The following text has been automatically reproduced by an Optical Character Recognition (OCR) algorithm. It may not have been checked over by human eyes. For matters of precision please consult the original pdf.

Risked democracy

The delay involved in the publication of lectures or seminars has strange effects: what comes late and in a different time to its own is research and words which were caught up – more so than the books – in the historical circumstances of their elaboration; and the text that is finally published, with the reflections of the author and the remarks of the audience, carries something of the historical situation that produced it. This documentary dimension is sharper still when there appear together works undertaken in the same period by two thinkers between whom, at the time, no debate took place, and who appear to have been totally unaware of each other. An outline appears, in the background of their preoccupations and intellectual trajectories, which we could call, following Frédéric Worms, a specific ‘moment’ in which political history and the history of thought are mixed.

Thus, in 2008, the traces of two research paths that were very unlikely to meet were published, and their conjunction is striking: on the one hand, we have the series of classes given by Michel Foucault between 1982 and 1983 at the Collège de France, under the title The Government of Self and Others;1 and, on the other hand, we have the seminars conducted by Cornelius Castoriadis between 1983 and 1984 at the École des hautes études en sciences sociales (EHESS), in the context of his vast cycle ‘what makes Greece’, entitled The City and Laws. [2] The pure coincidence of the publications brought to the surface what might seem, over twenty-five years later, also to be a coincidence:

that two French thinkers should, in the same period, have felt the need to explore, each in his own way, the Greek corpus and the question of democracy; and that they should have returned to neighbouring texts (the tragedy of Ion in Foucault, and of Antigone in Castoriodis), and to common figures (first of all, and above all, that of Plato, the proclaimed adversary of democracy, which the one and the other question, as we shall see, in a roundabout way).

So, we have a number of coincidences. It is obviously possible to reconstruct for each philosopher the different path that brought him to the vicinity of Athens. In Castoriadis, the reflection gathered in The Imaginary Institution of Society (1975) is continued from 1979 in an examination of the link between the Greek polis and the creation of philosophy as the opening of a space of thought linked to the experience of a particular relation, in the human world, between the imminent organization in the city and the disorder that continues to underlie it, which it knows it cannot entirely avert. Foucault’s bringing to light of the motif of governmentality would, from 1978, involve a vast restrospect, from the period which was most familiar until then (between the classical age and modernity) back towards medieval thought, towards the Church Fathers, and then towards Classical Greece. Nonetheless, these different returns only reinforce the suspicion that there is meaning in the coincidence – the impression, to use a Greek expression, that the latter is as much blind automaton as it is tukhè – a concept which Aristotle used to describe coincidence, inasmuch as it allows human activity and agrees with it, in politics in particular, and weaves with them a sensible practice.

The questions could therefore be the following.

What does this concern for Greek democracy, shared by Foucault and Castoriadis, tell us about the singular historical moment in which they experienced it, each in his own way? And what lessons can we learn, in this moment which is ours, from their respective research?

It would be easy to conclude: return and continuity.

Return, in authors that had somewhat returned from their radical wanderings in the 1970s, towards the more traditional examples and problems of political philosophy, whose influence on modern writers Castoriadis ceaselessly insists upon, in order to claim to follow them (Jean-Jacques Rousseau) or to distinguish oneself from them (Benjamin Constant): ‘the invocation of ancient democracy has indeed played Risked democracy Foucault, Castoriadis and the Greeks

Mathieu Potte-Bonneville

a fundamental role in the fight against monarchy and for the establishment of citizens’ rights’ (CL, p. 31). Continuity, therefore, of a preoccupation with intellectual sources whose contemporary regimes are still proclaimed, and with regard to which historical knowledge has, for twenty-five years now, continually progressed. (Philipp Raynaud, in his preface to Castoriadis, insists on the progress of archaeology regarding the birth of the Greek city, and we could say as much for the philosophical exploration of concepts mobilized by Foucault).

Such a diagnosis might be unfaithful to the teaching that the two authors claimed to offer, and that consisted in insisting not on the continuity of political philosophy but on the effects of rupture, of discontinuity, which exploration of the Greek sources makes it possible to uncover, with regard to our present. In answer to the search, in Foucault, for a ‘historical ontology of ourselves’ (which the first lecture of GSO already mentions), we find, in Castoriadis, the concern to disengage from retrospective illusions vis-à-vis the Greek corpus:

The ancient constitutions serve as screens upon which are projected the ideological needs of the present, and thus by the same token a whole collection of important aspects disappear – important not from the point of view of exactness, let us say ‘philological’, but really from the point of view of significations. (CL, p. 27) If we follow this lesson, the reading of these texts changes: we see less a stage on the way to the slow reinstallation of political philosophy in its timeless space and canonical references, and more the play of a double discontinuity or of a double interval. An interval, first of all, in this strange period at the beginning of the 1980s, characterized by what Michel Feher suggests calling an ‘interreign’. Prior to that was the exhaustion of the communist and revolutionary lexicon in which the radical experiences of the 1970s were formulated, and towards which Castoriadis and Foucault always remained sceptical (not only for doctrinal reasons, but because it seemed to them that this interpretation was profoundly inadequate for describing the newness of social movements born of May 1968). Further down the line, from the second half of the 1980s, lay the establishment of democracy as an evident horizon of the new world order, and the tension [mise en tension] between democratic demand and the republican model; in other words, the affirmation that democratic societies must protect themselves from individualistic or communitarian temptations by becoming vigilant with regard to objective and stable forms of life in common (dignity of the law, abstract citizenship, nationality).

The texts of which we are speaking are to be found in the interstice between these two crushing bodies of reference, in a moment where the experience of dissidence in the Eastern Bloc made the motif of democracy emerge as a critical motif – posing the problem of knowing whether and how the democratic demand can also play a questioning role in the West; in a moment, therefore, when the affirmation of democracy can no longer be satisfied with the opposition, imposed by Marxism, between ‘formal’ democracy and ‘real’ democracy, but where at the same time it is less a question of invoking democracy as a principle than as a practice, which no principle could replace and which demands, on the contrary, the questioning of all principles. As a result, from that very singular moment of our recent past (an almost invisible moment, since it is so easily forgotten in the teleological reading of history), the question comes back to us, challenging the identity of our very present: what should be done, in this period in which the signifier ‘democracy’ has been compromised in imperial adventures and neoliberal globalization, but also in which what Jacques Rancière calls ‘the hatred of democracy’ could not take the place of politics? What should be done with democracy?

I will try to support a simple hypothesis: what we learn from the crossed reading of Foucault and Castoriadis is that there is only a risked democracy, where the verb ‘to risk’ means at the same time ‘attempt’ and ‘threat’ or ‘hazard’. There is no democracy that is not committed to inventing its own institutions and procedures, on the basis of a radical uncertainty – this is what we learn from Castoriadis; but there are no institutions or procedures that can exempt citizens from an exercise that is always in excess of the rules that contain it – and this is what we learn from Foucault. Between the one and the other, therefore, appears the outline of an idea which goes against the foundational movement by which political philosophy, regularly, goes from politics as a contingent activity towards ‘the political’ which is supposed to give it its foundation and dignity.


Let us begin by remarking that Foucault and Castoriadis seem to borrow, with regard to Greek democracy, certain profoundly divergent reading strategies. The central object for Foucault, as we know, is parrêsia, that attitude seen successively in the courage of Pericles and the insolence of Plato; an attitude that he describes at the same time as a manner of speaking the truth, the taking of risk regarding oneself, the constitution of a relation to oneself centred around that very risk, and the attestation of a free act. The manner in which the examination of parrêsia seems today to integrate naturally into our knowledge of Foucault’s work hides perhaps the signification and the radical nature of the gesture that consists, in Foucault, in making that dimension appear, and of thus destabilizing the established understanding of Greek democratic mechanisms. One must remember here that if Foucault used the term ‘archaeology’ with regard to his own research, he did so by playing against each other the two possible etymologies of this word: the archè of the philosophers (as timeless origin giving its foundation and its justification to experience) and the archive of historians, as a material whose historicity, multiplicity and absence of hierarchy profoundly question the identification of a possible foundation.

Thinking, in Foucault, is therefore to grasp the central significations of the terms that philosophers claim to raise to the dignity of an essence, but in order to make appear in the centre of that centre an invisible dimension, considered to be secondary, which will allow the philosophical categories to submerge in a history that they do not master.

This is precisely what is achieved in the lectures between 1982 and 1983, on at least two levels. First, in relation to the idea and word ‘government’: the government, in the vocabulary of contemporary philosophy, as in public space, is a quasi-synonym for executive power, to the extent that the examination of the manner of governing seems entirely contained and explained by the institutional context in which power is exercised. The displacement operated by Foucault, by examining the way in which the government of others implies a government of oneself, consists, on the contrary, in undoing that subordination, and in showing that the art of governing (and of governing oneself) is under-determined by the constitutional system in which it is exercised. Second, it is inside this strategy, present since the very first works on governmentality, that the examination of parrêsia finds its true meaning. It has to do less with completing the traditional understanding of Greek democracy, defined by the notions of isonomia and of isegoria, by means of the adjunction of a third term, than with ‘de-completing’ or de-totalizing this understanding itself. Parrêsia, indeed, has a status that is profoundly heterogeneous with that of the isonomia and of the isegoria. It is not a determination of similar level. As Foucault remarks in his commentary on Euripides’ Ion, ‘parrêsia is, in a way, a discourse spoken from above [d’au-dessus], which comes from a higher source than the status of the citizen, and which is different from the pure and simple exercise of power’ (GSO, p. 104).

