---

S'asseoit dans sa salle de séjour devant sa télévision au son baissé, ne dit rien... Revenir dans la pièce principale de sa maison, souffrir d'un manque, se promener dans la pièce, avoir mal à gauche, attendre... Tu ne penses pas, tu regardes, tu enregistres. Se lever sans but défini, se lever tout simplement sans le vouloir, jouer avec un fusil à lunettes jouer, avec un fusil à lunettes, imaginer derrière les murs une mer de cocotiers, monter dans sa voiture pour aller draguer au bois de Boulogne, appeler le 0142171793, faire des recherches sur la révolution, aller jusqu'à un radiateur... Nous prenons garde à ce que le côté obscur des événements ne nous submerge pas. C'est vraiment sûr ? Peut-être ! Comment faire échapper le goût de la réalité au réalisme ? Ne compte pas une distance dans l'espace : celle à qui j'écris n'est pas une jeune fille. Peine d'un oedipe mal soigné, j'aurais aimé rendre à chacun ce qui est à chacun. C'est en méditant sur la page que certains y parviennent. Retirer ton texte. Je ne cherche pas à m'engager, mais à me dégager. Imaginaire : une lettre dans une langue humaine... L'écriture est un désespoir. Lettre retranchée du zéro. Se souvenir de la voix. Une lettre à propos imprévu. On pourrait imaginer une lettre ! C'en est fini. Réfléchir vite n'est pas réfléchir. Juste aligner son écriture sur la vie. Il y a longtemps que je m'en torche, il ne faut pas donner une couleur vert dollar aux phrases, un début est autre chose qu'une origine. Certitude sensible, désespoir de la modernité, conditions de jugement du goût, nécessité immanente, déception du sublime, aptitude à juger, philosophie de l'art, tradition aristotélicienne, réalité mutilante et mutilée, loi du moindre effort, réconciliation de l'esprit et de la nature, raison. Littérature tournant politique de l'esthétique. Imitation de la nature. Indifférence à la perfection devant la beauté. Liberté de la volonté, pulsions inconscientes, sentiment du sublime : la mort noire mène son troupeau de danseurs colorés.

---

It might be only by the virtue of metaphors that I speak in a conference about the chaos theory : I'm not a mathematician and my personnal use of chaos, fractals, non linear models and complexity concepts is certainly incorrect. You try to analyze our world, I'm trying to create one world among a multitude of possible ones. The best would be for me to explain what I try to create and show my work.

I like the infinity of shapes in regard to the multiplicity of things. What concerns me, is the infinite !... Not duration, eternity; but the inexhaustible : infinite.

The infinite of creation : infinite in litterature. And, for that reason, the new possibilities that computering offers to litterature.

First chapter : one purpose

The litterature I like is a litterature which refuses stability, which does not close up the time of its narratives in one unique line, but allows an infinite of them. That is what I call a chaotic, non-deterministic litterature, as opposed to a teleonomical litterature, oriented to an unique aim and in which each fragment depends directly on that unique aim.

To illustrate that affirmation, I have prepared for you a generator of novel in english. This generator is poor, because its english dictionnary is poor, I apologize for that but all my other exemples will be in french...

As it is easier to speak about well known things than about unknown ones, I will first define this litterary approach only by negatives proposals :

— it is not a random litterature; or if there are some random aspects in it, it is only because "the instability of non-random dynamic leads to the appearing of random" (Y.G. Sinaï)
— its texts cannot be read two times in the same way
— its texts cannot be memorised
— its texts cannot be studied : no princeps-editions, no various readings, no manuscripts
— its texts have no memory or, more exactly, if they have one, their memory is really different from that we are accustomed with
— preserving its texts has no sense.

The artiste always fight the chaos.

In the traditional view of litterature, he tries to master it by maping one line, something like an algorithm between some of the spots which set it up : litterature is deterministic... I use to say : fractal. But that is another subject...

With the computer, litterature can work the chaos in the way of chaos itself : taming the order of disorder. Making of litterature a written work where causes, seeming to be originally independent, mix their effects to built a totally new meaning.

A litterature which is sensitive to contexts, to its lecture's media, to its lecture's time, not closed in the fixity of a written apparatus : a really interactive litterature...

Let us speak about theory, just some theory...

I think that there is a difference between meaning and coherence. Meaning is related to the real world and to this world's actions. Meaning also needs coherence but coherence does not need meaning. For instance a discussion in the real world has to be coherent and mean something, that is to be congruent with the context where this conversation takes place; a conversation in a fiction has only to be coherent because its meaning is created in one possible world among others.

So I can define a fiction as a coherent trajectory drawn into a space of available "informations".

In the "fractal" litterature, the writer thinks that there is only one trajectory and all his work is to avoid all possible deviation — or more likely to make believe that he acts so.

