Literature from Page to Interface: The Treatments of Text in Christophe Bruno's Iterature

Literature from Page to Interface: The Treatments of Text in Christophe Bruno's Iterature


Søren Pold explores the ways in which Christophe Bruno’s Iterature expands the notion of literary form and shows what happens when words are no longer only part of a language.

Lori Emerson:

Marjorie Perloff has long advocated rereading the development of 20th century poetry through modernist experiments of the historic avant-garde - using works such as Mallarme’s as touchstones rather than as examples of literary aberrations.

Lori Emerson:

All three ebr reviews (by Chris Funkhouser, Adalaide Morris, and John Zuern) on the Electronic Literature Collection touch on the ways in which e-writing builds on a rich tradition of bookbound, constraint-based writing.


1. Ontological unrest

In the late 1980s and early 1990s, digital hypertext literature was broadly considered a reckoning with the book. In continuation of a post-modern and post-structuralist tradition, hypertext was seen as an exhaustion of, and a potential break with, the literary forms of the printed book, a view that was reinforced by hypertext theory. Already before the general appearance of digital hypertext literature, there was a widespread theoretically based understanding of the crisis of the book, the late age of print, the end of the Gutenberg galaxy, and so on.

However, the book lives on, and neither the web nor the e-book seems to be able to kill it. Consequently, the many predictions of the book’s demise put forth by digital literary theorists ten to fifteen years ago seem overrated and hysterical today, and classical literary scholars have wiped the sweat off their foreheads. Nevertheless, I suggest taking the predictions seriously as a sign of a more profound ontological unrest in the field of literature and the very ontology of the text that is far from over, even though the book seems to be surviving. What is more, I believe this kind of unrest is vital for the continuing development and timeliness of literature as an art form. Even though the book does not appear to be dying anytime soon, changes are taking place in our literary culture that have major consequences for the role of writing and reading in our current digital media reality.

In a broad sense, books and literature have always been about literary culture and society: from the close connections between the printing press and national languages and cultures in the 16th century, to the articulation of freedom of speech and of the press during the Enlightenment, to relations between book production and assembly-line capitalism or the close connections between the author function and authority, as outlined by Barthes, among others (“The Death of the Author” ).

For ages, the privileged medium of literature has been the book, and the book has been of key importance for the organization of knowledge - indeed, for the very concept of knowledge, organization, and so forth . However, while books are not becoming insignificant or superfluous anytime soon, we still have a new dominant medium for the organization of knowledge, culture, and society. Digital literature consequently has a role to play as a form of media-art that makes us aware of what is happening to text as a material and concept, to reading and writing, and to the material basis of text in the ongoing process of digitization, networking, and mediation, and how these material and formal changes correspond to social and cultural changes. The concept of text is currently undergoing dramatic changes, and most text is now produced and read at the networked interface. Text in contemporary society has become increasingly kinetic, electrified, spatial, and more or less cybernetically controlled by, for instance, commercialization in our postmodern urban environment and on the web. As pointed out by Walter Benjamin, such a development has influenced printed literature already with Mallarmé and up through the historical avant-garde (e.g., Aragon and Dada). Benjamin writes about this and Mallarmé already in Einbahnstrasse, and later he was inspired by Aragon when writing Das Passagen Werk ((Benjamin, Tiedemann and Schweppenhèauser 1980)).

Contemporary poetry can also with advantage be discussed in connection with the formal, material, and technological developments of text. In a fairly recent article, Marjorie Perloff criticizes the diminishing role of form and formal constraints in contemporary poetry. Her argument is in general that today’s free verse has freed itself from all the formal constraints that have normally defined poetry, rhymes, rhythm, meter, and caesurae. It is just lineated prose, which designates “an ironized narrative or, more frequently, the personal expression of a particular insight, presented in sometimes striking figurative language”. She does not believe in a return to classical forms, but nevertheless she wonders whether poetry is becoming expendable: If it loses its formal constraints, what would distinguish it as a genre and as an art form? Her strategy for the continuous relevance of poetry is not to take up the old forms, but enact new kinds of formal constraints in the heritage of the French-international group of writers, Oulipo (Ouvroir de littérature potentielle, “Workshop of Potential Literature”). Oulipo has from 1960 onwards worked to unleash potential literature by applying constrained writing techniques, often mathematical, algorithmic, or lexicographic, or by otherwise treating language and meaning in formal, structural ways . She emphasizes how the formal constraint in Oulipo writings “creates a formal structure whose rules of composition are internalized so that the constraint in question is not only a rule but a thematic property of the poem.” This approach to form she develops in continuation of the Oulipo law, which the Oulipian author Jacques Roubaud summarizes as “A text written according to constraint describes the constraint”.

When discussing how the computer and the web challenge our notions of literature, reading and writing, this formal perspective is important. We might not know that much about the changes brought about in our millennial tradition of reading, writing, and text to the point where we can already make clear assessments, but by experimenting with formal constraints and how these constraints become thematic properties, as suggested by Perloff, we ourselves can discover how text is treated in our contemporary literary machines. The heritage of Oulipo seems especially well suited, since it already has integrated algorithms, code, and formal logic into the world of literature. In his brilliant cultural history of code, Florian Cramer pointed out how Oulipo writers can be seen as forerunners to later and software art in the sense that they were not invested in the rationalist, modernist belief in the possibility of encoding art and aesthetics, which was predominant on the German scene in the Max Bense group:

(…) Oulipo considered formalisms and computations neither a poetic end in themselves, nor a philosophical-ideological base. Oulipo created a computational poetics as anti-computational poetics, using computational formalisms for the sole end of circumventing them artistically. (…) Oulipo made it their game to let imagination fight against self-imposed formalism and have it triumph in the end .