Where isonomia and isegoria appear to contain reflection on democracy in the circle of a double reference to the law and to rights, parrêsia brings into play the supplemental aspect of an attitude and a practice which no institutional framework could organize on its own, and which can, and which can only, vouch for itself through action – in other words, through history: ‘parrêsia, which is of course underpinned by isègoria, refers to something a bit different, which is actual political practice’ (GSO, p. 188). On the other hand, this attitude does not simply amount to a knowledge of truth, which it attempts to tell: it is not measured against the value of the truth of the enunciation that is supported


but rather the type of relation to oneself that is established through its formulation.

It is therefore a double displacement: of government as institution towards government as activity; and of Greek democracy as ordered system of rights and duties towards parrêsia as attitude towards a truth whose pre-existence in no way promises that one will appear to be ‘parresiastic’.

It is the play of mirrors of democratic modernity that is broken: its institutional definition of government, its juridical-formal conception of democracy, and the gesture that consists in justifying the latter in the name of the former, in affirming the sufficiency of modern institutions in the name of the truth and authenticity of the Greek model, as though the invocation of this founding model sufficed to call us democrats. To be democratic ‘in truth’ is, on the contrary, according to Foucault, to displace the examination of the reference to a true model of political order towards the problem of the relation to truth, which conditions the exercising of democracy. It is to suggest that a democracy is worthy because of the capacity of subjects to take risks in words. As such, the reflection on parrêsia is not only an archaeology of critique, going back as far as its Greek models (‘we could … see the appearance of a third figure of the dramatics of true discourse in the political domain, which is the figure of, let us say, “critique”’, GSO, p. 70); it is, just as much, an example of archaeological critique, regarding any pretension of democracy to founding itself in a regime whose truth or authenticity might serve as its model.

In a sense, Castoriadis’s approach can in this way seem a lot more traditional: it brings to the fore the Greek experience that Foucault leaves deliberately to the side. If Castoriadis evokes parrêsia in passing (‘the obligation to say frankly what one thinks regarding public affairs’), it is in order to underline immediately the fact that this outspokenness ‘obviously is not guaranteed by the law, but … is considered self-evident for everyone’ (CL, p. 84), to the extent that parrêsia is considered to ‘cover more or less the same semantic field and the same political function as the isègoria’ (p. 289n). Above all, parrêsia is inscribed on the grounds of the ‘formal and informal institutions of the city’ that the laws contain and that is embodied in the principle organs of democracy (ecclesia and boulè) as well as in its procedures (for example, ostracism). In response to this choice of an institutional reading comes the echo of a long development that Castoriadis devoted to Clisthenes’ reform; that is, to the reorganization of the administrative division making it possible to divide transversally, through territorial units and social groups, in a way that tears the egalitarian definition of citizenship from the hold of geography and of lineage.

The question adopted by Castoriadis is therefore that of the institution, of the way in which the city is established and affirmed collectively like a political community in action, opposed to the pre-political forms that constitute the tribe, the family or the village, in a sort of reactivation of the Aristotelian investigation.

However, this return must not create an illusion:

in Aristotle, the ‘jump’ from the village to the city is interpreted and justified by the specific end that the latter pursues (not living, but living well), an end that appears retrospectively as the cause tending to actualize itself through the successive forms of human community, to the extent that the polis appears less like an event than an accomplishment. In Castoriadis, the institution of the city is the extreme opposite of such a search for essence – and the mutation to which he subjects the word ‘institution’ is related, in its radicality, to the forcing that Foucault impresses on the idea of government. Where philosophy defines institutions as collective rules whose stability founds the political community, by giving a civil translation to the natural needs of man, it is defined by Castoriadis as the whole of the representations of systems, together, by which a society attempts to give itself a figure – that is, tries to overcome in an imaginary way the fact that no single natural or transcendent reason justifies its existence or its continuation. This was already the lesson of The Imaginary Institution of Society: all of the necessities ordinarily invoked to explain that a society exists (whether these necessities are biological, economic, etc.) are still not enough to justify the fact that we live together, that we form a ‘we’. Collectivity finds its source in a contingency and a vertiginous absence of foundation; hence the fact that all societies produce an ‘imaginary supplement’, guaranteeing social existence on a superior authority (tradition, divinity or meaning of history); hence also the fact that communist society, which sought to be the pure translation of material necessity, should have produced a radically transcendent and proliferating imaginary – this is the lesson learned from totalitarianism. The institution is therefore not a framework, but (to borrow a phrase from Foucault’s Madness and Civilization) a ‘profile against the void’, which leaves a question open: what form could a society take when, rather than repressing its own contingency under an imaginary transcendence, it assumes its own institution, its own self-creation, in the manner of a ‘we’ which would claim to belong neither to any before, nor to any other, to no being, nor to any ‘it’?

