Martin Rosenberg


I wish to confront the claims made for hypertext systems as vehicles for avant-garde expression--for tactical resistance by artists against cultural forms of domination, and, implicitly, for the "liberation" of the consumers of avant-garde art from those forms of domination. I am moved, in particular, to respond skeptically to the tropes from physics and mathematics used by participants of the discussion list Technoculture (from February through May, 1992) to describe the potential of hypertext to offer resistance and liberation. What follows, then, is a critique of the correspondence, implied by these tropes, between specific laws of physics, associated with the terms geometry, contingency, nodes, entropy, and non-linearity, and any property of hypertext that would validate its use as an avant-garde medium.

According to Hayden White, a trope is a turn of phrase linking an abstract concept to the physical world, thus establishing a correspondence between the physical world and human ideation. Tropes are "inexpungeable from discourse in the human sciences" (1978: 1-2). Every trope, therefore, becomes a fiction, the authorship of which all writers must deny, in order to preserve the truth-content of their discourse.

For Jacques Derrida, tropes reflect a tolerance for a "provisional loss of meaning" to arrive at "what is proper" (1974: 45). Tropes, then, demonstrate their truth content only by assuming some essential connection between word and thing: "Like mimesis, metaphor comes back to physis, to its own truth and its presence. Nature always finds in it its own analogy, its own resemblance to itself, and finds increase there only of itself" (45). But this implied self-evidence becomes problematic when the fiction of essential language no longer disguises how tropes allow for "an inevitable detour," a "horizon of circular appropriation of the proper sense" (73), and this problem becomes especially acute in trans-disciplinary borrowings. Scientists resort to tropes from cultural phenomena to make their descriptions of physical phenomena accessible; artists and social philosophers resort to tropes from physical phenomena to similarly explain phenomena in their domains.

These borrowings become complicated further by the relative status of each discipline. For example, the borrowing of tropes from physical phenomena by the arts and social philosophy marks their marginalized position, while the borrowing of cultural tropes by scientists marks their cosmological reach, reflecting the domination of the sciences across the range of social discourses. Any physics trope will be more true than any cultural trope. As Michel Serres writes of the poverty of the arts:

Science is on the side of power, on the side of effectiveness; it has and will have more and more credit, more intellectual and social legitimacy, and thebest positions in government; it will attract strong minds--strong in reason and ambition; it will take up space. (1990: 4)

Here Serres emphasizes the legitimating power as well as the fictiveness of tropes from science. Before we proceed, it will be important to recognize that the motive for constructing such correspondences lies with the will to power--a power that the avant-garde does not possess without recourse to the tropes that possess it.


An environment for software design with pedagogical and aesthetic implications, hypertext has attracted numerous enthusiasts seeking to construct "exploratory" and "constructive" (Joyce, 1988) texts in Storyspace, Intermedia, HyperCard and other systems. These enthusiasts seek to transform pedagogy and art through tactics made possible by the inherent capability of hypertext to alter writing and reading in ways barely imaginable except for certain recent critical theorists.

For example, many premise these claims for hypertext on its remarkable capacity, as discussed by George Landow (1992) and Jay Bolter (1991), to "literalize" the post-structuralist notion of intertextuality. By literalize I refer to the capacity of hypertext to offer "non-linear" (Landow: 24-5) access to units or "lexia" (Barthes) of text or information, thus dispersing the logic(s) of narration and argumentation by constructing divergent or recursive patterns. Further, these patterns explode the relationship between writer and reader by making the role of the reader more participatory, even subversive. As Jay Bolter writes:

"The reader may well become the author's adversary, seeking to make the text over in a direction that the author did not anticipate....the computer therefore makes visible the contest between author and reader that in previous technologies has gone on out of sight, "behind" the page (1991: 154)."

Part of the charm associated with this new "writing space" has to do with the capacity of "wreaders" to jump through links from lexia to lexia, forwards, backwards--at the will of the reader in control of the cursor, a wreader who can freeze the text, who is aware of a Home button, as in HyperCard, or the mapping features of Storyspace, who can gain an instantly transcendent perspective of the wreader's trajectory. This control over experience, exemplified by the "function" of transcendence to detach the observer from the phenomena being observed, reminds one of the control over the natural environment enabled by the calculus of Leibnitz and Newton.

