Foucault's philosophy is often presented as an analysis of concrete “dispositifs” or apparatuses. But what is an apparatus? First of all, it is a skein, a multilinear whole. It is composed of lines of different natures. The lines in the apparatus do not encircle or surround systems that are each homogenous in themselves, the object, the subject, language, etc., but follow directions, trace processes that are always out of balance, that sometimes move closer together and sometimes farther away. Each line is broken, subject to changes in direction, bifurcating and forked, and subjected to derivations. Visible objects, articulable utterances, forces in use, subjects in position are like vectors or tensors. Thus the three main instances Foucault successively distinguishes - Knowledge, Power and Subjectivity - by no means have contours that are defined once and for all but are chains of variables that are torn from each other. Foucault always finds a new dimension or a new line in a crisis. Great thinkers are somewhat seismic; they do not evolve but proceed by crises or quakes. Thinking in terms of moving lines was Herman Melville's operation: fishing lines, diving lines, dangerous, even deadly lines. There are lines of sedimentation, Foucault says, but also lines of "fissure" and "fracture." Untangling the lines of an apparatus means, in each case, preparing a map, a cartography, a survey of unexplored lands - this is what he calls "field work." One has to be positioned on the lines themselves; and these lines do not merely compose an apparatus but pass through it and carry it north to south, east to west or diagonally.

The first two dimensions of an apparatus or the ones that Foucault first extracted are the curves of visibility and the curves of utterance. Because apparatuses are like Raymond Roussel's machines, which Foucault also analyzed; they are machines that make one see and talk. Visibility does not refer to a general light that would illuminate preexisting objects; it is made up of lines of light that form variable figures inseparable from an apparatus. Each apparatus has its regimen of light, the way it falls, softens and spreads, distributing the visible and the invisible, generating or eliminating an object, which cannot exist without it. This is not only true of painting but of architecture as well: the "prison apparatus" as an optical machine for seeing without being seen. If there is a historicity of apparatuses, it is the historicity of regimes of light but also of regimes of utterances. Utterances in turn refer to the lines of enunciation where the differential positions of the elements of an utterance are distributed. And the curves themselves are utterances because enunciations are curves that distribute variables and a science at a given moment, or a literary genre or a state of laws or a social movement are precisely defined by the regimes of utterances they engender. They are neither subjects nor objects but regimes that must be defined for the visible and the utterable with their derivations, transformations, mutations. In each apparatus, the lines cross thresholds that make them either esthetic, scientific, political, etc.

Thirdly, an apparatus contains lines of force. One might say that they move from one single point to another on the previous lines. In a way, they "rectify” the previous curves, draw tangents, surround the paths from one line to another, operate a to-and-fro from seeing to speaking and vise versa, acting like arrows that constantly mix words and things without ceasing to carry out their battles. A line of forces is produced "in every relationship between one point and another" and moves through every place in an apparatus. Invisible and unspeakable, this line is closely combined with the others but can be untangled. Foucault pulls this line and finds its trajectory in Roussel, Brisset and the painters Magritte and Rebeyrolle. It is the "dimension of power" and power is the third dimension of space, interior to the apparatus and variable with the apparatuses. Like power, it is composed with knowledge.

And finally, Foucault discovered lines of subjectivation. This new dimension has already given rise to so much misunderstanding that it is hard to specify its conditions. More than any other, this discovery came from a crisis in Foucault's thought, as if he needed to rework the map of apparatuses, find a new orientation for them to prevent them from closing up behind impenetrable lines of force imposing definitive contours. Leibniz expressed in exemplary fashion this state of crisis that restarts thought when it seems that everything is almost resolved: you think you have reached shore but are cast back out to sea. And as for Foucault, he sensed that the apparatuses he analyzed could not be circumscribed by an enveloping line without other vectors passing above and below: "crossing the line," he said, like "going to the other side"? This going beyond the line of force is what happens when it bends back, starts meandering, goes underground or rather when force, instead of entering into a linear relationship with another force, turns back on itself, acts on itself or affects itself. This dimension of the Self is not a preexisting determination that can be found ready-made. Here again, a line of subjectivation is a process, a production of subjectivity in an apparatus: it must be made to the extent that the apparatus allows it or makes it possible. It is a line of flight. It escapes the previous lines; it escapes from them. The Self is not knowledge or power. It is a process of individuation that effects groups or people and eludes both established lines of force and constituted knowledge. It is a kind of surplus value. Not every apparatus necessarily has it.