It is in this perspective that the reflection on Ancient Greece comes into play: it is not a question of elevating Greek democracy to the level of the foundation of modern systems. It is, on the contrary, a question of examining Greek democracy as an example of a society that confronted the dizziness of its own absence of foundation, and of asking of a society made to maintain itself, when it rests precisely on the idea that the way of life depends on nothing other than the will of its citizens:

Now, from what does this people of equals in the eyes of power and of the law pose and utter rights?

The greatness of democracy consists in recognizing this fundamental philosophical fact: it poses and utters rights from nothing. … nothing: this means that the law cannot be deduced from anything else, that it is not the commentary of the Decalogue nor a consequence of the theory of Plato on being. By the constitutive act, the people are self-instituted as legislator; this act describes the forms in which the legislating activity must be accomplished to be valid, forms that are valid for as long as that selfinstituting act lasts. (CL, p. 203).

From this perspective, the reference to the laws profoundly changes meaning; the latter appear less like the base from which the action and the management of the collectivity become actually possible, than like systems aiming to introduce an order inside a contingency which is, knowingly, strictly impassable. This decentring brings back into question, no less than in Foucault’s work, the idea according to which Greek democracy could constitute a mode: for this it would need to have constituted a stable ‘state’ in the manner of a thing on which it would be possible to lean:

this is, of course, a more than criticizable view, strictly metaphysical in the worst sense of the term:

Greek democracy is at no single moment a state of things but rather an historical process by which certain communities are self-instituted … as communities for free citizens. The process of democracy … is at no time a ‘constitution’ given once and for all. (CL, p. 41)Let me summarize. If, upon a first reading, the approaches proposed by Castoriadis and Foucault seem opposed to a ‘legal’ or ‘extra-legal’ reading of ancient democracy, these interpretations seem in reality instead related and complementary. Related is the manner in which they discourage any attempt to take Greek democracy as a model, by gleaning a radical critique of the idea of model from the interior examination of this democracy: the parresiast cannot allow himself truth value for what he says to guarantee what he says ‘in truth’; the democratic process is that of a community which is instituted in the eclipse of any model. These readings are also complementary, in the sense that, according to inverse trajectories (Foucault ‘descends’ from institutional frameworks to the ordinary activity of the citizen; Castoriadis ‘comes back from’ the system of laws to the absence of any base upon which they could lean), they discover finally two symmetrical insufficiencies: in Foucault, the principles do not suffice to guarantee that one will behave democratically; in Castoriadis, it is not the insufficiency of principles that is revealed, but rather the insufficiency as principle, which democracy is indeed obliged to count on. From both sides, the origin of democracy is by itself open upon history.


This strange intertwining reverberates through the strategies of reading that our two authors adopt, when the concern for defining democracy in another way brings them to pass through neighbouring references:

let us say, quickly, that in their reading of the ‘obliged passages’ of the Greek corpus, Foucault and Castoriadis meet in questionings that have nonetheless radically different styles. I will refer to two examples of this convergence.

The first example concerns, of course,

Plato. A history could be written on the ambiguous space occupied by Plato in the debate on democracy (from Spinoza to Badiou or Rancière); an ambiguous space, first of all, for the adherents to a moderate democratic model. On the one hand,

Plato is in a way the model of models.

He is the one who claimed to found the political order on the reference to a rational transcendent and fixed norm (an operation that Jacques Rancière calls, in Disagreement, ‘archipolitical’ and that according to him secretly haunts any attempt to reduce politics to a pedagogy consisting in bringing citizens to the recognition of the validity of the political order which is imposed). On the other hand,

Plato is of course a counter-model, in so far as the rigorous development of his ontological options made him adopt anti-democratic positions. Hence the fact that the adherents to an ‘open society’ against its enemies (to borrow the phrase from Karl Popper) should have come to define democracy against Plato, without ceasing, however, to borrow from him the gesture that consists in refusing the radical historicity of democracy, in order to found it upon something other than itself. If we admit, on the contrary, that Foucault and Castoriadis intended to propose a radically historical conception of democracy – that is, one that exists only in the immanence of its self-institution and in the vigilance of its practices – we see that the ‘problem of Plato’ presents itself to them in a symmetrical and inverse way. It is a question, on the one hand, of skirting what, in Plato, refers to the ontology of the eidos and of participation, of leaving to the side the philosopher of models, and, on the other hand, of detecting in Plato’s anti-democracy something other than a simple authoritarian temptation, of which our democracies should be very wary, as though this opposition sufficed to justify them and to define them.