The problem with this claim, and its valorization of the creation of and navigation through multidirectional flows of documents and/or "writing spaces," lies with the explicitly geometrical dimension to the environment of hypertext. "Lexia" and links--all the connections possible between cards, buttons and fields (as in HyperCard), or the "container-and-spaces mapping of Storyspace" (Joyce, electronic mail, 6/4/92)--create rhetorics entrapped in the necessarily logocentric geometry of regulated time and space. (We might note that rhetoric, in its classical sense, depends much on geometrical tropes such as "stasis.") In other words, we need to confront the geometrical nature of hypertext systems and their rhetorics by noticing: first, how in physics (and mathematics) the terms non-linear and multi-linear are equivalent; and second, how these equivalent terms, as they apply to hypertext systems, correspond to the indifference toward the direction of time's flow that calculus and other geometries exemplify, and render problematic, by their spatialization of duration and its implicit linearity into discrete units. Now the term non-linear does have a slightly altered significance in a form of chaos theory, with reference to recursiveness in the history of dynamic systems. But I think that it is safe to say that the uses of the trope non-linear in statements about hypertext don't involve this sense, although perhaps they should, insofar as that recursiveness is a characteristic of self-organizing systems.

At issue here is the hegemony of the human experience of duration by mechanical clocks and by calculus, solidified during the First Industrial Revolution, and first diagnosed by Henri Bergson in Time and Free Will (1890) and Matter and Memory (1896). Historically, the avant-garde derives much of its transgressive energies by resisting this hegemony, and its obsessive celebration of contingency, exemplified by the works of Duchamp, Beckett, Cage, Pynchon, Pilobolus, and by African American classical music, symptomizes that resistance and its energies. If we consider hypertext seriously as an avant-garde medium, it should stand scrutiny in terms of this ideological struggle between geometry and contingency.

Theorists of hypertext hope that the contingencies of choice made available through the operation of buttons opening windows into other texts, as they are found along the otherwise stable linearity of a single primary text, in effect liberates the users (the "wreaders" or "riters") from the confines of the relentless linearity of narrative, or from the linear mono-logic of academic argumentation. But, in theories of the avant-garde, the enemy is not linearity, but the non-linear perspective of geometry; not the prison-house of time but the fiction of transcendence implied by the indifferent epistemological stance toward time. We need to confront the fact that, for any avant-garde moment, the problem lies with the capacity of geometry, as it constitutes the structure of any medium, including hypertext, to seduce, in Baudrillard's (1988) sense of the word, any particular strategy of resistance to its frame. Geometry can do so simply by promising to the observer functional transcendence.

Ilya Prigogine and Isabelle Stenger (1980; 1984) have analyzed rhetorically the ideological dimension of two oppositional discourses within the physical sciences: "Being" (premised on geometry as the precise, non-linear representation of events, and associated with the field of dynamics) and "Becoming" (premised on statistical formulations of contingent events irreducible to certainty, irreversible with respect to duration, and associated with the field of thermodynamics). By applying Prigogine and Stenger's ideological critique of epistemological assumptions in physics to the polemics of avant-garde discourse, including hypertext systems and their uses, we may observe how the possibilities for contingent non-linear movement within hypertextual systems, valorized by recent proponents, merely enact simulations that remain complicitous with the logocentric structures that regulate our thoughts as well as our lives.


In a recent posting on the e-mail discussion list Technoculture, Stuart Moulthrop constructs an opposition between savvy hypertext designers and the naive hypertext readers they wish to influence, to initiate (4/1/92). This statement reveals the conceit that the environments enabled by hypertext may in fact constitute a virtual ground for a kind of warfare. I want to reflect on just what that warfare may be like: to what extent the battlefield may be designed by the combatants; and to what extent the combatants are already inscribed before they even design the field and begin the struggle. Savvy, the combatants may inscript the hypertext; naive, they may be inscribed by the hypertext. Without doubt, whether scripting or inscribed, all combatants remain determined by fundamental structures of thought that have geometry as one origin.

My rhetorical thrust here is to bring the war back home, to demonstrate how, so far, proponents of hypertext as a transformative medium have yet to frame their discussions in terms of past debate over the "status" of the avant-garde moment, as it affects subjectivity, a moment that embraces both transgression and liberation. Further, I wish to demonstrate how hypertext, as an avant-garde medium, remains inadequate as a vehicle for that transgressive and liberating moment. Here I am limiting my focus to investigating how hypertext may influence aesthetically the processes of subjectivity.

