming, creating) that reside in the multitude.” [Hardt and Negri, 2000: 357]
Here, Hardt and Negri speak about a virtual as if it were already in being 'residing' somewhere in the multitude, instead I think that it is produced and reproduced through the becoming of the multitude.
Multitude comes into being in the virtual.
Virtuality is an assembling of powers to act.
On the contrary, when the passage from the virtual through the possible to the real is made, thus when an action is accomplished by a machine or a chain of nodes, the potentiality dissipates or dissembles and is reduced into a singularity. This sounds counter-intuitive, but I maintain that a network, if it is net-working, does not accomplish/perform/cause any single thing, if we take it as a result or a consequence or a product. This would mean a meltdown of multiplicity.
Network needs to continue to fold along its edges and turns the nodes inside out outside in and is turned outside in inside out by their agencies. In the overlappings joints underlappings of edges, surfaces behind beyond between the folds are created, these lines are virtualities in potency. (This recursivity generates another improbability to fully comprehend and represent a network.)
How is then the virtuality created and how is it undone?
First, an example of what net-work is not. Transport for London, an immensely complex system of buses, trams and trains daily moves millions of people via a work of thousands of technical machines and humans. But, what happens in this “network” is that one boards a train “in-order-to” move from point X to point Y. Everything else that happens in transit – breakdowns, delays, funny announcements by the driver, unexpected encounters, etc. – is an undesired contingency or chaos. This type of system, I claim, does not act as a network since it does not produce potentiality, but instead is all about actualisation. This is because it is ideated as a tool, as an instrument, a “means” for transport. The implication is that “insofar as the tool is a tool, it is quite invisible”. [Harman, 2011: 38] This is not a problem of perception, but of the fact that a tool is relegated to the background and thus made subject to domination. Subject acts upon something else – an object. In a network, there is no domination, but only inter-agency, thus networks cannot be instruments.
Net-work is an activity of the creation of the potentiality for anything/anyone to connect to anything/anyone, this puts all networking actors on equal footing or status. According to Deleuze and Guattari's rhizomatic principle of connection: “any point of a rhizome can be connected to anything other, and must be.” This latter “must” is not a diktat for everyone to be perpetually connected to everyone, but underlines the principle as a fundamental right.
What type of networking allows for this right? Alexander Galloway differentiates three types of networks: a) centralised, that is hierarchical and bureaucratic linkage of nodes to a single central hub; b) decentralised, a multiplication of centralised network, with many hubs and their own arrays of dependent nodes; c) distributed network, which “has no central hubs and no radial nodes. Instead each entity in the distributed network is an autonomous agent”. [Galloway, 2004: 33] It is clear that only the third type comes close to this condition, but does not necessarily match it. I will get back later to this.
Heterogeneity of connections is the second condition of the rhizome. Connections are not all identical, contrary to how it looks like on many network graphs. Every single tie is a singularity, there cannot be two identical ones. At a first glimpse, the difference can be interpreted by saying that some ties are weaker and others stronger. But 'weak' and 'strong' are misleading, because they evoke 'power plays', so dear to Latour but also Foucault. Networks are not driven by power relations, if we consider them as power of something over something or similar, because there is no subject nor object. Relations in networks are ties of common labour and affect and desire.
To describe this infinite gradation of ties, I would like to use words like 'thick' and 'subtle', or 'dense' and 'thin', but they are not sufficient either. Let's think about net-working as weaving, literally, not metaphorically, thus knitting or interlacing of (diverse) yarns. An infinitely wide array of yarns of different textures widths resistances colours warmths. These diverse material and sensual properties suggest the heterogeneity of every actor's investment into each tie. The connections are flows of labour, language, passion and desire. All at the same time and in different nuances of each. Connecting is not merely a linguistic act, it always possesses an irreducible and uncountable surplus of sensibility [Berardi, 2012: 121], or I would add, sensitivities.
In weaving networks, connections are something different than conjunctions. As Berardi explains, “[w]hilst conjunction means becoming-other, living, and the unpredictable concatenation of bodies, connection means the functional interoperability of organisms previously reduced to compatible linguistic units.” [Berardi, 2012: 123]
If networking means becoming-other, who/what is the subject that emerges?
Deleuze and Guattari speak of subtracting one from the multiple <n-1> as a process of subjectivation in a rhizome. [Deleuze and Guattari, 1987: 6] In my view, a networking subject resonates with the network, she/he/it is in the process of becoming-many and making-many, and is all but 1. Networking subject is less than 1 and more than 1. 3 /4, 2 1 /2, 15, 0, etc.
Multiplicity is not about either addition (n + 1 + 1 … - capitalist accumulation) or subtraction (n – 1 -1 …. - control and command), there are no such mathematical operations in net-working. “The foundation of the production of multiplicity lies just as much in overcoming the additive logic of counting (up) as in rejecting the one, which emerges only in the (dis-)counting from the multiplicity. “ [Raunig, 2013] What is important is that both the network and and all the networking actors are multiplicities.