This is why, from one text to another, we see the outlines appear of two versions of what we should call an ‘anti-anti-Platonism’: two ‘roundabout’ readings in which the Platonic reference is reinterpreted, torn from those mirror games to become not a model or a counter-model, but a paradoxical illumination of democratic experience.

This reading is only sketched in the text of Castoriadis that is available to us. (The central seminar of 8 June 1983, in which he dealt with this question in conclusion to his year of classes, has unfortunately been lost.) The elements available to us make it possible, however, to guess his reading strategy. First, Castoriadis returns to the classical reading of the Republic and the Laws, according to which Plato would have sought to found the city on a radically transcendent order: ‘it is just such an absolute that Plato seeks, a measure of the law, a norm of the norm, an extra-social standard of society’ (CL, p. 206). But he adds immediately: ‘the genius of Plato, obviously an immense genius, was therefore to find and to make explicit the only other term of the alternative, the only one which contrasts with democracy, that is, theocracy or, if we wish, ideocracy, but it is the same thing’ (CL, pp. 206–7).

Praise for the genius of Plato is only paradoxical here in appearance: what Castoriadis suggests is that Plato did not want democracy because he saw in it a radical experience that ordinary democrats most often refuse to see. In other words, it is not a question of defending democracy against Plato, but of defending democracy according to Plato against those who would be tempted to confuse it with some model of society – as though there were in the anti-democratic Plato a more acute and sharper consciousness of the radical immanence of democracy that is not the case in his moderate adversaries. Second, to this abstract analysis Castoriadis joins a historical hypothesis: if Plato did not tolerate democracy it is because it did not tolerate itself (in the double sense in which it could not stand itself and became intolerable to him). This thesis refers in Plato, on the one hand, to the Peloponnesian War, to which I will return later; on the other hand, it refers to the condemnation of Socrates, whose questioning represents a passage to the limit from the free democratic confrontation of his opinions. The Socratic elenchos is at the same time brought into question from democracy (where everyone can freely question everyone) and a bringing into question of democracy, by demonstration that ‘strictly no one knows the meaning’ (CL, p. 212).

Castoriadis concludes: ‘democracy must be able to assume the risk of this demonstration. And most of the time it did assume this; it accepted the sophists, the philosophers, etc. But it did not accept Socrates.’ As such, the Platonic search for a callipolis would constitute the resuming and inversion of the Socratic hybris, which consisted in exercising one’s democratic right to control and contesting the opinion of others, but without proposing anything in its place. Thus, writes Castoriadis, ‘one places oneself outside of the game of the city, one transgresses one its fundamental unwritten laws but that is no less the most important of them all’ (CL, p. 211) – with the law consisting in bringing back through the exchange the immanent order of opinions which interweave the existence of the city. In other words, not only does the negative conception that Plato makes of democracy outline what it effectively and radically was (a state without a model); it is also an inheritor of democracy when the latter collapses from exercising itself to the full and from no longer believing in itself.

This motif of inheritance is found precisely in the place that Foucault assigns to Plato, through his reading of the history of parrêsia. One finds in his commentary, first of all, another way of tracing the oblique vis-à-vis the traditional reading of the Platonic model. Where Castoriadis reads, in Plato’s ‘ideo cracy’, the underside of a democratic self-creation that the latter recognized and refused, Foucault insists on the irreducibility of philosophical practice, as Plato conceives it, regarding the contents and norms that it takes as objects, which it endeavours to know and to apply. The reading of Letter VII, and of Plato’s judgment of Dionysius of Syracuse, is strategic here:

it is a question of underlining, against the haste of Dionysius to raise himself to truth and to transcribe it in a treatise, that the philosopher assumes an attitude or an activity, an occupation (Foucault underlines the word pragmata) which conditions access to the truth, and whose truth itself cannot exempt the thinker.