However, Cramer points out that Oulipo did not yet connect their algorithms to a cultural situation created by computing, as did later and software art. Today, PC and WWW have become part of our everyday culture, and their interfaces have become important aesthetic and cultural forms or media for writing and reading . Therefore we need to re-think the concept of digital literature through the interface as a concrete, technological, and theoretical framework in order to be better guided towards understanding digital literature in its current cultural environment. Instead of utopian ‘beyond-the-book’ thinking, we can with the interface as framework understand literature within a real and existing representational form which exists next to the book and even incorporates formal elements from the book . Therefore, I suggest looking at the interface as a new medium for the understanding of digital literature in order to understand more precisely how digital literature critically explores its own forms.

I will defend this perspective by looking at a specific writer and his work. The Paris-based, French writer and net-artist, Christophe Bruno, has explicitly linked his work with Oulipoan experiments (visited 3 August 2006). and his general poetics can be described in continuation of the Oulipian understanding of constraints in the sense that he plays with these in order to describe them and discover their aesthetic, cultural, and political meaning. He has made a collection of works with the heading and address It is a collection of literary experiments with the networked text of the WWW through various Google hacks, that is, Google ‘front ends’ that turn a Google search into a literary engagement with the text of the web. Bruno’s thesis is that through the web, and especially through our ability to search and monitor it thoroughly by means of Google, we are heading towards a global text that among other things enables a new form of textual, semantic capitalism, which he explores in his work. In the following, I will introduce some of his most important works on the Iterature website in order to discuss the transformation of texts, reading, writing, and capitalism currently being generated by Google and similar services as well as by general developments on the web commonly discussed under headings such as web2.0. Christophe Bruno writes about these issues himself in various introductions and essays on and notably in (Goriunova et al. 2004) and (Goriunova 2005) I will return to web 2.0 below.

2. Critical plot machines

Christophe Bruno’s Iterature stages our search for meaning and coherence on the web and in general through literary machinations or plottings. If a novel sets up a plot to form its world in a certain way and thereby makes us aware of certain connections when we read the novel and explore its plot, the machinations of Iterature are certainly plots in this sense, though like most modern novels, the plots are not always coherent but only present the sometimes melancholic, sometimes paranoid, sometimes utopian urge for this coherence.

Iterature is a still growing collection of pieces or documentations of performances, currently about a dozen, which are all written with the text from the web as material. All the pieces use search engines, primarily Google, to get hold of text floating around the web and use it as raw material for various re-workings, cut-ups, algorithmic text generations, visualizations, and so forth. Most of the pieces allow for user input, which is typically used as input for a Google search. The result of this search is then used to generate an output, which is a literary adaptation of the search result. In short, the pieces write the networked text, or set up plots that let the reader experience how the network gets connected.

The most technically simple piece, “Life,” is a staging of the basic narcissistic search, where one searches for one’s own key characteristics. It explores a narcissistic cybernetic process - an autobiography-on-demand - where one, via input in an empty text field in a predefined text written by Bruno, constructs an autobiography in the left window of the screen. Guided by the already written text around the empty fields, you typically start with your name, and date and place of birth, but further down you have to fill in more private and peculiar things, all of which trigger Google searches in the right window of the screen.

Figure 1. Christophe Bruno’s Life (by courtesy of Christophe Bruno -

During the process of searching for your own autobiography, or perhaps more precisely the text, context or pretext of your life, you might discover an encyclopedic narration through the databases of the web with your own life as the leitmotif. In spite of its technical simplicity (as just a simple front end for Google), “Life” highlights how the most personal and identifying words live their own life and form new contexts on the web. In this sense it can be experienced as a work about identity in network society, where one can explore how one’s life is linked to the web or how the text of the web has already woven itself around it. It is well-known from literary autobiographies and literature dealing with the grammatical first person that the written self is never self-identical, but always displaced and disfigured through the retrospective process of remembrance, the arranging of things in meaningful and narrative ways, and the structure of writing and language. However, in “Life” this condition is demonstrated in an unusually straightforward way, through one’s life and the events characterizing it, literally becoming nodes in a network. “Life” cannot be read as one linear text; rather it is a web to explore, to re-read, and it is not held together by one perspective but is spread out in many perspectives - one’s life and its context seen by many others. As such it demonstrates and radicalizes Rimbaud’s famous quote, “je est un autre”. The autobiography is no longer written in the self-possessed first person, but in the third person and in fact in the plural, and if you delve into this process, you potentially discover yourself to be like ‘several others’.

Even though “Life” is rather simple technically and merely stages the process of narcissistically searching for the traces of one’s own life, it becomes clear from “Life” how writing, reading, and searching are intricately connected. In addition, questions about who the author is and who controls the text come to the fore. Even though it is clear that Christophe Bruno stages the plot machines, that the user writes the input, and that the more or less anonymous masses write the output we get through Google, Bruno’s work experiments with plotting how this mass or web of text is acting and being controlled.

Google’s potentially growing control of the web is the focus of another work, “The Google AdWords Happening,” which does not work directly with the search algorithms of Google, but instead focuses on how Google makes money from our search inquiries. Google’s AdWords are the small “sponsored links” on the right side of the screen that appear with your search results and on web pages that license Google to handle their advertisements. Bruno started four campaigns on the search words “symptom,” “dream,” “mary,” and “money,” with small poetic messages and a link to his web site, and his campaign was shown 12,000 times in 24 hours. However, it was quickly censured by Google because the click-through rate was too low, which led Google to (automatically!) deem them irrelevant. Still, the happening revealed an interesting list of the estimated price of using different words as ad-words, which demonstrates that words get a price tag calculated on the basis of Google’s immense overview of our search behavior. This process demonstrates, as Bruno himself points out, the evolution of generalized global semantic capitalism Bruno points this out on the project webpage (visited 3 August 2006) and in (Goriunova et al. 2004) that is fully governed by supply and demand as calculated through the search engine. However, the search engine itself is not a commercial good on this market: it is the very technology that enables the market. Generalized semantic capitalism is, according to Bruno, generated and served by Google. As exemplified by the happening, it even performs its own effective censorship, which is not based on politics or ethics in a traditional sense, but on the effectiveness and flow of the market. The censuring of the AdWords Happening also shows that this semantic capitalism is potentially misled through ambiguities and the poetic use of language, and as a matter of fact Google even knows this somehow - hence the censorship.