Foucault designates the apparatus of the Athenian city-state as the first place of creation of a subjectivation: according to his original definition, the city-state invents a line of forces that moves through the rivalry between free men. From this line on which a free man can have command over others, a very different line separates itself according to which the one who commands free men must also be master of himself. These optional rules for self-mastery constitute a subjectivation, an autonomous subjectivation, even if it is later called on to furnish new knowledge and inspire new powers. One might wonder whether lines of subjectivation are the extreme edge of an apparatus and whether they trace the passage from one apparatus to another: in this sense, they would prepare "lines of fracture." And no more than other lines, lines of subjectivation have no general formula. Cruelly interrupted, Foucault's research was going to show that processes of subjectivation eventually took on other modes than the Greek mode, for example in Christian apparatuses, modern societies, etc. Couldn't we cite apparatuses where subjectivation no longer goes through aristocratic life or the aestheticized existence of free men but through the marginalized existence of the "excluded"? The sinologist Tokei explains how freed slaves in a way lost their social status and found themselves relegated to an isolated, plaintive, elegiac existence from which they had to draw new forms of power and knowledge. The study of the variations in the processes of subjectivation seems to be one of the tasks Foucault left those who came after him. I believe this research will be extremely fruitful and the current endeavors towards a history of private life only partially overlap it. Sometimes the ones subjectivized are the nobles, the ones who say "we the good..." according to Nietzsche, but under other conditions the excluded, the bad, the sinners, or the hermits, or monastic communities, or heretics are subjectivized: an entire typology of subjective formations in changing apparatuses. And with combinations to be untangled everywhere: productions of subjectivity escaping the powers and knowledge of one apparatus to reinvest themselves in another through other forms to be created.

Apparatuses are therefore composed of lines of visibility, utterance, lines of force, lines of subjectivation, lines of cracking, breaking and ruptures that all intertwine and mix together and where some augment the others or elicit others through variations and even mutations of the assemblage. Two important consequences ensue for a philosophy of apparatuses. The first is the repudiation of universals. A universal explains nothing; it, on the other hand, must be explained. All of the lines are lines of variation that do not even have constant coordinates. The One, the Whole, the True, the object, the subject are not universals but singular processes of unification, totalization, verification, objectification, subjectivation immanent to an apparatus. Each apparatus is therefore a multiplicity where certain processes in becoming are operative and are distinct from those operating in another apparatus. This is how Foucault's philosophy is a pragmatism, a functionalism, a positivism, a pluralism. Reason may cause the greatest problem because processes of rationalization can operate on segments or regions of all the lines discussed so far. Foucault pays homage to Nietzsche for a historicity of reason. And he notes all of the importance of epistemological research on the various forms of rationality in knowledge (Koyré, Bachelard, Canguilhem), of socio-political research into the modes of rationality in power (Max Weber). Maybe he kept the third line for himself, the study of the types of "reasonable" in potential subjects. But he refused essentially to identify these processes in a Reason par excellence. He rejected any restoration of universals of reflection, communication or consensus. In this sense, one could say that his relationship with the Frankfurt School and the successors to this school are a long series of misunderstandings for which he is not responsible. And no more than there are universals of a founding subject or exemplary Reason that would allow judgment of apparatuses, there are no universals of the disaster of reason being alienated or collapsing once and for all. As Foucault told Gérard Raulet, there is not one bifurcation of reason; it constantly bifurcates, there are as many bifurcations and branches as instaurations, as many collapses as constructions following the cuts carried out by the apparatuses and "there is no meaning to the statement that reason is a long story that is now over." From this point of view, the objection raised with Foucault of knowing how to assess the relative value of an apparatus if no transcendental values can be called on a universal coordinates is a question that could lead us backward and lose its meaning itself. Should one say that all apparatuses are equal (nihilism)? Thinkers like Spinoza and Nietzsche showed long ago that modes of existence had to be weighed according to immanent criteria, according to their content in "possibilities," freedom, creativity with no call to transcendental values. Foucault even alluded to "esthetic" criteria, understood as life criteria, that substitute an immanent evaluation for a transcendental judgment every time. When we read Foucault's last books, we must do our best to understand the program he is offering his readers. An intrinsic esthetics of modes of existence as the final dimension of apparatuses?