How does this movement proceed?
When many yarns connecting different points intersect, pass above and below each other, thus interlace or interlock, some are wefts the others are warps. “[Rhizome] is composed not of units but of dimensions, or rather directions in motion” and “[u]nlike a structure, which is defined by a set of points and positions, with binary relations between the points and biunivocal relationships between the positions, the rhizome is made only of lines...” [Deleuze and Guattari, 1987: 21]
They do not intend to say that there are no nodes, but that the patterns of interlacing are those which create the tension of the network, which resides in the middle (milieu), a field among the points nodes actors. This middle belongs to no one, and it is of everyone. It may look or even act as a unit, but it is many. Everyone are contributing to it and making it, and everyone is drawing from it and is being drawn by its in-tension. The middle is the common.
Through these movements, subjectivation takes place as a motion between singularisation and multiplication. Singularisation of every conjunction. Multiplication of every actor. Singularisation of the middle through the in-tension of actors. Multiplication of the middle through continuous weaving of the lines. Networking is a process of becoming else than one, more than and less than one.
But network, as we saw, is far from flat, symmetries and asymmetries abound in the relationships between nodes as they pass through the middle. Nevertheless, there is a potential for equality, for each node to conjoin with any other. This must be exercised, conjunctions have to be started, maintained, and restarted. Network is always in restarting, always starts from all of its beginnings, and tends to and fro all its endings, towards and through the middle. This restarting is a quest for equality, which maybe cannot ever be fully achieved. But it is more than worth to keep trying.
Against all the storms and aggressions. “[R]hizome may be broken, shattered at a given spot, but it will start up again on one of its old lines, or on new lines” [Deleuze&Guattari, 1987: 9]. This is not so sure. Network may start again or it may as well not. Actors and actions can get included and excluded, appear and disappear, and this always affects the whole of the network, even if it eventually survives and persists. In the shattering network might have been co-opted and instrumentalised by a sovereign power, something from the outside, and thus it becomes something different.
This shattering, the undoing of the network, how does it take place? Through any operation that aims to
erase the potentiality of any single node in a network to conjoin to the other(s). This is an operation of unstitching, of cutting, of destroying, ultimately of killing out the yarns and/or the node. This is accomplished through monodimensional investment of power on behalf of one or more (unweaving) actors who act upon an other or others. Thereby the nature of the network is destroyed.
This is an action of drawing the yarns in a linear action <chain of action> that moves inconsiderately of the tensions of other actors and their interconnections, without waiting or allowing for response, and, by doing this, it breaks or strains the ties and actors. What remains is a 'line of cutting' – a disjunction – which leads to an end of the network. A dead end. A goal. An outcome. A coming out of the network.
Any operation of moving actors along a single direction, making them work along one line, aligning them, whatever that intent or interest might be, is probably a symptom of gradual or discreet or exponential erasing of potential, flattening down or ironing of the multiplicity of folds and realising them on a grid (or a map or a spreadsheet or a graph). When any such operation becomes real, there is no space for a refolding reverse movement. The communication has ended, the signal is dead. On one side of the line remains nothing, while profits are harnessed on the other.
But, to get there, a great energy effort has to be dispended and the disjoining actors lose at least part of their overall potentiality to act further.
What is bewildering is that this undoing of networks is channelled precisely through a network, so it seems suicidal. Not really, because the undoing actors are dissociating themselves from the network as they go.
A realising action transforms segments of the network into a chain of command. This modus operandi is always grounded in codifying procedures protocols rules , which are as strict as possible. Paradoxically, these protocols still rely on communication, but they enact only connectivity (not conjunction). Disjoining actors make close-mesh nets of meaning, tight chains that resemble shields or armours, certainly not light adaptable floating fabrics. Thus they are slow heavy less robust, their way of being is actually a continual self-shattering. They do not reproduce themselves, they can only incorporate other vital parts of the network.
Disjoining is based on “functional interoparibility” of language (of the code or protocol), on the reduction of network actors to “compatible linguistic units”. Every networking actor is reduced to unit and made exchangeable, functionalised, made pliable to instrumentalisation.
In terms of time, a chain of command&control moves only from past towards the future, which is calculated or postulated, it is a time of realisation of a particular project. In this, ever grander parts of this mechanism become obsolete, lose their function, recede into background, as they played their role. So, it is a mechanism of dispersal of nodes and yarns: n – 1 – 1 – 1 -1 … all the way until one goal is reached, until it becomes real. Then this goal is left in the past, and a new goal is projected and so on. It is always one mission after another, a linear sequence of ones which relies on the process of unifying of actors towards the imposed goal.
Instead, a network does not know where it goes, because of its loose yarns left swinging in the wind, because of its radical openness towards the inside and the outside, its not having boundaries. Because each and every actor is singular and multiple at the same time.
