Reconfiguring Hypertext as a Machine:
Capitalism, Periodic Tables and a Mad Optometrist.

Belinda Barnet

"There is no universal capitalism, there is no capitalism-in-itself; capitalism is at the crossroads of all kinds of formations" (ATP, 20).

We might say a similar thing about hypertext. That viewing it as a "thing-in-itself" with a liberal-democratic ideology blinds us to the movement of parts, to the forces and elements which are directing this show.

I don't mean to be a party pooper. Having worked with computers and hypertext for some time now, I am already convinced of the importance of new writing technologies, already convinced of the need to locate and open points for political intervention. It's just difficult to believe that hypertext will leap tall hierarchies in a blink of phosphor. Liberation claims pervade the field of hypertext theory: common theoretical approaches tend to envision hypertext as embodying the iconoclastic radicalism of the poststructuralists. There is something both familiar and irresponsible about all this - we have been there already with the introduction of broadcast television and the liberation claims which accompanied that. TV didn't make education more democratic and accessible, it didn't increase our collective IQ. It became the midday drug of choice for the children of the baby boomers. Utopian thinking is a function of molarity, an overcoding which neglects pragmatism and the movement of parts: the gentle death of revolution, as Brian Massumi would say. I would like to briefly explore the stream of hypertext theory, not simply to criticise, but to locate the points where the flow turns into wild whorls and eddies, to exacerbate these problems and encounter hypertext as an interrelation of conflicting forces: a machine. I'd like to suggest that literary hypertext as a unified, liberating phenomenon is an optical effect.

Hypertext as it is commonly defined is a means of organising information in a 'nonlinear' fashion, consisting of chunks of text ("lexias") connected by links to other lexias in a networked manner. The term refers to both the system and its contents. Theorists began exploring hypertext from a literary perspective in the late 80's and early 90's, claiming that the interactive nature of hypertext invites us to reconfigure our conceptions of 'text', 'narrative' and 'author' (Landow & Delaney, 2) in a fashion more suited to the nature of the medium. Hypertext shifts the responsibility of construction partly to the authors who write the links and partly to the readers who activate them. It also encourages connection across disciplinary boundaries, abandons print-based conceptions of fixed beginnings and endings and challenges narrative form based on linearity due to its dispersed, networked nature. Hypertext heralds a new form of writing: instead of the linear, passive narrative of the book and Codex Culture, we have the multilinear universe of the networked system.

Poststructuralists, it seems, have been going on about this for decades, confined to the world of print. For Derrida, Foucalt and Barthes, what was "only figuratively true in print becomes literally true in the electronic medium" (Bolter, 158). Landow, Bolter, Lanham, Joyce and a handful of other hypertext theorists point out that the Barthesian Text, a text which writes itself across the interface between the body and the unconscious as a living network, "the blocks of signification of which reading grasps only the smooth surface, imperceptibly soldered by the movement of sentences" (Barthes, 13) is finally realised in the new medium. Similarly, Derrida seems to be yearning for silicon. He writes of a "differential network, a fabric of traces referring endlessly to something other than itself, to other differential traces" (Derrida, cited in Landow, 59). Hypertext and poststructuralism seem to be speaking the same polylogue. In fact, contemporary critical theory "promises to theorise hypertext, and hypertext promises to embody and test aspects of [critical] theory" (Landow, 40).

So far so convenient. A problem arises, however, when theorists attempt to extend the qualities of this semiotic view of communication and the hypertextual network out onto society and our subjectivities. From the networked text emerges the symbolic subject: hypertext theorists attempt to construct the reader, author and society as themselves heterogeneous, polysequential texts. For instance, hypertext as a Derridean network admits of an "attenuated, depleted, eroding or even vanishing subject" (Landow, 75) who 'logs in' to a hypertextual matrix and exchanges informational patterns across the interface. For Bolter, hypertext even promises to realise a social network more suited to the way this fragmented consciousness works, to the process of subjectivation itself. As subjects in the "late age of print", ours is now a floating consciousness, "a network of signs, of which the computer is the embodiment" (208), and due to the influence of technology we are beginning to "function in a hypertextual network of affiliations". According to Bolter, hypertext as a system "has become the social ideal" (233). Something inside me feels a bit queasy as I read these words. A subject, a society... a whole universe emulating a system of links and lexias?

Perhaps this is where my feeling stems from. Claims such as these, investing themselves in "propositional statements that ascribe agency to technology itself'" (Grusin, cited in Snyder, 122) and generalising outwards to the subject, to society and politics, prescribe to a form of technological determinism. We "seem to think the cyborg can lead our dance into liberation barefoot and alone" (Joyce, 196). Technological determinism attaches a certain ideology--whether liberatory or Luddite--to the medium. Such claims seem ignorant of the fact that ideology is a human concept which is brought to bear upon technology, and similarly, that the technology itself is literally and conceptually programmed by the subject, by the interlocking institutions of author/reader/text, by the limits and directives of the technology, by the immanent desires of capitalism.