Figure 2. Extract from Christophe Bruno’s AdWords Happening (by courtesy of Christophe Bruno -

The AdWords Happening explores what happens when words are no longer only part of a language (the structuralist “langue”), functioning according to a particular speech act or communicational situation (the structuralist “parole”) with a sender and a receiver, but are also becoming indexes to a database of the global text of the web, in this case Google. Instead of the rather hierarchical and homogeneous advertising structures of earlier portals and media, where comparatively large advertisers dominate with univocal messages, Google has, with its avoidance of the commercial portal format with the dominating banner ads and an abundance of highly visible services that was common before Google, moved advertising further from the broadcasting mode to a more networked and individually targeted format. We are used to words transforming into logos, slogans, brands, and trademarks which to some extent control our language and semantics, but Google and the web offer a much finer-grained and effective commercialization and control of language and semantics. Instead of having to create brand recognizance on a global scale and influence the way individual consumers react to words and carry out semantic processes, advertisers can now work in much more detail with specialized, local words and contexts. They do not have to implant themselves in the minds of everybody in order to be recognized instantly in case someone should want a soft drink or a MP3 Player; but they can market themselves ‘just in time’ via Google AdWords by making sure that they are at the top of the results when a consumer expresses his desire through a search entry. This method might spare us some stupefying exposure to mass marketing, but it will simultaneously aim at rendering our semantics more ‘effective’ in relation to its functional and marketing goals. Such a method is, in short, the effect of generalized semantic capitalism and its promotion and reinforcement of the commercial semantics of logos, trademarks, domain names, and so on. Past work that has dealt with the commercialization of words on the web is Etoy’s Toywar, which explored (and was caused by) the new discourse economy of the web fuelled by the commercialization of domain names and search words. Recently, Ubermorgen released the very sophisticated and intelligent “Google Will Eat Itself,” which “generate[s] money by serving Google text advertisements on a network of hidden Websites” and uses the money to buy Google shares, thus cybernetically turning the semantic capitalism generated by AdWords back into the capitalization of Google. The project currently predicts that taking over Google will take 202.345.127 years (, 3 August 2006). In such functional terms, semantics should be predictable, controllable, and thus univocal, which rules out irony and the equivocal explorations of language normally found in poetry and poetic uses of language. Therefore, it is significant that the AdWords Happening got censored, since it is evidence of a new function of language which AdWords exemplify but which also operates on a larger, though less controllable, scale with the Web itself and its mediation through search engines and other major entrances.

Figure 3. Christophe Bruno’s Epiphanies (by courtesy of Christophe Bruno -

3. Reading and writing the global text

As the above pieces show, Iterature is critical of the globalized text stitched together by Google from the web, and especially of the control and exploitation that Google enacts with its generalized semantic capitalism. However, Iterature is also a poetic play on the global text and the urge to obtain meaningful glimpses into it. We do not only search for clearly defined answers, but also for answers to questions we did not know existed - concerning the surprising, unknown, strangely meaningful, even the sacred and metaphysical. Pieces like “Epiphanies” enable a flaneur-like search that generates a text with formal similarities to James Joyce’s epiphanies such as they are described in Stephen Hero as a “sudden spiritual manifestation” in everyday surroundings and speech. What is more, the actual quoted epiphanies in Joyce’s text have the same “…” which typographically marks omissions in a quotation and which one also finds when Google displays its search results with bits of the text from the referred websites. “Epiphanies” directs the searches, scans the resulting text from Google, and serves it as an epiphany, which it occasionally might be. At least sometimes a “sudden spiritual manifestation” of meaning is revealed by the way the piece creates new coherences out of pseudo-anonymous found text. As in Joyce, “Epiphanies” seems to be about how a spirited reading can actually perform coherence and create meaning in a contingent and disparate situation or text. It does not suggest a preconceived metaphysical coherence, but shows and enacts how coherence is potentially constructed in the situation, in the reading. It thus demonstrates a meaning that does not stem from the predetermined, but thrives on the equivocal, the implicit, the accidental, and the associative. The epiphanies can also be contemplated in the piece titled “Fields,” where they appear as floating text fields instead of static texts.

Figure 4. Christophe Bruno’s Gogolchat (modified by jimpunk) (by courtesy of Christophe Bruno -

In another piece, “Gogolchat,” developed out of the same basic techniques, the web text is fed through a character by the name of Gogol. Besides being the Russian author of Diary of Madman, Gogol is a pseudo-anagram on the word Google, and Gogol/Google is the rambling resident of “Gogolchat”. As the polite but quite disturbing host of this chat, he communicates with whomever present about whatever they write - his speech is generated by algorithms similar to those in “Epiphanies” and “Fields.” Parts of the text one writes in the chat are used as input in a Google search which generates Gogol’s writings. This recylcing creates a communication situation where Gogol often responds to what is said by changing languages if one writes in another language or by chatting about the same subject and keywords - for example, by suddenly introducing one of the guests of the chat in the middle of what appears to be self-centered stream of consciousness. Such a situation is quoted below. In this sequence Gogol picks acquires my name from “unbehagen” (Christophe Bruno’s pseudonym) and finds a greeting from an old online posting of mine, which in this context reappears in Gogol’s discourse:

unbehagen : do you know soeren pold ?
Gogol : human beings … suspected that he was pleased being alone at YMCA … who likes The Who, Bukowski, Dostoyevsky, Gogol, Dickens and … …
Gogol : Turgenev : Virgin Soil : … Gogol’s Boulevard became such a place for me … to the un­righteous, And is also Gogol’s feeding pig … Or like me, working in a room, alone, Watching out … … The departure point of Romanowicz’s novels is her World War II concentration …
unbehagen : chatting alone with gogol
Gogol : … Dr … … Upgrade Bulletin Dear All, we are glad to inform … … Sincerely, Soeren Pold, Lecturer and researcher in literature and multimedia aesthetics … … … … …
impun< : Il etait une fois la revolu.mp3
yuban : prof. pold we know who you are

