PMC Logo

    Node 1: Charting

    The *system* is the art, not the output, not the visual screen, and not the code. I want to let the data express itself in the most beautiful possible way.

    --Net artist Lisa Jevbratt, in Alex
    Galloway's "Perl is My Medium"

  1. From its very inception, hypertext has had the question of its ontological difference from analog text as one of its core themes. Indeed, from the earlier wave of critics such as George Landow, Michael Joyce, Jay David Bolter, Stuart Moulthrop, and Jane Yellowlees Douglas to the more recent work of Raine Koskimaa, Terry Harpold, Espen Aarseth, Mark Poster, and N. Katherine Hayles, virtually the entire history of hypertext criticism and hypertext itself has played out in terms of this very question. Generally organized in units called nodes or packets and interconnected through links--a syntactic, structural, and distinctive feature anticipated within the visionary labor of Vannevar Bush and Ted Nelson--hypertext is stationed upon the problem of itself as a discrete form of textuality.[2] Despite its claims for difference and the claims of a great deal of hypertext criticism for the same, I must say from the outset that it is not possible to locate a strict or fundamental difference in the metaphysical sense: this mode of distinction must always be fated and any binary that is constructed between the analog and digital is bound to be unraveled or dissolved. There cannot be a metaphysical or ontological difference between the analog and the digital, and yet it cannot be denied that something different happens when one works with, even performs, hypertext: the difference this difference makes is the problem that concerns me and hypertext itself.

  2. Up to this point, the question of what constitutes a difference between the analog and digital--with regard to language, text, material substrate, modality, reader, or author--has been answered at length in practical, rather than theoretical terms. While a certain reduction is required to do so, we can discern a significant divide within critical commentary thus far between those commentaries holding that the digital constitutes an epistemological break, and those holding that the digital extends, amplifies, or overlaps with the analog, or even that these categories are not adequate to describe textual properties that extend across media. Whether the line between the two is fixed, fluid, or obliterated, the two sides share the same inclination toward practical, functional standards. So, the question of the difference of digital textuality has tended to produce a standard litany of responses, whether in the mode of elegy or encomium:

    • Different media produce different readers, different reading environments, and different reading practices;
    • The book retains a kind of democracy by virtue of print technology and public libraries, while the computer is technologically and economically elitist; or, the digital retains a kind of democracy by virtue of its circumvention of the modern institutions of publishing and circulation, while the book is bound to the elitist institution of the school;
    • The modern figure of the author is no longer a tenable idea in the face of WYSIWYG editors and web rings; or the author persists as an author-function, a juridical category preserved by the renewed attention to copyright and the ownership of digital information;
    • The digital text is non-linear, open while the analog is closed, and interactive; or, the analog is itself non-linear and interactive, from the I Ching and "Choose your own adventure stories" through to artists' books and the novels of Julio Cortazar, Italo Calvino, and Milorad Pavic;
    • Computers have displaced, even killed off, the cultural authority and relevance of the book; or, the beauty and sensuality of the book can never be equaled by the flat pixels of the screen because the book maintains voice, presence, and materiality;
    • The analog book is the repository of canonical cultural value; or, despite its connection to the archive, the digital book can never be a repository at all, much less bear the weight of culture--it is too ephemeral, too closely aligned with the dot-coms, too prone to fluctuations and arbitrary standards of evaluation and appreciation.

  3. Critical treatment of the discrete and particular qualities of digital textuality is by this point quite extensive and even ubiquitous: it has been played out in such widely-divergent forums as the notable "Culture and Materiality" conference at UC Davis (1998), online forums at FEED and Wired magazines, chat settings, academic syllabi, and mainstream newspapers. These debates may not reiterate the exact terms that I have outlined, but they share a fundamental set of criteria: authorship, reading, the physicality of the book, the materiality of language, data access, utility and ease of use, speed and temporality, narratological form, and cultural value. As the noted hypertext critic and author Michael Joyce remarks on the distinctiveness of electronic textuality and his critical project that culminated with the recent Othermindedness: "[my work has been] an attempt to isolate a distinctive quality of the experience of rereading in hypertext. The claim that hypertext fiction depends upon rereading (or the impossibility of ever truly doing so) for its effects is likewise a claim that the experience of this new textuality is somehow not reproducible in the old" ("Nonce" 586). In the end, reproducibility is the de facto or most significant criteria for the distinctiveness of hypertextuality for Joyce; that is to say, it is the irreproducible and even unfixable effect that makes difference paradoxically manifest. He goes on to claim that "It is not a literary stratagem but a matter of fact that the particular experience of the new, albeit parallel textuality of reading hypertexts is somehow not reproducible in the old" (588), but the general differences in hypertextual writing and reading ("wreading") practices that he describes, signified as well with shifts in his own prose, are not obviously "new," and rereading as such can easily be named as inherent to language processing itself. Without a precise neurological map of cognitive functions, in fact, the irreducible difference of rereading hypertextually cannot be situated as a "matter of fact" at all. It is more compelling and accurate to argue, as he hints, that "differences show as differences are allowed" (587), differences which he locates in the practice of (re)reading hypertextually. Moreover, his emphasis on the uniterable, untranslatable "experience of this new textuality" highlights what for me is a crucial component of the performance of hypertext: the connection and interaction between the user-operator and the machinic-operator, both language processors, but of a different order.

  4. Within a different critical context, Mark Poster, although not over-invested in the idea of specifying an epistemological break, nevertheless suggests that the analog and the digital belong to fundamentally different material regimes of authorship and that the emergence of digital writing was anticipated by Foucault: in both are the author's presence and reference to a founding creator eliminated.[3] In Poster's analysis, books offer a technology of the analog because they reflect and reproduce the author. Moreover, technologies affect practices, and a shift in the material mode of inscription from paper to the computer thus elicits a re-articulation of the author-function (and, more widely for other critics, a re-articulation of the meaning of literacy).[4] But his more extensive claim holds that the differences between the analog and the digital can be delineated in terms of copyright and ownership, spatial fluidity, the materiality of the medium, and a shift in the trace (What's the Matter 78, 92-3, 100). With respect to the last, to argue for a shift in the trace is to say that with digitalization, the material form of language changes: electric language severs, in the last instance, reference to a phonetic alphabetic code that Poster reads as analog and not digital (81-2).[5] Alphabets, though, are themselves digital--Greek letters, for example, are units that do not bear resemblance to either sounds or things--and thus the binary Poster establishes begins to founder. So, how exactly has the material form of the trace changed and become destabilized in the transition from print to digital? How exactly can one register the difference between analog and digital through the material dimension of language or linguistic systems of reference?

  5. One great utopian promise of much hypertext criticism has been that the reader is in charge of ordering the information in front of her on the screen in a manner quantitatively and qualitatively distinct from the page and in a manner that constitutes authorship in its own right. Such a promise is worthy of further scrutiny not simply on the basis of its untenability, which Aarseth has exposed in his typology of cybertext by noting that reading and writing, using and developing, are spatially, temporally, physically, and epistemologically distinct activities. This illustrative promise of difference is worthy of further scrutiny, though, because it stations, receives, reads, and classifies the digital in terms of the analog. It in fact preconceives, preorchestrates, and preordains hypertext in terms of analog textuality. It traps the digital within the purview of print, and, without a mode to emerge on its own terms, digital textuality stands to be erased from its very beginnings.[6] Why, after all, should the digital exist in the world of the analog? Such an imprisonment dissolves its difference in the world of the analogical; it calls an end to the performance of its difference before it is permitted to announce itself as such. This is not a call, I should note, to exaggerate context: I have no particular stake in the historical bounding of digitization evinced by the claim that the category and mythology of the author in the modern period is bound to print technology. But it is to say that the quest to situate metaphysical difference and sameness alike--George Landow's vision of hypertext as performing, if not the literal death of the author then at least a literal evacuation, for example--cannot provide the terms we need to think about difference.[7] The problem is a difficult one, and it is not for nothing that hypertext theory often breaks down and dissolves into almost-impossible and nonsensical abstraction at the point at which it attempts to make clear distinctions between page and screen. Witness Mark Amerika on the experience and "being" of hypertext: "Rather, hypertextual consciousness will not have been a book (real or potential) due to its mediumistic discharge into the foundation of cyborgian life-forms whose 'archi-texture' is the deterritorialized domain we call virtual reality" (<http://www.grammatron.com/htc1.0/book.html>). A certain covering over of a conceptual gap is almost inevitably found within the claims for the special status of hypertext. The search for difference has produced valuable heuristics and compelling insights along the way, from Poster and Hayles (both following in part Friedrich Kittler's Discourse Networks, 1800/1900), and from Jay David Bolter's early analysis of the ever-alterable digital writing space, to Steven Johnson's more recent analysis of the empirical component of writing and the somatic adjustment to the machine. But the question of a theoretical difference, of a difference in kind and not in degree, is as yet unanswered.

  6. The problem of ontological difference can be initially displaced with an investigation of the ways in which hypertext fiction and hypermedia (primarily net art) have themselves handled the problem of their own difference, how they have imagined themselves as a distinctive form of textuality, precisely because they are strongly concerned with both theorizing and aestheticizing themselves, unlike primarily communicative and informational modes of writing (e.g., CNN.com). Digital textuality, or what I am calling hypertext, functions partly by creating itself as a discrete textual object, by referring to itself as itself.[8] Instances of the use of self-referentiality as such a stylistic and thematic marker are too numerous to catalogue in their entirety, but examples can be found in Matthew Miller's Trip ("No leads, no help, no future, no way.... We had no money, and almost no direction.... Better to know where you're going than to know where you are"); M. D. Coverley's "Fibonacci's Daughter" ("and you, dear reader, did you expect a map?"); Shelley Jackson's Eastgate novel, Patchwork Girl ("I can see only that part most immediately before me and have no sense of how that part relates to the rest"; "I sense a reluctance when I tow a frame forward into view.... I will show you the seductions of sequence and then I will let the aperture close"); Jane Yellowlees Douglas's Eastgate novel, I have said nothing ("He can't seem to get the narrative order of events quite right"); Linda Carroli and Josephine Wilson's water always writes in plural ("But I fear that waiting will be extinguished by the pursuit of pure speed, flat and undiscerning"); or, Judy Malloy's l0ve 0ne ("the room appeared to have no exits").