In other words, where Dionysius claims to reduce Platonism to doctrine, Plato reminds us that the truth is inseparable from ethics. Nietzsche entirely rejected the Plato of ‘hinter-worlds’, but showed his admiration for the style of Plato. Foucault shows, for his part, that to attain the transcendence of the hinter-worlds supposes a stylization of existence which is in no way given by the latter, an exercise of philosophy of which nothing, outside of the conduct of the philosopher, guarantees victory. Although these remarks only indirectly concern democracy (in a mode that I will indicate in a moment), they seem to come from the same preoccupation as Castoriadis’s. In both cases it is a matter of reading, in the philosopher who is known for having entirely submitted the practices to a norm of truth which overhangs it, the affirmation of an irreducibility of practice – irreducibility perceived and refused according to Castoriadis, in a political order, and claimed, according to Foucault, in the philosophical order. Yet, and this is the second fundamental element of Foucault’s reading, this philosophical order is in no way foreign to the political horizon: not only will ‘the test of philosophy’s reality with regard to politics … not take the form of an imperative discourse in which men and the city will be given constraining forms to which they must submit for the city to survive’ (GSO, p. 255 – we could not better brush aside the vision of Plato as a defender of models, and of the ideal city), but this demand of the relation to self is in a sense nothing other than the democratic demand, in a way enveloped and folded back in the figure of the philosopher.

Witness to an exemplary transition, of the failure of democracy and its disintegration in tyranny, Plato becomes the one in whom truth-saying, a condition for common speech, becomes the solitary duty of the adviser and critic of the Prince. It is again a question of inheritance – the life of Socrates plays the role here not of a crisis which would have encouraged Plato to radically reject democracy, but of a mediation which displaces the exercise of parrêsia outside of the field of shared speech: [the parresiast] is no longer simply, solely or exactly that citizen among other citizens and a bit in the forefront of them. He is, you will remember – we saw this with Socrates – a citizen, of course, like the others, who speaks like them, who speaks the language of everyone, and yet who holds himself, in a way, aside from them. (GSO, p. 341) Such a reading has two principal effects. On the one hand, it has to do with making the freedom of thought an avatar of the freedom to speak. On the other hand, it makes the subject of the philosophizer not an autarchic self, but an implicated subject who exists only if he risks confrontation with power. Where Castoriadis treats philosophy and politics in parallel, the one and the other supposing that ‘chaos and cosmos coexist in nature and in the human world’ (CL, p. 8), Foucault makes the philosophical parrêsia a relay of political parrêsia: The disappearance of democratic structures does not mean the total disappearance of the question of political parrêsia, but clearly it greatly restricts its field … And as a result, philosophical parrêsia, in its complex relationship with politics, can only assume greater importance. (GSO, p. 342)


We find these complex relations between the demands of shared speech and the constitution of the subject in the second example that I wish to examine: the readings that Foucault and Castoriadis give of tragedy. If the reading of Plato focused on the very definition of democracy, tragedy obliges our two authors to confront the general question of history, its signification and the manner in which the actions of men are tied up with the fate that carries away the community. It is a question which, traditionally, leads philosophers to increase the value of the dimension of the meaning (what does tragedy teach us regarding the events of which men are the actors and the victims?), to the detriment of the event that tragedy constitutes in itself (what is a tragedy, as historical and civic practice, and what does the institution of this practice have to do with the birth of democracy?). This amounts, says Castoriadis, to seeking the political dimension of tragedy ‘in the political positions of poets, which amounts to transforming tragedies into thesis plays’ (CL, p. 139). There is an increased value, we could say, of the tragic utterance regarding the enunciation and of the manner in which, on the stage of Ancient Greece, a new manner of saying and of presenting oneself as actor of one’s speech arose. It is on this very terrain of the statement that we find Foucault and Castoriadis.

Thus, the long reading of Euripides’ Ion is an opportunity for Foucault to deploy an actual genealogy of citizen speech. If the matter is to understand how Ion succeeds in achieving the right to speak, the stages of this conquest appear like so many avatars of the act of speaking. We find, first of all, the truth

saying of the oracle, as a speech that is at the same time transcendent and masked – Foucault’s analyses quite directly echo, on this point, those of Marcel Détienne in Les maîtres de vérité dans la Grèce archaïque (a reading whose importance for Foucault has not been sufficiently underlined, and which can be detected between the lines of the 1970 ‘The Order of Discourse’). The speech of the oracle constitutes the very model of speech reserved, where truth is authenticated by its inaccessibility to those who do not have the necessary status for speaking it. This transcendence is reversed, second, in the truth-saying of the humiliated woman.

This time, it is the powerless speech of the weak that acquires the capacity to turn against the powerful and to denounce their injustice:

this complaint of injustice hurled against the powerful by someone who is weak, is an act of speech, a type of spoken intervention which is recorded, or anyway perfectly ritualized in Greek society. (GSO, p. 133) We cannot help thinking here of the intervention written by Foucault, in the same years, at the foundation of an international Committee against piracy, to help the boat-people: ‘Who then nominated us? No one. And that is exactly what gives us our right.’ [4] The cry and the imprecation seem to be the condition of democracy – more precisely, the foundation of a democracy that does not allow, for speech, any other foundation than itself. It is on this basis that the parrêsia of Ion can finally appear, and the guarantee that the god (Apollo) gives him appears finally less like the restoration of a transcendent justification than the ratification of a right which men, first, by themselves, seized: ‘the cry of humans was needed to extract from the silent god the discourse which will rightly establish the power to speak’ (GSO, p. 152).