I'll state my thesis explicitly. The current definition of hypertext relies on the structure of the networked system and its contents (ie: hypertext as a thing-in-itself), and often envisions this system as the embodiment of radical poststructural theory. Consequently, it appears to be at war with the Line and Print Culture as we conceptualise them. This is the optical effect. This is the utopian dream we need to depart from. In some cases this perspective is then generalised outward, producing a society which is the specialised offspring of hypertext and poststructuralism, and a subject more concerned with extracting signals from the noise than it is with the apprehension and engaged desiring-production of itself and its environment.

I would like to note here that I am not arguing for the abandonment of literary perspectives: quite the contrary. The insitutions of thought we have inherited have influenced the development and use of the technology. They are part of the assemblage, and thus an opportunity to rock the sedentary order from within. It's just that we need to stop playing 'snap' and 'perfect match' with poststructuralism and computers, to start using theory as a machine which can intervene in and, to some extent, produce the flows. We need to open our theorising to the outside world, to encourage theoretical mutation. If we encounter a perfect match, this means that we have done nothing to induce change and movement. We've just been playing with tracing paper, trying desperately to make old systems manifest as relevant and self-same. In a word, unthreatening.

Mutations. Movement. Let's meditate on this from a different perspective. If we are to view hypertext as a productive landscape without prescribing to a framework which fundamentally cedes control and revolutionary capacities to the system, if we are to open points for intervention by recognising the many contributary flows expressing themselves as this technology, we need to "re-embody reading as movement, as an action rather than a thing... [to find] a way of reading sensually" (Joyce, 11). We need to recognise that the subject determines the shape of the technology as much as it determines her, and that this is not just a matter of counting links or hurling hosannas at some "embodied multiplicity".

Hypertext is a moving thing, an assemblage of directed flows: a machine that we must connect to if it is to function as a hypertext. It is also moving on more than just the level of nodal leaps: the contributary flows, the 'elements' of hypertext are shifting as well. More about this soon. For now, we need to understand that it isn't going to destroy print culture, the Line, logocentrism, repression or save gay whales as a thing-in-itself (ie: a system of links and lexias). We also need to understand that it isn't going to combat the military infotainment complex and make for a more democratic, healthful society, because this is one of the places where the technology has emerged from: military science and capitalism. We can intervene and change things if we recognise our part(s) in this, if we map the various bits that are factored into its form.

The first place we might start is by redefining the term "hypertext" as a thing which acts only in relation to other machines and view it in functional terms: that is, what it can produce. Johndan Johnson-Eilola and Stuart Moulthrop have already formed offshoots from the hypertext canon in this direction, and I owe much to their ideas. Viewing hypertext as a machine involves entirely rephrasing our question. Let's not ask, "is hypertext like this thing/that thing/ a Derridean network/a rhizome/liberatory/a Barthesian Text/a poststructural whatever". Let's not overcode, play with cookie-cutters, make the edges neat. Better to ask, "what else can I plug it into?", "(how) does it work?" and "which disparate elements are expressing themselves as this thing?"

Here's a periodic table of hypertext. As a machine, hypertext at once comprises and is in departure from previous generations of technologies, the demands of the workplace and educational settings, military research, the mass media, programmers and their pets, Microsoft and the corporate players, the perpetually mutating desires of the web and its puppeteers, intellectual and artistic fashion, technological limits, the limits of capital. It arises from the connective synthesis of eye-to-screen, from the institutions of author/reader/text, from the behavioural grammar which emerges across the point'n'click interface, from the immanent desires of the machine of capital. It is an assemblage arising from the interrelations of a field of forces, these forces. It is inherently unstable and in a process of perpetual change. This is the ontology of hypertext. It is a system of directed flows. If we know where these flows and their intersections are, what they are emerging from, we can muddle them. I'm going to begin my explorations from the periodic table listed above. Bit by byte. To read hypertext as a smooth, unified object with a liberal-democratic ideology says more about political composition of the critics concerned than it does about hypertext. I'd like to play with just one moving part here that I think is particularly important. Capitalism. The limit of society, the limit of hypertext. It astounds me that so many writers in the field (including myself) have fostered such a romantic intellectual relationship with hypertext to the neglect of what seems so glaringly obvious: all technological developments are overcoded by the images of consumption and production. It sounds so simple, so pedestrian, so Disney. And that's just it. This is where we need to be conducting strategic interventions.

