Still speaking in a low voice, the stranger said, "It can't be, but it is. The number of pages in this book is no more or less than infinite. None is the first page, none the last."--Jorge Luis Borges 1


What, until recently, has been called for alternately empirical and mystical reasons the book is entering a distinct epoch in which it will no longer be possible to limit the range of a material body of writing by enclosing it within a published volume, as, for instance, something we could call a definitive or even standard edition. As early as the beginning of the eighteenth-century Jonathan Swift had already conceived of the marriage between the book and machinery, projecting the idea of the Gutenberg invention towards its (techno)logical, evolutionary ends.
2 With the advent of hypertext and of the World Wide Web this marriage seems to have been at last consummated, linking together both the means, medium and matter of publication as something like an open, universal "mechanised" text. Moreover, this marriage has linked the traditional domain of the book to the entire field of techno-mechanical production and re-production, by means of extensive, interconnected computing networks.

The historical advent of the World Wide Web in late 1990 opened the possibility that any electronic text could, conceivably, be directly linked to any other electronic text. Invented by Tim Berners-Lee, a computer scientist at the Centre Européenne pour la Recherche Nucléaire (C.E.R.N.), the World Wide Web was originally conceived and developed for large high-energy physics collaborations which require instantaneous information sharing between physicists working at different universities around the world. Berners-Lee, along with a colleague at C.E.R.N., Robert Cailliau, also established fundamental protocols such as URLs (Universal Resource Locator), HTTP (Hypertext Transfer Protocol) and HTML (Hypertext Mark-up Language), which provide the basis for electronic hypertext. Anticipating this development as early as the mid-1940s, the American science administrator Vannevar Bush envisaged a form of electronic, interactive text archive, or prosthetic memory, which he termed the "memex." According to George P. Landow, Bush's memex is essentially a "poetic machine," which works on the basis of analogy and association, describing a cybernetic interface between poetics and technology which echoes the poetic science of Giambattista Vico and anticipates the mind ecologies of Gregory Bateson.

In the 1960s Bush's ideas were taken up by computing engineers like Theodor H. Nelson, who coined the term "hypertext" to describe a similar type of interface between techne and poiesis. In Literary Machines, Nelson defined hypertext as "non-sequential writing-text that branches and allows choices to the reader, best read at an interactive screen. As popularly conceived, this is a series of text chunks connected by links which offer the reader different pathways."
4 For Nelson, a "hypertext" isisi not determined by its assumed physical or conceptual boundaries, but by communications between texts, between textual structures and between textual elements (where every apparent "limit" suggests, instead, the possibility of a further linkage).

The interlinking possibilities of electronic texts have doubtless had profound effects upon the conditions of reading, particularly in the case of such authors as James Joyce, Geoffrey Chaucer or William Shakespeare, whose works have been subject to prolonged and intense scholarly and genetic analysis. In the case of each of these canonical figures computing technology has been brought to bear upon the received notion of the "definitive text," underlining the fact that in each case there already exist either variorum and composite texts, or distinctly different versions of texts published independently of one another and read contemporaneously.
5 In regards to Chaucer, this has extended to the application of advanced programmes for genome analysis to the eighty one extant versions of the 'Introduction to the Wife of Bath's Tale' in an attempt to map the text's numerous morphogeneses from the time of its composition in the fourteenth-century to its present reception. For Shakespeare, this has meant the transferral of existing debates surrounding the various Folio and Quarto editions to the domain of computing technology and hypertextuality, just as with Joyce it has brought about new developments in manuscript and notebook analysis, and in the conception of synoptical rather than "definitive" texts of Ulysses and Finnegans Wake.6


According to hypertext theorist Jay David Bolter, tradition has it that the "written text" is a "stable record of thought, and [that] to achieve this stability the text has to be based on a physical medium: clay, papyrus or paper; tablet, scroll or book."
7 Indeed, certain empirical approaches to the materiality of the "book" argue that "not only is the text [...] caught in the materiality of the book, it is also tied to the book's paper, cardboard, ink, and glue to the historical and economic conditions of its production and distribution."8 Such an approach considers the advent of electronic texts as representing a crisis and a threat to the integrity and meaning especially of literary texts and of literary genres. J. Hillis Miller, commenting upon the fate of the novel, suggests that electronic texts are given:

a strange new historic placement in the cyberspace of today. A date of original publication is indicated, and that is about all. The novel exists not as embodied in material form, or at least not material in the fixed way of a printed book. It exists as a large number of bits of information, zeroes and ones inscribed as magnetic differences on a hard disk or on magnetic tape or as minute scratches on an optical disk or as electronic pulses on the wired and wireless transmissions of the Internet.9

For Hillis Miller, a "text" suspended in cyberspace is thus "detached from its local historical context" and becomes "a text in the context of an enormous and incoherent abundance of works of all kind-verbal, pictorial, and auditory-on the Internet." Moreover, "this transformation is occurring even though it is still a primary goal of literary history and literary criticism in the modern languages to understand and interpret the culture of the book."

