«The artists try their willingfull mastership of the unwillingness as what distinguishes them from dilettantes.»

Theodor William ADORNO, Aesthetical Theory


Systemic art triangle

To a certain point of view, all work of art is specific. However if it was so, it wouldn't be any other thought on art that this one which would oblige us to admit its unicity and its ineffable appearence. Because, at the same time, all work of art is - and in an unique movement - exhibition of a context that allows itself to be. This visible contradiction proceeds from the fact that all work of art is the objective moment of a dynamic system, the peculiar representation in a definite context of the interactions of all the variables that contribute to its being : the form is the content of the tensions that appear in the evolution of the system, an instable stability. The same and the different. It is why art - as says Adorno in his Ästhetische theorie - may not be thought in terms of communication, has nothing to say, and can never deliver messages : a dynamic space may only be define under the point of view of an inner observer bound to that space, and not from its outside. In a proper sense, art is an in-formation, an internal construction of in formations free for all other finality : it is by itself an homogeneous system of signs.

Even if it is not here the place to develop that theme, taking it into consideration is essential to the understanding of the historical changes of the work of art notion, as this of its creative posture. Art always participates, in the same manner and without any separation, to the subjective, the objective and the cultural fields as to the whole instances through which, in a definite moment, all these components appear. This cluster of interactions define what may be - in a some metaphorical and simplifying way - called the «systemic art triangle».

Figure number one

The artist shapes the balance of the tensions which he strives to make objective. If there is some rationality in the work of art, it is in the conscious thinking of these tensions : that rationality is inner to the work whose greatest purpose is to impose itself not as representation but rather as the invention of a specific system of signs - one sui generis language -, adequate to the moment, by this way made visible, out of a complex environment.

Maybe it is why too many aesthetic approaches were not convincing enough because they don't completely analyse the whole consequences of such a system.

Information age

The age of information may be defined by different displacements and innovations which at the same time affect several variables of this systemic triangle. Technical innovations, but also social transformations and symbolical changes each of them producing different consequences on all the others. It is not surprising that art, system of information without any pragmatical purpose, will be interrogated, in its most fondamental concepts, by the setting up of a culture where information and communication are the first industrial material. Such is the context in which the mutations of the contemporaneous artistic creation, in its search of an adequate mastering of the tensions which try to dominate it, have to be understood.

The age of information is the age of communication systems which set up the technical treatments of information and also of communication. Therefore, what distinguished it is what may be called a complete marketable communication. All its numerous consequences may be classified in different types of questions representatives enough to exhibit the principal axes of the main question.

The first one, wich lays the foundation for the information age is of a tech-symbolical order.

- at the symbolical level, it consists of considering that all the information classes, including the most complex, may be represented by a hierarchy of N-articulations with the simplest one based on a single binary opposition 0 -1. This symbolical system is the simplest, therefore the most abstract, never invented.

- at the technical level, it consists of the invention of the electronics, and all the others things which came from there, which allows, in a more efficient and quicker way, digital data treatment.

Art of the information age betrays what founds it

So, art is provoked on its own ground : permanent place of invention of non-pragmatical information systems, it is challenged, as never before in its history, by the pragmatical use of abstract information systems with no primary links to any reality. Artistic creation can only try to invest this place which traditionnally was its one. Out of this attempt it may be deprived of its soul and lose its justification. To say this differently, contemporaneous art, even if it has always been concerned by technic of which it ever knows how to isolate some elements to divert them to its own purposes, is now in front of a specific technic of which it can no more isolate the aspects it wants but to which it has to confront as a whole. In the digital field art is, indeed, completely technical and - as it was shown by Pierre Lévy -, by some aspects of its algorithmic approach, the technic seems to follow an artistic comportment. A digital image, for instance, is not simply a new technic of creating images, but a completly new image. It constraints art to a new conceptualisation of the notion of image itself. The french producer Michel Jaffrennou, who realised the movie Peter and the wolf in digital images and with whom I actually work on our generative opera Blue Beard says that «in the digital image all pixel is visible». Thus he shows that in that kind of images, at the level of the visible, the presence of the digital is very strong, and modifies deeply all the perception modes : «It is why visibility of the image becomes a legibility» (Gilles Deleuze in Pourparlers). And what is legible in the digital image is the whole of the technical layers which made it. This intrication of visible and legible, among different other constraints, is what the contemporaneous artist has in his work to think. So, all the first exemples of image creations using new technologies as a simple tool, and that have not seen such a deep change, have been artistic failures. There is no middle way : or art put the total questions of digital image or digital image is a simple fabrication technic for the cultural market, that means a sub-product of art aimed to its commercial vulgarisation.