So, a network lives in a state that most intimately weaves past (previous conjunctions of nodes, which create potentiality, the in-between tension of the mesh), present (the provisional position of nodes and of the yarns in between them), and future (made of virtual movements, picking-ups of old threads, makings of new threads, and so on) into a continuum. Anything must be compresently possible for a network.
Looking at Murillo's paintings of the child beggars of Seville, Hegel exclaimed: “We have the feeling that for a young person of this type any future is possible”. [quoted in Ranciere, 2009: 14] This is how one feels whilst net-working, whilst becoming multitude.
Whereas mechanic action works by calculating future and then realising it, “the virtual is never fully realised” [Lévy, 1998]. Better said, it should not be fully realised. Great care has to be taken not to fully realise a collective potential of a network. That does not mean that everything stays in a cloud of fuzzy potentialites, in some kind of a stationary state of indetermination. Potentialities are always partially realised, they need to be, because weaving relies dispenses immense effort. But, the potentiality is reproduced and built up through weaving, by the creation of new twists of yarn, new foldings, joining of new actors, regeneration of the old ones.
That is what the 'networking' or net-work (verb) mean, and this is a “great and unceasing effort” [Law, 1999] of upkeeping and maintaining a network.
To keep up in this, networks should be able to foster and take care of relationships of identification and solidarity, not alienation and instrumentalisation. Along its yarns needs to travel a sense of commonality and it has to be somehow internalised by the actors [Naess, 1989: 172]. Net-work is the place where 'in unity diversity' is being made. This unifying diversity may be the immanent end of a network, if we can say there is one. And this unifying diversity cannot ever be fully realised, but must always be driven by desire. Other name for unitary diversity might be solidarity. “[S]ocial solidarity is not an ethical or ideological value: it depends on the continuousness of the relation between individuals in time and in space.” [Berardi, 2012: 128] Continuousness of the relation. Desire for relation. This is what the net-work is made for and of.
On one start (end), many open endings are in motion. On the other end (start), many open beginnings are in motion.
Network stands in the middle of the infinity of resonations of vibrant matter on one side and the infinity of transformed vibrant matter on the other side. It is a prism of the multiplicity, with no outside (horizon) and no inside (centre or end), a virtual compresence of multiple pasts presents and futures. Virtual as actual power of wish dream desire. And, desire is always the production of the real [Deleuze and Guattari, 1984: 26], the incessant flow between the virtual the possible the real.
“There is only desire and the social, and nothing else”. [ibid.: 29] The bridge between the desire and the social is the becoming of net-work.
And this conjoining is never smooth. We daily bump into numerous firewall, closed doors, no-access signs, places where networks have been brought to dead ends. Maybe these once-networks can be reopened, there must be potentiality left. The same applies to the contrary, we may be trapped in a net, a fortress of power, a vampire castle [Fisher, 2013], thus a multitude must be reassembled to radically open it. For this, we need help and solidarity. There shall not be outsides nor insides.
Where we encounter loose and open threads, we may pick them up and conjoin the multitude. Learning to gather with the other or the others who are weaving, help them to continue the net, solidarise with them.
We can also begin a net-work with an other weaver. Two ends make a beginning of the multiplicity.
Then, when a network starts resonating, acquiring a rhythm or a sensibility, while it is happening while it is becoming, network(ing) has and shall not have an end(ing).
Berardi, F. (2012) The Uprising. On Poetry and Finance. Los Angeles: Semiotext(e).
Deleuze, G. and Guattari, F. (1987) A Thousand Plateaus: Capitalism and Schizophrenia. Minneapolis / London, University of Minnesota Press.
Deleuze, G. and Guattari, F. (1984) Anti-Oedipus. Capitalism and Schizophrenia. London: The Athlone Press.
Fisher, M. (2013) Exiting the Vampire Castle. The North Star. 22 November 2013.
Galloway, A. (2004) Protocol: how control exists after decentralization. Cambridge, MA: MIT Press.
Galloway, A. and Thacker, E. (2007) The Exploit: a theory of networks. Minneapolis: University of Minnesota Press.
Hardt, M. and Negri, A. (2000) Empire. Cambridge, MA: Harvard University Press.
Harman, G. (2011) The Quadruple Object. Winchester, UK; Washington, D.C.: Zero Books.
Law, J. (1999) Traduction/Trahison: Notes on ANT. Centre for Science Studies, Lancaster University.
Lévy, P. (1998) Becoming virtual: reality in the Digital Age. New York: Plenum Trade.
Morton, T. (2010) The Ecological Thought. Cambridge, MA: Harvard University Press.
Naess, A. (1989) Ecology, Community, and Lifestyle: Outline of an Ecosophy. Cambridge, NY: Cambridge University Press.
Rancière, J. (2009) Notes on the photographic image. Radical Philosophy 156, July/August 2009.
Raunig, G. (2013) n-1. Making Multiplicity. A Philosophical Manifesto. Eipcp.net.
plus a networked conversation thread with Marika Troili
plus a networked conversation thread with Marika Troili
To be continued...