Nostalgia for the "text" as artefact belies a central confusion in much of what continues to pass for "literary criticism," between what we might call bibliographic and linguistic codes, for example: a confusion that often persists in the synonymous use of terms like "book" and "text." This can be seen as representative of an empirical tendency still current in some areas of literary scholarship, where "materiality" and historical context remain conceptually fixed outside any discourse, including critical discourse, which would not respect such boundaries or which would challenge the certainties implied by them. Joyce himself parodies this preoccupation with the artefactual value of the book at length in Finnegans Wake in regards to a certain letter, "discovered" by a hen in a dunghill in an advanced state of decomposition. This letter, which is said to "belong" to A.L.P., is subjected to extensive "genetic" analysis by a grave Bròfessor, and posed as evidence during various "inquisitions," but nevertheless remains indecipherable (due not only to its decomposition, but to the physical "damage" wrought upon it by the hen, the Bròfessor, and the general process of its exegesis, not to mention the fact that the letter is to start with also a text):

Well, almost any photoist worth his chemicots will tip anyone asking him the teaser that if a negative of a horse happens to melt enough while drying, well, what you do get is, well, a positively grotesquely distorted macromass of all sorts of horsehappy values and masses of meltwhile horse. Tip. Well, this freely is what must have occurred to our missive (there's a good sod of a turb for you! please wisp off the grass!) unfilthed from the boucher by the sagacity of a lookmelittle likemelong hen. Heated residence in the heart of the orangeflavoured mudmound had partly obliterated the negative to start with, causing some features palpably nearer your pecker to be swollen up most grossly while the further back we manage to wiggle the more we need the loan of the lens to see as much as the hen saw. [FW 111.26-112.02]

The desire to decipher "all there may remain to be seen" (FW 113.32-33) from this tea-stained letter, suggests a desire, as Jacques Lacan might say, to gain knowledge about the real, revealing that beneath the desire to address the artefactual value of the book is also hidden a desire to situate the meaning of the text in the material reality that is supposed to frame it. This hermeneutic recovery is shown, however, to be a mirage. As Joyce puts it: "Closer inspection of the bordereau would reveal a multiplicity of personalities inflicted on the documents or document" (FW 107.23-26), suggesting that the singular identity of the letter is not only question,
11 but a product itself of hermeneutic "infliction":

Anyhow, somehow and somewhere, before the bookflood or after her ebb [...] wrote it all, wrote it all down, and there you are, full stop. [...] [B]ut one who deeper thinks will always bear in the baccbuccus of his mind that this downright there you are and there it is only all of his eye. Why?
Because Soferim Bebel [...] every person, place and thing in the chaosmos of Alle anyway connected with the gobblydumped turkery was moving and changing every part of the time: the travelling inkhorn (possibly pot), the hare and turtle pen and paper, the continually more or less intermisunderstanding minds of the anticollaborators, the as time went on as it will variously inflected, differently pronounced, otherwise spelled, changeably meaning vocable scriptsigns [...] riot of inkblots and blurs and bars and balls and hoops and wriggles and juxtaposed jottings linked by spurts of speed. [FW 118.11-30]

But even if we are to take into account the different possible factors that confer meaning upon the "book" as an artefact, or even upon the printed word as signifying in a way distinct from, say, a word displayed on a computer screen-that would still not mean that the text can be considered simply as the shadow or trace of an idea "already shaped," as it were, by an historical context within which the "technology of the book" would be imbedded. Even adopting a more or less socio-empirical view of literate cultures, what we would call textual structures can be seen as actively determining so-called "ideas" just as powerfully as the "primal structures" that are considered as shaping language itself (as in McLuhan's dictum: the medium is the message).
12 And this would suggest, quite trivially of course, that what Hillis Miller persists in calling the "culture of the book" and its grounding in certain "historical contexts" is a basic confusion: on the one hand, of textuality with the normative structures that it in fact determines and to which it is made to appear subject, and, on the other hand, of the signifying "materiality" of the text with the artefactual value of the printed book.