The difficulty of such an exercise of definition is therefore that the concepts wich describe the digital art should not be isolated. Each of them is strongly linked to all the others upon which comprehension and extension depend. Digital art now has indeed to face all the aspects of an information system with the aim of following the principles which made artistict legitimacy, to objectify it out of its peculiar field.

Information age art betrays its foundations : the most symbolic image would certainly be this strange and false Moebius band that represents the drawing hands of M. C. Escher, and where in an infinite loop, the drawing creates itself.

Escher reproduction

In that sense, all digital artistic attempt is global, it enlists for its own purposes simultaneously and in an inseperable way all the concepts coming from the digitalisation of information : setting an image on the web has an artistic meaning if, and only if, the setting of that image responds to all the symbolical constraints of the net. So, Reynald Drouhin, a young french artist, grabs sounds, pictures, texts, videos on the net to create his own work - www.ensba.fr\alteraction - or, in another way, Fred Forest, an artist who works on what he calls the art of communication, has sold by auction in 1997 the access code to a digital image. The buyer of this code possesses this work as a real one and makes what he wants with it. Thus the image is no more an image but becomes a demonstration of artistic change of the concept of net. In order to see all the consequences of that action in the artistic field, it is very interesting to know that the company which bought that code, considering its free quotations on different television channels, multiplied by eighty the amount of its primary investment. Without its inscription inside the globality of such an action, the digital images on the net are only pictures and do not question the net in the systemic rules of art point of view.

An hypertextual approach based on numerous links would be more convenient to present these concepts. Different attempts in this way does exist, for instance the CD-ROM Actuality of the virtual published in 1996 by the National Museum of Modern Art from the Centre of Industrial Creation of the Georges Pompidou Centre in Paris or also on varied websites like Instruments and shapes of the interactivity made by Douglas Edric Stanley (http://www.labart.univ-paris8.fr). But a linear presentation, with a deductive approach, is not really congruent which isolates each concept in its own territory. They, fundamentally, are systemic concepts - that means, each of them is made of classes of strongly linked sub-concepts.

Digital art is an art of techniques

What distinguishes digital art is first the use of a technical treatment. So it seems congruent to analyse it through its instrumentation. All the concepts of the digital art may be so, at a first level, assembled into six classes of operations : hypermediatisation, conceptualisation, generation, performance and memorisation. Each of these classes includes some secondary concepts some of which interact in other operations.

Hypermedia

Digital art is the simulated moment of an absent matter

Because digitalisation - a symbolical treatment - is the heart of digital art, this art took its origin from a rupture : it is an art without matter. That doesn't mean that in its outside manifestations it is realized out of any matter. Indeed it is recorded on different materials and, to become visible, has to invest the matter of different spaces - screens, environments, volumes, and so on. Above all, the lack of matter of the digital art shows that, by nature, all its productions have to be thought out of a pragmatic connection to the matter. In that sense, digital art takes a form only through simulacra : digital art is the simulated moment of an absent matter

This situation put it in a paradoxical position similar to these of theatre or opera : when published one play doesn't differ from all the rest of literature, it is only a possibility of play which took really a meaning in the different actualisations which are the performances which make it always new. One written play is the prototype of a performed play. In the same way, one generative novel exists only by and through the displaying apparatus which shows it. So, according to its place of exhibition, it can assume different appearances : public exhibitions, individual readings or performances. To give a concrete example, my generator Three Mythologies and one blind poet, does actually exist under three shapes : a performing one for the IRCAM in 1997, a private one on micro-computer for individual reading and a public one on the net for the website of our university. Each of these shapes presents really different characteristics. However, it is exactly the same poetical generator of which none of the elements of its program have been changed.

Visibility of a digital work is a potential one

The visibility of a digital work is a potential one, as Pierre Lévy says, it is «a numerical reservation of sensorial and informational virtualities which only actualises in interaction with human beings» (Cyberculture, p. 173). So, digital work never depends on its visibility which can takes different shapes. Its visibility is only a surface, like a skin under which the spectator doesn't know what really exists, exactly as, between the performance of an actor and the text he is playing, a kind of separation takes place.

