Between Madness and Literature by Michel Foucault from a Philosophical Point of View of Language

Michel Foucault’s work in the sixties is marked by two important works – The Order of Things and The Archaeology of Knowledge – but his research spans further topics. Among these, two are particularly interesting: his studies on the history of psychiatric thought from the Renaissance to Freud, and his work on literary works, which include two essays on Roussel and Blanchot. The psychiatric and literary experiences share a sense of being outside, a dehors, to use Blanchot’s expression, and each has its own way of doing it. The discourses of mentally-ill people are considered to be outside of the space of Cartesian reason, that is enlightened and positivist. A certain type of literature experiments being outside semantic schemes, as it pushes writing to its limits. From these two sides, Foucault experiments with the meaning of experiencing difference: the language of the mad and that of literary people are two different ways to bring to light an originary language, untamable, that is before any taking of the floor – any speaking out – happens. The study of madness, psychiatry and its history, and in parallel with the study of a certain kind of literature, with its respective languages and discourses, meant, for Foucault, understanding what it means to be outside of the order of discourse widely considered reasonable.


Introduction
A key difficulty encountered by those who first read Michel Foucault's works in the 1960s was not understanding what type of writing his was, to which order of discourse it belonged. Foucault's writing is "mad," in the sense that it is outside any scientific and literary genre. Was Foucault a philosopher, an essayist, an epistemologist, a historian of science, or what else [1]? Putting aside these impossible classifications in Foucault's case, the issue of language -of writing in general and of literary writing in particular -represented a "space" that the philosopher found hard to define. What type of language did Michel Foucault study? From Madness and Insanity: A History of Madness in the Classical Age published in 1961 through his lectures at the Collège de France in the early 1970s, Foucault studied the mad, their discourses and those delivered by the doctors across five centuries. Both the mad, the doctors, and their discourses represent for Foucault an outside and an inside of the the discourses of science, and moreover in a confrontation at a distance with the difference of madness, a madness that the discourses of medical knowledge named and explained. Not only that, they put it in its place. This is where Foucault draws madness close to literature. Some literature is, in fact, a "disalienated madness" [4], which cannot be confined to a madhouse, since it is made of "doublings of figures" (Roussel). This literary writing was not just a game and a literary experiment. It was also, and especially, a way to open up the signs and take them beyond their signifying function [5]. This literature was, in the 1960s of Foucault, "the madness of the outside" (Blanchot), when things can be said, but cannot be thought of as they are uttered. An example is Borges' Chinese Encyclopedia, which opens The Order of Things and makes Foucault laugh. But if I speak and cannot think of what I say, what am I? If I move in a literary scene (Roussel, Blanchot, Borges, Artaud and many others), I am not raving mad. If I am otherwise a normal man, then I am anormaux. It is worth repeating that we are on neighboring territories, because the two dimensions of mental illness and a certain type of literature are similar only for the discourses that they deliver. They are both outside the order of discourse of scientific knowledge, although under different titles. This is how Foucault studied the history of psychiatric thought, from leper hospitals to Freud's research, following a pace of this history that consisted also of interruptions, jumps, and sudden forward and backward leaps. In parallel, the non-works of literature coincided for him with acts, events/enunciations, where what counted was the gesture, not the person that made it, not the author who made it. Many were the incomprehensible, unnamable, un-assignable (at least temporarily) words in this semiotic horizon. It was almost like the language twisted against itself (Joyce). The sign opposed a resistance, and he who knew how to practice it, moved to the edge and played on the borders of the outside (Blanchot), because the signs that he used could not be immediately recognized. In the face of this word, it is even possible that some dispositifs of power can change, because transgression [6] -even when only literary (Sade, Artaud, Nietzsche) -can cause a general movement of reversal that breaks with the "old." For Foucault, at the basis of these breaks are events, a "solitary movement of singular precociousness" [7]. The switch to history, to a collective strategy, is hard, because everything seems temporary, isolated -an accident, an event, where the bringing back and integration deployed by the dispositifs of power are always right around the corner [8,9]. It is thus necessary to look always, again, for another outside. The work of detachment, what Foucault will name sé deprendre, has to begin all over again, both in psychiatric studies and in the field of an archeological and genealogical philosophy that knows how to dialog with certain literary forms.