We can see how critics have found it inherently difficult even to identify the forces contending in this warfare within hypertext, as in this telling passage from George Landow's superb investigation of the relationship between hypertext and recent (read: avant-garde) literary and cultural theories. Attempting to link the multi-directional nature of hypertext and its effect on narrative and on time itself, Landow draws on the literary criticism of Gerard Genette in order to compare this capacity for multi-direction to the readers' experience of the characters in Proust's meditation on time, la Recherche de temps perdu:

Whereas Genette's characterization of the Stendhalian oeuvre captures the reader's experience of the interconnectedness of [Michael Joyce's] Afternoon and other hypertext fictions, his description of temporality in Proust conveys the experience of encountering the disjunctions and jumps of hypertextual narrative. Citing George Poulet's observation that in la Recherche de temps perdu time does not appear as Bergsonian duration but as a "succession of isolated moments," he points out that similarly, "characters (and groups) do not evolve: one fine day, they find that they have changed, as if time confined itself to bringing forth a plurality that they contained in potentia from all eternity. Indeed, many of the characters assume the most contradictory roles simultaneously" (216). In other words, in Ý la Recherche de temps perdu readers find themselves taking leaps and jumping into a different time and different characters. In a hypertext narrative it is the author who provides multiple possibilities, by means of which the readers themselves construct temporal succession and choose characterization....(1992: 114)

The problem, however, with Genette's reading of Proust's Recherche, and Landow's appropriation of that reading to make a claim for hypertext, arises with the fact that Proust's characters experience time as leaps, as a "succession of isolated moments." They move forwards and backwards from one singular moment to another precisely because the hegemony of clock time and of calculus causes them to fail to sustain the intuitive experience of Bergsonian duration. The immersion in pure and irreversible contingency, according to Bergson, constitutes the fundamental ground of human consciousness (and which bears a genealogical relationship with Nietzsche's prior "Eternal Return"), the loss of which, according to Bergson, has had unfortunate psychological and socio-cultural consequences:

In vain, therefore, does life evolve before our eyes as a continuous creation of unforseeable form: the idea always persists in that form, unforeseeability and continuity are mere appearance--the outward relection of our own ignorance. What is presented to the senses as a continuous history, would break up, we are told, into a series of states. (1911: 30)

Those consequences are the direct result of the mind's entrapment by a "latent geometry, immanent in our idea of space, which is the mainspring of our intellect and the cause of its working" (1911: 210).

For Bergson, Immanuel Kant's a priori categories (of time and space as "coordinated" in a relational grid) were not a priori at all, but ideological constructs superimposed upon the manifold of human consciousness to aid in the activity of that consciousness in the world. Irreversible, contingent duration, then, can only occur by disrupting this "latency," by demonstrating how the human mind can experience duration completely independent of the precisely uniform repetition governing the trajectories embedded in spatio-temporal geometry:

For the system of to-day actually to be superimposed on that of yesterday the latter must have waited for the former, time must have halted, and everything become simultaneous: that happens in geometry, but in geometry alone. (1911: 277)

Liberated human consciousness, in this sense, means liberation from a geometric ideological construct that disguises the nature of human awareness in order for it to better plot industrial schedules, the trajectories of cannonballs, the circumnavigation of the globe.

In fact, one way to discuss the relationship between romanticism and the avant-garde is to point out the romantics' fascination with the fragment, and the energies released by the accompanying fragmentation and transcendent liberation of what Lacoue-Labarthes and Nancy call the "system-subject" (1988: 34-5). For these theorists, the system-subject refers to the self subjugated by Kant's a priori categories; the sublime moment its liberation; the fragment the vehicle for that liberation. As a system, the subject believes that there is a cogito, a transcendent je independent of objects of perception. From the perspective of Western epistemology, this je has had autonomous existence since at least Descartes, as Dalia Judowitz demonstrates (1988). In the energized moment of fragmentation, subject and object are no longer distinct, and time takes on supple forms. If continuity exists between romanticism and the avant-garde, it lies with the polemical will to fragmentation. Therefore, our crucial question addresses whether the "lexia" of hypertext and the fragments symptomizing the romantic transgression of that system-subject function in the same transgressive way.

In his works Proust and Signs (1977), and Bergsonism (1990), Gilles Deleuze demonstrates the centrality of Bergson's theories of contingent duration, and its extrapolation in his theory of creative evolution, to Proust's aesthetic, which suggests that the translation of Ý la Recherche de temps perdu should not be Remembrance of Things Past, but In Search of Lost Time. Accepting this reading, then, Proust's multi-novel sequence becomes the search for momentary insurrections of an experientially contingent duration subjugated by the regimes of modernity, enabled only by a precise sensory stimulation, and compromised inevitably when that moment (tea and cake) becomes associated with the regulated frames of memory and of the habitual rituals of daily life.