“jimpunk” is a net-artist that once ‘hacked’ “Gogolchat” by filling its interface with visual junk from the web: obscure images, animations, links, and even sounds. This hacking job was recorded and is now a permanent part of the “Gogolchat” experience. Sometimes jimpunk’s visual interference completely takes over the interface and renders the communication hardly readable. Still, Gogol rambles on - even if he does not get input. The searches that generate his output are supplemented by some default key words that make Gogol go on and on in a monomaniac way and give him a preference for slightly subversive issues (sex, terrorism, religion, pop groups and even literature…):

Gogol: Cliquez ici !Lien commercialArabe-Arabes … Arabe-Arabes … Bienvenue sur Arabe-Arabes … … … HARD-PUTE … La référence des sites sur le sexe … … … Femmes-Black … … < … … Femmes B …
Gogol : paranoia, paranoia … … Even if this is just some nutcase trying to scare people, it doesn’t help my sense of paranoia … … TIR sector contains an …

Both the interface and Gogol are so noisy and intrusive that “Gogolchat” creates its own kind of discourse. Often Gogol steals the agenda and the users continue to discuss his output, or the users actually end up trying to communicate directly with Gogol, which on the other hand is seldom satisfying. This scenario is not unlike discussions on mailing lists or chat forums with many users, where there can be many threads that entangle the discussion. In comparison, Gogol is almost totally promiscuous and constantly weaves the text of the web into the discourse with its flow of text, system messages, links, graphics, and buttons. A special discursive situation is created, influenced by the constant intervention or chatter of the web - like an amplification of the typical situation of trying to do focused work on a networked PC with e-mails, Messenger, advertising, and the constant possibility of checking the millions of persuasive blogs and websites competing to divert our attention. Instead of a traditional understanding of communication as the passing of a message from a sender to a receiver, “Gogolchat” demonstrates a highly mediated communication situation in which the medium becomes a personalized participant in the communication through the Gogol character and through jimpunk’s interface hacks. Like agents and advertising, it constantly interferes, guided cybernetically by the current act of writing and interaction, though Gogol is much more surprising and witty than commercial attempts to create cybernetic agents such as Microsoft’s Clippy (even though Clippy’s interventions are often absurd as well). Since its first appearance in 2005, Gogol has had physical brothers and sisters in “The Human Browser” project, which is a series of performances with an actor, wireless internet, and a PDA. The actor talks to bypassers, but her speech is generated by a Google hack similar to “Gogolchat” and fed to the actor through wireless headphones and text-to-speech software. Until now the PDA has normally been handled by Bruno. “The Human Browser” is reminiscent of “Gogolchat” but has the added dimension of the user being able to talk to an actual human being in a public space, which is a very strange and funny experience. Bruno has also worked consciously with site- and situation-specific issues such as performing “The Human Browser” outside polling stations at the confused EU constitution vote in France in May 2005. From the perspective of this article, “The Human Browser” demonstrates how interface literature can function as performance through an incarnated interface.

Figure 5. The result of a blog generated in Christophe Bruno’s “Dreamlogs” (by courtesy of Christophe Bruno -

With “Gogolchat” Bruno stages web text as the lines of a fictional character. In “Dreamlogs” he continues to experiment with using autobiography as a format to create formal coherence in web text. “Dreamlogs” is a piece that uses the blog format to create a special combination of searching, reading, and writing. Blogs or weblogs appeared on the web at the end of the 1990s; in 1999, the popular Blogger software appeared, which is software that helps the user to write and publish a blog, and since then a number of alternative software and services have emerged, and Blogger has been bought by Google. A blog consists of postings that are temporally categorized and can also be categorized according to self-elected groupings or “tags,” with the newest post on top to emphasize the dynamic character of the blog. Often blogs offer their readers the possibility of leaving comments and TrackBacks, which is a technology that supports bidirectional links and thus enables a communicative social space, the so-called blogosphere. Of course, blogs contain many different kinds of content, but often they contain personal comments on news, developments on the web, and life in general, sometimes with a professional view such as blogs related to academic research of various kinds. Blogs are often driven by a personal voice that comments and reflects on events in a genre that combines diary, autobiography, blatant self-promotion, and essayistic writing. Often the writing is short and rich with links, so reading a blog is a mixture of traditional reading and surfing - a way of keeping updated with the networked text. Writers of blogs are also often writers that read other blogs and participate in the dialogical blogosphere. In short, the distinctions between reader and writer are diminishing since blog software and services make it easy to publish, make comments, and configure and rearrange one’s reading material through hyperlinks and web feeds. Web feeds or RSS feeds enable feed readers or some browsers to show an updated list of the headers of a blog or a news site which links directly to the relevant articles. For further discussion of blogs, see Gurak, Laura J et. al (eds.). Into the Blogosphere: Rhetoric, Community, and Culture of Weblogs. June 2004 ( Ultimately, blogs reuse the writings of other blogs either openly as quotations or as content theft, which is when the content of select blogs is republished in order to earn money from, for instance, Google AdWords.

