"Because it is sweet to hear [during¹] the night a concert of flutes which is performed in the distance and of which it reaches me only some scattered sounds that my imagination helped by the sharpness of my ear, I believe that the concert performed close up has its price. But can you believe it, my friend? It's not this one, it's the first one that intoxicates²."

Diderot is addressing his friend Falconet here, initiating with this metaphor a series of letters about posterity. The flute concert heard at night in the distance metaphorizes the voice of posterity. From its snippets, we reconstitute the fullness of the melody, we imagine what "the concert that performs up close" can be, what it might have been. A strange present that defines an irreal: this close-up concert that we lack, to which we have no access, "does have its price", but it's not so precious, because deep down it remains external, foreign to us. The other concert, the one we refigure for ourselves, in ourselves, because it "is largely the work" of our imagination, is the concert that intoxicates us: it is the potential concert of posterity.

"But will you believe it my friend? It's not this one, it's the first one that intoxicates." Diderot addresses Falconet in the future tense: "will you believe it". No doubt this future tense is essentially rhetorical; no doubt it can be explained by the epistolary context: will you believe it, my friend, when you receive my letter, when you read these lines? Nevertheless, the other concert is projected into the future, virtualized by this future. Falconet makes no mistake, moreover, responding to Diderot:

"The comparison you make of the concert is very sweet, it is pleasant and seductive; but to make it just, you would have had to say, I hear a concert and I believe there will be another after me³."

Diderot superimposed two presents, potential and unreal. Falconet re-establishes them in the logic of a succession. What intoxicates me is not only the reconstitution, from the flute concert I am now listening to in the distance, of a presence in the fullness of a bygone concert; it is also the projection, from the concert now being performed before me, of future listeners who will perceive snatches of it in the future and in turn reconstitute today's concert. Posterity is part of a chain, and the exhilaration it brings is above all that of the prospect of "it will have been". The intoxication of the concert of posterity is not conjugated in the present tense, but in the future past.

Diderot and Falconet rework here an example that comes from Augustine's Confessions. Confronted with what Ricoeur would call the aporia of time, Augustine gives the example of a voice that resonates and that I listen to:

"Behold, for example, a bodily voice begins to resound, it resounds and resounds again and ceases to do so; already it is silence, this voice has passed away (praeterita est), there is no longer a voice. It was future before it was heard, and could not be measured, since it was not yet; and now it cannot be measured any more, since it is no longer. It could be measured while it resounded, for then it was a measurable thing. But even then, it was not stable (non stabat), it went and passed (ibat enim et praeteribat); And isn't that why it could be measured better? For while it passed (Praeteriens enim), it extended (tendebatur) over a space of time that made it measurable, the present having no extent⁴."

I will not repeat here Ricoeur's analysis of the Augustinian contradiction between a present that ontologically has no extent and the experience that nevertheless imposes itself on us of this extent, the necessary experience by which we measure time and grant it a duration⁵. Diderot pulls the Augustinian example in another direction, and draws our attention to another paradox: while it is through its sensible continuity that we can apprehend time as time, grasp its notion and experience it, this experience of time appears, against all expectations, profoundly and even doubly discontinuous. A voice that is heard is never full and continuous, but a succession of interruptions and restarts, "scattered sounds" that imagination alone "manages to bind together". On the other hand, the present of sound is a double present, that of the "concert that performs far away" and that of the "concert that performs close up": but the co-presence of these two presents is phenomenologically unthinkable.

.

Also for Diderot, who here is merely forcing the line in the Augustinian example, the true experience of time is an experience of the snippet: "a few scattered sounds that my imagination aided by the finesse of my ear, succeeds in binding together, and of which it makes an ongoing song". The experience is not one of a continuous song: it first takes in the incomprehensible, the heterogeneous, the fragmentary. Only then does temporality unfold subjectively, both as the intrinsic temporality of the song, to which imagination gives its "semantic relevance⁶", and as the extrinsic temporality of posterity, through the linking of this concert with the other, then of this other with the one I imagine of my own posterity. This double temporality, I must and I cannot think its copresence; the two presents of actuality and posterity, of the near and the far, of the link and the snippet, are in the present but are not copresent.

Snippets of time, projected into the future anterior: that time will have been mine, and will only be grasped as my time in the refiguration that my posterity will make of it, if it comes to pass. The perforated character of time⁷, its dimension if we may say preterite, can thus be formulated both as immaterial materiality and as discontinuous continuity.