Bergson meant for his theory of evolution, a dialectic of contingent duration and systemic memory propelled by a living force called Älan vital, to apply to intellectual as well as physical systems--thus Proust's critical concern with the symptoms for the "evolution" of the characters in Recherche. I wish to hammer home this point of the role that contingency plays in transgression--defined here as this exploration of the desire for liberation and for evolution (broadly construed)--because I argue that Bergson's critique of duration had a seminal influence not only on a genealogy of the avant-garde that stretches from Proust and Duchamp in any number of directions, and in particular to Beckett, Cage, and Pynchon as well as to performance artists. Bergson's ideological critique of time and space had similar impact on the hegemony theories of Gramsci, and therefore on Althusser, and thus as well on Foucault's notion of discursive formations and Deleuze and Guattari's notion of a War Machine.

Perhaps equally important for this discussion, Bergson's critique of the human cognition of duration becomes a crucial reference point for Ilya Prigogine's ideological critique of the distinction in the physical sciences between the reversible, geometrical frame identified as "Being," and the irreversible, contingent perspective identified as "Becoming" (1980).

But to return briefly to the passage by Landow: in Recherche, the characters' experience time as discrete, discontinuous units, as jumps and leaps. The reader experiences as well the units of narrative and of discrete representations of character. Landow refers to these as analogs of the discontinuous "lexia" of Afternoon and of other hypertext systems. I argue that, if Landow wishes to valorize these qualities of hypertext, then Landow's own comments underscore precisely the failure of hypertext to serve as a medium for liberation, as avant-garde artists and theorists have defined that struggle for liberation since the turn of the century. But let's look at time, and the question of non-linearity, from the perspective of Ilya Prigogine.


In order to participate in the warfare described above, we must assume naively that we have freedom to wander the topology, to define our own territory, and to design the conditions for the advanatgeous waging of war. Yet, if we confront what kind of grounding hypertext systems may have, we need to look at two things: 1) how the relations between cards, buttons and fields (as in HyperCard) constrain the rules for that warfare by the limits to its logic and/or rhetoric; and, 2) how the structures of thought brought to this virtual battleground (yet real because it brings structures of thought to the surface of a thinker's attention by projecting those thoughts onto a screen and allowing the thinker to "witness" them), constrain even our ability to manipulate its supposedly distinct logic and rhetoric. So, even the designers of hypertext must necessarily participate with the same credulity with its geometry as the "interactive" wreaders.

Let's think about our naivety (for I participate in this credulity along with many of the other contributors represented in this collection). This is what I call the naive line of reasoning: the properties of hypertext--lexia, links and maps, or cards buttons and fields--enable non-linear access to information, and non-linear autonomous construction of narrative forms from pre-existing options; or, these properties enable non-linear access to different questions, and to options that aid in the generation of thought (Michael Joyce's distinction between "exploratory" and "constructive" systems). Logic, which governs academic discourse, is linear; non-linear thinking becomes therefore associated with creativity, as in associative leaps. Logic is hierarchical; non-linear association is smooth: Here we literalize Deleuze and Guattari's distinction between arboreal and rhizomatic structures--linear/hierarchical is phallogocentric; smooth/non-linear is nomadic (1987: 25), and so on.

According to this line of thinking, then, linear thinking represents the prison-house and non-linear thinking offers liberation from that prison: freedom of thought becomes literalized in a virtual environment in which the programmer takes the initiative, and victory occurs when this virtual garden wages war on the structures of thought brought to this virtual reality by the "naive" interactive consumer of hypertext--or so we would like to believe.

Let us confront what we mean by non-linearity, by analogy, to the source of our recent cultural valorization of it--specifically from the concept of non-linear dynamics in recent chaos theory; and earlier, the non-linear symmetries of quantum field and quantum electrodynamics. Let us also recall the pop-culture texts, Capra's The Tao of Physics (1972), Zukoff's Dancing Wu Li Masters (1974) and other attempts to exfoliate Carl Jung's appropriation of Niels Bohr's work to posit culture-universals by reference to Eastern religions. In other words, the valorizing of non-linear thinking depends to a large extent on the distinction between what the popular understanding construes as old and new physics, and the popular appropriation of tropes specific to that opposition.

I would suggest that this distinction is a false one, and that our pop-culture's celebration of non-linearity is unfortunate. It depends on the incorrect premise that the so-called new physics constitutes a new and better knowledge than the old physics. In fact, if we accept the insights of Bergson and Prigogine, no fundamental epistemological shift really occurred between classical and quantum physics, and between quantum mechanics and non-linear dynamics, and its most familiar manifestation in chaos theory, fractals. Two underlying assumptions unite all these very different analyses of physical systems: time is as symmetrical as space; and geometry can represent, with certainty, the physical world.

As it is construed in physics, non-linearity is coextensive with a geometrical representation of physical reality. To anticipate my argument, anything produced out of a systemic relationship between lexia and links, cards, buttons and fields, also participates in the same geometrical episteme that produced Newton's laws and classical stasis theory, Feynman diagrams of sub-atomic particle interactions, formal logic, computer languages and the fractal scaling of seacoasts, black holes and chess.