In Bruno’s “Dreamlogs” the user is first asked to write some words, and these yield a number of search results from Google. From these search results the user then chooses new words to generate new search results, and so on, in a process that does not end until the user writes a word which does not yield any results or chooses to stop and press “save”. Saving generates a blog consisting of texts and images related to the search words, with the latest ones at the top. The result is a blog which, consistent with the conventions of the blog genre, produces expectations of some sort of personal and temporal coherence in the text, though in this case the text is taken from the web without any coherence besides that created through the search and choice process described above. Nonetheless it is interesting to both read and write “Dreamlogs” if one engages some hermeneutic effort in the project, and it frequently generates interesting juxtapositions and findings. Besides, it explores the kind of reading that blogs generally support: an associative, often superficial reading or scanning via search engines, links, and the occasional finding. As a piece of literary art it is remarkable that it enacts reading, writing, and publishing in the same process. At its best, “Dreamlogs” is a literary tool to write a meaningful search through the text of the web.

4. Interface text

In the introduction, I briefly discussed the notion that the book was in a crisis due to the new possibilities offered by interactive hypertexts. The book is still the primary medium of literature, but as Christophe Bruno’s work demonstrates, substantial changes are taking place in the concept and materiality of text as well as in the culture of reading and writing, since it is increasingly connected to networked interfaces. These changes are significant and of great magnitude for current literary and textual culture, even though major parts of literature will continue to use the book in the foreseeable future. How these changes will influence literature in book form is a question that is beyond the scope of this article. However, media changes are complex cultural processes, and it is never just a question of exchanging one technology for another. It is worth mentioning that many literary theories, concepts, and techniques of relevance to digital literature have been developed in connection with print technology: the importance of poststructuralist theory in hypertext, William Gibson’s ‘invention’ of cyberspace in a printed novel, Oulipo’s algorithmic experiments in book formats, etc. A number of books currently discuss literary history and theory encompassing the digital media perspective to the extent that it may be possible to claim that literary history and the literary field is being rewritten from a digital perspective. Examples of this include (Hayles 2002), (Cramer 2005), and (Pold 2004).

In digital aesthetics we can now trace a development from an emphasis on the forms of reception of the individual user or reader by discussing degrees and forms of interactivity to discussions of the materiality of digital art and how this materiality can lead to new art praxes, distribution forms, and so forth. E.g., remix culture, bastard pop, mash up, cut-up videos, and on a larger cultural scale the battle over copyright versus copyleft or creative commons, software patents and free software/ open source; see, e.g., (Lessig 2004), (Stalder 2005) and (Nielsen and Pold 2005) Perhaps this gradual shift from the interactive “open work” (as Umberto Eco famously described it in 1962) to a more material and cultural openness can in fact be understood as a continuation of the poststructuralist concept of text and intertextuality. In short, there seems to be less emphasis on the openness of the structure of the single work and more on the cultural and economic function of the work. These aspects can be viewed as two sides of the same coin. And of course the interactive open work is not something that we have moved beyond, but perhaps interactivity in interfaces is, as argued by Lev Manovich, increasingly seen as a tautology, since “[m]odern HCI is by its very definition interactive” ; or perhaps the ‘ideology’ of interactivity - such as in the concept of interactive fiction - has been overblown to a degree that already in 1997, according to Espen Aarseth, it deserved a total reversal: “…interactive fiction is perhaps best understood as a fiction: the fiction of interactivity.” .

Aarseth instead proposes the concept of “ergodics,” but this concept is still preoccupied with the structural openness or dynamic character of the artifact - that every user can chose his own path in the work - and it does not fully grasp the cultural, intertextual openness that I propose here, though Aarseth does point at multiuser texts such as MUDs and MOOs. Lev Manovich proposes concepts such as “modularity” and “post-media aesthetics,” which seem more to the point , but I propose to go further back to the concept of “text” as described by Roland Barthes, since this concept has been, and I believe still is, central; despite the fact that it was developed before the personal computer and the internet became standard, it seems well suited to reflect the changes that are taking place in the concept and materiality of text.

In 1971 Roland Barthes presented a new concept of text for literary theory in the short article “From Work to Text” , continuing the work of theoreticians such as Eco, Kristeva, Foucault, and Derrida. Literary theory should, according to Barthes, move from the study of substantial, unique, and original works to the study of text as a methodological field - from the study of the author’s work to the study of the text as a material that is governed by a metonymic process, an activity of associations, contiguities, and cross-references. Whereas the work is an object of consumption and consumer culture, the text “decants the work (…) from its consumption and recuperates it as play, task, production, practice,” so that reading and writing in Barthes’ terms become linked together “into one and the same signifying practice” .

In 2000 N. Katherine Hayles commented that Barthes’ prescient essay has been highly influential, but also problematic in the sense that its generalized concept of text has led to a blindness towards the specificity of media. The poststructural and semiotic concept of text has been applied to all kinds of cultural products and across all aesthetic fields without resistance from the material dimensions of various media. Hayles suggests that “[p]erhaps now, after the linguistic turn has yielded so many important insights, it is time to turn again to a careful consideration of what difference the medium makes” and proposes a medium-specific analysis . But perhaps we can still use Barthes’ concept of text to discuss digital literary works and their medium-specific materiality; in other words, use Barthes’ concept with new medium-specificity?