  7. In its tendency toward self-referentiality and self-ironicization, hypertext participates in the stylistic, linguistic, and formal games played out in what is variously categorized as the literature of chaos, meta-fiction, or postmodernity: Julio Cortazar's and Ana Castillo's chapter orderings in Hopscotch and The Mixquiahuala Letters, respectively; Donald Barthelme's interruption of Snow White with a questionnaire for reader-response; the novelistic fragments in Calvino's If on a winter's night a traveler; the problem of closure in Jessica Hagedorn's Dogeaters, and others; linguistic hybridity and fragmentation in Joyce's Ulysses and Finnegans Wake; the self-referentiality and attention to the mechanical process of narrative transmission in Art Spiegelman's Maus I and II; and the often-cited meta-criticism of Borges's fiction. Because hypertext emerges out of postmodern fiction and uses a similar set of symbols, it is unlikely that its allegorizing structure and systems of reference would be materially different. It is not simply that hypertext is inherently about itself in a postmodern or metafictional mode, however, but that it has constituted itself around the problem of its difference; self-referentiality is not just another or exchangeable move in the game, but a necessary move.[9]

  8. Katherine Hayles, following an unpublished MS by J. Yellowlees Douglas, similarly remarks upon the distinguishing rhetorical and formal properties of hypertexts, a category that she outlines so that it includes the media of print and the computer. She delineates hypertext in terms of three central components: "multiple reading paths; text that is chunked together in some way; and some kind of linking mechanism that connects the chunks together so as to create multiple reading paths" ("Transformation" 21).[10] Acknowledging that the distinction between print and electronic texts is not inviolate, she goes on to note that "the boundary is to be regarded as heuristic, operating not as a rigid barrier but a borderland inviting playful forays that test the limits of the form by modifying, enlarging, or transforming them" ("Print" 6).[11] Artists' books--one of her primary examples of texts that illustrate a formal connection between print and electronic hypertexts--and children's pop-up books are in fact able to stretch the medium of print to its limits, but they are not able to exploit the resources of language in the way that code is able to do. In contrast, Hayles comments on the "significant" differences in narrative between print and electronic texts with a claim that does not necessarily exclude print artists' books such as Tom Phillips's A Humument: A Treated Victorian Novel from the category of hypertext: "In electronic hypertext fiction, narrative takes shape as a network of possibilities rather than a preset sequence of texts" ("Transformation" 21). My argument here is that different modalities of textual performance must necessarily lead to the classification of print-precursors as precisely that: precursors and not hypertext per se. Digital textuality is able to achieve a spatial and temporal fluidity precisely because it is able to activate and manipulate the resources and complexity entrapped within language itself. Within analog text these spatial and temporal resources remain present, but only as potential and possibility.[12]

  9. My thesis thus proposes that hypertext must be conceived in terms of performance and that approaching the problem of a difference between the analog and the digital must be done in a mode through which digital textuality can emerge on its own terms.[13] To that end, this essay proposes a theory of practice for hypertext by articulating its form and aspect of performance, a performance that functions to separate the digital from the analog. To link hypertext writing to the play of performance is also to allude to the mechanism of high-performance computing, the linking of computers and computer networks for the purposes of performing complex tasks. It is also to speak of this writing as a map that produces its object rather than one that replicates pre-traced structures. Such a focus on writing and textuality need not evade, obscure, or evaporate the materiality and material substrate of the text, as Mark Hansen argues in Embodying Technesis. The materiality of the medium and the technological substrate, primarily the chip, cannot be over-emphasized, but it is important to remark as well that we are still dealing with texts whose materiality lies in the modality of its own structure and performance, in its code. In this sense, we do not have a hyper-discursivity, but a materiality of hypertext that itself cannot be fixed, especially insofar as there is no "tape" per se. Echoing Clement Greenburg's construal of modernist art, isolating the medium tends in the last instance both to revive the distinction between matter and information and to locate materiality and specificity in the physical components of the medium. However, the machinic component of the text cannot be disregarded or distilled: all texts are performative in some way but this does not mean that there is a not a significant change when the medium changes. As Anne-Marie Boisvert similarly notes, "in the reading of hypertext, the necessary, if not enforced relationship with the machine can't be long forgotten" ("Hypertext").

  10. Put more directly, both operator and machinic processor are crucial components of the performance of the system. The performance that encompasses user and the machinic system is an interactive one and to some degree collaborative. Further, the performance collapses processing and product, ends and means, input and output, within a system of "making" that is both complex and emergent.[14] My task in this article is thus to articulate a mode of understanding hypertext in terms of two components of performance: that of the user and that of the system. The latter suggests the processing done by the computer, which itself performs or is even performative, and the former suggests the performance of the user who operates as a functioning mechanism in the text, an idea whose genealogy includes performance art's situation and inclusion of the viewer within its boundaries, as well as the literary theorizations of the reader in terms of interaction, encounter, agonistic struggle, dialogue, and experience.[15] As Jim Rosenberg notes of the synergy of agent and the constructivism of code: "the code might act as a *coparticipant* in the constructive act... [but] one constructs with and against and amongst code" (qtd. in Calley, "Pressing").[16] In this sense, the interactivity of the viewer is a functioning instrument in the work. We can say, then, that the experience of digital textuality is different from that of analog. In that it bears a certain similarity to the temporal and empirical structures of performance art, digital textuality is itself a "happening."[17]

  11. The difference as such between hypertext and text, therefore, is not ontologically discernible and is locatable only in effect. Indeed, it is precisely that which cannot be revealed in the analog sense: its difference cannot be located in analog code, but only in digital. To conceive of this difference within the discursive frame of the analogical, in other words, is to frame it in terms under which it cannot emerge. The texts produced from HTML coding manuals--including those produced in other digital platforms and with other manuals and coding languages--are linked neither metaphysically nor ontologically, but through embedded codes of practice, codes that ascribe a certain relation among them on the basis of their performance. Hypertext optimally performs a different order of code, then, one that cannot be demonstrated metaphysically, but that can be analyzed in terms of complexity and emergence, that moment when the system programs and operates itself. Complexity appears as a discourse and occasional metaphor within hypertext criticism (e.g., in the rhetoric of dynamic systems, breakdowns, and so forth), but we need to move beyond this rhetoric and address complexity and emergence as paradoxically concrete. Complexity and emergence are not metaphors in my analysis but are instead scientific phenomena--aspects of hypertextuality and thus an inherent part of a logical system. Neither is quantifiable, which lends an even greater force to my locating them as non-locatable systemic components.

  12. In a complex system, the addition of discrete units does not equal the combined effect of the units; the sum is greater than the interactive parts. When discrete computers are linked to form a complex system, one cannot know in advance what the networked system will do. It is also impossible to predict in advance what the effects and significance of one alteration to the system will be. All one can know is that the system will be different. As John Holland, the inventor of genetic algorithms, notes of complex, generated systems: "The interactions between the parts are nonlinear; so the overall behavior cannot be obtained by summing the behaviors of the isolated components... more comes out than was put in" (225).[18] Emergent properties, however, produce recurrent and persistent patterns in generated systems, as with weather patterns (42-5, 225-31).[19] These properties and behaviors are internal to the system itself, and they are capable of producing auto-generative moments of self-organization, i.e., systemic states or systemic output that emerges without external input. "Evolutionary computation," or genetic or automatic programming, is the means by which this mode of artificial intelligence is achieved (Tenhaaf).[20] A recent Katherine Hayles article points the way toward articulating the relationship between performativity and complexity in terms of emergent behavior. In "Simulating Narratives: What Virtual Creatures Can Teach Us," she also reads Poster's manuscript on analog and digital textuality and transposes textuality into virtual realities. For Hayles, analogical relations are structured on a depth model; that is, the analogical requires links between the surface and depth units (13). For the analogical, complex codes produce a simple surface, and here we might think of the mythology of the Author that holds that a kind of complex interiority lends the text its depth. For the digital, on the other hand, a complex surface is produced by underlying simple models.

  13. There are moments, then, when a complex system formulates itself into an operating system, when the system becomes so complex as autotelically to run itself, or to program itself to solve problems. That a system whose future state is unpredictable and indeterminate until it actually emerges and comes into being should bear a certain connection to hypertext has been provisionally suggested by Hayles in a different context: "The actual narrative comes into existence (emerges globally) in conjunction with a specific reading" ("Artificial" 213). More apropos to my analysis, however, is her suggestion in the same article that a hypertext program is a "self-organizing system" capable of undergoing "spontaneous mutation" autotelically or collaboratively with other users (218). While she notes that print texts might require a similar syntactic organization, she also notes a difference in degree by extolling the "pay-off in redescribing spaces of encoding/decoding through the dynamics of self-organization [which] is obviously greater for electronic media rather than for printed words. When the words have lost their material bodies and become information, they move fast" (215). The difference in degree is reiterated in her claim that reader, technology, and text are all mutually and simultaneously constitutive "in a deeper, more interactive sense than is true of print texts" (214). The notion of self-organization, though, achieves its critical apotheosis in her analysis of "flickering signifiers" in How We Became Posthuman, wherein she articulates the differences in the material functioning and appearance of language:

    [Flickering signifiers are] characterized by their tendency toward unexpected metamorphoses, attenuations, and dispersions. Flickering signifiers signal an important shift in the plate tectonics of language.... When a text presents itself as a constantly refreshed image rather than a durable inscription, transformations can occur that would be unthinkable if matter or energy, rather than information patterns formed the primary basis for the systemic exchanges. (30)

    Such "metamorphoses" and "transformations" can be conceptually reprogrammed to include emergent behavior, which, like complexity, is a manifestation or quality of a system that cannot be thought of as summated as a whole or in terms of its component parts. It is that which cannot be fixed with any degree of totality, precision, or accuracy, that which cannot really be captured at all. Because it is not possible to locate the moment that brings together the computer units to produce something new, the quantum shift that changes the structure and system, complexity is itself not locatable. Nor can complexity be metaphysically demonstrated; it exists only in action, in performance, in terms of the influence of one component part over another. To remove its performance is to take away its difference; it is as if the computers were returned to their discrete units. That difference, the performance, is the trace, a moment in which hypertext itself performs. The operative difference of hypertext can only be revealed in the performing and tracing of itself, in its own instantiation. This, then, is the trace performance of hypertextuality--an argument in terms of performance rather than metaphysical difference.