A digital work is multiple

That characteristic of the digital work can be called hypermediatisation : the work is potentially multiple, virtual, therefore out of location. So, being not assigned to any location, it can, so long as all the locations, to which it is allocated, do accept numerical in formations, invest all of them. Also, for the first time in history of human creation because its materialisation is only a surface out of visibility, a work can, by the net, exhibit itself, without any loss, on each point of our planet. While being one alone work, it can also exhibits itself differently depending from each place of displaying. That situation asks questions to the notion of culture itself, therefore the notions of universal and totality.

A digital work has the capacity of ubiquity

The digital work has the capacity of ubiquity. Moreover, all the works present themselves simultaneously, without erosion on each place of the universe, their manifestations have therefore the capacity of creating an event for the whole human community of which they unceasingly reorganise the landscape of significations. It participates to the shape of the rhizome : digital work can spread and reproduce itself by diffusion on all the nodes to which it has access. It is not surprising that the net was the natural space of aiming and expansion of the digital works. Theoretically, the net has even the capacity of being one unique large work of art.

For the same reasons, the aspect of its access to visibility depends only on local and temporal decisions taken by one spectator or from one given place. The separation between text-picture-sound-music becomes a theoretical one. The same program can be at the origin of completely different or mixed surface effects.

A digital work summons multiple sensorial potentialities

Thus, hypermedia works out the old dream of a total art : it unifies what was disunited before. As, for instance, the opera which, in its ways, tries to reach that aim by a simultaneous presentation in the same location, for the same purpose, of elements of heterogeneous components, a digital work always summons, in the same location, the potentialities of multiple sensorial events.

But contrary to the opera, it summons them from one unique management centre : it is no more a simultaneous co-ordinated presentation, but merging and inter-operability. In the hypermedia, sound is, in a circular way, a part of the text as the text is a part of the picture... More precisely even, because these varying sensorial manifestations are nothing else than various codifications of digital data, «a sound - parodying Gertrude Stein - is a text is a picture is a music is a movement is an image is a sound is a picture is a...» and - maybe soon - a smell

Figure number two

In our digital opera Blue Beard, for instance, music, text and scenarios are not the junction of three different artistic wills made by a foreman who is the director, but the simultaneous production of three indivisible generators, one text-generator, one scene-generator and one music-generator all conducted by the same program and by the same intentions.

Of course, such an approach is not without incidences on the definition itself of these media and on their modalities of reception : what becomes one text if it is also picture and music, what is stage-craft, and so on... All these queries are today in the heart of the work of many digital art artists. Nowadays works appear which, as soon as they are conceived, are the result of tight collaborations.

Modelling

Modelling, foundation of digital art

Modelling is the foundation of digital art. How, indeed, in a system which is essentially information, an object may be conceived which has no shape or which, more exactly, may take all the possible shapes. An object the shape of which is only a symbolic one. Digital art wants what has never been; but all that he is was before it. It has to construct the appearances of what, one time, was.

Digital pictures plays with their appearances

The notion of matter is a good example of that problematic : how digital art, immaterial by nature, may represents the matter, or how may a digital picture objectifies itself without using one presentation of matter ? The matter of digital image is a theoretical one. At the contrary, for instance, of the cubist picture which minds so much of matter that, in its works, it wants to put real pieces of it, the digital picture knows, by definition, that it can only play with its appearance. The matter of digital art is built two times. First by its creator who has to accomplish a theoretical conception, then by its reader who faces illusions of matter which displays itself and that he cannot read except through that convention he can accept or reject : the digital picture always tests the cooperative will of its reader, the prime confidence that the spectator gives to all message. Art is mimesis : from the conscience of imitation, he has necessarily to achieve the reasoned construction of it, that means to translate it into concepts.

Digital matter is a concept-matter

So, all digital matter sets up the concepts which found it : it is built memory of repetition and differentiation of all the matters invoked. Even with ordinary scanning, the final picture has passed through numerical filters, which change it and also through the parameters that its maker uses. Digital matter is an ever stronger image. Thus it is a critical matter, something like a thought-matter. All the other matters known before concentrate in it. It is both a deeper and idealistic matter. Digital matter is a concept-matter.

A digital image is more than a picture

A digital picture is always more than a picture.

Digital art is an art-concept

Digital art is an art-concept, not one conceptual art which thinks itself through the discourse which brings it, but an art built on the conceptual analysis of the systems of signs which become its object. This doesn't mean that art doesn't think itself before, but that digital art, refusing all non-formal approach does, in the thought it is building of itself, let no more place to ineffable.