Digital art and literature such as Christophe Bruno’s works described above share a number of characteristics with the concept of text, whether they express themselves through alphabetical text or other multimedia. In addition, digital artifacts consist of text in the form of source code, and the struggle over whether this source code should be proprietary, patented, trademarked, unreadable or open, readable and (re-)writeable open source is a struggle that could be (partly) understood in Barthesian terms. See, e.g., (Cramer 2001) and (Bertelsen and Pold 2003). With its focus on text as an intertextual flowing material and its integration of reading and writing, “Dreamlogs” is almost a literal staging of the concept of text as introduced by Barthes. It points to a development where the poststructuralist concepts of text and intertextuality are carried out with a remarkable literalness. Instead of being applied metaphorically to works by Joyce or Proust, it is literally true for Christophe Bruno’s works that “the Text is experienced only in an activity, in a production. It follows that the Text cannot stop (for example, at a library shelf); its constitutive moment is traversal (notably, it can traverse the work, several works)” . Just think of “Epiphanies,” “Dreamlogs,” or “Gogolchat,” which also have the associative, rambling, differentiating character that Barthes invokes by comparing text to an idle stroll:

“the reader of the Text might be compared to an idle subject (…): this fairly empty subject strolls (…) along a hillside at the bottom of which flows a wadi (I use the word to attest to a certain alienation); what he perceives is multiple, irreducible, issuing from heterogeneous, detached substances and levels: lights, colors, vegetation, heat, air, tenuous explosions of sound, tiny cries of birds, children’s voices from the other side of the valley, paths, gestures, garments of inhabitants close by or very far away; all these incidents are half identifiable: they issue from known codes, but their combinative operation is unique, it grounds the stroll in a difference which cannot be repeated except as difference.”

“Gogolchat,” “Dreamlogs,” and “Epiphanies” are literally texts “entirely woven of quotations, references, echoes: cultural languages (…) antecedent or contemporary, which traverse [them] through and through, in a vast stereophony.” In fact, one does not read Iterature as much as play it in the polysemy of the term Barthes suggests: the text plays, the reader plays at the text, and he plays the text in the sense of executing it. However, one should be aware of reading Barthes too literally, since he is not pointing to a new object (and of course could neither foresee the web nor imagine Iterature), but to a methodological field. The work still exists (as a concept, a metaphor, or an ideal) as “the Text’s imaginary tail”: “the work is held in the hand, the text is held in language” .

How should we understand this textual character when considering Iterature? Iterature’s enactment of the Barthesian concept of text is literal, not theoretical, metaphorical or ideal in any sense. Consequently, it does not fulfill any utopian notions of a potential anti-authorial liberation of text from the homogeneous work. The intertextual weaving of references is in fact carried out by computers on the web, and it is not an idle reading freeing us from the fixed paths of the work. We must take into account that the text of Iterature is controlled and generated by algorithms and computers mostly beyond Bruno’s control - some at Google, some at the websites that Google indexes, and some at - and these literary pieces are something other than traditional search results or web pages. In fact, Bruno’s algorithms serve to make us realize and witness the instrumentalization of text on the web and at Google; they also show us how Google’s index and search algorithms turn all of the web into one big global pool of text that can be analyzed and quantified, taken apart, measured, and tracked in many ways before being synthesized again by being stitched or woven together anew by links, indexes, mappings, search engines, and other text generators in, for instance, dynamic web pages. When we do a normal search, we do not fully realize this instrumentalization, but Bruno’s algorithms stage and narrate this as autobiography, epiphany, and chat communication. In this way, his algorithms make us marvel at the global text so that we can read it and read it in new and interesting ways, often through chance encounters and strange cut-ups; but it also lets us deconstruct the phenomenon when we try to figure out how this text is generated - for instance, in an attempt to discover who or what is writing.

The global text might pose problems for the passive consumption that characterizes the consumer economy, as can be seen from the textualization (in Barthes’ sense) of music through remix culture and bastard pop (cf. ), but it can also feed its own new economy, which is suggested by “The Google AdWords Happening” and Bruno’s subsequent concept of a generalized semantic capitalism. The semantic capitalism of Google AdWords is created through intertextual linking and association structures that surf smoothly on the global text, but the semantics are still one-way: click here, buy this! If more open poetic semantics are introduced into this, the direct marketing value drops - the system becomes sloppy, which is detected by Google’s algorithms. Introducing a ‘textual’ reading into the marketing scheme leads to censorship, and the fact that even this censorship is enacted by an algorithm only adds to the critical irony! In fact, “The Google AdWords Happening” illustrates how Iterature tackles the global text of Google by using the system in a non-intentional and open way. Instead of obeying the rules, it plays them and thereby lets the reader enjoy the algorithms while simultaneously deconstructing their functionality.

5. The global text or web 2.0

Iterature reflects a situation in which web text is beginning to act as a generalized global text technically removed from its borders and framings such as pages, authors, readers, linearity, stability, or beginnings, and ends. The text has become a system by technical means. Iterature explores this system by staging it and using the system in non-intentional ways, thereby openly demonstrating this new character instead of just masking it as functionality - for example, as contextualized advertisements, links, indexes, collages of texts functioning as summaries, and so forth. The duality played out by Iterature of, on the one hand, textual systems or tools for text and, on the other, new textual ways of publishing, reading, and writing is not only a feature of Iterature, but a general new trend that is increasingly acknowledged under headings such as Web 2.0 or social (networking) software. See (O’Reilly 2005) and (“Wikipedia”) (“web 2.0” and “social software,” retrieved 18 Jan. 06).