    14. Jasper Johns, 'Flags' (1965)'
    Jasper Johns, Flags (1965)
    oil on canvas
    © Jasper Johns/Licensed by VAGA, New York, NY

  14. The exemplary illustrative device for digital practice is the anamorphic, a visual trick of perspective based on hidden codes and structures of signification. Apart from its theorizing by Lacan and Zizek, anamorphosis has been important for hypertext critics such as Aarseth because of its reliance on visual perception, labor, and the active production of the text. For my purposes, Jasper Johns's anamorphic painting Flags (1965) is the ideal visual emblem for the trace performance of hypertext.[21] Johns's painting of the two flags (one orange, black, and green, and the other monochromatic until the red, white, and blue colors are optically projected) is an illustration of the trace in that the anamorphic allows for meaning, the second flag, to appear. To perform Johns's painting and allow it--Flags--to emerge, one cannot hold both objects, both flags, in view simultaneously or analogically. Meaning happens in the exchange, but the exchange can never be fixed--it just happens. Meaning exists in the interplay between the two flags. As with Nam June Paik's multi-screen video installations, Johns moves into the realm of the untotalizable: neither a stable spectatorial position nor a fixed meaning is available. To fix on one image, one flag, one screen, one layer, is to exclude the others. Although the perspectival optics of the postmodern aesthetic require that the reader-viewer hold all of the various fragmented semiotic parts in her mind before she assembles them into a whole, we do not have a consciousness or mode of perception that would allow us to view the work as a complete whole. Flags is in fact a proto-hypertext, situated in a space between text and hypertext and gesturing toward a hypermedia effect. To say that hypertext is an effect is to name exactly the play that Johns knows: both flags cannot be held in the same moment of the sign. One flag must be there opaquely for the other to emerge; one flag cannot come into being without the other; one flag is marked only by losing the other. This is the performance aspect and modality of hypertext.[22] It cannot be denied that something different happens when we work with hypertext, but we cannot fix what that something is--it exists as effect, as the trace. To describe it verbally is to destroy its effect, again because it cannot be placed within the analogical, but only in the mode of its performance--its location, not locatable in the metaphysical sense, is thus under erasure. The nodes that follow in this article--Combinatorial Writing, An-anamorphosis, and Linking--will be a continued displaying and situating of this new aspect of performance in the digital terms of hypertext. However, given the temporal acceleration and mass diffusion of hypermedia production, notwithstanding the collection work performed by journals, meta-lists, and installations, my analysis of digital practice cannot claim to be totalizing, comprehensive, or even complete. True to my own thematic, such a clear picture of the state of digital textuality can only be an unrealizable fantasy.

    15. Node 2: Combinatorial Writing

    And so I spent whole days taking apart and putting back together my puzzle; I invented new rules for the game, I drew hundreds of patterns, in a square, a rhomboid, a star design; but some essential cards were always left out, and some superfluous ones were always there in the midst. The patterns became so complicated (they took on a third dimension, becoming cubes, polyhedrons) that I myself was lost in them.

    --Italo Calvino, The Castle of Crossed Destinies

  15. Combinatorial writing in a digital environment often involves the use of new technologies to literalize, make visible, or otherwise animate the themes and stylistic features of contemporary writing.[23] Perl scripting is a dominant mode of generating these texts practically and theoretically produced on the fly, and prominent examples include the cut-ups of Dadaism and William S. Burroughs, the permutational play of Raymond Queneau's Exercises in Style, a story in about one hundred variations, and Italo Calvino's tarot card literature machine in The Castle of Crossed Destinies, a text that partly informs Solitaire (Thorington), which is at once game, program, and story.[24] These literature machines are essentially machines for generating texts according to a pre-existing code or procedure. But what words, phrases, events, or combination of words, phrases, and events, is necessary for a story to emerge, or for the system to change? I emphasize "story" here because much permutational or combinatorial writing--after the early experimentation with animating and translating the work of those such as Burroughs, Tristan Tzara, and the Oulipo movement--has been in the mode of narrative. These machinic story programs, also framed as participatory, collaborative, and interactive, function by addition and accretion.[25] One enters the database, alters the material therein, and leaves behind a record of the visit in the form of an added phrase, an added word, an added twist in the narrative, or a completed and signed story. The interactive text network Assoziations-Blaster is one such example: "Anyone, including you, is allowed to contribute to the text database," it advertises, "so with your contribution you can help to build up a non-linear map of all things that exist." As a contributory visit to one of these sites can attest, language and narrative gaming in the particular context of hypertext fiction--assembling words and phrases into a whole--can be read as terroristic in the Lyotardian sense that every combination of phrases, every reading and writing, stands to cancel out other phrases and produce a kind of homogeneity, singularity, and univocality as a result (often puerile in content).[26] But such combinatorial language gaming also operates on the principle of complexity: it, too, is a system based on accrual whose principal is that of the network and whose outcomes are unknowable and not univocal or totalizable. The semiotic effects of addition, accretion, and linking in such a system will be unpredictably magnified.[27]

  16. When André Breton began l'écriture automatique, he could not have predicted what the outcome of his performance was to be.[28] Automatic writing is Roland Barthes's scriptorial project made visual, performative in that the text comes into being at the moment of its digital birth.[29] Michael Joyce draws a comparison similar to mine between hypertext and the performative: "Electronic texts present themselves in the medium of their dissolution. They are read where they are written; they are written as they are read" (Of Two Minds 235). While Aarseth takes issue with this mode of collapse of reader and writer by insisting on their ideological, epistemological, and geographic separation, the first part of Joyce's claim strikes the chord of performance, as well as of dissolution, destruction, and a failure of realization and completion. In these terms we can also understand the moment of the digital text's emergence--its coming into being at the moment of its performance. Such an understanding of language almost released from the subject has resonated strongly within hypertext criticism. Indeed, it is the condition of possibility for the argument in favor of the liberatory potential of hypertext, which is imagined to follow in the wake of Barthes's reading of the text as "that social space that leaves no language safe or untouched, that allows no enunciative subject to hold the position of judge, teacher, analyst, confessor, or decoder" (81).[30] Hypertext has been read as the fulfillment of the promise of contemporary critical theories of the death of the author, the network, the supplanting of the work by the text. Although skepticism about the rhetoric of the exemplum is necessary, the link between high theory and digital textuality is in fact already embedded in hypertext and media art, as it is Bill Seaman's Red Dice, which recasts Stéphane Mallarmé's "Dice Thrown Will Never Annul Chance"/ "Un coup de discussion jamais n'abolira le hasard" and combines techno-soundtrack, spoken text and images of old technology so as to produce and meditate upon the problematic of "new writing," "computer-mediated poetic construction."[31]

  17. The explanatory system of reference has thus necessarily expanded beyond and prior to hypertext criticism and theory, and digital textuality has been conceived as a "docuverse" (Nelson), montage (after Eisenstein), collage (Landow, Jameson), "an evolving virtual electronic collage" (Gaggi 138), recombination or "utopian plagiarism" (Critical Art Ensemble), assemblage (Talan Memmott), and as a "virtual graft" (Bill Seaman). Nearly all suggest the impossibility of synthesizing the parts into a complete and totalizable whole capable of being apprehended by the mode of perception and consciousness available to us now. However, the new media technologies have brought us to a point whereby collage is not simply a "feeble name" for the assemblage of discontinuous parts--as Jameson suggests in the context of his reading of Nam June Paik's video installations, which he uses as an illustrative example for the geometral optics of the postmodern aesthetic, practiced by viewers who try impossibly to "see all the screens at once, in their radical and random difference" (31).[32] Collage, also, is too material for a postmodern aesthetic and digital textuality alike. Complexity, in my analysis, is not a substitutive metaphor for collage but an inherent part of the system of hypertext itself. In this sense, it speaks to the liminal moment we inhabit between the consideration of hypertext as a genre, in terms of its formal and stylistic properties, and the consideration of new computer and scientific technologies and ideas, both as they are incorporated into electronic writing and as artifacts that themselves have effects and properties, such as autonomous behavior, that are inherent to the system of hypertext.

  18. John Cayley uncannily invokes the themes of performance and complexity with respect to compositional programming: "It points to an area of potential literature which is radically indeterminate (not simply the product of chance operations); which has some of the qualities of performance (without departing from the silence of reading)" ("Beyond" 183). Computer-generated and processed texts, for Cayley, allow for an innovative and even subversive departure from the standard node-link model of hypertextual composition--an escape into potentiality.[33] Cayley persuasively argues that the "digital instantiation" of his work makes for substantive, "non-trivial differences" between his text-generation procedures and those of Emmett Williams, Jackson Mac Low and John Cage, all of which achieve a relative fixity through print: "any aleatory or 'chance operation' aspect of such work is only fully realized in a publication medium which actually displays immediate results of the aleatory procedure(s). Such works should, theoretically, never be the same from one reading to the next (except by extraordinary chance)" ("Beyond" 173). The use of transformational or generative algorithms in his work results in texts that, in a significant sense, program and emerge from themselves. As he says of one component of Indra's Net, chance operations and the accrual of data input mean that "the procedure 'learns' new collocations and alters itself" (180).

  19. What is it about hypertext, then, that lends itself to the discussion of accrual, connectionism, the combinatorial, networks, patterns, the scriptorial, classification?[34] As Ted Nelson writes of "transclusion," the document per se is made up of the sum of parts materially located in different documents. The terrain of the document is marked by "transclusive quotation"--additive, inclusive content blocks assembled together and treated in the moment of reading as if they were isolate, closed or shut off from other, similar documents.[35] That is, hypertext works by connection, assemblage, and combination--by connecting content blocks, phrases, phrase regimes, nodes, computers, programs, and lines of code. It is not about signification but mapping: not ordering, tracing, and fixing, but transmission, relay, and movement. Revolutionary becoming, one of the great emancipatory promises of hypertext, has thus been bound to the combinatorial, to connection, variation, movement, and invention (Deleuze and Guattari 77, 106). It is not accidental or incidental that one of the operative concepts here--connectionism--is itself connected, as Paul Cilliers has shown, to a Saussurean concept of language, because connectionism as a paradigm for complex systems, like language for Saussure, functions in a relational mode, by the position of nodes in relation to other nodes, or signs in relation to other signs. Systems must have rules in order for patterns and significance to arise, and patterns can only be traced through the establishment of differences among the components of the system and the elimination of that which is "superfluous." A hypertext system, then, is paradigmatic; not all of the parts are necessary to the system, and the aesthetic whole can be sustained even through the destruction of a singular part (a node, perhaps) because the pattern rests with the code of production. Within a syntagmatic system, on the other hand, there are internally coherent but not necessarily linked patterns, and the removal of a singular part would affect the overall aesthetic pattern because the organic totality of the work would be entirely disrupted.[36] Hypertext does not adhere to a fixed, rule-based system; rather, it takes on the quality of disturbed, deferred, bifurcated movement. In that its performance is that of the trace, emphasizing not only the play of difference, but also open systems, feedback loops, a flattened network, links, and the interval between links, its dynamic is more différance than difference.

    20. Node 3: An-anamorphosis

    A text is a text only if it hides the law of its composition and the rule of its game from the first glance, from the newcomer. In any case a text remains an imperceptible text. The law and the rule do not dwell in the inaccessibility of a secret, put simply, they do not deliver themselves up to the present or to anything that could rigorously be called a perception.