However, as long as "the text" was seen to be married to physical media, the majority of readers and writers took for granted three crucial attributes: that the text was linear, bounded and fixed. Generations of scholars have internalised these qualities as the rules of thought, and they have had persuasive, and pervasive, social consequences. Nevertheless, these rules of thought have come under increasing scrutiny during the course of the twentieth-century, most recently in the form of post-structural theory and the advent of hypertext. As the critic Richard Lanham has noted:

It was establishing the original text that the Renaissance scholars thought their main task, and generations of textual editors since have renewed their labours. The aim of all this was to fix the text forever.

The continuing controversy over the Hans Walter Gabler edition of Joyce's Ulysses makes abundantly clear just how intense the desire to "fix the text forever" can be.
14 Gabler's conceptualisation of the editing process may have occurred "independently of his decision to use the computer,"15 but the controversy over his "synoptic" version of Ulysses offers a clear illustration of the way computers can revolutionise our understanding of text, but also how they can become the tools of a form of technological nostalgia. Every word in Gabler's synoptic text was written by Joyce himself, and yet the final "reading text" is a text no one ever wrote: it had never existed prior to its publication. In this sense Gabler's Ulysses recalls certain mathematical formulae which produce structures that, while in some cases resembling phenomena in the "real" world, have no counterpart in that world.


Gabler's use of the computer to "restore" Ulysses to the form of its "total conception," belies a tendency within some areas of textual scholarship to pursue a form of encyclopaedism, constructing upon the assumptions of a possible totality of authorial intention something like a codex, or signatura rerum. The archival desire which underwrites this project also extends to the totality of authorial identity, and to the projection of this encyclopaedism onto the body of artefacts and testimonies which circulate around the proper name of the author. Indeed, since its inception Joyce studies has been effected by issues of legality and ownership. As William Brockman has pointed out, "amongst these should be included the aggressive acquisition by American Universities of library collections during the 1950s, the beneficence of donors, the tastes of private collectors, the sporadic restrictions placed upon access by the Joyce estate, and the intentional destruction of documents"-all of which has strongly influenced the disposition of Joyce studies.

As Brockman suggests, one result of Joyce's "lifelong peripatetic style of residence," and of his abrupt departure from Paris in 1939, was that there remained after his death a legacy of letters and manuscripts scattered throughout Europe.
17 In the 1940s Joyce's papers were distributed amongst a variety of private owners, beyond the knowledge, or at least control, of Joyce's immediate family, and for the most part unavailable to scholars. Sales in 1924 of most of John Quinn's library (not including a manuscript copy of the eighth draft of 'Circe'),18 and in 1935 of manuscripts in the possession of Sylvia Beach, brought a limited number of papers onto the open market. Among the items offered for sale by Beach was:

a selection of typescript pages [of Ulysses] with autograph corrections by the Author [...]. These pages may be acquired separately by those who might like to enrich their copy of Ulysses with a little manuscript of Joyce.19

Whilst Harvard had acquired the manuscript of Stephen Hero in 1937, there was no effective institutional collecting of Joyce's papers until the 1950s. Even after Joyce's death in 1941, libraries remained ambivalent about collecting Joyceana, despite Judge John M. Woolsey's decision lifting the United States ban on Ulysses in 1933,
20 while the upheavals of the Second World War, combined with issues of propriety and public responsibility, necessarily limited availability. But over the next fifteen years, during the height of collecting by American universities, Joyce's papers gained sufficient value to subject their acquisition to contention and dispute. It was this market, as Brockman points out, that resulted in their complicated distribution among libraries today.21

In 1979, three scholars-Michael Groden, Danis Rose and David Hayman-completed work on one of the major efforts of publishing in the history of Joyce Studies, aimed at resolving at least some of the difficulties caused by the distribution of Joyce's notebooks and manuscripts. The James Joyce Archive in 63 volumes was, for a time, seen as solving the more immediate problem of accessing the many disparate library collections, although it soon became apparent that the sheer size and cost of the archive was itself prohibitive. Moreover, the archive did not in fact represent the entire body of Joyce's extant papers, many of which still remain in private hands. By 1984 the question of accessibility had become of pressing importance, with the publication of Gabler's three-volume "Critical and Synoptic Edition" of Ulysses by Garland. Gabler, with the assistance of ten researchers at the Institute for English Philology, at the University of Munich, and supported by a grant of US$300,000.00 from the German government, produced what was for a time hailed as the definitive text of Ulysses, with no fewer than five thousand "improvements." In 1986 Gabler's Ulysses: The Corrected Text was published in one volume, with all other editions being withdrawn. It was hailed as a massive success, not least by Anthony Burgess, who was confident that it was far superior to the original 1922 edition or to the Bodley Head editions of 1934 and 1961.