This "aporia of time", which Ricoeur draws from his reading of Book XI of Augustine's Confessions, is resolved, according to him, by relating it to an inverted aporia in Aristotle: this is the thesis of time as "something of movement", τί τῆς κινήσεως⁸. It does not immediately manifest itself to him as an inverted aporia, but first as an "inverted image⁹", for Ricoeur begins by pairing with Augustine not the Physics but the Poetics of Aristotle. It is on this inverted articulation that the stroke of genius of Time and Narrative rests. Aristotle comes to repair Augustine, as it were: the play of intrigue, the work of fiction supplements the original aporia of time, which the Confessions address. And indeed, Diderot in his own way abounds in this sense: the snippets of the flute concert heard intermittently in the distance only take on meaning and consistency from the moment my imagination reweaves the musical sequence, reinvents the composition. The experience of reality is first and foremost the experience of heterogeneity and discordance, which the mind reverses into "the preponderance of concordance¹⁰", by "the configuration of the plot¹¹", or at least here of the musical composition.

The weak point of Ricoeur's imagined coupling, of which he warns us from the very first pages, is the absolute heterogeneity of the texts he relates, from the sensitive experience of time in the Confessions on the one hand, where there is no question of plot, of the structuring of plot in the Poetics on the other, where there is no question of time¹². What does this heterogeneity mean? It's not a question of the theme of the two texts, or the philosophical orientations of the two authors, or even of time and narrative. On the one hand, what Ricoeur reads in Aristotle as the exposition of the principles of "mise en intrigue¹³", mobilizes a freedom of "choosing" supported by an ethic. On the other hand, what he isolates in Augustine, radically separating him from his mystical foundation as well as from his inscription in a theology, is a sensitive experience based on a knowledge of nature. Ricoeur's Augustine is a Kant of pure reason, while his Aristotle makes practical reason speak for itself. The pairing of the two, highly improbable yet absolutely necessary, is the pairing of the Third Critique, from which to establish a faculty of judgment and, more specifically, an aesthetic judgment. The weak point exhibited is therefore not a weak point: on the contrary, it constitutes the post-Kantian foundation of the hermeneutical gesture.

Let's recall how Kant dramatizes this pairing of heterogeneities, at the end of the introduction to the Critique de la faculté de juger :

"The understanding is legislative a priori for nature, as an object of the senses, with a view to a theoretical knowledge of it in a possible experience. Reason is legislativea priori for freedom and its proper causality, as suprasensible in the subject, with a view to unconditioned practical knowledge. The domain of the concept of nature under the first legislation and that of the concept of freedom under the other legislation are, despite all the reciprocal influence they may have on each other (each according to its fundamental laws), entirely separated by a great chasm, which disjoins the suprasensible from phenomena¹⁴."

That Augustine's approach, in Ricoeur's reading of it, consists in delineating how the understanding legislates for a theoretical knowledge of time in a possible experience of it, and from the difficulty of this possibility, is clear enough. With the Kantian model in mind, Ricoeur's proposed approach to Poetics also becomes clearer. This approach is resolutely anti-structural: no, Aristotle did not seek to identify the typical structure of tragedy, and it's not a system of forms we're talking about. For Ricoeur, there is a "dynamic mark¹⁵" that the word poetic imprints on the entire treatise: it is not, with the muthos, about a narrative made, but about an arrangement to be made of the narrative; it is not, with the mimesis, about imitation, but about mimetic activity:

"action appears as the "main part", the "intended goal", the "principle" and, if we may say so, the "soul" of tragedy. This quasi-identification is assured by the formula: "It is the plot that is the representation of the action" (50a1)¹⁶."

In other words, the Poetics, and behind it the literary text in general, will be read as a combinatory of deliberations on the action to come, of imaginations of the distant concert. Mimesis is seen not so much as the actualization of a representation as the projection of a posterity. From this perspective, literature is always metaleptic: it doesn't say things that are happening; it shows how they could be said¹⁷. Not only does Poetics, when it defines the parts of tragedy, in fact define the parts of the art of composition, but narrative fiction, when it unfolds its episodes, at the same time unfolds choices of action from a combinatorial whose system is revealed by its implementation in the narrative. This is why every plot is set in intrigue, every action is the implementation of deliberation and choice. In this sense, Poetics is thought of as a Critique of Practical Reason, in which Kant proceeds by antinomies and resolutions just as Ricoeur, in Time and Narrative, clears up aporias to then elucidate enigmas.