Notes on reason and madness in History of Madness in the Classical Age
Madness has been the object of many discourses, which Foucault studies and divides into periods. In his reconstruction in History of Madness, Foucault claims that the mad, at least since the 1750s, had to be excluded and their speech controlled. Only reason can speak and this is how madness is named, identified, and circumscribed. Of course, so pigeonholed, madness contributes to define the limits of this same reason: it is the exteriority, the outside, the other of reason. In any case, only one of them -reason -speaks, speaking also of the other one. And reason speaks about itself, about what it is and what it is not, of what others are and are not, being the only one entitled to do so. In all these cases, discourse is already an instrument in the hands of a power: the power of reason, and a psychiatric power too, if only in its early stages.
But how had people spoken until then? According to Foucault, the Middle Ages held together the scientific, allegorical, mythological and poetic discourse, as well as that of magic and chemistry (alchemy). Again in the Middle Ages, the world had long ceased to be a cosmogony, a tapestry where man could read the signs sewn by God, because Babylon and its punished sins had rendered everything "opaque." The men who built the Tower wanted themselves too, and not only God, to be able to name things. Then like now, it was not a matter of simply designating things, but also of creating something, or re-creating something that already existed by designating/ nominating it. We know that God punished the men of Babylon for their arrogance and for building that Tower that pointed to the sky. And he punished them by making the signs and the traces that He has always left across the world, opaque and not immediately legible. Foucault maintains that the medieval men then tried to put back together a possible interpretation. Six centuries later, Ferdinand de Saussure would write about the arbitrary relationship between the concept and the acoustic image. He would note that certain correspondences/transparences cannot be sought anymore, because sign, meaning, and the world we refer to are different things, and nothing is transparent in these linguistic relationships. To the contrary, we have leaps of semiotic atmospheres.
While Medieval people were still convinced that the great jigsaw puzzle of opacity could be read through different languages belonging to the same leggenda (in the sense of what can be read) [10], the Âge classique imposes itself and imposes the singular language of reason, of science, and Cartesian knowledge to everyone. This is how man will start differentiating between scientific, allegorical and mythological discourses. Based on these rigorous languages, those with the power will be able to start serializing, categorizing, naming, identifying, creating taxonomies (maybe we have never stopped doing it), to make other women and men into classifiable phenomena. Given this semiotic premise, the mad of the Âge classique are all pigeonholed in a template populated with cases of abnormality. According to Foucault, a "tragic awareness" of this madness developed since the very beginning, that is, in the early Renaissance. Nature could not be trusted anymore, because it was suddenly populated by monsters, including the mentally ill, which could come out anywhere. In his reconstruction, Foucault claims that that is when the mad, the beggars, the vagabonds, the libertines, the blasphemers were all gathered at the Hôpital: a lager. It was a social problem: order had to be made. Undoubtedly, that Hôpital was filled with chaos: the homosexuals and debauched laid next to the mad. However, that was the price to pay to the Cartesian method, when a truth must be established that let only the reasonable and rational be free, whereas many and varied persons ended up together in a big hospital. Vagabonds cannot roam the streets, even better if everyone moves to the city from the country, because the country is still too wild and we must discipline, build the citizen, and create demography. Individuals must be controlled in order for them to be sent to the factory to work, already serialized, from the 1800s onward.
All the mad, indiscriminately, are empty, negative, and unreasonable [11]. Their bodies and minds (spirit) are too entangled and there is no physical cure that can go with some moral consolations. In the early Âge classique, this is how one could be innocent and should be cured, but chaos reigns and then the same innocent individual is guilty and must be punished. Thus, takes place the quick shift: reason on one side, madness on the other. Finally, with Pinel and Tuke, madness is medicalized, made into a scientific object. The mad are examined as if they were phenomena, in the same way objects are examined. These mad objects (which lack the qualities of the rational and reasonable subject) are our "objects", because they are in our "possession" and, as such, belong to us in every sense. Then, perhaps, we lock them in a drawer, but it is something we decide to do. These prisons/drawers are ours: we built them to put in our objects, our possessions, our mad, and our physicians. They are the drawers of our desks, on which we compile our archives and note that someone is mentally ill. In Foucault's reconstruction, at this point this kind of madness is very close to us, because it teaches about us, who are the mentally sane. We know ourselves through this madness, that is, through the modalities with which some discourses are made that let us know that someone is mad and maybe we, too, are not so well.