    --Jacques Derrida, Plato's Pharmacy

  20. In the "Conclusion" to his book, Cybertext: Perspectives on Ergodic Literature, Espen Aarseth introduces the tension between the metamorphic (a text in which there is no final revelation or state of knowing) and the anamorphic (a text that presents an optical illusion as a vital hidden principle waiting to be discovered by a user that influences and produces the outcome, such as Jasper Johns's Flags, William Scrots's Edward VI, 3D pictures, and some ASCII art). Complex in its internal variations, Aarseth's typology of cybertext includes interactive fiction, synchronous and asynchronous chat settings, and print hypertexts such as the I Ching, all of which are distinguished by the work the reader must perform in order to "traverse" the text. Because the anamorphic forces the reader to perform the qualitative work of orienting the text, it occupies the space of hypertext. Within the anamorphic text, there are hidden codes: its structures of signification and perception are concealed and a certain perspective is required in order to make its form and content visible. The anamorphic is a visual trick of perspective, and its regimes of perception and imperception are optical. The mode has been an important one for much hypertext fiction and criticism because of its emphasis on visual perception, labor, reader-construction, and the production of the text. But the anamorphic does not really belong to the order of the code; it belongs to a slightly different episteme than that of the fractal, for example, which is essentially a digitized code, a code of information made visible.[37]

  21. The anamorphic, on the other hand, is a physical instantiation of what is done in the act of reading, searching for "the one right combination," the singular figure in the carpet, bringing it into focus, reducing complexity into perceptible patterns.[38] Reading, animating, and righting the perspective of an anamorphic text, bringing it into being--and this is not bound to a particular medium--involves not only going, doing, choosing (or "narrative drifting" as Mark Amerika puts it), but also searching and finding, structures of reading and navigation that replicate the structures of gaming.[39] Like Borges's novel-labyrinth The Garden of Forking Paths in the story of the same name--an important point of reference for Stuart Moulthrop and many first-generation hypertext authors and critics--all possible outcomes are imagined to be coded into the hypertext and traced by readers who can discern the structural logic of the system. This is the model of Borges's book, one of endless possibility, in which anything that can happen, does, and each possible plot outcome is pursued and multiplied into a seeming infinitude. Like knowing to go to the left in certain labyrinths, the key to reading the anamorphic text is meticulously to decode the system, to discern patterns and fault lines, and to attempt to bring the picture into focus.

  22. To read and to see is to attempt to impose a certain performance on the system; it is to engage with the system such that it performs and produces a coherent and legible output. In this sense, and in that it contains at least two mutually exclusive pictures and perspectival positions within its frame (e.g., Hans Holbein's painting The Ambassadors <http://www.artchive.com/artchive/H/holbein/ambassadors.jpg.html> contains a "correct" picture of the ambassadors or of the skull, but not both simultaneously), the anamorphic has a strong connection to Eduardo Kac's holopoetry. Kac identifies the primary formal quality of his holopoetry as "textual instability," "the condition according to which a text does not preserve a single visual structure in time as it is read by the viewer, producing different and transitory verbal configurations in response to the beholder's perceptual exploration" ("Holopoetry" 193).[40] Such visual and verbal instability, whereby "the linguistic ordering factor of surfaces is disregarded in favor of an irregular fluctuation of signs that can never be grasped at once by the reader," is achieved through what Kac terms the "fluid sign," which resembles the anamorphic in the description of its operation: "[A] fluid sign is perceptually relative.... [it is] essentially a verbal sign that changes its overall visual configuration in time, therefore escaping the constancy of meaning a printed sign would have" (193-4). The perceptual change for Kac, however, is one achieved through time rather than dimension. Holopoetry strives for temporal mutability, so it is not a true anamorphic, but Kac's theorizing of fluidity and the impossibility of a stable perceptual position does speak to the hypertextual process of construction, making, and interactive performance. With its dense textual and iconic layers and its abstraction of geometric, mathematic, lexical, and iconic arrangement, Talan Memmot's Lexia to Perplexia similarly invites such a performance. Lexia to Perplexia presents the reader-user with difficulty, entanglement, abstraction, confusion, unreadability, even obfuscation. Rather than moving into clarity and visual focus with each link, the text assemblage gains a greater opacity and density and moves from signal to noise. To achieve this effect, it utilizes punctuation that intrudes upon the word ("Exe.Termination"), embedded commands and command structures ("PER[(p)[L(EX)]]ia"), and a creolized, mechanized language characterized by syntactical and semantic errors. In that it moves from encryption to an even greater encryption, the central trope of the text assemblage is interference, the mechanism by which chaos is produced and the text paradoxically emerges. In this sense, the text partly thematizes decomposition, incompleteness, the gap, a mode of perception not yet achieved, the mechanical and operational failures of code, and digital texts that do not "work."

    23. Talan Memmott, 'Lexia to Perplexia'
    Talan Memmott, Lexia to Perplexia

  23. My central anamorphic text, Johns's Flags, itself fails within a digital environment. That is, the effects of Flags are not translatable to the screen, a setting where it does not work and cannot be brought correctly into focus.[41] This failure, though, is less illustrative of mechanical or material failure than it is of conceptual difference. When Johns's painting is projected in a digital environment, in other words, it produces a historical difference between the analog and the digital, a trajectory from the painterly to the hypertextual. Johns's anamorphic image extends beyond and above the physical limits of the painting and produces an illusory effect of depth and dimension. Depth in this instance is a trick of perception. That is, Johns's is already in some sense a flattened or postmodern anamorphosis, not the depth and dimensional model of the early modern anamorphic of Holbein, but a surface model that does not play with volume: not anamorphic, but an-anamorphic.

  24. An-anamorphosis--the digitized version of anamorphosis--paradoxically references the anamorphic but flattens out its volume.[42] It simultaneously succeeds and collapses, and it contains within its collapse the trace, remainder, ghost image, negation, and evacuation of the anamorphic. Anamorphosis is a matter of correcting or adjusting one's spectatorial position so as to locate a correct perspective. An-anamorphosis, on the other hand, presents us with a spectral image of collapsed depth--a smooth, flat, discrete surface rather than a modernist shattered surface that betrays an underlying depth.[43] The hidden code suggests a depth model but it remains a projection and illusion of depth. An-anamorphosis illuminates the operative codes of the new media in that it, like hypertext, does not, and indeed cannot, articulate a border or attain a perfect realization. Neither an-anamorphosis nor hypertext can be full or finished but are instead incomplete--an axiomatic principle for computers that alludes also to the condition of digital textuality. This claim does not suggest that one cannot make sense of a hypertext, but that hypertext makes the blocking of knowledge manifest in its embedding of a range of unknowable forms, blind passages, and visual aporias within the code, from the cracked screens of Moulthrop's Victory Garden, the blank screens of Jane Yellowlees Douglas's I have said nothing, the dead ends of Matthew Miller's Trip, through the undecipherable linguistic and iconic layers of Memmott's Lexia to Perplexia. The anamorphic, the figure in the carpet, the "law of its composition and the rule of its game," is offered and then withdrawn. The anamorphic does not just fail; it is withheld and sabotaged.

    25. Node 4: Linking

    Linkages are very, very quick, you know.

    --Jean-François Lyotard, "Links, The Unconscious, and the Sublime"

  25. Links and linkages suggest connections, signification, conversation, intertextuality, and even context, insofar as context, as Lyotard notes, "is the result of a series of linkages" (111). They range from ostentatious to demure, frenetic to sedate, unenclosed to hidden, anchored to ambient, animate to inanimate, associative to disassociative, random to ordered, and syntactic to paratactic. They leave behind the traces of their presence, their performance, recorded in history trackers (also a formal feature of Eastgate) and emblematized by the after-effects glow of fading pixels. In a felicitous link, meaning is constructed so that the next link might be submitted or followed, but links can be coded both to open up and to simulate the end of the play of signification.[44] As part of an early attempt to theorize the function, linguistic meaning, and philosophical import of the link, Stuart Moulthrop asks: "In what sense is a dynamically computed, implicit link analogous to turning a page?" ("Beyond Node/Link"). The answer is that it is analogous only up to a point, insofar as the page has historically been the organizational paradigm of codex. So, too, is the link the primary quality, device, mechanism, formal feature of hypertext, even as it operates in different modes: contextual (emergence or disappearance dependent on the state of the user or the system); "discovered structures"; "computational nodes"; and "virtual hypermedia."

  26. Eastgate's "guard fields," for example, allow the writer to create dynamic and conditional links, which are somewhat interactive in that they guide the reader and are dependent on prior user choices.[45] They allow the writer to assign priorities to the various links and thereby define and force paths and control the reader's access to the text. Because they preserve the architecture and structure of the text (with narrative as a specific consequential effect), guard fields often produce a sense of gaming, such that navigating becomes akin to finding hidden objects and surmounting obstacles so that a higher level might be reached.[46] Within more complex coding systems, however, the link opens up the potentiality of hypertext: it is one means by which it can be truly innovative and sever the ties both to the form of the page and the historical category of print literature. While many hyperfiction and hypermedia artists want to move beyond the node-link model, especially in light of the expansion of coding systems beyond the relative simplicity of HTML, Marjorie Luesebrink argues that "it is precisely the link (and the varieties thereof) that provides the most fertile ground for literary expression... the hypertext link enables the spatial and temporal aspects of multilinear electronic texts to function as an erasure of hierarchies.... Links have just begun to provide us with a vocabulary of new literary gesture and movement".[47] According to Luesebrink, the link traverses space and time and has its own syntax, which we are still in the process of creating and revealing. I also want to retain the node-link model within critical view because the link is both the mechanism for the performance of hypertext's difference and the means by which that difference is recognized.

  27. Form and content achieve a near-perfect suture in the first selection in the Eastgate Web Workshop: Judy Malloy's lOve One, a first-generation hypertext composed in the generic form of a diary, with linked entries that enforce both non-sequentiality and the illusion of sequentiality. Its genre, that is, allows for a kind of retroactivity whereby a doer may be placed behind the deed, and causal triggering mechanisms established. But its most compelling and emblematic structural and thematic feature is its links: Malloy uses images of cathode-ray tubes that refer literally to the text displayed on the screen. Akin to reversing a garment's seams, the cathode-ray tubes are a visible manifestation of the technological substrate of the text.[48] Cathode-ray tubes, in fact, literally convert the electronic to the visual in that they are animated by electrons that are first pulled in all directions in order to produce a concentrative beam that leaves behind a trace of its presence by marking the spot at which the electrons hit the screen. In this structural aspect, lOve One is in line with other hypertexts' foregrounding of the mechanism of their own performance, with code, directories, speed, waiting, trips, archives, paths, and bifurcations figuring as thematic and graphic elements. "Waiting will not be permitted to bring the nuanced possibilities of in-between," announces Water always writes in plural, thereby thematizing the link, and its own interlinked, interconnected, even inter-networked, parts and the necessity of both piecing together and reading the space between those parts in order to move forward. Such a textual suturing must necessarily involve textual haunting, a reappearance of the paths taken and not taken: "my parts will remember me," promises Patchwork Girl, not just about its re-created Frankensteinian monster, but also about itself.