All these systems of thinking assume that physical events are reversible with respect to time, so that the laws governing the motion of the solar system and the trajectory of sub-atomic particles make sense whether we view those events moving forward or backward in time. Further, with geometry it remains possible to reduce all events to simple immutable laws that are true for all time, that exist transcendent to the events that they describe. As Prigogine argues, serious problems exist with this perspective. In order to understand why this abstract investigation of the geometrical conception of reality is important to hypertext, let's briefly examine Ilya Prigogine's critique of that conception.


The value of Prigogine's work (alone and with Stenger) lies not just with his claim for locating the "becoming" of time's arrow even in the reversible worlds of quantum electrodynamics and low-temperature physics, but with his confronting the ideological dimension to certain epistemological assumptions concerning physical systems--an ideological bias towards a mechanistic model of physical events. This bias within the discipline of physics remains so dominant that Prigogine himself, though his research was in the physics of self-organizing systems, was awarded the Nobel Prize in Chemistry in 1977.

Irrespective of his reputation as a theorist, his career has lasting value simply for having brought this issue (which really belongs to the philosophy of science) to the surface. Further, a long genealogy exists of attempts to construct relationships between physical and intellectual systems, and I'm drawing on that tradition from a skeptical perspective in order to ask questions about our claims for hypertext. First I wish to address what Prigogine argues are two fundamental assumptions at work in physics.


According to this asssumption, Laws of Nature are transcendent to nature, and are reversible with respect to time. This means that Time has no real existence with respect to physical laws: it remains a constant, a function but not a presence. The purpose of physics is to discover the laws of nature in their simplest formulation as they govern the behavior of physical systems. The prime symptom of this assumption is the reduction of those physical systems to geometry: as in calculus, which spatializes time with respect to the history of a system; Feynman diagrams, which represent the symmetry and reversibility with respect to time of sub-atomic particle interactions; Einstein's General Relativity, which, for example, captures the effect of gravity on time-space geometry, and the effect on light's trajectory as it passes through that geometry; and John Wheeler's Geometrodynamics, which theorizes how black holes in time-space geometry could actually become "worm-holes" into another quadrant of that same geometry (universe), or "white holes" into a potentially infinite number of parallel time-space geometries (universes). Classical stasis theory, the fractal scaling of seacoasts, and chess--a metaphor used by both Feynman to describe the laws governing quantum interactions, and by de Saussure to describe the principles of langue-parole, and which is deconstructed by Duchamp in his treatise on endgame strategy--all share the geometric frame.

The field of dynamics studies matter and its interactions with the aim to "reduce" those phenomena to the simplest formal description possible. The greatest proponent of this view, Albert Einstein, put it this way:

In mechanics the future path of a moving body can be predicted and its past disclosed if its present condition and the forces acting upon it can be known. Thus, for example, the future pathes of all planets can be foreseen. The active forces are Newton's gravitational forces depending on the distance alone. The great results of classical mechanics suggest that the mechanical view can be consistently applied to all branches of physics, that all phenomena can be explained by the action of forces representing either attraction or repulsion, depending only upon distance and acting between unchangeable particles. (1960: 62; 65)

Leibniz and Newton developed infinitesimal calculus as a quantitative tool to freeze dynamic forces into a potentially infinite series of instant still frames. By providing a way to perform this operation, calculus enabled scientists to discover what Prigogine calls the lawful, deterministic and reversible nature of those forces (1984: 60). Calculus enabled a mastery over the natural world precisely by allowing scientists to deduce, from an initial state, the series of inertial or accelerating states that a planet, cannonball or other object pass through--not only into the future, but into the past as well.

We have already noticed how the series of lexia in a hypertext system seem to possess the properties of calculus, by enabling users to leap forwards or backwards, lexia by lexia. Each lexia remains separate, autonomous; the wreader's experience of the sequence of lexia remains discontinous. But this doesn't cover the profound dislocations made possible by leaps through nodes or links to other texts that embody other linearities. However, by referring to Einstein's General Theory of Relativity, and John Wheeler's Geometrodynamics, by constructing correspondences with these methodologies from physics sharing the geometrical assumptions of calculus, we will be able to demonstrate that the possiblities for transforming the processes of reading and writing, inherent to hypertext, are constrained by those very assumptions. The shared dependence on the geometrical frame by these disparate theories provides a crucial hint at the limits for hypertextual expression.