In 1988, however, an otherwise unknown Joyce scholar, John Kidd of Boston University, published an article entitled 'The Scandal of Ulysses' which severely criticised Gabler's methodology.
23 One of the principal points of contention between Kidd and Gabler was the efficacy of working from facsimile editions of Joyce's manuscripts, rather than directly from the originals (a task which poses extreme logistical difficulties). This was a problem that the Joyce Archive proved insufficient in solving due, among other things, to the importance of the many different coloured pencils and crayons which Joyce used to indicate corrections or deletions. There was also the problem of the weakness and often near illegibility of Joyce's script, and the difficulty in differentiating, in black and white facsimile, between Joyce's markings or deletions and those of his copyists, typists and proof readers-problems which were even further exacerbated in the case of Finnegans Wake by the fact that much of the text had been dictated, and that it had been composed over a period of seventeen years during which time Joyce changed his methods of working from his notebooks.24

In 1990 the bulk of Joyce's work temporarily moved out of copyright. But in 1993 Britain and the United States revised their copyright laws in line with broader European standards, retrospectively extending the term of copyright protection from fifty to seventy years from the author's death. However, during the three years in which Joyce's work fell out of copyright control, Donald Theall at Trent University established online versions of Ulysses and Finnegans Wake, while revised editions of Joyce's work were produced by several major publishers. Amongst these was the international publishing house Penguin which, following the 1984 controversy over Hans Walter Gabler's "corrected" text of Ulysses, chose to re-issue instead the 1961 Bodley Head edition. Soon after, Gabler's major rival, John Kidd of Boston University, himself lost favour with critics, resulting in the indefinite suspension of plans by Norton to publish his three volume "definitive" text of Ulysses, initially scheduled for 1992 release, but first deferred and then later abandoned altogether.

The situation soured further in 1994 with new threats by Joyce's grandson, Stephen Joyce, to issue suits against academics who chose to quote from Joyce's private correspondence. Already Brenda Maddox's Nora: A Biography of Nora Joyce (awarded the Los Angeles Times Book Prize for Biography in 1988, and subsequently, in 2000, adapted for film by director Pat Murphy) had been withdrawn by its publishers, Random House, and reissued with substantial emendations. On this occasion the Joyce estate had taken offence at Maddox's reference to Joyce's "pornographic" and scatological letters in the last chapter of her book.
25 Similarly, at the XVth International James Joyce Symposium, held in Zürich in June 1996, Stephen Joyce vowed that he would "prevent the genetic scholars from 'mucking up' Joyce's texts with new editions 'just to build reputations.'"26

These concerns aside, there remains the problem of access to the disparate and extensive collections of Joyce's notebooks and manuscripts. Joyce himself often joked that his books would keep the professors busy for centuries, something which seems at times to be born out by the academic concern for textual minutiae. But after the excesses of the 1980s, Joyce studies has for the time being ceased to be dominated by the Alexandrian project of resurrecting an Ur-text of Ulysses or Finnegans Wake: "Ur greeft on them!" (FW 241.31). On the other hand, the challenges of the James Joyce Archive, and the attention aroused by the Gabler-Kidd scandal, have prompted many Joyce scholars to look more deeply into the significance of manuscript analysis and its impact upon Joyce studies generally.

In the early 1990s the widespread availability of hypermedia software ;seemed to provide possible solutions to many of the dilemmas posed by the existing distribution of Joyce's papers and to the ethical problems that had arisen during the 1980s over editorial practice. It also seemed to offer a possible bridge between traditional manuscript analysis and current trends in textual theory, a way of re-consolidating the Joycean project, and of combining the enthusiasm of commercial investors (particularly in the area of CD-ROM development) with the interests of scholarship. With the rapid growth in virtual technologies, there has also been a renewal and extension of existing theories and practices of "genetic" criticism, particularly at the Institut des Textes et Manuscrits Modernes (I.T.E.M.) in Paris, and among various individual projects internationally.
28 Amongst other things, the marriage of textual genetics and hypertext promises a means of tracing the formation of received versions of Joyce's work and of the "unreceivable" Joycean archive which haunts them, without necessarily resorting to the archaeological precepts of Ur-critics like Gabler and Kidd. But as Jacques Derrida has warned:

there is an incessant tension here between the archive and archaeology. They will always be close the one to the other, resembling each other, hardly discernible in their co-implication, and yet radically incompatible, heterogeneous, that is to say, different with regard to the origin, in divorce with regard to the arche. 29