Digital art is a modelling art

Digital art is an art of the model, at the mathematical sense of the word. To actualise, it needs absolutely before to conceive one formal abstraction. All its possibilities will then follow.

Such is the meaning of the virtual : one spring of possible objectifying actions, one endless principle of actualisations, all of them written down into the model.

Yet, the great difference which separates this notion from its traditional philosophical meaning is that, in that case, the model is not related to ideas but formal, calculable, operative : this model has to be described in formal languages external to its proper object. This principle attends the temptation of infinite regression : one model may always be taken as one concept for another model of a superior level. Two virtual universes may so remain unknown from each other except if one model of a superior level defines their connections. The map is never the territory, the model doesn't exhaust the object it is modelizing. As shown by the Mechanical models of Claude Cadoz and Jamel Nouiri, almost perfect engines except that they never will be complete - unless to materialize themselves into real engines, what will be giving up their numerical aim.

Digital art depends always on selecting operations. If it naturally aims to simulation, it contains also the risk of simulacrum, the inversion of the idea of model itself. To built a model is to conceive the concepts useful for a pragmatic representation of one world. Therefore - unavoidably - to take some distances with regard to the concrete realities of that world. The worlds of the digital art may not be only abstractions of worlds because their creator doesn't create the result of the actualising but models. So in the models of the digital world, it is this creator who, because he translates his ideas of one world, modelizes in part himself. He has to be able to define, in a technical manner, at the same time, how he conceives his art, the resources he needs to do that, the place he takes and his relations with the different modes of display. In spite, sometimes, of the appearances of identities in the results, it is what fundamentally separates on one hand concrete music - or the contemporaneous practice of sampling in the techno-music - to which the composer takes some sensitive fragments of music that to mount them into his own work; and on the other hand electro-acoustical music, where sounds are defined only by groups of parameters open to different calculations.

Generation

Digital art aims to repetition

Model and mould have the same etymological origin : they are both the materialisation of concepts of objects intended to their mass production.

If according to Gilles Deleuze, «repetition is said from elements which are really distinct and which, however, have strictly the same concept» (Différence et répétition, p.26), the digital art aims to repetition. But, that repetition is of the order of the concept, not of the surface. What is implicated is the distance between concepts and surface. The creator who originates a concept of work can not master all the parameters which will make the work itself. He works on an idea of work, but of a work he wants not to realize directly because what interests him, it is the relations to his concepts. The representations of the digital art require to be not fixed, otherwise they are only week imitations of works, only simple objects of archives. And the fact that the productions of a model can be completely forecasted is in discrepancy with the necessity of permanent changes. Without this possibility, without a paradoxical amount of random, the digital art has not a real meaning. One permanent and fixed digital work like one serigraphy or one monotype is in contradiction with the matter itself that it shapes : that means in part the time, in part the space, in part the system of signs itself. In that sense, what is the most productive, is not the repetition of the same, but this one of the very same which, under the reminders of identity, shows inequality.

Digital art plays the same and the different

It is through the dynamical variations of the same and the different that the digital art acts on the minds of the people who perceives it. In the models of the digital, time and random have a very important place. Random, because it is at the opposite of the formal modelling, is maybe the obvious manifestation of the simulacrum generated by the model. In that case, the problem of reproduction is no more set because each moment of the production is, by itself, one re-production. They can not be copies. The copies of the numerical world are copies of the same under the different. What the spectator sees, is the rhythm, the semantics of the same and the different.

From its beginning, the work is thought as an open class, generic, of copies each of them being yet an original. The works of Maurice Beanyoun - Is god flat ?, Is the devil curved ? or The tunnel under the Atlantic - for instance, are of that kind : the spectator is placed in position of infinite repetitions of events and each micro-event is always different. More, at a certain level, these three works are identical - they allow to dig in the cultural matter - and deeply different because it is the relation to that cultural matter which, each time, is set in a new action.

That necessity of variation finds its best example in the notion of generation : one generative work is a work which, each time, re-produce itself identically and, each time, at the level of its actualisation, is displayed in an original manner.

The numerous creations of virtual worlds are of that kind : a virtual world indeed is a world which simulating the accidents of life has to be able to generate new situations depending from its unforeseen events.

The simulations of artificial lives - although built on very different algorithmic approaches - are also of that sort : they create works which evolve and modify themselves without any other intervention of their creator than the conception of the model which generates them. The numerous Genetic Images of Karl Sims (Particle Dreams, Genetic Images, Primordial Dance, Panspermia, and so on)

Figure four

or the installation Mutation of Yichiro Kawaguchi are rather good examples of that kind of works.