Einstein's Special Theory of Relativity demonstrates mathematically that observers are not stable with reference to objects of perception. For example, the Lorentz transformation computes, with reference to an observer in an inertial (geometric) frame, how objects shrink or human beings (as in the Twins Paradox) may "age" less quickly as they accelerate to the speed of light. The General Theory of Relativity demonstrates that space, as represented by the geometry of Newton, also lacks stability: given certain gravitational conditions, time-space geometry may curve, as revealed when light appears to bend in the gravitational field surrounding a local star such as our sun.

But so far, these "revolutionary" discoveries do not in any sense invalidate geometry as the premise for the dynamic perspective in physics. Einstein (and Minkowski's) innovations simply adapted geometry to account for these distortions by encompassing them in a four-dimensional, space-time geometry. As the Nobel Laureate John Wheeler puts it, "Space is different for different observers. Time is different for different observers. Space-Time is the same for everyone" (in Hayles, 1984: 47).

What becomes invariant here is not time and space, as represented by a geometrical matrix, but pure geometry itself. Geometry represents differing events relative to particular observers in terms of a postulated stable field. As Lawrence Sklar discusses in his works on the philosophy of space-time, three possible interpretations of the status of geometry emerge from this problematic: first, a substantial "ether" underlies geometry, energy and matter; second, geometry and therefore gravity only emerge heuristically in the extended space between distant bodies and their relative observers; and third, that in a cosmological sense, there is only geometry, gravity and radiation, and that not only bodies but their relations and transformations are products of pure space-time geometry itself (Sklar, 1984; 1985). Now we have hinted that Bergson demonstrates other possible formulations of the experience of duration that cannot be represented even by fourth-dimensional geometry (Einstein and Bergson debated each other directly on this subject: see Bergson's DurÄe et simultanÄitÄ, 1922). Some seventy years later, Prigogine adopts this angle of attack in his critique of time-reversible and time-irreversible systems. But the third option spelled out by Sklar becomes of special interest to our discussion of hypertext, because it situates John Wheeler's theory of geometrodynamics, which provides the most precise analogy to Ted Nelson's concept of the docuverse (1981; 1987).

John Wheeler addresses black holes, which are anomalies in the time-space geometry of Einstein and Minkowski. Now most conceptions of black holes sees them as points of infinite density, as singularities with infinite gravitational pull that distort the time-space geometry in their environs. Wheeler theorizes that these black holes may also indicate not just a point of singularity, but a portal, a vortex that may serve as a node through which matter may enter and, in some mysterious way, exit. This portal enables two possiblities: first, a black hole may become a "wormhole" that allows jumps from one place, through a "hyperspace," to another location in the same geometrical grid; second, and more interesting, a black hole may have a reverse function, so that mathematical representations of "white holes" have been developed to describe how matter emerges from out of a node in space-time geometry. This theory evolved even to the point of postulating that a black hole may function as a white hole emerging in a parallel universe. In this extension of General Relativity, there could be any number of parallel universes connected by any number of gravitational anomalies in their respective time-space geometries.

Wheeler took this concept further by attempting to explain matter itself as densely-curved time-space geometry that traps radiation in a standing wave: nothing exists except energy and geometry. One implication of Wheeler's attempt to subsume everything extant into geometry is that Geometrodynamics constitutes a theory that represents all forces and processes in terms of an invariant, ontologically stable frame that dissolves the distinction between form and essence. At its moment in the history of field theories, Geometrodynamics was thought of as the ultimate hegemony, the beginnings of Einstein's dream: a Grand Unified Theory of Everything (Geometrical). It has had, however, no experimental verification, and therefore constitutes a dead end in theoretical physics. But perhaps because of that fact, Wheeler's theory provides an excellent model for Nelson's "docuverse" and its rhetoric of texts and links.

Recent theorists of hypertext, and of the rhetorics possible from observing reading and writing within its environs distinguish between text and hypertext as a field of words, paragraphs, and pages on the one hand, and all those as well as nodes and links on the other. As Stuart Moulthrop puts it:

Hypertext rhetoric must take into account more than just the ordering of language into structures and genres inherited from orality or print literacy. It must also address a more complicated meta-management in which the user modifies ordering processes themselves...a secondary literacy. (1992, unpublished)

The movement through a link or node from one text, arranged as a sequence of lexia, to another part of that text or to a completely different text that is also a sequence of lexia, corresponds with the movement or trajectory of particles through a black hole that could serve as a worm hole or as a white hole to another time-space geometry. Therefore the laws governing properties of Wheeler's multi-universe and Nelson's docuverse must include those governing black holes as local disruptions of the geometrical field as well as the operations of the field itself, and those governing nodes as local disruptions of the linearity of an individual lexia as well as the operations on the lexia itself.