This notion of "divorce with regard to the arche" presents a central paradox in this relation-genetics and hypertext-as describing a type of quasi "archaeological" mechanism, sifting through the probable ruins and simulacra of "the Joycean text." At the same time, this relation can be seen as generative. Not with the aim of derivation or of "restoring" an Ur-text, but of setting the possibilities (and "plurabilities" [FW 104.02]) arising from this co-determination to work simultaneously, as a movement of desire and the impossibility of its realisation: a simulated archaeology, producing and organising simulacrum artefacts of its own interminable process. That is, as a kind of hermeneutic monstrum. According to Derrida:

one perhaps could say that the movement of any archaeology [...] is an accomplice of [...] reduction [...] and always attempts to conceive of structure on the basis of a full presence which is beyond play.30

At the same time, the co-implication and incompatibility of the archive makes such a reduction impossible-or rather, hypothetical, as the self-interested (solipsistic) perpetuation of two "systems" of mutual determination. Joyce's machine, as Rabaté points out, is a paradox, which has no other aim "than what it accomplishes itself in running."

Similarly, in the genetic process of reduction and recuperation, what have heretofore been considered distinct, if problematic, features of Joyce's writing (textual deviations, omissions, printer's "corrections," manuscript variations, Joyce's own amendments and possible oversights, etc.), would thus be brought within one another's sphere of signifying influence, as it were, as "material" parts of a single, if contradictory, hypertextual apparatus. This apparatus, far from introducing a further "meta-level" of complexity into the Joycean text where it wasn't there before, would instead open itself to those significatory forces operating already between the so-called text and its proto- or avant-textes which had previously been elided through, among other things, editorial practice and the availability of appropriate technological resources.

Such a generative text of Finnegans Wake has (at least in part) been signalled under the generic title Work in Progress. Work in Progress is, in one sense, the title given to Joyce's several published pieces between 1924 and 1938 that were either left individually untitled, or else appeared under autonomous titles such as Anna Livia Plurabelle, Haveth Childers Everywhere, or Tales Told of Shem and Shaun. In another sense, "work in progress" defines Joyce's writerly condition between 1923 and 1939.
32 Work in progress in the latter sense, documented as it is from the time of the so-called Finn's Hotel vignettes in notes, drafts, fair manuscripts, typescripts, and proofs, has posthumously and only recently come to public notice, and it is only still beginning to have an impact on Joyce scholarship. Inroads into a "genetic" understanding of this "contonuation of the word in pregross" (FW 284.21) were first aided by the publication of the James Joyce Archive and began by way of transcriptions and analyses of the Finnegans Wake notebooks. One of the Archive's compositors, Danis Rose, has, since the mid-1970s, been involved (with John O'Hanlon in Dublin) in the establishment of a genetic and critical edition of Finnegans Wake. The publication of this four-volume edition of Finn's Hotel/Finnegans Wake, however, has been indefinitely delayed due to legal problems arising out of a conflict with the Joyce estate.

In 1995 Rose published a monograph titled The Textual Diaries of James Joyce,
33 in which he attempts to outline the chronological arrangement of documents used in the composition of Work in Progress and Finnegans Wake, thereby giving the basic arrangement of his four-part critical edition of the Wake, in an effort "to lift, or try to lift, the cloud of ignorance under which for the most part these matters are being debated."34 As against the chronological arrangement of Rose's Diary, however, Joyce's work in progress is shown to have emerged non-linearly. That is, in relation to the "completed" Finnegans Wake, hence exploding a popular fallacy that Joyce's writing followed a clear and singular line of development over the seventeen year period of the Wake's composition.35

For Rose, Joyce's notebooks are indexes to the textual realm of Work in Progress/Finnegans Wake-an idea first put forward by Rose himself in his The Index Manuscript,
36 but which in the Diary he outlines more emphatically, stating that:

it is solely through the index connection [of notebooks and Finnegans Wake text] that we can analyse the larger patterns of Joyce's work in progress.

Rose's non-linear view of the genesis of the Wake, and his constructivist approach to how this genesis effects our engagement with Finnegans Wake as a "stable" text, suggests not only a possible hypertextual approach but a solicitation of hypertext. In this sense, the hypertextuality of Joyce's "work in progress" is not founded upon an empirical logistics, but to a textual condition.


In 'The Garden of Forking Paths,' the Argentine writer Jorge Luis Borges describes a book by a fictional author Ts'ui Pên which, like a textual Chinese box, suggests a potentially infinite number of simultaneous narratives bound within the material finitude of a single volume:

In all fictional works, each time a man is confronted with several alternatives, he chooses one and eliminates the others; in the fiction of Ts'ui Pên, he chooses-simultaneously-all of them. HE CREATES, in this way, diverse futures, diverse times which themselves also proliferate and fork. Here, then, is the explanation of the novel's contradiction.38

As with Rose's Index Manuscript, Borges's fictional text suggests a type of writing in which an economy of ideogrammatic summarisation and bifurcation operates a structural "apparatus" whose contradictory nature can only be "explained" by something equally as complex and contradictory. The solicitation of this hypertextual apparatus, however, is located in the structure of possibility itself, and not merely within the fiction's thematic organisation. Moreover, as with Joyce, possibility in this sense stands in advance of narrativity, as a condition of the text "in progress," hence describing a type of signifying programme which also underwrites the basic mechanisms of textuality.