In the litterature I want to produce, I let the generator lay down its own trajectory and, eventually, all the trajectories it may lay down while crossing the space of "informations".

What these informations are is not really important. Imagine that they are crumbs of something which may have a meaning. For instance : "crash|es", or "fur|s", or "the bulk of the central ridge", or "[1|m_bulk] [7|a_central¶m_ridge]", or anything that may be registered in the space of informations datas and that respects the rules of the program.

All these informations registered in the space of a novel constitute an incoherent space of meaning. If analysed in terms of coherence, something like a space of unstable coherence.

To generate in this space means to find one possible trajectory between all the informations.

The result produces the text :

"8 - 15 - 20 - 13 - 2 - 9 - 4 - 11 - 16 - 17 - 22 - 29"

The trajectory is always influenced by the information from which it starts and by the state of the information it reaches.

Third chapter : technology

These generators are based on what I call a semantical grammar.

Each time a data is choosen, there is a probabilty of choices which depends on :

— the number of data available in the data base corresponding to the node of the graph being expanded
— the state of the text's memory
— the contextextual representation of the data

It is why I call this model of litterature a "chaotic litterature" : we start from an unstable space where they are some probabilities of coherence but, once a trajectory begins, it is impossible to predict exactly where it ends though all the trajectories are coherent ones.

And, as the reader projects the text in the coherence of his world and keeps it in his memory, there is no possibility of reversibility : once a first text is written all the other texts are read through it, even if two readers will read to different texts of the same novel.

The question I was asked was: "How can digital poetry as language art in digital media deal with the demands to represent and reflect the specific mediality : by working in a "radical" or "concrete" way with programming languages and the codes upon which they are based; or by focussing on an aesthetic of the screen which takes the perceptions and activities of the user as its point of departure?"

To try to explain what kind of "specific mediality" implies digital poetry, and more generally digital literature, I would focus my five minute speech on two main points :

The first one is that the screen is not the only media for text in digital literature, a computer indeed is a structure made of three obligatory components which are relatively independent :

_the first one is the heart of the computer, a totally abstract machine which, using different kinds of algorithm, one that can work on any information made of sequences of zeroes and ones. It transforms this information in various ways.

_the second one is the input interface which, depending on the tools used, translate any kind of information in a language that the brain can understand. Input interfaces may be camera, microphones, infrared cells, etc. There is so no limit to varieties of inputs.

_the third one is the output interface which, depending also on the technique used can translate any kind of information produced by the computer brain in information for any kind of media. The screen is one of them, but is only one among many possibilities. And if it is important for digital literature, it is not the only way to mediate digital literature, sound is another very important one but movement can be also one. In that point we reach a deeper question which is : what do we call literature and what do we call poetry. Today the answer is not so easy. But I don't have time enough to analyse it here.

That point is very important because we have to consider the screen only as one among many other possibilities used to mediate digital literature. In my personal work, I try to explore all these different possibilities. For example, I call interaction the communication between two computers producing both some effects, a text generator working and exchanging information with a music generator to produce something new with or without a screen. That point answers to a part of the question : digital literature is always, but in different ways, related to languages and codes even if that link is not directly used by the writer itself and, for instance in text-generation, it is not so easy to separate the work on natural language of the work on the digital one.

The second main point is that a screen is nothing.

When I said that a screen is nothing I really said "nothing material" : a screen is just the place where at one peculiar moment light may be projected. That seems to be of no importance. On the contrary, it is a fundamental remark, it implies indeed that :

_what imports is not the screen itself which is just a surface used by light at a specific moment and in a specific context, but the light itself and the information it may contain.

_a screen is a virtual space, that really means that everything has the capacity of being a screen : human body, mountains, water, walls, air, clouds... even the sky... and the moon if we had projectors enough powerful.

_a screen is a temporary place, that means that it is naturally not made to keep permanent information.

_a screen can be a place among other to perceive and sometimes experiment interactivity.

These four points can be considered as the basis for an aesthetic of the screen. Indeed they open a great wide range of possibilities  : to play with literature as with moving pictures, - films in the best cases, movies in the others - ; to relate literature to time, that is not only to movement, but to time as ever changing infinity of potentialities ; to use literature as a process, as a show, as an environment, as an installation, as an architectural design, etc. ; to use e-literature as a component of other shows, for instance dance, events, theatre, music, etc. In all these cases, the main question is how to keep to literature is own aims, what the aesthetic of literature can bring to theses contexts and what the other way of expressing something bring to literature.

In conclusion, I would insist that the real problem does not concern techniques and media but our conception of literature and maybe, essentially, because poetry is more concerned by forms and more systemic, our modern conception of poetry. In that case indeed, the screen is nothing more than one of the numerous possibilities of outputting literature from the computer and more generally of outputting literature out from six or five centuries of books. I think that the real and interesting problem is what set of really new possibilities computers offer to literature and what kinds of literature can be created from them.