Figure five

All these works are generative in the sense that their variations are infinite : it is completely impossible that the result could be the same twice. The generativity is so defined as the process of materialisation and achievement of incomplete elements of different levels corresponding to one conceptual description and driven by one central model. Generativity questions, in a radical way, the notion of the author, more exactly, of the relationships between the author and his work, and between the work and its reading for, in most of the traditions, the work is always read through the personality of the author himself.

A generative novel, for instance, is a novel of which the pages are written nowhere before the precise moment of their displaying on a screen which can be of different types : collective screen versus individual screen, isolated screen versus screen on a net, and so on... The material that the writer works is the dictionary, the rhetorical rules, the syntactical rules, the more or less complex representations of universes : all this built the model of the novel to write. The written pages are strictly dependant from that material. However, the writer can not foreseen them in the final shape of their displaying. He knows approximatively what can be foreseen, he knows that they can describe such or such event, speak of such or such character, but he can not in any way says what will be their exact content. Out of all reading, the generative novel may produce infinitely texts by itself or, on the contrary, produces pages only when a specific demand is made. Depending on various strategies which are part of the true conception of the work, its pages can be kept or destroyed.

Page of An unfinished debate

An infinite debate, for instance, is so programmed to destroy itself as soon as it will have written three hundred thousand pages. The readings of these pages are so very varying and different for one reader or for another depending on the context in which this reader has been set in contact with the work and especially depending on the moment of that contact : it is completely impossible that, even on the net - except if they are together in the same time and in the same place, in front of the same screen - two readers may read the same page of the novel. One reader may never, except if he had the possibility of recording it, read two times the same page. Two readers do never read the same novel. While reading one common work, each reader is the reader of an unique work.

The last word belongs to digital work

For the same story they are as many different views as different readers. In another one of my generative novels, ROMANS (Roman) - NOVELS (Novel) -, four generators exchange their in formations and, by their transversal readings, they create new novels about these characters that the writer himself had nor foreseen nor, exactly speaking, programmed.

The author exists : without him, ROMANS (Roman) would not be possible, and it is really the author who - according to Pierre Lévy - defines an horizon of meanings to the work, but what is moving, is the relation between the author and the work. Not because in the classical conditions of creation, hazard had never nothing to say, but because until now - as shown by the surrealistic strategies of choice - the author had always the last word. In the digital work of art, the last word belongs always to the work because, when its manifestations are put in action, the author, except if he changes the model, is only in the position of a reader who can no more change the result beingactualisedd. What he attends, is then something like the objectifying of the ideal aiming of his work.

A digital work is a process-work

Digital work is a process-work. A flood-work : it integrates time and context as fundamental parts of the artistic expression. To the fractal conception of the work of art in which structure is the most emblematic representation, the conception of an artistic shape like a deterministic repetition of strongly assembled fixed elements, a digital work of art substitutes chaos in its physical meaning, that means the constant possibility of local divergences which make the final result of this complex process completely unforseenable.

A digital work asserts the temptation of infinity

A digital art of work asserts the temptation of infinity : its process would never have to end... Just like life...

Performing

The digital artist shapes worlds

Maybe, this demiurgic claim founds all digital work of art : to simulate the life processes. Avoiding the religious vocabulary, Pierre Lévy speaks of worlds engineers but they are false ones. The artist creating digital art formalises only his worlds. Not because his imagination, different from others, is more abstract but because what he creates are possibilities of interactions into different systems which sometimes seems to be self-driven. If life is also essentially information and if digital art is information about information, is it so surprising ?

Digital art assumes the risk of the event

Maybe, more than the temptation of creating life, what digital art wants to conceptualise first is the pair of game and risk. Fascination of real time : digital art is an art which, like dripping or some others uses of lyrical abstraction, but with really different purposes inside the systemic art triangle, assumes the risk of events. As with television, the work occurs in direct and the spectator attends that occurrence. But this real time is this one of reception, not the electronic one of physical processes.

Digital art is spectacular

This real time is the one that the spectator can seize not the true speed of the physical engine running : the reality of that time is itself part of the scenery. The digital work leads to the perpetual motion in which texts, music and images, between simulacrum and simulation, always rebuild themselves : above all, digital art is spectacular.