Only Freud will begin claiming that there is something profoundly true in the mad, which belongs to all of us. Again, Freud will say that there exist some familiar, sleeping forms: the same voices for everyone, and strange lights. It is as if the mad showed us our youth, being the mirror of the distortions of a civilization that has become modern. The instincts, perversions, pain and violence, the selfishness are ours. They are our mad side. But before reaching Freud, Foucault reminds us that Pinel -who took the important step of declaring that the mad are not monsters, but persons in need of a cure -did not free anybody, precisely because he claimed that madness is a sickness that must be cured. Pinel as a physician was a man of power who practiced a discipline that had its own exclusions, "declassifications," and desk drawers. However, Pinel's time is already the time of the man-madness-truth tryptic; not the time of the truth/errors, world/ghosts, being/not-being, and night/day dichotomies anymore. Foucault identifies a three-partite anthropological articulation that ripens in Pinel's hospitals of the 1800s, on city and country roads, in the literary works of all the authors that we have already mentioned. It may be 1961 and Foucault is speaking of madness, but The Order of Discourse of 1971 is ready, already.
In any case and to stick with our theme, at the time of The History of Madness Foucault attempts straying in the literary field more than once. Diderot's Le neveu de Rameau (1891), whose analysis occupies the introduction of the third part of the History of Madness, is a lonely man who makes continuous pantomimes. He is vain, full appearance, immediacy. He is a raving man, who interrogates us, even if he represents "all the elements that form a wordless dialogue between day and night […] in the burgeoning transcendence of any act of expression, from the source of language itself […]" [12]. Roussel, Joyce, Foucault: "He is mad because that is what people tell him and because he has been treated as such" [13]. Rameau's nephew is lyrical: he imitates everything, knows all the languages. He is one and nobody because he is everyone. He is a man who, in the end, is left on his own, locked up, remaining there with an empty smile on his face that will frighten us. Foucault: "To be oneself that noise, that music, that comedy, to realize oneself as both a thing and an illusory thing, and thus to be not simply a thing but also void and nothingness, to be the absolute emptiness of the absolute plenitude that fascinates from outside, to be the circular, voluble vertigo of that nothingness and that being […]" [14]. In the end, Rameau's nephew will be an object in the hands of the physicians and his lyricism will be explained and normalized in the medical discourse: his will be an organic problem and, as such, medicalized. And yet, like in a concave mirror, this monster of a nephew shows what and, most importantly, who is normal. In a short time, Sade and Baudelaire will be "declassified" too, because their discourses are perplexing. Sade will be classified as a pornographer. He will be identified and taxonomized: his work will be considered obscene, perverse, and deviant. The danger that he represents must be softened, thus he must be locked up and condemned.
Are le neveu de Rameau, Nerval and Hölderlin, Nietzsche and Artaud mad? Absolutely not, not in an important phase of their lives, because they have their own way to tell their experience, which for Foucault is symbolic. It is almost as if they established the sense of how the experience of life is changing in general, between the end of the 1700s and the beginning of the 1900s, thus also the experience of life of those who really experience madness and, especially, of those who watch and listen to it, deciding whether one is mad, or not. Madness and literature stare at each other from their respective sides, experiencing each its non-sense. We are on the outermost edge of "two movements of poetic conversion and psychological evolution" [15], which imply, for Diderot in 1791 and Foucault in 1961, the inebriation of sensible things, the enchantment of immediate things, a painful irony, a certain loneliness, that of the neveu de Rameau and his madness that belongs to us. However, there is not only Diderot. Let us repeat and anticipate a theme that we will return to: Nerval and Hölderlin speak about man's senseless secret, about "his first morning," a "young light," a "starting again," an outdoor (an outside), which is crammed with items that are not mediated by any reason and are very attractive. We will return again to how these literary experiences happen like a tragic and painful "laceration," even if always in the full light of the possibility that we might start all over again (Nietzsche). In short, we are dealing with dangerous experiences: "The moment of the Ja-sagen, of the embrace of the lure of the sensible, was also the moment they retreated into the shadows of insanity" [16].