As a result, the citizen’s truth-saying will be struck by a sort of constitutive ambiguity: it appears like a statutory privilege, sanctioned by a divine guarantee which will give it its power and make it the depository of truth and of justice. But it appears also like an exercise which nothing guarantees definitively that it will be exercised in the right way, and whose peculiarity is to be shared between its competing citizens: ‘the use of parrêsia presupposed a series of problems, or rather exposed the person who resorts to parrêsia to risks and dangers” (GSO, p. 156). The structure of parrêsia that Foucault brings to light is that of a position that is always in excess of the right and status that it can insist upon, and in that way always likely to be contested. This structure does nothing other than deploy, in the conflicting space of the democratic deliberation, the contradiction between the different modes of enunciation successively brought to light in the tragedy of Ion.It is precisely this contradiction, not between the transcendent laws of fate and the actions of men, but between the power upon which political speech can insist and the absence of transcendence which is peculiar to it, that Castoriadis brings to light in the reading he proposes of Sophocles’ Antigone. This reading does not actually dwell on Antigone and on the superior demands that she claims to assert, but rather on Creon and on the error that he committed. At the centre of the interpretation that Castoriadis proposes (and that he makes ‘the political lesson of the play’, CL, p. 145) is the argument that the son of Creon, Hemon, repeats in order to try to make his father change his decision: ‘we must listen to the point of view of the other, and … no one is ever right alone [phronein monos]’. Castoriadis underlines the fact that this argument is used by the one who (precisely because he is his son) cannot directly contest the authority of his father: ‘a son does not tell his father that the latter has made a mistake’. It is therefore at the moment when the authority of Creon, his statutory right to exercise power, is recognized, and by the one who is not in a position to deny it, that the condition of intersubjectivity of political speech is recalled. (Castoriadis comments: ‘even if one is right in one’s reasons, to listen only to the reasons one has is already to be wrong.’) As such, the tragedy constitutes an element of the democratic institution because the latter demands that individuals, at the very moment when they are emancipated from all transcendent authority, at the moment when they acquire power on themselves, continue to control themselves against a background of chaos.

We can see the convergence at work with the reading of Foucault: not only do the two authors displace tragedy from its ‘said’ towards its ‘saying’ (adopting, regarding the latter, a point of view that we could call ‘pragmatic’), but they also bring to light the complementary and contradictory conditions of speech and of democratic decision. There would be no democracy if men could not claim a relation to the true and a capacity to decide to be independent from all forms of transcendent discourse – even if, as Castoriadis remarks ironically, such a statement of fact resembles a sort of defence of Creon and of tyranny (CL, p. 145). But no democratic speech can take its authority from a foundation which would exempt it from opening up to the possibility of another speech, by claiming a sovereign and exclusive right to truth.

On this point, the Foucauldian analysis of parrêsia intersects exactly with what Castoriadis considers to be Creon’s error: [parrêsia] is a discourse spoken from above, but which leaves others the freedom to speak, and allows freedom to those who have to obey, or leaves them free at least insofar as they will only obey if they can be persuaded. (GSO, p. 104)

Democracy risked

To conclude, first of all we can see how Foucault and Castoriadis both define democracy as an exercise.

If there is no democracy without institutions and without procedures (Castoriadis), none of these procedures could exempt citizens from inventing the way of making them play and of being defined by the way in which the citizens situate themselves in relation to them (Foucault). In short, democracy is not a regime that exhaustively defines the fundamental laws and the statuses that it distributes. Second, this exercise can be defined as the constitution of speech acts, and of a subjectivity that is susceptible of taking charge of them:

the subject of democracy does not exist before it, but is defined by the manner in which, in the immanence of history, it effectively does politics, and in so doing produces itself. In this regard, we could perhaps say that Castoriadis and Foucault are not on the same level.

With Castoriadis, the subject of democracy is first a ‘we’, a collective affirmed outside of any superior reference (‘we are the instituting body, we are the source of the institution’, CL, p. 200). With Foucault, the attention is, rather, focused on the democratic ‘self’, on the particular type of individual subjectivity that democracy brings forth, and that it needs.