The Kantian antinomy of practical reason gives Ricoeur the general model for determining the praxis of the Aristotelian tragic hero. For Kant, what is at stake is the union of virtue and happiness: yet the pursuit of happiness hardly accords with the practice of virtue, it is not the cause of it; and conversely the implementation of the sovereign good is unlikely to accord happiness and virtue. To resolve this antinomy, Kant proceeds as he did for the antinomy of pure reason, deconstructing the notion of event, which he breaks down into phenomenon and noumenon. For the same event, we can in fact consider the acting being both as a phenomenon in nature, determined by natural causality, and as a noumenon, pure intelligence "which escapes the determinations of time" and asserts itself as "free with regard to all laws of nature¹⁸". If I start from the pursuit of happiness, i.e. the satisfaction of desires, in all cases, whether I consider action in the natural chain of phenomena or as an object of abstract deliberation for the subject, it will never necessarily enable the acting subject to attain the Sovereign Good. Conversely, if I start from the opposite pole, the pursuit of virtue, in the sensible world experience clearly shows, alas, that not all actions are motivated by virtue. On the other hand, if I place myself in the intelligible world of choices which, virtually, present themselves before the decision and engage an abstract combinatorial of possible actions, then the search for the sovereign good is possible, and this search will have a connection, more or less direct admittedly, but a connection with the search for happiness¹⁹.

The question of virtue and happiness is not of direct interest to Ricoeur. On the other hand, the Kantian division of the event underlies the general interrogation of Time and Narrative. To maintain a virtuous causality in the determinations of practical reason, Kant is obliged to divide the event into phenomenon and noumenon, into experience of what happens in nature and the speculative choice that presents itself to the subject before acting. In other words, the event is both an experience and a choice. As experience, it is plunged into the Augustinian aporia of time that passes; as choice, it is converted into the speculative structure of mise en intrigue.

So, strictly speaking, there is no longer an event. If we enter time, the deliberation at the crux of the plot vanishes into the flow of experience and the event becomes unrecognizable as an event; if we leave time to consider the structure of deliberation, this structure becomes a logical form, detached from the experience of time. This problem is directly linked to that of the consistency of time as time with holes in it: there is a compactness to the event, which is not what passes, but what happens and is maintained as coming, whereas intermittence constitutes the touchstone of the phenomenological experience of time. This compactness of event is the very compactness of plot, which etymologically designates not the end we're aiming for, nor the detail of a narrative, but the entanglement, the density of entangled actions and intentions, what Raphaël Baroni calls "narrative tension²⁰". However, unlike the plot, the event carries this compactness phenomenologically, as a non-theorized, non-modelizable experience, as the randomness of nature: in the post-Kantian perspective, this density and randomness, this evidence and heterogeneity, are irreconcilable. This is why, in Husserlian phenomenology (and consequently in the hermeneutics and narratology that draw on it), Ricoeur points out that the event is reduced to almost nothing:

This is why, in Husserlian phenomenology (and consequently in the hermeneutics and narratology that draw on it), Ricoeur points out that the event is reduced to almost nothing.

"In order to begin his analysis of retention, Husserl gives himself the support of the perception of an object as insignificant as possible: a sound[...] To this minimal object - the sound that lasts - Husserl gives the strange name of Zeitobjekt, [...] tempo-object²¹. "

The already simple example of Augustine is further stripped down, reduced to an almost duration-free experience. Husserl intends to grasp time at the moment of its emergence in phenomenological experience, whose metaphor here will be the advent of pure sound. Sound thus reduced to its "now" constitutes a tempo-object: it has the punctuality of the object, yet "comprises a longitudinal intentionality", i.e., the virtuality of an extension in length, in duration: extension towards the past, that is, retention, which is in it the trace of what has preceded; extension towards the future, or protention, which is the projection of what will follow. We recognize here, in a highly refined and disembodied form, the pregnant instant theorized by Lessing and Diderot²². But whereas classical aesthetics proceeds by condensing a narrative into a moment, Husserl emphasizes the originary character of the tempo-object, the paradox of this ceaselessly begun origin, which is also an end:

"It begins and it ceases, and the whole unity of its duration, the unity of the whole process in which it begins and ends, 'falls' after its end into the ever more distant past." [24] (37)²³

We see the tempo-object coming as that which, as soon as it has happened, will fall²⁴ immediately into a past that will irretrievably recede: it will have been; what allows the moment to persist as time-consciousness is this aura-été, which makes the future anterior the form of the temporal inscription of time. In Husserl, however, the narrative dimension of the pregnant moment falls away: there is no plot; Ricoeur will reintroduce it in the form of the historicity of the Heideggerian being-for-the-end (zum Ende sein) in the following chapter²⁵. We should, however, ask how this narrative dimension fell: to do so, a return to Diderot is in order.