The History of Madness was not appreciated by some psychiatrists, who deplored that someone could claim that their retreats had an obscure origin in the leper hospitals [17]. We have already recalled that, in spite of their merits, according to Foucault even Tuke and Pinel cannot be ascribed to a hagiography of psychiatry. For Foucault, madness is a cultural product and, for psychiatry, things went one way, but could have gone another way, as well: a roll of the dice -as we will see -a mode to conceive history marked by unforeseeable events and ruptures, which coincided with the birth of the bourgeois society and its ensuing, wide exclusions. At the beginning, psychiatry was, with Pinel and Tuke, one of Positivism's forms that corresponded with a sort of apotheosis of the figure of the psychiatrist. In Tuke and Pinel's retreats, a rigorous morality applies, where no diagnoses are made, only observations of the behaviors of the sectioned patients. For Foucault, here it is all about knowing how to deal with the mad and there is nothing metaphysical in the patient-physician relationship, but rather a full-fledged political confrontation. Madness attempts to make its voice heard, but science classifies it like as a sickness, and achieves this by indulging its moral sensitivity. Since that time, many steps have certainly been taken, as we at least have tried to rely on the medicine-verification model [18]. Is Foucault an anti-psychiatrist? Maybe. Certainly, mental illness is not only a natural occurrence to him; it is also the effect of a specific medical interpretation. Psychiatry advances/imposes such interpretations, which are historically based on a power relation with the patients, is a power that makes institutions -and not only them -work. And yet, if this is how things stand, if psychiatry was and is a power, then its choices can be questioned and subverted [19].
Nevertheless, what procedures do men/women undergo to become physicians, as subjects of conscience? Simultaneously, on what basis do men/women become patients as objects of conscience? For Foucault this question is not only about repressive systems, with respect to which we would be passive, but about processes of self-formation: the mentally ill and the sane self-form with respect to the psychiatric power. The same goes for the physician with respect to the patient. It is a power fight over yielding the power of delivering a certain type of discourse. Is psychiatry willing to ask itself about how its 'truth' has affirmed itself through history, under sometimes violent circumstances? Does it make a genealogy of its historical processes that have led it to, today, deliver certain discourses, which profess the truth about who is mentally ill? To work on one's history effects the liberation -albeit partial -of thought, including the psychiatric thought, working silently on what we think, because those who are defined as mentally ill do, indeed, think silently, but the same goes also for the physician who observes the patient's behaviors and listens to his/her discourses [20].  (1973)(1974) and Les Anormaux (1975), was not easy, because Pinel is not at all bad. He is a physician who stands before the mad and tells his assistants that these people are not criminals, nor possessed by the devil, nor wild beasts to chain and lock up, but sick people, and it is necessary to treat them with care and humanity. But we have seen that for Foucault, Pinel's progress still carries within itself the discursive mesh of a series of passages that reduce the other to nothing, but a sick person to cure and, in a Positivist sense, a scientific object to analyze like any other phenomenon. We will have to wait for Foucault to read Freud and even he will not be enough for the philosopher, because his psychoanalyzes are induced self-confessions. In any case, madness remains the result/discard of the Cartesian cultural gesture of separating reason from non-reason, being mentally sane from being mentally insane. For Foucault, psychiatry does not liberate, it excludes. Lepers were too excluded by means of specific rituals of purification. They were only one step ahead of the beginning of psychiatry's history, which was grafted onto their historical branch via an event, an Hôpital, a Cartesian distinction between being and not being.
Is Foucault's reconstruction of the history of madness perhaps too ideological? Does it wink to Surrealism? Is it too not historical enough, anti-psychiatrist, anti-Cartesian, and anti-Enlightenment (these are the accusations that were leveled at him and that he took in serious consideration) [21]? Maybe, but this is not what matters today. What matter are his observations on the discourses of psychiatry, powerful discourses that forced many different ones into a silence that stretched to Freud's time, even if, to Foucault, certain tales of the self (the dialog and the psychoanalytic encounter) look a lot like self-surveillance. In any case and at every moment in history, psychiatric time needs its own microphysics, which is only possible starting from the discourses that the psychiatrists make and let others make, or not: from a psychiatric point of view [22], society, power and normality are important stakes. Is Foucault anti-psychiatrist, post-romanticist, and an idealist? Foucault signaled a discursive shift: the mad's discourse, which was silence, was replaced with the medical discourse, a kind of "transcription" of madness in the language of mental illness. This affects us, if nothing else, for all the times we speak of medicalization, a concept that touches us throughout our lives, from the moment of birth to that of death.