  28. Terry Harpold comments upon the theme of navigational paths in his extended discussions of links and endings. A "hypertextual detour," he suggests, might be articulated as

    a turn around a place you never get to, where something drops away between the multiple paths you might follow.... Doing something with the hypertext link substitutes narrative closure for the dilatory space of the gap between the threads. It disavows the narrative turn, and fetishizes the link. I want to stress this point, because it seems to me that the instrumental function of the link exactly matches the psychoanalytic definition of the fetish object.... For the pervert, the missing phallus is still there: no lack, no gap, no cut. To read the link as purely a directional or associative structure is, I would argue, to miss--to disavow--the divisions between the threads in a hypertext. "Missing" the divisions is how the intentionality of hypertext navigation is realized: the directedness of the movement across the link constitutes a kind of defense against the spiraling turn that the link obscures. ("Threnody" 172-3, 81)[49]

    Harpold's psychoanalysis of electronic textuality draws a parallel between the pervert and the inefficient or even weak reader of hypertext: if the neurotic overrides or denies the effects of castration trauma by finding an object to fill in the lack, such a reader would perform as the pervert if she were to override, deny, or miss the gap in the link and try instead to substitute linear narrative. So Harpold suggests that we have to acknowledge that the ineradicable gap ("missingness") forces us into a different narrative strategy, and if we deny that difference and supply the narrative turn, then we use the link-as-fetish to fill in the lack.

  29. Extracted from his article, "The Contingencies of the Hypertext Link," "doing something" might be translated for the purposes of my thesis so that it becomes part of the context for performance, for my claim that links somehow always present us with the failure of realization signified by the older 404 messages. Linking, then, is both complex and a performance of complexity. Along with Harpold, Luesebrink argues that the link "represents a rupture in the ontological world." Finally, prefacing his own analysis of "breakdowns," Moulthrop glosses Harpold on the "deeply problematic nature of links" and the inevitability of their failure ("Pushing Back" 664). Despite his interest in the failure, brokenness, and incompleteness of the system, however, Moulthrop does allow for an ultimate and singular realization of the link, in that he will say that it does arrive, even if it fails to arrive at its intended or predicted destination: "only one possibility is realized, and likely as not it will not be what the reader anticipated" ("Pushing Back" 665). This moment of possibility, while conceived in breakdown, is simultaneously the spectacular moment of the birth of digital text, "the point of impact" ("Traveling").[50] The crash, then, is a productive one: out of the ashes of the wreck, "new order" may emerge ("Traveling").

  30. Links, and the phrases, nodes, words, icons, and images that are linked, realize their meaning in relation to each other. Thus there is no inherent, originary, final, totalizing meaning behind their ordering--meaning only comes once they are assembled. Meaning, in fact, is a reverberation of the effects of linking. This illustrates the trace performance of hypertext once again. The patterns and the system become not only complicated, but complex--that in which one not only gets lost, but also vertiginously loses stable and totalizing perception. The patterns they form are thus those of reiteration, recurrence, and frequency, on the one hand, and dissolution, disintegration, termination, on the other. During the time lag that occurs before a link is actualized, that interval or period of waiting while a page loads from the top or fills in an outline, it is usually possible to make out the text that is emerging, and yet one might get it wrong. In the moments of waiting, as one waits for speech to emerge through a stammer and wants to speak for, to fill the gap and complete the utterance, there is an implicit invitation for the link to be written for, to be written through, to be reloaded, to be completed. The condition of the link is such that it is not occasionally broken--it is always broken and almost anti-presence, high-speed network connections notwithstanding. Links, in other words, are not stable, set systems that can entirely emerge from themselves. It makes intuitive sense, then, that hypertext should contain gaps and ellipses apart from the link, even that ellipses themselves should function as links, and that it should never come to rest with a period, a note of finality, or a demarcation of the end.[51] Hypertext does not, and indeed cannot, articulate a border or attain a perfect realization. It is never full or finished. It is in fact incomplete.[52]

  31. Without a complete systems crash, hypertext by its very nature cannot come to rest partly because kinetic modality, or movement, with its emphasis on dynamism, process, fluidity, metamorphosis, transformation, is one of its defining formal properties. André Vallias, for example, writes of "continuous mutation" as the distinctive quality of digital media, derived from the progressivist movements of R&D and the haunting specter of an inevitable future obsolescence. However, Vallias's commentary on the "permanent process of making and remaking, of endless 'work in progress,'" of "instability" and "vertigo" is even more apropos with respect to my commentary on performance as the operative difference of hypertext (152).[53] That is, what the <entropy8zuper.org> designers call "the physical need for wonder and poetry" is not fixable or locatable in, but required and produced by, and in movement with, the digital object, particularly with the movement and motion facilitated by new media technologies and specific software programs like Flash [54]. Again, the hypertext system itself performs, but the human operator is another component of its performance. To return to my central visual examples: looking at a painting by Jackson Pollock is not absolutely different from looking at Jasper Johns's Flags, but Johns forces the viewer into a certain performance, into the fluctuation of the trace that his painting performs. One of Jasper Johns's flags must necessarily recede in order for the other to emerge, but the realization of the work can never be a perfect one, and in this respect it illustrates the way that hypertext code functions because hypertext itself cannot achieve a complete or finished realization. As a complex system, hypertext is internally inconsistent, and it gives a new resonance to Peter Lunenfeld's claim that electronic textuality is in a state of "unfinish" and Michael Joyce's claim that "electronic text can never be completed" (qtd. in Moulthrop, "Traveling"). In such a system, systematization itself is impossible.

    32. Department of English
    University of California, Santa Barbara
    raley@english.ucsb.edu

    Talk Back

    COPYRIGHT (c) 2001 RITA RALEY. READERS MAY USE PORTIONS OF THIS WORK IN ACCORDANCE WITH THE FAIR USE PROVISIONS OF U.S. COPYRIGHT LAW. IN ADDITION, SUBSCRIBERS AND MEMBERS OF SUBSCRIBED INSTITUTIONS MAY USE THE ENTIRE WORK FOR ANY INTERNAL NONCOMMERCIAL PURPOSE BUT, OTHER THAN ONE COPY SENT BY EMAIL, PRINT OR FAX TO ONE PERSON AT ANOTHER LOCATION FOR THAT INDIVIDUAL'S PERSONAL USE, DISTRIBUTION OF THIS ARTICLE OUTSIDE OF A SUBSCRIBED INSTITUTION WITHOUT EXPRESS WRITTEN PERMISSION FROM EITHER THE AUTHOR OR THE JOHNS HOPKINS UNIVERSITY PRESS IS EXPRESSLY FORBIDDEN.

    THIS ARTICLE AND OTHER CONTENTS OF THIS ISSUE ARE AVAILABLE FREE OF CHARGE UNTIL RELEASE OF THE NEXT ISSUE. A TEXT-ONLY ARCHIVE OF THE JOURNAL IS ALSO AVAILABLE FREE OF CHARGE. FOR FULL HYPERTEXT ACCESS TO BACK ISSUES, SEARCH UTILITIES, AND OTHER VALUABLE FEATURES, YOU OR YOUR INSTITUTION MAY SUBSCRIBE TO PROJECT MUSE, THE ON-LINE JOURNALS PROJECT OF THE JOHNS HOPKINS UNIVERSITY PRESS.

    Notes

    1. This piece could not have emerged without the maieutic aid of Russell Samolsky, who withstood and responded to more questions than I could possibly enumerate. Karen Steigman helped me to find valuable material, online and in print. Jennifer Jones and Timothy Wager read and commented on an earlier version, entitled, "How to Make Things With Words: Hypertext and Literary Value," which was delivered as a talk for the Transcriptions Colloquia in the English department at the University of California, Santa Barbara (<http://transcriptions.english.ucsb.edu/>) in May 2000. I am grateful to Alan Liu and Bill Warner for the invitation and the necessary inspiration. Students in my Electronic Literature & Culture (undergraduate; Spring 2000) and Hypertext Fiction & Theory (graduate; Winter 1999) classes at the University of Minnesota made a willing, enthusiastic, and provocative audience for parts of the thesis. Another section of the article was presented at the ACLA 2001, and I am grateful to Espen Aarseth and Mark Hansen for posing questions that re-oriented my thinking. Finally, Katherine Hayles's critical suggestions have been instrumental, and her latest NEH seminar, "Literature in Transition," provided the perfect environment to complete the revisions.

    2. See Vannevar Bush, "As We May Think," Atlantic Monthly (Jul. 1945); Ted Nelson, Literary Machines.

    3. Mark Poster, "What's the Matter with the Internet," UC Davis lecture, April 1998, published as the chapter "Authors Analogue and Digital" in his recent book, What's the Matter with the Internet? Poster's delineation of the digital text in this book echoes his earlier analysis of the "mode of information" in the book named as such, wherein he makes the following related claims: "Electronic communications are new language experiences in part by virtue of electrification. But how are they different from ordinary speech and writing? And what is the significance of this difference?" (Mode 1); "I do assert the emergence of a certain 'new'" (19); "After this point the natural, material limits of spoken and written language no longer hold" (74); "Digital encoding makes no attempt to represent or imitate and this is how it differs from analog encoding" (94), a claim that is a precursor to the discussion of waves and grooves in the Matter book; and, finally, "Electronic language, on the contrary, does not lend itself to being so framed. It is everywhere and nowhere, always and never. It is truly material/immaterial" (85).

    4. In spite of, and at times because of, the burgeoning academic fields devoted to computer-mediated communication, cyberculture, internet studies, and the digital humanities, there is still a commonly held assumption that digital literacy is incompatible with traditional literacy practices, and even that digital culture has trained students out of the market for print-bound literary texts in particular. In this sense, a facility for reading computer screens is imagined as both unrelated and opposed to knowing how to read a poem or novel.

    5. By contrast, Paul Levinson claims in general terms that the phonetic alphabet was the "first digital medium" in that the bits, the letters, correspond to sound and not things (Soft Edge 11-20).

    6. Michael Joyce suggests that digital textuality ultimately resists such entrapment: "Electronic text--the topographic, truly digital writing--even now resists attempts to wrestle it back into analogue or modify its shape into the shape of print. Its resistance is its malleability" (Two Minds 237).