Digital work has a talent for ubiquity

But, because the digital art can be actualised in very various places, under different and simultaneous modes of distance and nearness, it is a show without stage. Jump cut / Faust - show produced by Marianne Weems from the U.S. Builders association - is a good example of that. On the same stage, but with multiple screens, in various kinds of uses, the show burst in different changing scenes, their divergent modes of reading obliging the spectator to a constant zapping. So he has always to implicate that centrality of point of view which, until today yet, define the reading of works. Here, the centre is everywhere. On the net, the stage is the whole world. This affirmation is not a metaphor : today, the same work can be simultaneously displayed under different shapes on all the screens of the earth. A digital screen is nothing else than the visible skin of an infinity of others, the digital work has, by nature, the power of ubiquity. In a way, it tends to surround its spectator, to immerse him into a creative universality where «the observer is himself a part of the simulacrum» (Deleuze, Difference and repetition).

The degree of attention being proportional to the degree of immersion, the more the spectator is inside the work, the more he takes care of it. Internet, for instance is a communication all to all, each person can alternatively plays all the parts. The large number of creation websites shows what is happening : creation and reading tend to converge, maybe to merge. To produce a digital work obliges to ask this question : how may the work build and integrate its spectator ?

Digital work overflows the world

These are the fundamental questions of interaction and interactivity. Before all, the digital work is a whole of mathematical variables. In that sense, it is open to any other homogeneous mathematical variables. Two works can interact. It is enough that some of these variables are coherent. Their interactions built a new work. In Three mythologies and one blind poet, for instance, the show was only depending on the possibility of communication, on a coherent semantical way, by choosing the common parameters and their running level, the music generator modelized by Jacopo Baboni-Schilingi and my own generator of poems. As for Blue Beard, the gathering of the three generators - music by Alexander Raskatov, scenery by Michel Jaffrennou, and text - is not much more complex : it is enough to know at which level and when these three generators will exchange in formations on their respective states and what kind of parameters will be used for that. Indeed, that way of working changes the traditional of cooperation between the creators. But, what is more interesting is that the digital work allows a very large universe of cooperations which permits integrating - as a true part of it - some of the formal parameters of its contexts. Built on models, it is available to exchange. Also many works can use their environment as a part of the work itself : light, temperature, noises, presence of the spectator, live cameras, and so on. The choice is only depending on the intention and on the imagination of its creator. In Three mythologies and one blind poet, the Midy instruments, played by musicians, had a part in the interaction : when its interprets were playing the work, the digital work was playing its interprets.

Digital work risks interaction

The digital work can also ignore its context, live its life alone. But, in a certain sense, it misses one of its principal aims which is to simulate risk and game. It is like any other performing approach. Because the purpose of interaction is there of prime importance.

Digital work builds its spectator

For the same reasons, the digital work has the capacity of building a theoretical spectator, a spectator who is part of itself. It is only a matter of parameters : it is enough to put among the variables of its model, one model of spectator. Of course, the part of the spectator, by the reflexions on perspective or on scenery for instance, has always been taken into account by the artistic creation, but here, to model one spectator is not a theoretical matter, it is a formal concrete one because that means with which real parameters the presence of that spectator would be integrate into the mathematical model of the work. Often, for instance, some critics asked me why my generative novels are not interactive. Good question... Because they are. But they are interactive in two very precise manners. First that generator never writes a page without a clear request, the spectator is so inside the model as a starting factor and decides of his own reading rhythm. Then these generators are interactive in a finite sense. For instance in my detective novel Pray for murder, the reader can choose at which moment of the action he wants to be. But he does that as a blind man. Indeed, he doesn't know the novel structure and has only the choice between 1 and 10 possibilities, what is completely arbitrary. He can choose the beginning , the middle or the end of that novel, but as the generator generates the story second by second and on a great number of points of view, even that power is a false one. The interactivity is always a disappointing one. The generator plays with that spectator who is a part of its parameters what, of course, is contrary to the ordinary meaning of interactivity. For most of the critics indeed, interactivity is the possibility that the spectator has to participate to the creation of the work.

Nothing more false. Interactivity and generativity are like cat and mouse where the author playing with his spectator take the absolute right to mock him. The interactive spectator is not out of the work but, inside, as part of the model, exactly as an other :  the work thinks and builds him. All other mode of interactivity is an alibi because it lets its spectator believe that he has some power on that work when he has only the power that the creator of the model agrees. Even virtual reality is a false and played one. If the industrial simulators - flight simulators, driving simulators, and so on  - aims to give to their users a maximum mastership on the processes, it is because they don't have the smallest artistical will. In that case, the author disappeared to let the place to an only objective relation. Simulation is a matter of engineers, their problems are never inside the systemic art triangle.