Capitalism has its own desires, and these desires are often coextensive with hypertext as it unfolds before us. Indeed, this is actually one of capitalism's desires: it wants to be immanent to everything we do, it wants to move our trigger-happy finger as it runs the mouse across those baubles and bright links. In this it is remarkably dexterous. We cannot locate it anywhere (it's Bill Gates, it's General Electric, it's...): it is a virtual body, a body without organs. It operates by locating and binding the disperse flows of reality to its body, by constantly redefining the limits of this body and consequently the limits of society relative to the situation. It is supremely pragmatic, engaged, flexible. It operates by cultivating difference, cultivating the "other", and then by becoming-other, by homogenising difference, passing through it and redirecting it in the aid of the collective. So difference is tolerated, but it is then redirected in the service of capital. Schizophrenic flows "are decoded and axiomatized by capitalism at the same time" (AO, 246). We have to admire this dexterity, this dual operation, at the same time as exploring its operations, utilising them and thinking strategically. It has two hands (or millions). Its remarkable capacity to reflexively snap up, slightly reconfigure and redirect the flows of life, this constant (re)circulation "constitutes capital as a subjectivity commensurate with society in its entirety" (ATP, 452). So, it's not a centralised thing. Its flows are schizophrenic, all over the shop. It's just that these flows are eventually flattened, axiomatized, bound to a desiring body. And this body is constantly changing form. We don't need to wheel out Derrida and press, "play" to demonstrate this. We just need to look out the window.

So how does this whole thing unpack itself across the interface? As we have previously observed, it means that hypertext is not a unified thing at war with the Line, the State, Codex Culture. Nor is capitalism. It means that hypertext has emerged from and comprises many different things at once, and that it will not partake wholly of any one particular ideology or agenda. There will be difference and threatening change, there will be artistic ventures happening at the fringe, but we need to think and act quickly with such strategies before they are snapped up and reconfigured by capitalism. Another thing that we need to understand is that on the level of the system, of links and lexias, code and microcode, hypertext is also composed within the bounds of the State. As Stuart Moulthrop has observed, at a systemic level it is unavoidably comprised of the geometries of science: linear or multilinear, lines are still lines, logos and not nomos, even when they are embedded in a hypertextual matrix. The Author has been there before us, has layed down the lines and the possibilities. We can go this way or that way in the same way that we can choose this product or that product in the supermarket. There is a choice, but not a contingent choice in this unfolding now. There is no 'becoming' on a purely systemic level. Hypertext is a machine, yes, but it's not a rhizome, as I have written about elsewhere. This is too romantic a notion as well. Labelling it 'rhizomic' and shelving it in Cultural Studies won't do. We need to be able to adapt our theory and practice to the situation as readily as the machine of capital, and our strategies will need to come from outside the bounds of the State, from the fringe that capitalism has not yet detected: from rogue desires, from our bodies, from the 'unthought' of technology.

The animal body, as it is progressively subsumed or spliced into the geometries of hypertext and its overarching desire for speed and efficiency (the retrieval of meaning), opens one possible space for strategic intervention. The different organs, the minds folded within the flesh, the embodied thoughts that smell like sweat and blood and shit and desire... these are the flows that the machine of capital redirects into a clean, seamless point'n'click. No dirt, no fuss. Capitalism desires that our desires be homogenised, stapled to its mutating body. It doesn't repress these desires. It isn't the government, the State, Big Brother. It isn't just the geometry of the system itself, the predetermined Lines: it operates also on the level of thought, of our behaviour as we read and write technology. It is a very perceptive machine, almost a thinking thing. It locates and redirects our embodied noise, cleans up the messy bits that produce our autonomy. I think of it like the conductor of an orchestra. We have the potential to strike notes out of tune, to fart and shift in our seats, to wave at the audience and sing whatever we want to, but we go with the flow. Capitalism makes itself immanent to the way we think, breathe, sit, click. We allow it to become immanent to the way we think, breathe, sit, click. The animal body, the person sitting at the screen with the mouse, bundled with the potential to move and think otherwise, shall be the site of our micropolitics. You see, there is a behavioural grammar involved in 'interaction'. This behavioural grammar is the domestication of difference, the integration of differential power relations into a habitual, seamless movement.

The user-friendly interface of multimedia-as-we-know-it is one of the strategies of capitalism. It operates through what Michael Joyce has called "the myth of emerging order" to redirect the schizophrenic flows of thought as we interact with technology. The myth of emerging order is the idea that each reading of a hypertextual document is a quest to progressively reveal a hidden meaning (191). The encyclopaedic interface hangs on the circuits of sight. On unreflective, spinal-reflex stimulus-response. The universe of seamless point'n'cick. The other senses of the animal body reside in this sequentialised framework, eternally the alternatives, railroaded, written over by the consumable interface. Sound, proprioception, movement, randomness, smell, touch, the play of our fingers across the keyboard: these are the stray intensities and desires that have been marginalised or redirected by the flow.