    7. Poster's argument in "Authors Analogue and Digital" (What's the Matter with the Internet?)--that the digital is not just supplementing but replacing the analog--differs from Landow's in that he is not using Foucault to talk about obliterating the line between the reader and author; rather, he wants to uphold a set of distinctions between modes of authorship, though it is implicitly the case that his sense of digital authorship would lead to a reconfiguration of the role of the reader. Bringing together the technical conditions of authorship with the theoretical question of authorship (although Espen Aaresth would have problems with this linking) lends itself to a rethinking of the figure of the reader and it also implies a dichotomous or even substitutive relationship between reader and author.

    8. Jurgen Fauth takes issue with the tendency of hypertext to be endlessly self-reflexive (which often is, as Bolter notes of Afternoon, "an allegory of the act of reading" [qtd. in Fauth]). For another articulation of the pitfalls of hypertext's pervasive self-referentiality of hypertext, see Robert Kendall, "But I Know What I Like," SIGWEB Newsletter 8.2 (Jun. 1999); also posted at <http://www.wordcircuits.com/comment/htlit_5.htm>.

    9. Such a meta-critical and self-reflexive mode is acknowledged within much hypertext criticism, as when Michael Joyce notes that "most hypertext fictions include these self-reflexive passages" ("Nonce" 591). Also see Greg Ulmer's "Grammatology Hypermedia" on "reflexive structuration, by means of which a text shows what it is telling, does what it says, displays its own making, reflects its own action" (par. 7).

    10. For the same claim ("multiple reading paths; some kind of linking mechanism; and chunked text"), also see Hayles's "Print is Flat" 5.

    11. Hayles articulates an 8-point typology of hypertexts in digital form as follows: "Electronic Hypertexts are Dynamic Images; Electronic Hypertexts Include Both Analogue Resemblance and Digital Coding; Electronic Hypertexts Are Generated Through Fragmentation and Recombination; Electronic Hypertexts Have Depth and Operate in Three Dimensions; Electronic Hypertexts Are Mutable and Transformable; Electronic Hypertexts Are Spaces to Navigate; Electronic Hypertexts Are Written and Read in Distributed Cognitive Environments; Electronic Hypertexts Initiate and Demand Cyborg Reading Practices" ("Print" 7-8).

    12. As will become clear throughout the article, this claim is not structured so as to disavow Espen Aarseth's typology of cybertext as much as it is to offer the beginnings of a parallel and alternative typology.

    13. In a different use of the performance paradigm, game-designer Brenda Laurel has employed theatrical metaphors in order to understand human-computer interaction and suggested that the ideal interface should correspond with the theatre. For Laurel, Aristotle's Poetics is the best functional metaphor for computer art, and a program needs to be thought of as a performance for the user, one that masks the technical component of production and its artificiality. The power of the theatrical metaphor for the interface is that the theatre is "fuzzy" and thereby necessarily involves repetition with variation and difference (23).

    14. Timothy Allen Jackson also comments on the new media aesthetic in terms of process and the dynamic quality of the system: "Such an aesthetic is projective rather than reflective, complex and dynamic rather than simple and static, often focusing on process more than product, and resembling a verb more than a noun" (348).

    15. For example, the theme of performance in relation to the reader could be linked to literary theories of reading proposed by Ingarden ("encounter"), Bakhtin ("dialogue"), Jauss ("convergence"), and Iser ("interaction").

    16. In a different forum, Rosenberg argues against Nick Montfort's review of Aarseth's Cybertext and subsequent alignment of hypertext with Chomsky's hierarchy of grammars (finite automata) by noting that one "simply cannot conclude that hypertext 'is' a finite state machine," whether or not it maintains a strict node-link model ("Positioning"). Montfort gives a primacy to the computational power of cybertext machines, which he distinguishes on the basis of their ability to calculate.

    17. On temporality and the digital aesthetic, see Marlena Corcoran, "Digital Transformations of Time: The Aesthetics of the Internet," Leonardo 29.5 (1995): 375-378. On anti- or non-object art, see John Perreault, "Introduction," TriQuarterly 32 (Winter 1975): 1-5; and Tommaso Trini, "Intervista con Ian Wilson/Ian Wilson, An Interview" Data 1.1 (Sep. 1971): 32-4. Conceptual art also separates out product--"art"--from material substance. For critical commentary on conceptual or dematerialized art, see Alexander Alberro and Patricia Norvell, eds., Recording Conceptual Art (Berkeley: U of California P, 2001); Alexander Alberro and Blake Stimson, eds., Conceptual Art: A Critical Anthology (Cambridge: MIT Press, 1999); Lucy Lippard, Six Years (Berkeley: U of California P, 1973); Tony Godfrey, Conceptual Art (London: Phaidon, 1998).

    18. Also see Holland, Hidden Order:How Adaptation Builds Complexity (Cambridge: Perseus Books, 1995). On complexity, see Paul Cilliers; Brian H. Kaye, Chaos and Complexity (New York: John Wiley & Sons, 1993); Roger Lewin, Complexity: Life at the Edge of Chaos (New York: Macmillan, 1992); and M. Mitchell Waldrop, Complexity: The Emerging Science At the Edge of Order and Chaos (New York: Simon and Schuster, 1992). The analysis of complex systems is sustained even through management theory: "the whole shows behaviours which cannot be gleaned by examining the parts alone. The interactions between the parts are crucial, and produce phenomena such as self-organization and adaptation" (McKergow 721-22). Mark C. Taylor's forthcoming book, The Moment of Complexity: Emerging Network Culture (Chicago: U of Chicago P, 2002) speaks to my concern with emergence as a component and effect of the hypertextual system, but it arrived too late to be incorporated within this article.

    19. My discussion of patterns vis-à-vis hypertext refers in a general sense to systems theory, a topic that has been explored by theorists of science and literature such as Hayles, Cilliers, William Paulson, and John Johnston. On systems theory, see What is Systems Theory?, <http://pespmc1.vub.ac.be/SYSTHEOR.html>; Cybernetics, Systems Theory and Complexity, <http://www.cudenver.edu/~mryder/itc_data/complexity.html>; A Curriculum for Cybernetics and Systems Theory, <http://www.well.com/user/abs/curriculum.html>; and John Gowan's General Systems HomePage, <http://www.people.cornell.edu/pages/jag8/>. On chaotic systems and chaos, see James Gleick, Chaos and Chaotic Systems, <http://dept.physics.upenn.edu/courses/gladney/mathphys/
    subsection3_2_5.html    >. Pamela Jennings touches on intedeterminacy, chaos theory, fuzzy logic, and open structures in relation to new media in "Narrative Structures for New Media: Towards a New Definition," Leonardo 29.5 (1996): 345-50.

    20. Also see <http://www.genetic-programming.org>, as well as Genetic Programming, Inc. <http://www.genetic-programming.com>.

    21. Jasper Johns, Flags is online at <http://www.walkerart.org/resources/res_pc_johns2.html>. For historical and technical analyses of the device of the anamorphic, see Jurgis Baltrusaitis, Anamorphic Art, trans. W. J. Strachan (Cambridge: Chadwyck-Healey, 1977) and Fred Leeman, Hidden Images: Games of Perception, Anamorphic Art, Illusion (New York: Harry N. Abrams, 1976).

    22. In the catalog for Jack Burnham's "Software" art exhibition at the Jewish Museum (1970), Ted Nelson also conceives of hypertext from its conceptual genesis as "writing that can branch or perform." See "The Crafting of Media," <http://www.sfc.keio.ac.jp/~ted/TN/PUBS/CraftMedia.html>.

    23. The phrase "combinatory literature" can be traced to François Le Lionnais's afterword to Raymond Queneau's Cent Mille Milliards de poèmes (Paris: Gallimard, 1961). It is later developed by Umberto Eco with respect to narrative forms, and by Oulipians such as Harry Mathews.

    24. The mathematical formulas of the Oulipo movement have been an influential genealogical and stylistic precursor for hypertext, particularly since some Oulipians were themselves inspired by programming and had already begun the work of using computers as an tool in literary production, and early hypertext work frequently cited and even programmed their work so as further to "animate" and apply their algorithms, e.g., Permutationen <http://userpage.fu-berlin.de/~cantsin/permutations/index.cgi>. The work of the French literary group, Oulipo, of which Queneau and Calvino were members, based its poetics on permutational possibility and procedure. In that both tend on occasion to work with elemental units--the Reader (addressed) and reader (actual) for Calvino's If on a winter's night a traveler, for example, or lipograms, or the famous bricks of Carl Andre--Oulipo could be profitably linked to the so-termed Minimalist artists that came to prominence in the 1960s and 1970s, particularly Andre, Sol LeWitt, Donald Judd, and Robert Morris. Also see Stéphane Susana, "A Roundup of Constrained Writing on the Web," ebr 10 (Winter 1999/2000) <http://www.altx.com/ebr/ebr10/10sus.htm>. For an English-language collection of their work, see Warren Motte, ed., Oulipo: A Primer of Potential Literature (Lincoln, NE: U of Nebraska P, 1986). On Calvino and the combinatorial, see Jerry A. Varsava, "Calvino's Combinative Aesthetics: Theory and Practice," Review of Contemporary Fiction 6.2 (Summer 1986): 11-18.

    25. In the larger project, I pursue the topic of computer and computer-generated poetry at greater length. After Rosenberg, see Philippe Bootz, "Poetic Machinations," Visible Language 30.2 (1996): 118-37; and Charles Hartman's memoir, Virtual Muse: Experiments in Computer Poetry (Hanover: Wesleyan UP, 1996), with a downloadable Mac platform program at <http://camel2.conncoll.edu/academics/departments/
    english/cohar/programs/    > With respect to the analogy between poetry and technology, Carrie Noland, in her genealogy from the nineteenth-century avant-garde French lyric poets to U.S. performance poetries (particularly those of Laurie Anderson and Patti Smith), wonders whether the "technopoems of the future" will embrace technology to the extent that they risk "losing the integrity of poetry as a language-based and voice-generated genre" (216, 15). Also see Rosenberg on Balpe generator poetry and the French poésie animée school (Jean-Pierre Balpe, Trois mythologies et un poète aveugle, an installation featuring poem-generating and music-generating robots [Centre Georges-Pompidou 1997]).

    26. Koskimaa also comments upon the intimate relation between reading and gaming: "the aspect of mastering a computer environment is an essential part of the hyperfiction reading experience, an aspect common with playing computer games."