The idea that other unrealised possibilities exist "in worlds of their own" (as it were) can be traced back to pre-Socratic philosophers like the Atomists or Parmenides.
39 And the idea that poetic "imitations" are imitations of the possible rather than the actual can be attributed to Aristotle, who insisted in chapter IX of the Poetics that:

it is not the function of the poet to relate what has happened, but what may happen-what is possible either in the way of likelihood or inevitability.40

After the Middle Ages, the notion of "possible worlds" moved away from the poetic, ideal and formal concerns of the Greeks and proceeded more along the lines of "scientific" positings of the actual plurality of worlds, from the Renaissance on through to the eighteenth-century and Leibniz's monadology.
41 The idea attracted renewed attention during the 1950s, when physicists working both on relativity theory and quantum mechanics began to posit the "existence" of other, "possible" worlds which connect to our own.42 Possible worlds here are considered to be parallel and alternative versions of our own world, where the individuals, temporalities, causalities or outcomes of situations, decisions or actions are, in some respects, different. In relativity theory, other possible worlds are conceived as existing side-by-side with our universe-in quantum theory, as occupying the same space in a ghostly manner. It is this last notion of possible worlds that comes to bear upon the precepts of genetic criticism and which is probably most useful to a consideration of hypertext and Finnegans Wake.

Interest in possible worlds also increased in the 1960s and 1970s when attempts were first made to develop computing programmes which could simulate realities (artificial intelligence, strategic systems, meteorological modelling). Many of these programmes owed their conception to the pioneering work of Alan Turing, whose 1936 paper 'On Computable Numbers, with an Application to the Entscheidungsproblem' provided a blueprint for the developers of the first electronic computers.
43 In a way that uncannily "describes" Finnegans Wake, Turing mathematically demonstrated what (in simplified terms) can be understood as the assertion that it is impossible to conceive of a computer more powerful than a "universal Turing machine." That is to say, the theoretical apparatus called a universal Turing machine is capable of "simulating" all possible Turing machines by means of a programmatics in which computing is linked to a general recursiveness (the Church-Turing thesis).44 As in Cantor's paradox, the Turing machine evolves from a "built-in obsolescence" to self-simulation along a supplementary metonymic chain of substitutions, recalling also Marshall McLuhan's comments on the simultaneous obsolescence and virtually infinite extensibility of the "book."

In his recent book, The Virtual Marshall McLuhan, Donald Theall has pointed out how such a convergence of ideas was bound to have an impact on Joyce's conception of Finnegans Wake.
45 Like Guillaume Apollinaire and Blaise Cendrars, Joyce was greatly impressed by the late work of Stéphane Mallarmé, particularly the typographically "concrete" poem 'Un Coup de dés.' The simultanéisme and concretion of Apollinaire's Calligrammes and Cendrars's Prose du trans-siberien have both been attributed to the influence of Mallarmé, and similar traits can be found in Joyce's work. What Theall identifies, however, is that the impact of Mallarmé can be traced through the implications of a poem like 'Un Coup de dés' in terms of contemporary mass media and communications technology (as in the 'Aeolus' episode of Ulysses).46 For Theall, the simultaneous and paratactic nature of fragmentary media messages reveals a hyper-textual network of otherwise disparate textual elements, an "idioglossary" (FW 423.09) of an encyclopaedic dimension:

in the Nichtian glossery which purveys aprioric roots for aposteriorous tongues this is nat language at any sinse of the world. [FW 83.10-12]

Such a glossary or encyclopaedia, as a type of hypothetical Turing machine, would re-determine the technology of the book as a matrix of all "possible worlds," beyond the notions of the book of revelation and the book of nature. As Mallarmé writes: "que tout, au monde, existe pour aboutir à un livre."

In the second of their two books dealing with capitalism and schizophrenia, A Thousand Plateaus, Gilles Deleuze and Félix Guattari describe Joyce's words as having "multiple roots," which "shatter the linear unity of the word, even of language, only to posit a cyclic unity of the sentence, text, or knowledge."
48 Subsequently, in Joyce "the world has become chaos, but the book remains the image of the world: radical-chaosmos rather than root-cosmos. A strange mystification: a book all the more total for being fragmented."49 Recalling those structuralist axioms which attempt to define foundational semantic units, this statement by Deleuze and Guattari mistakenly suggests that Joyce's language is merely a re-statement of linguistic normativity, even as his "words [...] shatter the linear unity of the word, even of language."