We know that in the 1970s and 1980s Foucault sought a return to the Aufklärung beyond its historical limits: a critical, archeological, and genealogical return to the time of the Luminares, because a different life is possible, because, if anything, the Greeks and the Romans, at least some of them, before the Christians, led lives that were ethically different from our present ones. But what kind of shift did Foucault attempt? Are not his Aufklärung and his critical detachment themselves constitutive of a new scene, with all the limitations that it entails [23]? Regardless of how we want to answer this question, already by 1961 Foucault had described a "madness of not madness," in the non-sense of a reason, a knowledge, and a medicine that impose themselves as dispositifs of power with catch-all pretenses. Then, next to it, we are left with the impression that the sense that we attach to things derives from a non-sense, which is of the world and of the same men and women that attempt to give meaning to themselves, walking on an abyss, the abyss of the non-sense that they find themselves facing.
For Foucault, let us repeat it, power and knowledge can constitute themselves in a moment, and therefore at the beginning of any time there is a roll of a dice, an empirical dimension that is necessary and not transcendental. There has never been, for Foucault, a transcendental subject. There was never a man, a woman, that had been always made in a certain way and would be like this forever. There has always been and there will always be a man and a woman of knowledge and power, who change at each historical passage and some passages are, as we were saying, a roll of the dice. This is where we run along the edge of the abyss, that is also connected with what might happen: Nazism, Fascism, the lagers, the atomic bomb, the "madness of not madness" taken to their limits. And yet, they are the mad ones, the people that we cure, medicalize, and confine in the madhouses. In Italy, Basaglia, first in Gorizia and later in Trieste, began opening some psychiatric hospital in the same years that Foucault was conducting his studies, from the end of the 1960s through the 1970s. Basaglia's operation was not easy, especially for the families of the liberated mad, who were again hit by their relatives' non-sense of the sense.

Why literature
Foucault writes for La  (Blanchot), a language that is an outside (la pensée du dehors, the precise and famous expression di Maurice Blanchot) that envelops us. Our word is not the beginning. It is not at the beginning. For Foucault/Blanchot, it is a certain surrealism, a certain aestheticism of the beginning of the century that matters.
In The Order of Things, the Renaissance, was for Michel Foucault, a time marked by a way of understanding language as governed by relationships of similarity and analogy: the inner microcosm corresponds to the macrocosm of the world via analogies, one resembling the other in the semiotic universe of the Renaissance. With the Baroque, these relationships of similarity are left behind and a crucial shift happens, as language is now considered a mathesis that allows things and names to be divided up and linked in a clear and distinct way. Then, in the 18th century, the signifier-signified relationship becomes problematic once again: a caesura intervenes between words and things. In any case, language has for a long time, since the Renaissance, lost its relation of similarity with something enigmatic, primitive, and shining, which coincided with an infinite opening up to the world. Foucault claims that no traces remain of that language today. This moment is when a certain literature that has its own semantic autonomy -not unlike a counter-discourse without representative pretenses -emerges, in a space otherwise dominated by scientific discourses. With it, a language that pulsates, that lives and breathes, makes its appearance once again with Hölderlin, and, for Foucault, Mallarmé and Artaud, who are the first among the others to engage with this language [24].
Let us now move, then, to Foucault's 1960s France and recall something we are now familiar with, that is, that Blanchot and Bataille intended language as a form of negation, a passage that Foucault appropriated. Blanchot and Bataille claimed that a refus, una dépense had to be opposed to an omnivorous dialectic of the philosophers, because things must be consumed for a real expenditure of energy (not only semantic ones). By doing so, one does not see goals and aims for themself. It could be that in this literary context -and not only in it -it is necessary to also work on psychic automatisms, like the Surrealists did, since it is possible to be manifold individuals without a definite identity. Certainly, according to Foucault [25], we should not apply a psychological reading to works of art (like Jaspers did, for example [26]) because that would end up, if nothing else, twisting van Gogh's art. Foucault notes [27] that if we want to adopt a psychological lens, then we should say, with Lacan, that the originary language, which brought together madness and reason, is always there, within us, even after reason and madness have been separated. And yet, since some rational limits have been set up, that language has trouble making itself heard. For many, it does not exist anymore, and thus, there is no possible accommodation, no exchanges, no communication between the madness of some and the so-called work of others (for example, some psychiatrists). Madness is the absence of work and so is some literature. Are Van Gogh's paintings works? In Foucault's reading they are not and this is not because van Gogh cut his ear off, but because his works, his paintings, represent a break with all that precedes them in a historical and dialectical sense. Van Gogh and his paintings mark an inaugural event.