Let's look at an example. A group of artists that disarticulate the seamless "don't worry, be happy - everything is under Control" interface (Fuller & Pope, 2) via the electronic magazine, I/O/D. "If we are locked in with the military and with Disney, they are locked in not just with us, but with every other stray will-to-power" (Fuller: interview, 2). Along with Adelaide based group Mindflux, these artists produce hypertext interfaces that involve sensory apparatus and navigational skills that have been marginalised as incidental in the "disabling interactive teletechnologies" of mainstream multimedia (Virilio, 3)."We are simply exploring the possibility of more complicated feedback arrangements between the user and the machine" (Fuller, cited in Barnet, 48). The reader must encounter the 'lexias' contained in the system via the stray flows, intensities, movements, stratas and organs that are not proper to the system but shift across the interface and the surface of her body.

In I/O/D2, the reader is called upon to converse with the technology outside of the domesticated circuits of sight, "dislocating the rigorous hierarchy of feedback devices" which privilege the sight-machine and disable contingent interaction in a technonarcissistic fashion (Fuller & Pope, 3). The written information is mapped across a sound-based interface, sensitive at every moment to the smallest movements of the reader's fingers on the keys and mouse: the screen itself is black, its swarm of links and hotspots dead to the eye. The reader's movements produce different bleeps and beats, each new track opening different entrances and exits through the information in dependence upon the fluctuating pitch and tempo of her music. Without the aid of searchpaths and bright links, she must move in a state of perpetual readjustment to the technology, attuned not to the information stored behind the interface, but to the real-time sounds her movements produce.

Interaction as conversation "is the difference between something that has a fixed grammar on the one hand and something that is continually and openly inventing its own logic on the other" (Fuller, cited in Barnet, 49). The electronic writing space is not inherently liberatory, but the "perpetual process of playing with process" (Fuller & Pope, 4) across the interface works to bring some real-time noise back into the game. According to Fuller and Joyce, the 'process of playing with process' simply means complicating the feedback arrangements between the user's body and the machine. Interactivity which calls upon a "mind folded everywhere within the body" (Badiou, 61) dislocates "the encyclopaedic organisation of data that preserves a point of privilege from where the eye can frame objects" by enlisting itinerant, diffuse desires in an extended period of readjustment to technology (Fuller & Pope, 3). There are no pre-ordained or privileged feedback circuits as the body is seen to comprise a myriad possible "elements or fragments of a desiring-machine" (Deleuze, cited in Grosz:1994, 168) with the potential to disrupt the flow, to proliferate.

Mainstream multimedia's desire for 'informational hygiene' would have us transcend this embodied flux and "bureaucratise the body into organs" (Fuller, 2). Information is fed through the circuits of sight in a Pavlovian field of buttons and bright links: interactivity is misconceived "as choice-making, when 'response' is a more germane concept" (Weinbren, cited in Joyce, 201). When the diffuse desire which thrives on disruption and alternative paradigms is written out in favour of informational hygiene, speed and efficient retrieval replace embodied conversation. "Disembodied [interaction] of this kind is always a con... the entropic, troublesome flesh that is sloughed off in these fantasies of strongly male essentialism is interwoven with the dynamics of self-processing cognition and intentionality. We see computers as embodied culture, hardwired epistemology" (Fuller, 2).

Another utopian dream? Nay, another opening, an opportunity to resist homogenisation, to encounter other universes of reference. A possible strategy. Like all other strategies, alternative interfaces are not an ideological overhaul enabled by the realm of techn�, but a space for "schizolupic break-out[s] from the bin" (Land, 482). Bifurcations are enacted on the micro level by desiring-machines, across an interface which seeks to dislocate intentionality in conjunction with the marginalised sensory apparatus of the reader, drawing other minds, other organs into localised conversation with the technological machine, if only for a moment. These words are also a strategy. As each letter falls behind the cursor, I realise that they are open to being subsumed and redirected in the service of capital. This is how the whole thing operates. We need to perpetually open new spaces as the old ones are snapped up. To note the interrelations of the elements and how they operate across life, the interface and everything.

The limits of technology. This is another contributary flow, an element. And by this I don't mean just the limits of what Windows '98 can do, of the speed limits, of the light barrier, of the current hardware.com. I mean also the interactions of previous generations of technology upon this particular generation, of a flow which extends back to when we first picked up a stick in the jungle to help us get ants from an anthill. The idea that technology is a tool to get things done. A thing which moves in generations of progressively more and more "efficient" tools towards an ideal: control of the environment. How does this operate? Its current "limits" shall also be defined, in part, by their former emplacement. The archeology of hypertext will give it a "spin defining the arc of its vector" (Massumi, 7). We might begin to explore this element by

Virilio, Paul. "The Third Interval: A Critical Transition" in Verena Conley (ed.) Rethinking Technologies. London & Minneapolis:University of Minnesota Press, 1993. pp. 3-12.