According Deleuze and Guattari, the vast growth in information technology and the electronic media has resulted in the idea that reality only exists through the global interaction of simulated possible worlds (informatics, speculative economies), in the form of abstract machines.
50 In place of the apparent "monotheism and the book"51 there are seen to emerge "collective assemblages of enunciation" derived from "principles of connection and heterogeneity: any point of a rhizome can be connected to any other, and must be."52 This rhizomatic text, rather than exceeding in any way the Joycean "book," serves to affirm the radical "hypertextual" nature of Joyce's project, in which the "word" links across the entire field of "language," whose "cyclic unity" it at once "shatters" and describes in the movement of its transversal.

In The Gutenberg Galaxy and The Global Village: Transformations in World Life and Media in the 21st Century, McLuhan similarly described a radical, global convergence of media (as a Joycean "chaosmos of Alle" [FW 118.21]), across the existing boundaries of sense and the senses (synaesthesia), in which language and the "book" are dramatically reconfigured. But while McLuhan is not credited by Deleuze and Guattari, many of the ideas which have since become foundational to the vastly over-categorised jargon of "rhizomatics" are derived, in a more or less direct genealogy, from the writings of McLuhan, and in particular his meditations upon the work of Joyce.

However this may be, the "Joycean tachyons, particles-holes, and quarks"
54 which organise the apparent "cyclic unity" of our hypertextual "radical-chaosmos," can be seen to bear direct implications for the way in which so-called "collective assemblages of enunciation" are in fact conceived and experienced. As hypertext develops further the use of hypermedia in contemporary life (in which anything that can be retrieved from an electronic archive can function as a text node, including digitised film, audio, complex imaging, generative and analytical functions and so on), the idea that the world is a network (or "web") of virtual realities is becoming more "concrete."


In Cratylus, Plato asks: "Does not the word 'techne' denote a possession or state of mind?" This notion of "states of mind" not only requires that consciousness operate on a level of possibility, but places such recent concepts as "artificial intelligence" within a context of technological artifice, as an operation of a certain programmatics or pro-gramme. That is to say, as an apparatus of possibility defining the structure of thought itself, which we might also consider as a form of text. In this way techne does not imply a prosthesis of mind, but "denotes a possession or state of mind" in which possibility is ubiquitous.

Among other things, this technics of "states of mind" raises ever more complex questions about the nature of "reality" and of "possible worlds." For Maurice Blanchot, these questions are also questions about the culture of the book-a literacy and literality of simultaneous world-states describing a virtually infinite hypertextual "chaosmos." As Blanchot argues:

if the world could be exactly translated and reduplicated in a book, it would cease to have a beginning or an end and would be the spherical, complete, boundless volume which every writer writes and in which he is written-it would cease to be the world but would, or will, be the world perverted into the infinite sum of its possibilities.

Similarly, Jacques Derrida describes this genesis of "hypertext" in terms of a failure of existing notions of textuality to contain significatory possibilities, suggesting that a text is always already in "interior" communication with what had previously been considered as constituting its "exterior." According to Derrida's by now famous dictum, il n'y a pas de hors texte:

What has happened, if it has happened, is a sort of overrun [débordement] that spoils all these boundaries and divisions and forces us to extend the accredited concept, the dominant notion of a "text," of what I still call a "text," for strategic reasons, in part-a "text" that is henceforth no longer a finished corpus of writing [...] enclosed in a book or its margins, but a differential network, a fabric of traces referring endlessly to something other than itself, to other differential traces.

This overrun can be said to effect the entirety of discourse, including philosophical discourse, in a way that renders traditional modes of inquiry problematic. Not only does this overrun affect the stability of an object of inquiry (even if a text is only nominally an object), as well as the limits of the genre of that object, but it suggests the impossibility of situating an object of inquiry independently of the mode of inquiry itself.

Derrida's formula does not merely relate to the way language operates in and of itself, but to the ways in which language is said to be manipulated (or programmed). In this way we might say that there are, in fact, no textual objects. In various ways this recalls Stanley Fish's notion of "interpretive communities," in which texts serve a dual empirical-psychological function (in what appears to agree with a broadly Lacanian definition). A text operates always within the field of the Other, as a locus of desire-a situation it shares with the Lacanian subject.
58 In this way the "dialectic of desire" described in Lacan's formulation of the "mirror stage" also functions as a model of interpretation.