Moving with Foucault to an exclusively literary side, Hölderlin, in his time, already resisted. Hölderlin resisted through his poetry, a poetic resistance, infused, however, with a divine violence that illuminates and incinerates [28]. We are with Foucault and the Hölderlin that he reads, at the limit. Blanchot (whom Foucault will follow almost to the letter), also reflected on Hölderlin, and wrote in La parole sacrèe de Hölderlin, in the book La part du feu [29], about a literary journey to an inner reality that can be in a relationship with what is sacred, here following Heidegger's lesson on silence. For Blanchot, this literary inner reality -and not only that -coincides with nothing reassuring, as its experience suspends the world and the self. Blanchot observes that, in Hölderlin, all this takes on an ambiguous and dangerous appearance, because it is like we are waiting for the dawn and the song of the Gods, while, at the same time, facing the loss, the ruin of the self 's word. As we have repeatedly pointed out, even if these are not experiences of madness, they are nevertheless painful dimensions that are difficult to hold, being at the very edge of what others would treat medically. We can, therefore, consider them as forms of alienation. The literary experience is a path that can be tread, which can lead to a new Zarathustra. Its journey and its outcomes too are painful, almost unbearable, and in any case they are not an experience for everyone. This is the origin of the clear political limits of such a proposal, the missed switch from individual experience to collective sharing, even though Nietzsche clearly changed the course of history, at least for a certain part of the Western world.
For Foucault, the fragmentary and incomplete nature of Nietzsche's writings represented the rapture and the shedding of the unitary system of so-called works founded on scientific discourses. For Blanchot, the work that Foucault mentions with Nietzsche, remains instead, albeit transformed, open, infinite, unfinished: Hölderlin, Nietzsche, and later Kafka and Mallarmé testify to it. Blanchot identifies an anchor in the unbridgeable gap between Hölderlin and Nietzsche's literary experience and their respective madness. He believes that Hölderlin paved the way for a new genre. Hölderlin announced the possibility to dissolve the artist's subjectivity, moving toward something deep that could not be reduced to the orders of the discourse of scientific knowledge, toward a nondialectical dehors that is an originary dimension. For Foucault, the dehors entails moving to a metaphysical void. For Blanchot, the dehors is a space, where one can play at differing the meaning of the words that one uses, if they know how to do it. Such words, if put down in a certain way, help one to put themselves at a small distance from reality and the world as they are given to us.
If we return our focus to Foucault, in the span of a few years, his positions will progressively adjust and, from Non du père to La pensée du dehors, they will move closer to those of Blanchot in a clearer and more definite way. Foucault detects a void, an absence, in language, what precedes the speaking out. Between 1963/64 and later, after 1966, in Préface a la trasgression [30], the introduction to Bataille's work, a certain way of doing literature becomes, for Foucault, a transgression of the limit, without being a Dionysian experience anymore, with no relationship with madness. In La folie, l'absence d'ouvre of 1972 [31], Foucault pronounces the definitive death of the homo dialecticus and between his death and the liberation of language we go from Nietzsche to Freud and finally Blanchot. Freud, in any case (and for Foucault in 1972) does not in fact consider the language of madness a deraison, but a reserve of meaning. It is not simply silence anymore, but a doubled-up word, folded on itself, with double meanings that go beyond the linguistic surface. According to Foucault, Freud reflects on an esoteric language, a code that loops back on itself. It is an absence of work, where one cannot be univocal. Perhaps, this is how, Freud and Foucault claim, we can disclose a certain closeness between literature and madness, thanks to these features that are essential to any linguistic code [32]. But, in reality, Freud, Blanchot, Mallarmé, and Foucault align on a new front. Because if Freud discovered the hidden meanings of language, Mallarmé took the same language to an originary dimension, which is not madness anymore for both of them, but a shared and originary dimension. Thus, under the sign of a certain interpretation of language, a new union is established not between madness and literature, but between a certain way of practicing psychoanalysis and literature. Here with Foucault we get close to Heidegger and Blanchot once again, acknowledging that language is something originary, a dehors, something that before speaking out is unrestrained, because it is an origin that we cannot dominate.