Now, we have mentioned that an adequate description of those disruptions in discourse can be found in post-structuralist theories of intertextuality. But what critical theorists have done by emphasizing those dislocations, and what hypertext theorists have done by emphasizing links and nodes as part of the rhetoric of a new literacy, is to ignore how, once the dislocation occurs, a normalcy emerges as the space-time black-hole adventurer, or the hypertext reader, acclimates to the new geometry or sequence of lexia. Like the characters in Proust's Recherche, hypertext users cannot sustain the experience of dislocation once it has occurred; unless, of course, as in the case of the unfortunate black hole jockey who accidentally picks a black hole singularity, never to be seen again, the hypertext user gets lost by passing, irreversibly, through one particular link.

It would be the goal of any avant-garde use of hypertext to find a way to sustain that experience of dislocation that would indicate the prolonged liberation from the hegemony of geometry and the constraints its rhetorics places on the user's awareness. In order to address this goal, let's return, through Bergson's notion of contingent duration, to the work of Ilya Prigogine.


According to this assumption, time is irreducible and irreversible; that is, because it remains inconstant, the laws governing its behavior cannot be simplified to simple, immutable laws. The prime symptom of this assumption about physical events with respect to the contingency of time's arrow is the recourse to statistical analysis in predicting the behavior of macroscopic systems like Brownian motion or heat in engines. Approximations become necessary since certainty, with respect to cause and effect, simply remains a pipe dream. As exemplified by equilibrium and non-equilibrium thermodynamics, one can never locate with certainty where on the new Ferrari rusting will begin, and when; one can never know precisely when, and according to what configuration, slime mold cells will spontaneously aggregate.

Prigogine's work, beyond ideological critique, has focused on locating to what extent physical systems heretofore privileged for Assumption #1, are vulnerable to questions confronted by Assumption #2--such as finding time's arrow in low-temperature quantum states, or even in sub-atomic particle interactions. Let's be more specific.

The search for certainty according to Assumption #1 was challenged decisively by the Second Law of Thermodynamics (given any isolated system, that system will move in the direction of greater entropy or disorder, ultimately reaching stasis, or equilibrium). Articulated by Ludwig Boltzmann's Order Principle, (given a closed system, that system will always choose the direction of greatest probability), this Second Law forces observers to recognize the roles that randomness and the irreversibility of time play in physical processes: a Ferarri is simply far more likely to turn into a pile of rust than a pile of rust will turn into a Ferarri. According to Boltzmann, the state of any system will remain perpetually contingent until it arrives at its state of stasis or equilibrium, when systemic fluctuation becomes no longer possible. Thus, Prigogine argues, came the birth of not the science of simplicity and certainty, but the "science of complexity" (1984: 104).

Ilya Prigogine won the Nobel Prize in Chemistry (1977) for demonstrating how the theory of non-equilibrium thermodynamics can explain processes that seem related formally to entropic processes, yet which generate order out of chaos. While randomness often seems a "threat," as when either a change in the "boundary conditions" or the "macroscopic parameters" of the Ferarri previously mentioned brings it to maximum entropy, for Prigogine, randomness also signals the moment when a physical, chemical or biological bifurcation point projects alternative futures or histories for that system. These futures may in turn indicate increasing orderliness rather than disorderly chaos. As Prigogine writes of the evolution of a chicken embryo:

It is a striking experience, especially for a non-biologist, to attend a movie describing the development of, for example, the chicken embryo. We see the progressive organization of a biological space in which every event proceeds at a moment and in a region that make it possible for the process to be coordinated as a whole. This space is functional, not geographical....the events are processes localized in space and time and not merely trajectories. (1980: xiv)

Yet, that moment may never be observable with any certainty, and therefore the specific pattern of self-organization can never be predicted, because the statistical nature of phenomena can never be explained with the precision expected of dynamic systems. It is the contingency of a system that enables alternative histories, alternative temporal linearities, to emerge. And it is this capacity of a textual system, with the aid of nodes, to not just project but to literalize alternative textual linearities, that theorists of hypertext point to as the essential property that enables avant-garde appropriations of the hypertext environment.


In his important essay "Reading From the Map: Metonymy and Metaphor in the Fiction of 'Forking Paths'" (1991: 119-132), Stuart Moulthrop demonstrates in the title the slippage of physics tropes between the ideologies of "Being" and "Becoming." On the one hand, the reference to Roman Jakobson's structural linguistic model of the linear and non-linear dimensions of narrative becomes linked precisely to the geometrical trope of the map: a grid is implied coordinating metonymy with reference to time, and metaphor with reference to space. On the other hand, Moulthrop's reference to Borges' short story "The Garden of Forking Paths" alludes precisely to the capability of a given hypertext system to generate any number of Prigoginian bifurcation points in the history of one text in that system, and therefore projecting, through a node or nodes, alternative narrative plottings.