But this opposition is relative: from the moment that the ‘we’ is freed from all transcendent norms, not only is the community of equals defined by the participation it allows of the ‘selves’ in the democratic debate, but it could not by itself put forward its unity outside of the effective confrontation of opinions and of the possibility for everyone to prolong, modify or contest the speech of the other. This is what, a contrario, the example of Socrates shows, where his hubris consists in mobilizing in order to show the emptiness of the doxa, the ‘self’ that democracy needs. Such a risk would not exist if the ‘we’ of the community could maintain itself outside of the exchange between subjects: ‘its truth, if there is one, is constructed (by democracy) through its confrontation, opposition, the dialogue of the doxai; and it could not exist if the idea, or rather the illusion, of a truth acquired once and for all became socially effective and dominant’ (CL, p. 211). Vice-versa, the ‘self’ which is constructed in the exercise of parrêsia is not an autonomous or monadic subject: not only does Foucault show how the question of ‘government of others’ assumes a ‘government of self’, and how ethics is thus enveloped in the exercising of politics (this is the frequent interpretation of his later works); he emphasizes how much the ‘self’ thus constituted is worked from the inside by the agonistic game in which it is involved, a game that at the same time founds and dismisses its capacity to speak the truth. The ‘we’ of Castoriadis and the ‘self’ of Foucault do not simply fit snugly one into the other; nor do they simply contradict each other dialectically. They exist only in the tension that links them, each time that the democratic exercise is undertaken.We are touching upon the last lesson to be gleaned from the confrontation of our two authors: what I propose to call the idea of ‘risked’ democracy. In Castoriadis, this constitutive risk of democracy comes directly from the absence of an exterior norm, which would contain the decisions taken within reasonable limits, as illustrated in the Peloponnesian War:

that democracy does not contain an assurance against its own excessiveness is shown in the form of an effective historical tragedy which will last for twenty-seven years, in the democratic city par excellence that is Athens. … The failure of democracy in and through the Peloponnesian War seems to show that the people are not capable of self-limiting, of posing and of saying rights, of correctly governing themselves. (CL, pp. 204–5) Castoriadis’s phrase, which introduces his reading of Plato, is ambiguous. It suggests that, if Plato was wrong to seek the essential norm capable of putting an end to this risk, he was right to think that hubris always haunts democracy (‘no one and nothing can secure us against ourselves’, CL, p. 205). If we stopped there, we could conclude that democracy must watch, above all, over its forms, and be all the more vigilant with regard to its institutions if the latter are alone in protecting us against chaos. (This is the reading that a certain number of French authors from the ‘republican’ movement glean from Castoriadis. Marcel Gauchet, for example, in whom the respect for objective forms of the Republic must be preserved, above all, because they preserve us from the dissolution tendentially brought about by democracy.) This is where the reading of Foucault goes further. For Foucault does not simply describe how the democratic procedures are instituted on the basis of a risk that is always present, or call upon us to confront this risk lucidly: he shows how this absence of guarantee, this precariousness, is not simply a fault in the background of institutions, but an internal condition of the democratic game, an element that is indispensable to its functioning.

The play of parrêsia is indeed inseparable from a tension since it articulates three dimensions: (1) true discourse is necessary to democracy; (2) democracy threatens true discourse (through the temptation of demagogy); (3) true discourse threatens democracy (by recognizing in some an expertise and a knowledge which gives them an ascendancy over others). Parrêsia cannot therefore regulate the exercise of democracy by confronting a whole series of disturbances that are born not from the absence of transcendent norms, but instead from its own norms. The risk exists that the people will no longer stand the ascendant critique exerted by a few. But the risk nonetheless exists that some will use their critical posture in order to raise themselves above others. There is a risk in allowing just anyone to speak; but there is just as great a risk in not letting just anyone speak. And there is a risk in lacking the necessary courage to speak the truth; but there is also an opposite risk in using truth to mask one’s absence of courage (which is done, often, in politics, by those who invoke ‘realism’ to avoid having to get involved against injustice). It is not only, therefore, that democracy is a thing that is as precious as it is precarious, as Castoriadis emphasizes; Foucault adds that this precariousness is in itself precious, since it makes democracy both possible and impossible as a game of truth.

Translated by shane lillisnotes

1. ^ Michel Foucault, The Government of Self and Others: Lectures at the Col ège de France, 1982–1983, trans.

Graham Burchell, Palgrave Macmillan, Basingstoke, 2010, cited in the text as GSO.

2. ^ Cornelius Castoriadis, La Cité et les lois, Le Seuil,

Paris, 2008, cited in the text as CL.

3. ^ Michel Foucault, ‘Face aux gouvernments, les droits de l’homme’ [‘Facing Governments, The Rights of Man’], Libération 967, 30 June 1984, in Dits et écrits, Vol. IV,

Gallimard, Paris, 1994, p. 707.; trans. at

Buy the newest RP in printDownload the PDF