Etymologically, the Latin interpretatio includes the meanings "to mediate" and "to translate," and this is perhaps one of the simplest ways of thinking the congruence of interpretation and textual production.
59 Further, it gathers together the various notions of difference between, and within, languages, and between language itself and its so-called exterior. In Derrida's view:

The interpretation or solving of the puzzles of the textual web only adds more filaments to the web. One can never escape the labyrinth because the activity of escaping makes more labyrinth.

All commentary, having in one way or another language itself as its object, succeeds merely in proliferating a discourse to which it already belongs-and this is perhaps the most direct way in which we can approach the inherently textual nature of "possibility." Moreover, it answers the more obvious questions of hypertextual cybernetics (in which the production of a hypertext or hypertextual transverse cannot be situated as an independent, decisive act, but rather as an "interpretive" one).

In the 1970s and 1980s, modal philosophers such as Saul Kripke, David Lewis and Nicholas Rescher, spent much of their time reworking traditional logic to accommodate the "existence" of other possible worlds, as well as thinking about the nature of possibility itself. For Kripke, possible worlds are stipulated rather than discovered, and, in pursuing Cantor's "naïve" principle (that from any predicate a set can be formed), developed a semantic interpretation of modal logic.
61 Further extending Cantor's principal, and drawing upon Bertram Russell's theory of types, David Lewis and Robert Stalnaker have asserted that the idea of a proposition is itself a set of possible worlds. More recently some literary and social theorists have begun to explore media and simulacrum in terms of possible worlds, in particular Umberto Eco, Jean Baudrillard and Paul Virilio. For Baudrillard, possibility can be thought in terms of simulation, where: "Simulation is no longer that of a territory, a referential being, or a substance. It is the generation of models of a real without origin: a hyperreal."62 Something of this idea was already current in the work of Pierre Klossowski and can be traced back, along with Lacan's "dialectic of identification," to the seminars of Alexandre Kojève in the 1940s.63 According to Klossowski: "La simulation étant l'attribut de l'être même, elle devient aussi le principe même de la connaissance."64

For many "possible worlds" theorists, possibility is ubiquitous and not necessarily or even primarily linked to myth, dream or illusion since, as Rescher perceives it, "the rational guidance of human affairs [also] involves a constant recourse to possibilities."
65 There is also a suggestion that modal philosophers were reacting largely to the role of the computer. In Homo Cyberneticus Hans Holstein argues that modal attempts to describe world states were required to permit the mechanical derivation of subsequent world states in problem-solving programs, where the computer was used to assist in human decision making.66 The establishment of world-states and rules of transformation was a preliminary to programmes which could explore the different possible consequences of decisions or the different possible future states devolving from the present world state. This is not to say that possibility is a function of a rational consciousness (or a function of intentionality), but of a certain techne of inscription, or pro-gramme, which only with difficulty can be thought of as an "artificial intelligence." The concept of possibility would thus no longer involve a coextensivity between the structuring logics of different "world states," or of a logos in whose shadow possibility would merely describe a form of "mimetic proliferation" of signs. Similarly, the concept of possibility would not dependent upon the "future possibility" of being realised or made manifest, whether temporally or spatially, as what we might call the reduction of a certain polysemy. At the same time, however, it would be impossible to do without these notions, whose reductional mechanics describe an axis or "horizon of possibility" from which terms such as simultaneity and simulationism obtain their meaning.

This is the basis of Stephen Dedalus's dilemma in Stephen Hero and A Portrait of the Artist as a Young Man, when he poses the Aristotelian question: is anything possible that does not actually take place? This question is partly addressed by Freud (although, as Derrida points out, unwittingly), in his critique of superstition in Studies in Parapsychology, when he raises the issue of chance in regards to the operations of the unconscious and psychoanalytic method.
67 There is an interesting admixture of pre-determinism and pre-destination in Freud's discussion, which might be said to programme Derrida's frequent use of the phrase "calculated and by chance."68 But this paradox is inherent to the structure of possibility itself, in that something implies the possibility of its being realised at some unspecified time or place, or in some unspecified way as a form of destiny to which it is tied regardless of whether such a possibility will ever be realised or not. In this sense possibility can be considered as operating at a remove from itself, as a constant deferral of itself, orientated by the promise of a future advent.69 At the same time this "forethrow" of possibility operates a mechanism of reversion, a prototypical cybernetic apparatus between psychological "affect" and material "causation," intentionality and chance, physis and techne, and so on. Moreover, in its deferral of totality in the assumption of a signified event or "realisation," this apparatus can be seen as operating textually, as so many chains or networks of signifying "substitution," describing, in fact, what we might call hypertext.

© Louis Armand
volume 4, issue 1, 2003