Yet, except for the (profound?) dislocations made possible by the leaps through the nodes of a hypertext, what really occurs is that the attention of an observer simply becomes shifted from one geometry to another. Now differences may exist between one geometry and another, but I've been arguing that no fundamental change can occur in the framing of human awareness by regulated space in time, except in that brief moment of nodal dislocation.

Further, no claim for the experience of dislocation can be made simply by asserting that the docuverse of hypertext reveals the simultaneity of all linear texts, except by suggesting that the functional transcendence enabled by assuming any geometrical frame itself constitutes a fundamental shift in human awareness to be valorized by hypertext theorists. Yet, surprisingly, William Dickey makes that claim in his essay "Poem Descending a Staircase: Hypertext and the Simultaneity of Experience" (1991: 143-152). Dickey's misreading of Marcel Duchamp's painting Nude Descending a Staircase again demonstrates the slippage that occurs when physics terms used by hypertext theorists are caught between the ideologies of "Being" and "Becoming." Duchamp read Bergson and other theories of duration carefully, and his work consistently assaulted the geometrical frame. Yet, Dickey writes:

As in Duchamp's Nude Descending a Staircase, we understand our experience as process rather than product, and in that process are all steps and all stances of the staircase's descent at the same time (1991: 144)

The problem with this reading of Duchamp's representation of a series of still moments of the nude's descent in two-dimensional space, is that it misses Duchamp's scathing critique of the hold that geometry has on the two schools of "avant-garde" art of his day: cubism and futurism. While Picasso's Cubism represents an object's three dimensions in a two-dimensional plane independent of the flow of time, and while Marinetti's Futurism attempts to capture time's acceleration in that same (necessarily geometric) plane, Duchamp spoofs them both by synthesizing them in a representation of the spatialization of time: an artistic calculus if you will.

Duchamp's scandalous work emphasizes geometry's complete hegemony over the modern Western subject, and we need to remind ourselves of that even as we reflect on the "trajectory" of Duchamp's career as an avant-gardist attempting to disrupt geometry's grip: in his work To stare at, with one eye closed, for over an hour, in his chess treatise Opposition et les cases conjugues sont rconcilies, which attempts quixotically to isolate the precise moment when contingency, in the form of a mistake, creeps into the precise formality of the endgame, and in Being Given: 1) the Fountain, 2) the Illuminating Gas, which celebrates sexual energies and the linearity of flowing fountains in a geometric space conscribed by a chessboard-like floor.

So, despite the attention paid to hypertext with respect to contemporary avant-garde literary theory, and to tropes from physics, we can see how the docuverse enacts, simulates in fact, an intertextuality that demonstrates, through its non-linear symmetries, how entrapped we remain within the geometry that dominates our awareness.

What I find much more interesting, and which in fact reinforces some of our intuitions about how the strategies of Deleuze and Guattari, specifically their notion of a micro-and-macro politics of creativity (influenced by Nietzsche, Bergson and Prigogine), is premised upon a liberatory state possible only with the death of Geometry. The RHIZOME Project is guided by this intuition. The fork-in-the-road, the bifurcation point in the history of a system, represents the moment when Nietzsche's Eternal Return acts upon a system, its always-already contingency destroying the sense of past and future. Made possible by geometry, this sense enables the evaluation of exigency with respect to expediency which is the cornerstone of classical rhetoric and of logocentrism. The destruction of this sense constitutes the horror of the avant-garde moment.

How we construct an environment in hypertext to horrify our users seems to be the task at hand, but we have a long way to go before we can do this. We have to be willing to become horrified ourselves, and then ask whether this kind of terrorism constitutes the tactics that we wish to employ on our virtual battlefield. We can sense how Michael Joyce's notion of real-time hypertext, with its analogy to avant-garde jazz ensembles, or Donald Byrd's collaborationist Awopbopaloobop Groupuscle, and especially the dislocations inherent to Virtual Reality systems, may provide possible ways to create terror.

Or, we may wish to take the next step and see contingency as Bergson, Whitehead, Fraser, Prigogine, and Deleuze and Guattari see it: the initial condition for all creativity in intellectual systems, and and for life in physical systems. Here we are back to valorizing creativity and freedom over non-linearity and system. But at least we are away from the model of hyper-freedom associated with Jung and the so-called New Physics, a "liberation" that celebrates a bankrupt transcendence complicitous with the prisonhouse of logos.