The second result of a philosophy of apparatuses is a change in orientation, turning away from the Eternal to apprehend the new. The new is not supposed to designate fashion, but on the contrary the variable creativity for the apparatuses: in conformance with the question that began to appear in the 20th century of how the production of something new in the world is possible. It is true that Foucault explicitly rejected the "originality'' of an utterance as a non-pertinent, negligible criterion. He only wanted to consider the "regularity" of utterances. But what he meant by regularity was the slope of the curve passing through the singular points or the differential values of the group of utterances (he also defined the relationship of forces as distributions of singularities in a social field). By rejecting the originality of utterances, he meant that the potential contradiction of two utterances is not enough to distinguish them or to indicate the newness of one in relation to the other. What counts is the newness of the regime of enunciation itself in that it can include contradictory utterances. For example, we could ask what regime of utterances appeared with the French Revolution or the Russian Revolution: the newness of the regime counts more than the originality of the utterance. Each apparatus is thus defined by its content of newness and creativity, which at the same time indicates its ability to change or even to break for the sake of a future apparatus unless, on the contrary, there is an increase of force to the hardest, most rigid and solid lines. Since they escape the dimensions of knowledge and power, lines of subjectivation seem particularly apt to trace paths of creation, which are constantly aborted but also taken up again and modified until the old apparatus breaks. Foucault's as yet unpublished studies on the various Christian processes will certainly open many directions in this regard. One should not believe, however, that the production of subjectivity is left only to religion; anti-religious struggles are also creative, just as the regimes of light, enunciation and domination move through very diverse domains. Modern subjectivations resemble the Greek subjectivations no more than Christian ones; the same is true of light, utterances and powers.

We belong to these apparatuses and act in them. The newness of an apparatus in relation to those preceding it is what we call its currency, our currency. The new is the current. The current is not what we are but rather what we become, what we are in the process of becoming, in other words the Other, our becoming-other. In every apparatus, we have to distinguish between what we are (what we already no longer are) and what we are becoming: the part of history, the part of currentness. History is the archive, the design of what we are and cease being while the current is the sketch of what we will become. Thus history or the archive is also what separates us from ourselves, while the current is the Other with which we already coincide. Some have thought that Foucault was painting the portrait of modern societies as disciplinary apparatuses in opposition to the old apparatuses of sovereignty. This is not the case: the disciplines Foucault described are the history of what we are slowly ceasing to be and our current apparatus is taking shape in attitudes of open and constant control that are very different from the recent closed disciplines. Foucault agrees with Burroughs who announced that our future would be more controlled than disciplined. The question is not which is worse. Because we also call on productions of subjectivity capable of resisting this new domination and that are very different from the ones used in the past against the disciplines. A new light, new utterances, new power, new forms of subjectivation? In every apparatus we must untangle the lines of the recent past from the lines of the near future: the archive from the current, the part of history and the part of becoming, the part of analysis and the part of diagnosis. If Foucault is a great philosopher, it is because he used history for something else: like Nietzsche said, to act against time and thus on time in favor, I hope, of a time to come. What Foucault saw as the current or the new was what Nietzsche called the untimely, the "non-current," the becoming that splits away from history, the diagnosis that relays analysis on different paths. Not predicting, but being attentive to the unknown knocking at the door. Nothing reveals this better than a fundamental passage from The Archaeology of Knowledge (II, 5) that applies to all his work:

“Analysis of the archive therefore includes a privileged area: it is both close to us and different from our current time. It is the edge of time that surrounds our present, overlooks it and indicates its alterity; the archive is what, outside of us, delimits us. The description of the archive unfolds its possibilities (and the mastery of its possibilities) starting with discourses that have just stopped being ours; its threshold of existence begins with the break that separates us from what we can no longer say and what falls outside our discursive practices; it begins with the outside of our own language; its place is the distance from our own discursive practices. In this sense it can serve as our diagnosis. Not because it would allow us to draw a portrait of our distinctive traits and sketch out in advance the aspect we will have in the future. But it releases us from our continuities; it dissipates the temporal identity where we like to look at ourselves to avoid the ruptures of history; it breaks the thread of transcendental teleologies; and while anthropological thought would examine the being of humans or their subjectivity, it exposes the other, the outside. Diagnosis in this sense does not establish the recognition of our identity through the play of distinctions. It establishes that we are difference, that our reason is the difference between discourses, our history the difference between times, our self the difference between masks.”

The different lines of an apparatus are divided into two groups: lines of stratification or sedimentation, lines of actualization or creativity. The final result of this method concerns Foucault's entire work. In most of his books, he determines a specific archive with extremely new historical means, the General Hospital in the 17th century, the clinic in the 18th, prison in the 19th, subjectivity in ancient Greece and then in Christianity. But that is only half of his task. Out of a sense of rigor, to avoid confusing things and trusting in his readers, he does not formulate the other half. He only formulates it explicitly in the interviews given alongside the publication of his major works: What are madness, prison, sexuality today? What new modes of subjectivation do we see appearing today that are certainly not Greek or Christian? This last question haunted Foucault until the end (we who are no longer Greek nor even Christian...). Foucault attached so much importance to his interviews in France and even more so abroad, not because he liked interviews, but because in them he traced lines of actualization that required another mode of expression than the assimilable lines in his major books. The interviews are diagnoses. It is like for Nietzsche, whose works are difficult to read without the Nachlass that is contemporary to each. Foucault's complete works, as Defert and Ewald imagine them, cannot separate the books that have left such an impression on us from the interviews that lead us toward a future, toward a becoming: strata and currentness.