Experimentation

The principal question is then : what happens in digital art that changes all the usual relationships between the work and its uses and why these changes ?

That question opens on two sorts of problems : the one concerning the link between work and its reader, and the one of the links between the work and the socio-economical situation.

Digital work always changes

To the first kind of problems, the answer generally made is the open work, the reference of which is Umberto Eco's book. But for Umberto Eco, the opening of the work proceeds from an interpretation, never from a modification of the work itself. In the open work approach, the work is fixed and its readings, alone, are open. As for the digital work of art, it is not only unstable, but it also accepts - is made for - constant changes. So, believing that between the changes of the work and its reading there is an other relation is a strong temptation. However, the digital art is not only more open than the non-digital work, but, by its constant movableness, it lends really not to hermeneutics. For interpreting the digital work, its reader has indeed to access not only to its surface but rather to what founds it, its mathematical model : an open work would be one where the reader could introduce personal parameters into the creator's model. This is impossible. Interactivity never leads to a deep modification of the work, it only allows to choose among the visible and authorized indeterminations of the model. If the steps of one dancer change the music, if the movements of the spectators change the projections of an installation, it is only because the model of the work includes the steps of that dancer in its own parameters, because this of the installation includes the movements and because, at this level, the author, while determining the boundaries, decided to let a place to choices. There is no example, because it is theoretically impossible, of a model which could change itself depending from the unforeseen choices of an undetermined interactor. The reader never becomes an author, there is no collective creation of the work. When it happens, like in collective writing, the approach is totally different and the numeric has there no specific part; most of the time the purpose is only to give to a group of authors - for instance on the net - one adequate writing tool. This defines none of the aspects of the digital art.

The hermeneutics of digital work is at a technical level

The hermeneutical attitude is also difficult because hermeneutics implies debates, that means one fixed element on which the debate can be. In the digital work, the only fixed corpus is generally inaccessible to the spectator model. It is like if discussing the meanings of a Faulkner's novel, its readers have to debate, not on the text of the novels because no reader has read the same, but on his work notes. The hermeneutics of digital work only has its place at its technical level.

Digital work is a work of repetitions

On the other hand, the digital work is a work of repetition. That doesn't mean that it repeats itself, but that, through the returns of the same, it is open to the building of the same and the different : «If repetition does exist, said Deleuze, it expresses at the same time one singularity against the generality, one universality against the peculiarity, one remarkability against the ordinary, one instantaneousness against the variation, one eternity against the permanence. In all respects, repetition is transgression.» (Différence et répétition, p.9).

Digital art incites to experimentation

So, in its returns, its differences and its repetitions, the digital art invites to experimentation. Not to the experience as defined by John Dewey, that subjective and global approach - but to the experimentation. That means the capacity of doing, from the observation of repetition, a selective test which built liberty. If the digital works asks to its spectator to act on it, it is not to give him the impression that he has the power, that he is the authority, the author in place of the author, but because, through the variations made by repetitions on the surface of the work, is the only possibility to deeply understand the aesthetic system of that work, to perceive how it works and, in that peculiar case, how it builds itself. Therefore what it means, on what possibilities it opens, how it changes the relations inside the art and with the world.

Some contemporaneous forms of art like pop art or, by some of its aspects, new realism may have given a similar impression - the recovery appearance of some works of Rauschenberg, the provocations of John Cage, the accumulations of Arman, for instance, while deleting their techniques let their spectators believe that they would be «able to do the same things» - but their bases are quite different. Reintroducing one visible techniques - something like the touch of Brueghel or Cézanne -, the digital art doesn't want to open on a ridiculous illusion of practice, it never says to its spectator to take the part of creator, but, by advancing the experimentation, it invites him to try to understand from the inside the conceptual steps followed by the author.

Digital art is a game

In this sense, digital art is as a game. A true player is somebody who assimilates the rules, not the one who conceives new games. He is somebody who knows the rules so well that they become part of himself. But the game is too much pragmatical : in the experimentation offered by digital art, nothing to win except experimentation for itself, more exactly the discovery of its pleasure. From the game, digital art preserves only the personal, individual and collective, absorption.