Web 2.0 is a rather loosely defined concept for a number of sites and services that use the web as a platform for communication, collaboration, and sharing. It ranges from business and company models that use the web as their platform and build on user-generated content (with Google as one of its key examples) to “a social phenomenon referring to an approach to creating and distributing Web content itself, characterized by open communication, decentralization of authority, freedom to share and re-use, and ‘the market as a conversation’ ” (Wikipedia, (Retrieved 18 Jan. ‘06)). Web 2.0 and social software ranges from establishing new kinds of technologies and businesses and hyping these as a new boom economy to cultural and social movements and practices creating digital commons, critical reflection, and public spaces. Consequently, in contemporary net culture we see the same mix of textual systems and textual use as pointed out above in relation to Iterature. This is also described in Jill Walker’s recent article on these new web developments, where she argues that hypertext has gone wild or “feral.” She does not explicitly mention the concept of web 2.0, but she mentions some of the same applications as O’Reilly, such as the photo-sharing website flickr ( and the social bookmarking website ( At around the same time as blogs became popular, peer-to-peer (P2P) networks such as Napster emerged and had great effect on discussions of copyright and file sharing. In pure P2P networks there are no central servers or routers; instead, peers act as both clients and servers, and therefore the network itself is emergent. She claims that pre-web hypertext was largely “a domesticated species bred in captivity,” self-contained and with authorial control intact. She also draws attention to the paradox inherent in George P. Landow’s13 Cf. (Landow 1992). convergence between hypertext and contemporary literary theory, and discusses the contradictions or even reversals of poststructuralist literary theory in hypertext with reference to intertextuality:

…[T]he concept of intertextuality and much other late twentieth century critical theory expresses an idea of texts as unruly and fundamentally beyond discipline. Much hypertext research, on the other hand, attempts to find ways to discipline and tame our thoughts, at the same time as it admits that our mind works associatively and that there are multiple ways of viewing connections in texts.

The kinds of feral hypertexts she refers to are “the large collaborative projects that generate patterns and meanings without any clear authors or editors controlling the linking” and “these systems work because they have simple but flexible ground conditions that establish environments that make emergent organization instantly visible” . One of the techniques often used - for instance, in flickr and - is user-created tags, the so-called folksonomies (a combination of “folk” and “taxonomy”), where users are free to create their own tags, attach as many as they like to each item, share them with others, and search for similar tags. In this way one can easily and effectively, but also idiosyncratically, create an emergent user-created taxonomy in a user-created database consisting of, for instance, photos (flickr), bookmarks (, or academic references (CiteULike- Walker describes how these are closer to associative links and Barthesian distractions than the traditional authored links of hypertext or the professionally cleared trails of Vannevar Bush, the grandfather of hypertext:

These links are not paths cleared by the professional trail-blazers Vannevar Bush dreamed of, they are more like sheep paths in the mountains, paths that have formed over time as many animals and people just happened to use them. Once formed, it is easier to use such a path than to blaze a new trail. .

According to Walker, an effect of folksonomy is that many tags are idiosyncratic and ambiguous, an outcome which is not avoided in an effort to distinguish between different meanings of the same word, as the ideal of the semantic web would have it . In fact, this open taxonomy also creates meaningful coincidences that have either poetic potentiality or significance on a larger social and political scale. An example of this is the “Bush” cluster in flickr: its neighboring tags are “protest,” “war,” “iraq,” “march” and so on, and in this way a certain context is created around the word “Bush” which reflects the political situation. There is also a function in flickr that makes it possible set up and define a group, which in many cases creates a themed photo exhibition out of the image collection.Good examples are the numerous groups of images that illustrate poems or quotations from literature such as Kafkaesque images that alludes to the atmosphere of Franz Kafka’s work (there are currently more than 4000 photos). Another interesting group of images are the images of writing machines. Bush clusters:, Kafkaesque:, Writing Machines: [all visited 3 August 2006]. Of course. this is not a fully bottom-up emergent structure, but still everybody can propose tags and groups, decide on the openness of the group, and, in the open groups, join and post what they wish. Nonetheless, the open folksonomic taxonomy potentially leads to heterogeneous collections and impossible classifications like Borges’s Chinese encyclopedia, which opens Foucault’s The Order of Things - classifications that are not ordered with stable relations and categories . As in language, semantics, and culture, the categories and their relations are dynamic, emergent, and created by context, contiguity and poetic plays on words, as well as on hierarchy, similarity, and the preconceived, lexical meaning of words. Such open and emergent dynamics are essential to the cultural effect and use of the web, to the development of a net(work) culture.

6. Iterature’s treatments of text

In the light of these developments, we have gained a more qualified understanding of Iterature. Iterature demonstrates and reflects on these developments, which are on the one hand technical in nature, and on the other hand are related to the culture of reading, writing, and publishing. Of course, the technical side implements and supports the cultural side, while the cultural side develops the techniques through use and renders them meaningful in a cultural execution of technical, formal algorithms. It is a dialectical development process, both in general (think of the development of hypertext and the web) and in particular with web 2.0: services such as, flickr and Google keep adding functionality by studying user behavior, perhaps starting with some playful ideas and treating users as co-developers in a so-called “perpetual beta” release process . But as argued by Walker and others, sometimes the technical side also reverses the cultural side, such as when the semantic web aims at specifying and standardizing language to get rid of ambiguity. In fact, has a small comment on semantic web and XML on its front page under the heading “XML: the end of iterature?” Perhaps if the semantic web really worked smoothly it would mean the end of the poetic displacements and ironies of Iterature. For a discussion of the semantic web, see (Berners-Lee 2001) and Wikipedia (e.g., “The Semantic Web,” “XML”).

Iterature consists of works that explore a new economy: how text is distributed, rewritten, and re-contextualized by search engines and the net. Bruno’s work uses the enormous database of text and the way Google makes it accessible. It uses the text of the web in algorithmic plot machines which stages in various ways what happens to text when it escapes the boundaries of the book and the page and flows kinetically and cybernetically in the network. Even though the text we read is primarily written by humans and not by computers, it becomes removed and alienated from its immediate context and put into a new one by Bruno’s search and cut-up algorithms. It is a literary staging of the text of the net and of the user’s search project, and through this literary re-contextualization and alienation the text is removed from its immediate and natural context, allowing us to see beyond the immediate utterance towards the enunciative situation and the textual character and its material and behavior.