Roussel, Mallarmé, Bataille, Blanchot accompany Foucault in his archaeologies of this language, at least for some time. But let us ask once again: why this focus on language? Certainly, Lacan's lesson, among the others, had its importance. But why Lacan? Because Foucault is interested in his psychoanalysis [33], a psychoanalysis that can look beyond the dimension of representation, proceeding in the direction of the limits of man and his being "finite." Death, Desire, Law-Language are the three keywords/cornerstones of Lacanian psychoanalysis. For Lacan, language is not a free expressive capacity. It is a "law" that governs man's thought and action, beyond their conscience and active mechanisms. This is where Foucault's research establishes the point of welding between literature and madness that we have been looking for. Let us read, then, Bruno Moroncini's words: "In other words, if I start reading a text by Maurice Blanchot (or Raymond Roussel, or Georges Bataille, or a poem by Mallarmé), what kind of experience am I having (in the sense of the Erlebnis of the phenomenological tradition, or the philosophy of life)? Am I experiencing the contents that have progressively accumulated in the subject and that are communicated to us in a nice form, or is it the experience of language as language, that is of that Language-Law that cannot be separated from Death and Desire, to which psychoanalysis leads us in its approaching the reason of the unconscious and of finitude?" [34].
Whether it acknowledges it or not, the I speak is inside a language that can spread itself infinitely. Raymond Roussel and Edgar Allan Poe know it: if you know how to play, this language can lead you to the infinite on its surface, breaking you, dispersing you, scattering you in a "naked" linguistic space. But one must pay attention, because this infinite spreading also concerns some discourses. These discourses, which purport to tell the truth on people and things, infinitely search for a truth, a certainty, a definition, that does not exist and has never existed, all the while counting who is in and who is out. In such cases, the I speak is a "crack" through which other, specific outsides form -the various "outsides" of the discourses of truth, the ones that are full of rules that exclude: the prohibited, the partage [35]. All these linguistic phenomena express a will of power. They are dehors the discourses that idealize reality -like the juridical and historical ones.
Madness too inhabits this dehors/outside of language, but it does it in its own way, a very difficult way to interpret. Death, the dark sides, the unsaid of discourse have their own I speak that we need to be able to listen to. It is not enough to label them as "mad", because they are not. They make up an important part of us and, in the end, are neither good, nor bad. There is, thus, let us say it once more, a sharing of a wild territory from different sides, a looking at it from different shores that must be kept separated: that of literature and that of mental illness [36]. In Blanchot [37] and Foucault's case, this wild dimension could, for some years, correspond to one of the findings of an archaeology of silence, a silence that once had a voice. For Blanchot, his dehors is not madness, but a way to distance oneself from reality and the world. It is an outside that is a sign that wanders the world and develops itself in a constant deferral. For Foucault, literary language is a language of literary fiction, which weakens the discourse of the Cartesian subject. Then there are other types of fiction, like those Bentham reflects on [38], but here we get to Discipline and Punish of 1976, almost ten years later and they are linguistic and symbolic fictions discussing the power relations and, perhaps, even the prison's architectural structure. In any case, the thread that goes from Blanchot to Bentham remains. In the end, almost at the end of the 1970s, Foucault will visit Japan where he will discover a different theater, even a different ars erotica, that will lead him to think of another semiotics, a semiotics of the scene, of the ceremony, of the ritual, of martial arts, all of them disciplines of the body that are different from those of the West. But this is a different story for a different paper [39].
Going back to the 1960s, literature was, for Foucault, transgression, resistance, contestation, a questioning of dialectics and of those who profess it. At that time, contestation could still realize itself, not only in politics, but in a literary space going from Sade to horror novels. Literature is a pretext, something nonhistoriographical that comes before the text, is not an academic discourse, and does not follow a method. The book dedicated to Raymond Roussel is Foucault's "secret garden." His step toward politics is drawing near, while at the same time still far away, because, in these years, Foucault practices mostly, not really an esthetic, nor a hermeneutics, but a philosophical-literary study standing between history and non-history, semiotics and noise, which is an archeology that is useful to take a step back and oppose those who adopt a strictly scientific method, thus counting who speaks and does not speak [40].