    27. In reference to the collaborative "International Internet Chain Art Project," Gaggi makes a related point concerning unpredictability, although his argumentative focus is the radically reconfigured notion of authorship that results from the critical and practical challenge of the ideas of individuality, autonomy, and genius: "Each participant alters, adds to, or comments on whatever he or she receives. Under such circumstances, surprise--as well as disappointment--is always possible. Although contributions are matters of conscious decisions individuals make--except when technical problems result in 'corruption'--those who create an image or text cannot predict or control what will happen to it" (139).

    28. For a discussion of the principle of "psychic automatism," Breton and Philippe Soupault's Les Champs magnétiques (1919), and Breton's Manifeste (1924), see Elza Adamowiccz, Surrealist collage in text and image: Dissecting the exquisite corpse (Cambridge: Cambridge UP, 1998) 5-10. It was not just Surrealism but Fluxism as well that maintained an interest in automatic writing. Jackson Pollock's painting would be another instance of the scriptor at work.

    29. As Russell Samolsky notes, the felicity or "efficacy of any performative is always, in some sense, the death of the author," a notion whose literal effects he demonstrates in the example of Kafka's relationship to the Holocaust (191).

    30. On the liberatory argument, see, as just one of many examples, Gaggi 103-5.

    31. Also, Noah Wardrip-Fruin links new media particularly to Deleuze and Guattari, A Thousand Plateaus (371). Seaman's Red Dice was exhibited at the Art Gallery of New South Wales (3-26 Mar. 2000) as part of the Telstra Adelaide Festival 2000. See <http://www.camtech.net.au/cacsa/program/2000.html>.

    32. Lacan treats "geometral" or "flat" perspective and the commingling of art and science in his reading of anamorphosis and the gaze; see The Seminar of Jacques Lacan: Book XI, trans. Alan Sheridan (New York: W.W. Norton, 1977).

    33. In the Montreal electronic art magazine, CIAC, Sylvie Parent provides another typology for electronic literature based on the variant processing of text by the computer: collaborative texts, text's new space, text generating programs, language atomization, and moving text. See <http://www.ciac.ca/magazine/archives/no_9/en/pers-intro.html>.

    34. On connectionism, see Cilliers, especially 25-47.

    35. Ted Nelson, Literary Machines 93.1 (1993 preface, unnumbered page 5).

    36. See Peter Bürger's schema for the organic and the non-organic artwork in Theory of the Avant-Garde, trans. Michael Shaw (Minneapolis: U of Minnesota P, 1984): 80.

    37. In a way this is a modern vs. postmodern distinction, but one does not have to choose one over the other.

    38. See Henry James, "The Figure in the Carpet" 381; and see also the passage that introduces the figure itself: "the thing we were all so blank about was vividly there. It was something like a complex figure in a Persian carpet" (372).

    39. With hypertext, "readership has been restored but not transcended" (Aarseth 94).

    40. Also see Kac, "Key Concepts of Holopoetry," ebr 5 (Spring 1997) <http://www.altx.com/ebr/ebr5/kac.htm> and his website, with examples of his work, at <http://www.ekac.org/>.

    41. I am grateful to Richard Helgerson for asking a variation of the question that produced this line of argument: does it matter that the Jasper Johns is a modern text not meant for the screen?

    42. My nomenclature here might appear to connect to the bilingual, Flash-intensive "electronic and interactive journal that examines the human condition in the digital age"--Chair et Métal / Metal and Flesh: The Digital Anamorphosis of the Universe <http://www.artsci.lsu.edu/fai/metal.html>--but the journal's subtitle does not have an immediately obvious significance. Rather, the concept of the an-anamorphic arose in a conversation with Russell Samolsky; from this collaborative moment, the concept and Jasper Johns as an illustrative example came into being.

    43. Martin Jay comments on anamorphic vision, which "helps us understand the complexities of a visual register which is not planimetric but which has all these complicated scenes that are not reducible to any one coherent space" (qtd. in Foster 84). Planimetry deals with the measurement of surfaces, so complexity of vision here has to do with volume. The postmodern anamorphic or an-anamorphic, however, puts us back in the register or space of the planimetric.

    44. For a discussion of linking in a hypertext environment, see George P. Landow, Hypertext 2.0 (Baltimore: Johns Hopkins UP, 1997) 11-20; Harpold, "Threnody" and "The Contingencies of the Hypertext Link"; and "Conclusions," Hyper/Text/Theory, ed. George Landow (Baltimore: Johns Hopkins UP, 1994): 189-222. Also, the hypertext author Jeff Parker is currently at work on a formal analysis of the "poetics of the link."

    45. See Mark Bernstein on "volatile" hypertext, "Architectures for Volatile Hypertext," Hypertext '91 Proceedings (Baltimore: ACM, 1991): 243-60; robin, "Hypertext for Writers: A Review of Software," EJournal 3.3 (Nov. 1993): <http://www.hanover.edu/philos/ejournal/archive/ej-3-3.txt>; the entry for "conditional links" in the alt.hypertext FAQ; Bolter on guard fields in Writing Space; J. Yellowlees Douglas on the seemingly contradictory harmony between varied and multiple links and reader expections in "Wandering Through the Labyrinth: Encountering Interactive Fiction," Computers and Composition 6.3 (Aug. 1989): 93-101; Robert Coover on the differences among links in FEED <http://www.feedmag.com/html/document/98.02nelson/coover1.html>; and Robert Kendall, "Hypertextual Dynamics in A Life Set for Two," Proceedings of the Seventh ACM Conference on Hypertext, <http://www.wordcircuits.com/kendall/essays/ht96.htm>. It is precisely guard fields that lead Raine Koskimaa to argue that the bricoleur metaphor so present within hypertext criticism is illegitimate vis-à-vis the Eastgate texts, primarily because the texts withhold information about their own structures. Without clear maps, Koskimaa suggests, we cannot properly speak of bricolage (VII). Finally, it is important to note that Storyspace, Flash, related authoring platforms, and HTML itself all to some degree limit the behavior of the hypertext system, but this is not to say that the system is either deterministic, with standardized and predictable behavior, or incapable of autonomous behavior.

    46. With software-produced and -mediated hypertexts, a variety of links and a kind of structure are possible, while with net-based hypertexts, there is presumably a lesser degree of structure, and links need not proceed from window to window, or lexia to lexia. See the description of the Storyspace software in the alt-hypertext FAQ. For a comprehensive student project on guard fields, see Loran Gutt, "Hypertext," <http://www.princeton.edu/~lzgutt/hypertext/paper3/node2.html>.

    47. See, for example, Stuart Moulthrop, "Beyond Node/Link," a node in The Shadow of an Informand: An Experiment in Hypertext Rhetoric; John Cayley's "Beyond Codexspace: Potentialities of Literary Cybertext" and Jim Rosenberg's "Structure of Hypertext Activity." Calls to augment and restructure the node-link paradigm were made by Frank Halacz in his keynote addresses at the Hypertext '89 and Hypertext '91 conferences <http://www.parc.xerox.com/spl/projects/halasz-keynote/review/>. Also see Steven J. DeRose, "Expanding the Notion of Links," Hypertext '89 Proceedings, ed. Norman Meyrowitz (New York: Association for Computing Machinery, 1989) 249-59; and H. Van Dyke Parunak, "Don't Link Me In: Set Based Hypermedia for Taxonomic Reasoning," Proceedings of Hypertext '91 (New York: Association for Computing Machinery, 1991) 233-42.

    48. Also see Ted Nelson on the cathode-ray tube (Dream Machines 84-5).

    49. Also see Adrianne Wortzel on the openness and dynamism of links, which she theorizes as the spaces that "allow chance, desire, and drive to drift unbound" (362).

    50. David Miall briefly argues that Moulthrop's claims for the difference of hypertext, notably the qualities of multiplicity and breakdown, are not tenable as such and that the particularities Moultrop ascribes to hypertextuality "replicate something we have always known about the form of literary texts."

    51. J. Yellowlees Douglas remarks briefly hypertext's presentation of "discrete pieces of information and ellipses" in "Wandering Through the Labyrinth: Encountering Interactive Fiction," Computers and Composition 6.3 (Aug. 1989): 93-101, <http://www.cwrl.utexas.edu/~ccjrnl/Archives/v6/6_3_html/
    6_3_6_Douglas.html    >.

    52. The allusion to Gödel's theorem with respect to hypertextual incompleteness is just that, an allusion, rather than an application.

    53. There is a conjunction between the two in that Vallias belongs to one branch of artistic production that has moved toward movement as both concept and integrative device, a "movement" to create what E. M. de Melo e Castro terms "videopoetry," which has been flourishing in Brazilian and Portuguese experimental poetry circles and shares the formal interests of the various North American experimental poets who work with software applications such as Flash to create poems experienced spatially, temporally, visually, aurally, and linguistically. See, for example, Thom Swiss, "Genius" <http://www.differenceofone.com/genius/genius.html> Similarly, Ladislao Pablo Györi's brief manifesto on "Virtual Poetry" calls for the production of poems "experienced by means of partially or fully immersive interface devices" (Visible Language 30.2 [1996]). The manifesto and examples of his work can be found at <http://megahertz.njit.edu/~cfunk/gyori.html>.

    54. Now authoring-design software applications such as Flash and QuickTime have taken design to the point whereby the text objects may be watched as film and media installations, e.g., the digital poems and works from Born Magazine <http://www.bornmag.com/> and the curated site Poems That Go <http://www.poemsthatgo.com/>, both of which rely almost exclusively on Flash.

    Works Cited

    Aarseth, Espen. Cybertext: Perspectives on Ergodic Literature. Baltimore: Johns Hopkins UP, 1997.

    Barthes, Roland. "From Work to Text." Image-Music-Text. Trans. Stephen Heath. New York: Hill and Wang, 1977. 155-64.

    Boisvert, Anne-Marie. "Cybertext." CIAC (2001). <http://www.ciac.ca/magazine/>.

    ---. "Hypertext: Feature on Electronic Literature." CIAC 9 (Dec. 1999). <http://www.ciac.ca/magazine/archives/no_9/en/dossier.html>.

    Bolter, Jay David. Writing Space: The Computer, Hypertext, and the History of Writing. New York: Lawrence Erlbaum, 1990.

    Borges, Jorge Luis. Ficciones. Trans. Anthony Kerrigan. New York: Grove Press, 1962.

    Calvino, Italo. The Castle of Crossed Destinies. Trans. William Weaver. New York: Vintage, 1997.

    Cayley, John. "Beyond Codexspace: Potentialities of Literary Cybertext." Visible Language 30.2 (1996): 164-83.

    ---. "Pressing the 'Reveal Code' Key." EJournal 6.1 (Mar. 1996). <http://www.hanover.edu/philos/ejournal/archive/ej-6-1.txt>.

    Cilliers, Paul. Complexity and Postmodernism. New York: Routledge, 1998.