Iterature’s output is no more alienating than the many textual screens greeting us in urban spaces or the text we normally get from a search engine or a dynamic 2.0 website, but Iterature’s literary staging forces us to reconsider the textual character and material more thoroughly, since the message is beyond the immediately decodable and recognizable. This is where Iterature unfolds as a critical insight into a networked and digital concept of text and into Barthesian textualization - a critical insight encompassing the character of all the text surrounding us that has escaped the page and the book. The internet and the automatic overview created through the linking of the search engines deliver the textual raw material; Iterature merely displays, displaces, and iterates this. By its very title and the omission of the “l”, Iterature invokes a position between literature and computerized iteration. Iteration in programming designates the repetition of a process from an initial set of conditions until a terminal condition is met, and it is a fundamental part of many algorithms, often expressed in loops . Iterature explores a border area between what is literature and text to the human reader and how computers treat text; an area we experience whenever we read and write in connection with machine readers and text processors on the internet such as search engines, machine translation, automated publishing and linking, dynamic web-pages, web 2.0, weblogs, social software systems, e-business systems, advertisements, marketing, and meta-tagging as it is envisioned in the semantic web project. is therefore an example of contemporary interface literature - a literary treatment of networked text that creates a critical experience and points to potentials other than the orderly, functionalistic ideas of the semantic web, or the semantic capitalism of web 2.0 and other attempts at domesticating and taming the world wild web.

Thanks to Stacey M. Cozart for proof-read and corrections.

Works Cited

Barthes, Roland. The Rustle of Language. Trans. Richard Howard. Berkeley & Los Angeles: University of California Press, 1989.

Benjamin, Walter, Rolf Tiedemann, and Hermann Schweppenhèauser. Gesammelte Schriften. Werkausgabe. ed. 12 vols. Frankfurt am Main: Suhrkamp, 1980.

Berners-Lee, Tim, James Hendler & Ora Lassila. “The Semantic Web.” Scientific American. May 2001 2001: 34-43.

Bertelsen, Lars Kiel, and Søren Pold. “Hvad Viser Skærmen? - Interfacet Mellem Referentiel Og Medial Transparens.” Kunsthistoriske Kontakter. Ed. Anne-Louise Sommer & Hans Jørgen Frederiksen. Aarhus: Aarhus Universitetsforlag, 2003. 97-116.

Bolter, J. David. Writing Space the Computer, Hypertext, and the History of Writing. Hillsdale, N.J: L. Erlbaum Associates, 1991.

Bolter, Jay David, and Richard Grusin. Remediation: Understanding New Media. Cambridge, Mass.: MIT Press, 1999.

Cramer, Florian. “Free Software as Collaborative Text.” Sarai Reader 01: The Public Domain. New Delhi & Amsterdam: Society for Old and New Media de Waag, 2001. 199-206.

—. Words Made Flesh - Code, Culture, Imagination. Rotterdam: Media Design Research, Piet Zwart Institute, 2005.

Foucault, Michel. The Order of Things: An Archaeology of the Human Sciences. Reprint ed. New York: Random House, 1994.

Goriunova, Olga. Readme 100 Temporary Software Art Factory. Dortmund: Books on Demand GmbH, Norderstedt. & Hartware MedienKunstVerein, 2005.

Goriunova, Olga, et al. Read_Me Software Art & Cultures. Aarhus: Digital Aesthetics Research Centre, 2004.

Hayles, N. Katherine. “Flickering Connectivities in Shelley Jackson’s Patchwork Girl: The Importance of Media-Specific Analysis.” Postmodern Culture 10.2 (2000).

—. Writing Machines. Mediawork. Cambridge, Mass.: MIT Press, 2002.

Howe, Denis, et. al. “Free on-Line Dictionary of Computing”. 1993-.

Landow, George P. Hypertext the Convergence of Contemporary Critical Theory and Technology. Baltimore, Md: Johns Hopkins University Press, 1992.

Lessig, Lawrence. Free Culture: How Big Media Uses Technology and the Law to Lock Down Culture and Control Creativity. New York: Penguin Press, 2004.

Manovich, Lev. The Language of New Media. Cambridge, Mass: MIT Press, 2001a.

—. “Post-Media Aesthetics”. 2001b. 13 March, 2006.

Mathews, Harry, Alastair Brotchie, and Raymond Queneau. Oulipo Compendium. Atlas Arkhive; 6. London: Atlas Press, 1998.

Nielsen, Henrik Kaare, and Søren Pold. Kulturkamp.Com - Mellem Åbne Værker Og Intellektuel Ejendomsret. Vol. 9. Århus: Center for Digital Æstetik-forskning, 2005.

O’Reilly, Tim. “What Is Web 2.0 - Design Patterns and Business Models for the Next Generation of Software”. 2005. 18 Jan. 06.

Ong, Walter J. Orality and Literacy. London & New York: Routledge, 1988.

Paulson, William. “Computers, Minds, and Texts: Preliminary Reflections.” New Literary History 20 (1989): 291-303.

Perloff, M. “The Oulipo Factor: The Procedural Poetics of Christian Bok and Caroline Bergvall.” Textual Practice 18.1 (2004): 23-45.

Pold, Søren. Ex Libris Medierealistisk Litteratur, Paris, Los Angeles & Cyberspace. Odense: Syddansk Universitetsforlag, 2004.

—. “Interface Realisms: The Interface as Aesthetic Form.” Postmodern Culture 15.2 (2005).

Stalder, Felix. Open Cultures and the Nature of Networks. Kuda.Read. Frankfurt am Main: Revolver, 2005.

Walker, Jill. “Feral Hypertext: When Hypertext Literature Escapes Control.” Proceedings of the sixteenth ACM conference on Hypertext and hypermedia. Salzburg, Austria: ACM Press, 2005.


Aarseth, Espen J. Cybertext Perspectives on Ergodic Literature. Baltimore, Md.: Johns Hopkins University Press, 1997.