This switch is not easy, we have already said this. Poe, Roussel, Blanchot split words. They embed them in different codes, from which the paradox can originate of a word that says what it says, while adding a mute surplus, that shows, almost silently, not only what it says, but also the code and codes that allow it to speak. In such a way, we move inside the existential folds of the word, where some words do not have a single meaning and what counts is not their verbal matter, but the game they play and the transgressions that they allow. These words do not hunt the truth, because they are not confessions, neither in Augustine's way, nor Rousseau's or Freud's. Here, for Foucault, we need a poet's talent, since writing, at the time of the first edition of the Archaeology of Knowledge, is something we lose ourselves in, we step back from, we play with, whereas the subject of knowledge of the same poetry must be destroyed and vanishes, since there are no authors anymore when we write and read (from Kierkegaard to some variants of the structuralism of the time). In any case, for Foucault some books are interesting, because, as you write and read them, you cannot say where they will lead you and they thus teach you what you do not know. These books are inventions that can transform those who write and read them. They can be real events.
Coming to the end of this review of Foucault's literary studies, let us insist again that a certain structuralism and the Heidegger of Being that manifests itself in language -as well as Nietzsche, Bataille, Canguilhelm, Lacan, Levi-Strauss, certainly Blanchot -have had an impact on Foucault's research, inducing him to believe in a language that precedes any scientific discourse. Literature admits this anteriority, it traverses it, opens itself to it and lets it flow, something that, at that time, the Nouveau Roman in France had been teaching for a long time. Literature grasps and practices an originary language and if someone expects to find their identity in this semiotic experience, then they will not. Writers, those who try to study language formally (the Russian formalism and more) [41], those who study the myths (Levi-Strauss), or those practicing a certain psychoanalysis (Lacan), all know it. The language that they contend with imposes to the self, not so much to keep quiet, as to retreat, if it expects to engage in discourses intending to represent the truth of something. It is a matter of stepping outside the discourses that make dialectics and existentialism, the same structuralism, and listening to a different way of saying things. This is how we can again create an opening (Heidegger), allowing the same language to spread itself infinitely, certainly in a void of words, for those expecting to say the truth. In this void, the signs come together, but they are also kept apart, because it is like they are dispersed, without a space that would let man and woman fold on themselves to seek their own truth. The language, that Blanchot experiments with, takes things to their limit, because it is a self-contestation, a void of consciousness, inside the noise of language, where words go after one another indefinitely, overcoming the same literary fiction. Raymond Roussel had allowed Foucault in 1957 to break with phenomenology. Foucault will then read Blanchot, Bataille, and finally Nietzsche, in an almost exclusively literary procedure with a sui generis philosophical outcome [42]. For him, language will then be a non-place, made of uncontrollable similarities, a neutral space where nothing can root, no one speaks, where all has already been said in a different and often ungraspable way. It is not easy to claim all this: it stands on the border of madness' territory.
In 1983, Foucault will declare that there exists a writing of the self, which has been practiced, following a certain old, pre-Christian morality. This writing of the self does not coincide with an obligation of truth that runs the risk of going on forever exhaustingly and without a real care of the self. Then, in Greece and Rome, we had journal entries, notes of quotations, soul searching, and correspondences, treatises, in a relationship with friends and teachers. But here we are beyond literature. It is 1981-1982 and these are the lectures on L'herméneutique du sujet, where Foucault reflects on a possible new ethical site that has, in fact, been possible A site, where man can transform himself, test himself, can take care of himself without being a subject who wants to know the truth, without establishing differences and placing outside and inside a madhouse, a prison, at the margin of society, those who have a different sense of things that does not coincide with the order of the discourse of Power.

Conclusions
This brief reconstruction of the important and complex themes of madness and literature in the philosophy of Michel Foucault shows that, in Foucault's analysis, they are neighboring territories. Both of them lay outside the prevailing order of discourse, experiencing the non-sense of the words that they pronounce, enunciate, and live. Finally, each plays its own game. Or, to better explain, if those who create literature may perhaps want to play with words, those who are put in a madhouse do not play anymore. Probably, they really are ill, but it can also happen that they have been labeled anormaux by those who have the power to do it and, then, reduced to silence and forbidden to wander (Nietzsche) the streets of the cities and the towns, or the country, that were once their home. Both madness and literature are dehors/outside. They are exposed to an outside, which is the outside that they live, because they found themselves excluded from the games -left out of them -or chose to keep away from them. Next to these two outsides, are the many outsides