    Deleuze, Gilles, and Felix Guattari. A Thousand Plateaus. Trans. Brian Massumi. Minneapolis: Minnesota UP, 1980.

    Derrida, Jacques. "Plato's Pharmacy." Dissemination. Trans. Barbara Johnson. Chicago: U of Chicago P, 1981.

    Douglas, Jane Yellowlees. I have said nothing. Eastgate Systems. <http://www.eastgate.com>

    Druckrey, Timothy, ed. Electronic Culture: Technology and Visual Representation. New York: Aperture, 1996.

    Fauth, Jurgen. "Poles in Your Face: The Promises and Pitfalls of Hyperfiction." Mississippi Review 1.6 (Sep. 1995). <http://orca.st.usm.edu/mrw/1995/06-jurge.html>.

    Foster, Hal, ed. Vision and Visuality. New York: New Press, 1988.

    Gaggi, Silvio. From Text to Hypertext: Decentering the Subject in Fiction, Film, the Visual Arts, and Electronic Media. Philadelphia: U of Pennsylvania P, 1997.

    Galloway, Alex. "Perl is My Medium--An Interview with Lisa Jevbratt." Rhizome.org 2 Feb. 2001. <http://rhizome.org/object.rhiz?2218>.

    Gleick, James. Chaos: Making a New Science. New York: Penguin Books, 1987.

    Gödel, Kurt. On Formally Undecidable Propositions of Principia Mathematica and Related Systems. Trans. B. Meltzer. New York: Dover Publications, 1992.

    Hansen, Mark. Embodying Technesis: Technology Beyond Writing. Ann Arbor: U of Michigan P, 2000.

    Harpold, Terry. "The Contingencies of the Hypertext Link." Writing on the Edge 2.2 (Spring 1991): 126-37.

    ---. "Threnody: Psychoanalytic Digressions on the Subject of Hypertexts." Hypermedia and Literary Studies. Ed. Paul Delaney and George Landow. Cambridge: MIT Press, 1991. 171-181.

    Hartman, Charles O. Virtual Muse: Experiments in Computer Poetry. Hanover, NH: Wesleyan UP, 1996.

    Hayles, N. Katherine. "Artificial Life and Literary Culture." Cyberspace Textuality. Ed. Marie-Laure Ryan. 205-23.

    ---. How We Became Posthuman: Virtual Bodies in Cybernetics, Literature, and Informatics. Chicago: U of Chicago P, 1999.

    ---. "Print is Flat, Code is Deep: The Importance of Media-Specific Analysis." Forthcoming in Poetics Today (Fall 2001). Unpublished MS 27 pp.

    ---. "Simulating Narratives: What Virtual Creatures Can Teach Us." Critical Inquiry 26.1 (Autumn 1999): 1-26.

    ---. "The Transformation of Narrative and the Materiality of Hypertext." Narrative 9.1 (Jan. 2001): 21-39.

    Holland, John H. Emergence: From Chaos to Order. Cambridge: Perseus Books, 1998.

    Jackson, Shelley. Patchwork Girl. Eastgate Systems. <http://www.eastgate.com>

    Jackson, Timothy Allen. "Towards a New Media Aesthetic." Reading Digital Culture. Ed. David Trend. Oxford: Blackwell, 2001. 347-53.

    James, Henry. "The Figure in the Carpet." The Figure in the Carpet and Other Stories. Ed. Frank Kermode. New York: Penguin, 1986. 357-400.

    Jameson, Fredric. Postmodernism; or, the Cultural Logic of Late Capitalism. London: Verso, 1996.

    Johnson, Steven. Interface Culture: How New Technology Transforms the Way We Create and Communicate. San Francisco: Harper, 1997.

    Joyce, Michael. "Nonce Upon Some Times: Rereading Hypertext Fiction." Modern Fiction Studies 43.3 (Fall 1997): 579-97.

    ---. Of Two Minds: Hypertext Pedagogy and Poetics. Ann Arbor: U of Michigan P, 1995.

    ---. Othermindedness. Ann Arbor: U of Michigan P, 2001.

    Kac, Eduardo. "Holopoetry." Visible Language 30.2 (1996): 184-213.

    Kellert, Stephen H. In the Wake of Chaos: Unpredictable Order in Dynamical Systems. Chicago: U of Chicago P, 1993.

    Koskimaa, Raine. "Visual Structuring of Hypertext Narratives." ebr 6 (Winter 1997/1998). <http://www.altx.com/ebr/ebr6/6koskimaa/6koski.htm>.

    Landow, George. "Hypertext as Collage-Writing." Lunenfeld 151-170.

    Laurel, Brenda. Computers as Theatre. Reading, MA: Addison-Wesley Publishing Company, 1992.

    Levinson, Paul. The Soft Edge: A Natural History and Future of the Information Revolution. New York: Routledge, 1997.

    Luesebrink, Marjorie. "Of Tea Cozy and Link." ebr 11 (Winter 2000/2001). <http://www.altx.com/ebr/riposte/rip11/rip11cov.htm>.

    Lunenfeld, Peter, ed. The Digital Dialectic: New Essays on New Media. Cambridge: MIT Press, 1999.

    Lyotard, Jean-François. "Links, The Unconscious, and the Sublime." Ellipsis 1.1 (Spring 1990): 109-129.

    McKergow, Mark. "Complexity Science and Management: What's in it for Business?" Long Range Planning 29.5 (1996): 721-722.

    Merrell, Floyd. Simplicity and Complexity: Pondering Literature, Science, and Painting. Ann Arbor: U of Michigan P, 1998.

    Miall, David. "Moulthrop: celebrating multiplicity and breakdown" (7 Aug. 1999). <http://www.arts.ualberta.ca/~dmiall/hyperead/moulthp1.htm>.

    Miller, J. Hillis. Fiction and Repetition: Seven English Novels. Cambridge: Harvard UP, 1985.

    Montfort, Nick. "Cybertext Killed the Hypertext Star." Rev. of Cybertext: Perspectives on Ergodic Literature, by Espen J. Aarseth. ebr 11 (Winter 2000/2001). <http://www.altx.com/ebr/ebr11/11mon/>.

    Morris, Robert. Continuous Project Altered Daily. Cambridge: MIT Press, 1993.

    Motte, Warren F., Jr., ed. Oulipo: A Primer of Potential Literature. Trans. Warren F. Motte, Jr. Lincoln, NE: U of Nebraska P, 1986.

    Moulthrop, Stuart. "Beyond Node/Link." <http://raven.ubalt.edu/staff/moulthrop/hypertexts/hoptext/
    Beyond_Node-Link30978.html    >. WWW version 1994.

    ---. "Pushing Back: Living and Writing in Broken Space." Modern Fiction Studies 43.3 (Fall 1997): 651-74.

    ---. The Shadow of an Informand: An Experiment in Hypertext Rhetoric. <http://raven.ubalt.edu/staff/moulthrop/hypertexts/hoptext/>. WWW version 1994.

    ---. "Traveling in the Breakdown Lane: A Principle of Resistance for Hypertext." Mosaic 28 (Dec. 1995): 55-77. <http://www.ubalt.edu/ygcla/sam/essays/breakdown.html>.

    ---. Victory Garden. Eastgate Systems. <http://www.eastgate.com>

    Murray, Janet. Hamlet on the Holodeck: The Future of Narrative in Cyberspace. New York: The Free Press, 1997.

    Nelson, Theodor Holm. Computer Lib. Tempus Books of Microsoft Press, 1974. Revised and updated with Dream Machines on the flip side, 1987.

    ---. Literary Machines. Ted Nelson, 1981. Edition 93.1, 1992.

    Noland, Carrie. Poetry at Stake: Lyric Aesthetics and the Challenge of Technology. Princeton: Princeton UP, 1999.

    Poster, Mark. The Mode of Information: Poststructuralism and Social Context. Chicago: U of Chicago P, 1990.

    ---. What's the Matter with the Internet? Minneapolis: U of Minnesota P, 2001.

    Rosenberg, Jim. "Positioning Hypertext in Chomsky's Hierarchy of Grammars." ebr 11 (Winter 2000/2001). <http://www.altx.com/ebr/riposte/rip11/rip11rose.htm>.

    ---. "The Structure of Hypertext Activity." <http://www.cs.unc.edu/~barman/HT96/P17/SHA_out.html>.

    Ryan, Marie-Laure, ed. Cyberspace Textuality: Computer Technology and Literary Theory. Bloomington: Indiana UP, 1999.

    Samolsky, Russell. "Metaleptic Machines: Kafka, Kabbalah, Shoah," Modern Judaism 19.2 (1999).

    Tenhaaf, Nell. "As Art Is Lifelike: Evolution, Art, and the Readymade." Leonardo 31.5 (1998): 397-404.

    Ulmer, Greg. "Grammatology Hypermedia." Postmodern Culture 1.2 (Jan. 1991). 19 pars. <http://muse.jhu.edu/journals/postmodern_culture/v001/1.2ulmer.html> and </text-only/issu e.191/ulmer.191>.

    Vallias, André. "We Have Not Understood Descartes." Visible Language 30.2 (1996): 150-7.

    Wardrip-Fruin, Noah. "Writing Networks: New Media, Potential Literature." Leonardo 29.5 (1996): 355-373.

    Wortzel, Adrianne. "Cyborgesian Tenets and Indeterminate Endings: The Decline and Disappearance of Destiny for Authors." Leonardo 29.5 (1996): 354-372.

    URLography

    Amerika, Mark. "Hypertextual Consciousness." <http://www.altx.com/htc/title.html>.

    Assoziations-Blaster. <http://www.assoziations-blaster.de/english/>.

    Carroli, Linda, and Josephine Wilson. water always writes in plural. <http://www.hypertxt.com/sh/hyper98/water/>.

    Cayley, John. Indra's Net series. <http://www.shadoof.net/in/inhome.html>.

    Coverley, M. D. "Fibonacci's Daughter." <http://califia.hispeed.com/Fibonacci/choice.htm>.

    Malloy, Judy. l0ve 0ne. <http://www.eastgate.com/malloy/>.

    Memmott, Talan. Lexia to Perplexia. <http://www.uiowa.edu/~iareview/tirweb/hypermedia/talan_memmott/>.

    Miller, Matthew. "Trip." <http://raven.ubalt.edu/guests/trip/>.

    Queneau, Raymond. "A Fairytale As You Like It." <http://userpage.fu-berlin.de/~cantsin/queneau/conte/conte.cgi>. (Temporarily offline due to copyright negotiations.)

    Thorington, Helen, with M. R. Petit and John Neilson. Solitaire. <http://turbulence.org/Works/solitaire/>.

LINKS: Non-Graphical Users See Top of Page