Aesthetic of digital art is an aesthetic of science

What digital says is, surely, the renunciation of representation, but also, and in the same time, the inscription of the life of the spectator himself as formal element of the manifestations of art. The work of art is no more an object of which only the external representations are catched, but a living process in the simulation of the true conditions of life. The dynamism of the digital art is never a multiplicity of representations. By means of experimentation, he wants to reach something like a scientifical approach of the aesthetical problems : the digital art is the art of the technique. Not as design, art serving technique, but art of the means of the technical advance itself and for itself. Another time, Gilles Deleuze : «The scientist brings out of the chaos some scientifical variables, the artist some artistical variety». The digital art wants to conciliate both. It wants to be at the same time, art of the technique and of the knowledge. What its creator hopes to modelize is, paradoxically, what he never seen. It is certainly why finding its balance is so difficult.

Memorising

At this point begins the second socio-economical point which questions the place of art inside society and, especially, its conservation.

If the aesthetical pleasure is lived through the mediation of an object, because this experience may be indefinitely renewed, the conservation of that object, as for a private than for a collective appropriation is absolutely necessary. What the museums say, forgetting the permanent changes inside the historical relations to what is today called an object of art, is with no doubt, come here in front of the objects, in the situations that we try to make the best, and live the aesthetical experiences that, so many people, before you, have lived. The experience can not be transmitted, but it can be reproduced. The mercantile investment in the work has no other justification : the object is expensive because it is unique and can not be replaced, the experience it allows can not be made without it. So this object presents an economical interest because that experience, renewable by each of them, can produce returns on investments. Especially if, even if the relation to the original is the strongest, an approaching experience can be also lived in front of a good copy of that object, as shown by the multiple realisations of virtual places - Lascaux cave, for instance. Theses bases justify the art business. And the banks locks up works of art in their coffers.

Digital art complicates these rules a little.

Fixed digital work is like any other object of art

First what it allows is experimentation. Not experience. That is very different. Essentially contextual, therefore made for installations and performances, the experimentation can be realized only in the context for which it has been built. What imports is the process, not the object which, for a while, gives a matter to that process. Nothing remains of that. There is surely always the possibility of memorizing a print of the process, to take a picture or to record it, but nothing more deceiving because when the digital art is fixed, it becomes as any other object and belongs to a completely different aesthetic approach. One videotape, even less one picture, from the Messenger of Catherine Ikam and Jean-Baptiste Barrière for instance, can never give the least idea of the situation that the interactions between the spectator, the falsely living face of the messenger, the changes of the music and the movements of the spectators create : all that this spectator experiments. What stays is only what has been changed in the individual critical memory.

Digital art has no value

More the less, in the infinity of original moments of production, when the numerical model may always create new ones, why to keep that track more than one another ? During the exhibition Les immatériaux in 1985, my poetry generator Renga wrote more than thirty-six thousands of poems. All of them where kept by the Georges Pompidou Centre, but what will be the interest of that keeping when the program may restart and produce a new infinity of others ? Why the number three-hundred and fifty would be more interesting than the number forty two thousands ? Why two seconds of the Tunnel under the Atlantic of Maurice Benayoun, recorded on October the 12 th 1996, would be more interesting than two other ones produced in 1999 ? Where is the real value of such tracks ? Where is the value of the digital art ? Modern art, with installations, or with Christo's wrappings, for instance, as yet asked these questions but because, all things considered, it created static objects of which fixed representations can be realized, they were relatively of less importance. What can not be then reproduced was only the precise context. A new wrapping can always be made... The digital art gives another dimension to that problem : it is completely impossible to only exhibit a work of digital art, it is always necessary to immerse in it, to navigate, to interact, to experiment its processes inside the exact environments made for it.

Flow

Before our eyes, the status of the work of art is changing.

For many centuries and still to a certain extent, the world was in what can be called the «religious posture» - a medium of transcendence where the object has no other value than to make signs for and to an invisible beyond... Then came the «aristocratic posture» where the work reflected the hierarchical importance of the person who ordered it, followed by the «middle-class posture» where being is confused with having and the value with the stock. Nowadays the work of art aims to take what can be called the «financial posture» : the work has no more value as an object, its value is only in the capture of its perpetual movements.

Digital art plays with life and death

The work of art is no longer in the object but in the possibilities of the processes that these objects allow, having has less importance than seizing. To produce is no more to reproduce but to re-produce : nowadays digital art work tries to provoke the experimentation of perpetual news events of which each spectator is the only one who can keep record of.

In his Aesthetic Theory, speaking of the works of art, Theodor William Adorno said : «Their proper life feeds itself with death». A digital work of art does not render the modernistic notion of art death but rather always plays with its life and with its death.