In Diderot, the metaphor of sound reappears a few years after his letter to Falconet, when he writes Le Rêve de D'Alembert. This was in 1769: the Salons, the site of the full development of the Diderotian instant, were no longer then the privileged locus of his creative activity, as they had been during the letters on posterity. Le Rêve engages in a thought experiment other than that of the creative moment superimposed on the instant of representation: the device, here, consists in contradictorily superimposing a materialistic system made (by Diderot, by Bordeu, both supposed to know) on a system being made, or to be made (by D'Alembert, by Julie de l'Espinasse). Metalepsis consists in deliberately confusing the process with the system. So music is no longer invoked this time as the music of fame, but as a materialist image of human thought in the process of happening:

"Diderot. [...] The sensitive instrument or animal has experienced that by making such and such a sound such and such an effect ensued outside it, that other sensitive instruments similar to it or other similar animals approached, moved away, asked, offered, hurt, caressed, and these effects became linked in his memory and in that of others to the formation of these sounds²⁶. "

Apparently, nothing here evokes time. Diderot places himself upstream of the emission of sound, at the moment of its "formation": a string is plucked and, immediately, makes a sound. This sound does not last; it manifests itself logically, as the virtuality of a sound. As for the constitutive linking of meaning, this only occurs after the fact, in memory, through the association of the effect caused with the combination of sounds that triggered it. In other words, a meaning is first produced only by chance, meaning is only established retrospectively.

It is the effect that makes meaning, and its recurrence that fixes the semantic link: a language is formed from the recording of events, materialized here by the verbs "approached, moved away, asked, offered, hurt, caressed". But the succession of these verbs in parataxis does not make a sentence, does not order a narrative²⁷. Strictly speaking, there is no time (in the sense of appreciating a duration), no narrative: events manifest themselves as verbal parataxis; the juxtaposition of imperfects is not syntactically articulated. A vertigo of universal synchronicity then opens up:

"And to give my system its full force, notice again that it is subject to the same insurmountable difficulty that Berkeley proposed against the existence of bodies. There is a moment of delirium when the sensitive harpsichord thought that it was the only harpsichord there was in the world and that all the harmony of the universe went on in it²⁸."

The absolute materialism Diderot experiences here is the inverted replica of the absolute idealism embodied by Berkeley: out of time, the thinking harpsichord thinks it is self-sufficient and can do without interlocution; because it can mimic all events within itself, it imagines itself to contain them. What happens when the sound is no longer a voice heard from outside, but music produced within me? When this music is not a melody, unfolding in time as the inflection of a song, but a harmony, i.e. a chord, a creation seized at its original point, and from the outset a polyphony, a concert, i.e. a world?

Diderot advances cautiously here. He makes no overt claim to propose an alternative theoretical model: he refers to an "insurmountable difficulty" and "a moment of delirium". But delirium is precisely, in Le Rêve de D'Alembert, the touchstone by which the dialogue expresses its materialist truth in the form of a mi-dire²⁹, i.e. as affirmation through denial, as backbiting to put it better. This is what D'Alembert's intervention in the dialogue will shed light on:

"D'Alembert. For example, we don't quite conceive, according to your system, how we form syllogisms, nor how we draw consequences.
Diderot. It's because we don't draw any: they're all drawn by nature. We only state joint phenomena, whose connection is either necessary or contingent, phenomena that are known to us through experience: necessary in mathematics, physics and other rigorous sciences; contingent in morals, politics and other conjectural sciences³⁰."

D'Alembert objects to Diderot's Aristotle: not his Poetics, but his logic and, with the syllogism, the theory of discourse that underpins it³¹. From the model of the thinking harpsichord, which is a model without time or narrative, events cannot unfold in the sequence of a discourse or reasoning. How can Diderot think the event if his thinking harpsichord does not deduce it logically, by syllogism, if it does not inscribe it in a causal chain that will at the same time be the chain of the discourse he will produce, through which he will configure it as discourse and as history?

Diderot rejects this solution, the collateral damage of which is the splitting of the event: in D'Alembert's system, inherited from Aristotle and formalized by Kant, on the one hand the event is chained in nature, on the other it is configured by discourse; on the one hand we experience it, on the other we could make a choice about it. For Diderot, nature alone draws the consequences, so there is no speculative autonomy for the subject. This refusal appears at first as a terrible restriction: no initiative, no creation for the subject, who "merely states joint phenomena".

But this conjunction actually opens up the essential dimension of the event, which is its simultaneous plurality, its complex harmony. Yet in the polyphony of the event, we don't hear everything: to the clarity, to the evidence of necessary connections, Diderot contrasts the contingent, even conjectural connections of action, particularly in the fields of morality and politics. These conjectural connections, open therefore to the supplement of the imagination, are bound to encounter D'Alembert's skepticism:

"D'Alembert. Is the connection of phenomena less necessary in one case than in another?
Diderot. No; but the cause undergoes too many particular vicissitudes that escape us, for us to be able to count infallibly on the effect that will follow. The certainty we have that a violent man will be irritated by an insult is not the same as the certainty that a body that strikes a smaller one will set it in motion. D'Alembert.
D'Alembert. And the analogy?
Diderot. Analogy, in the most compound cases, is only a rule of three that is executed in the sensitive instrument³²."

In contrast to the immediate and pressing necessity of physical phenomena are the vicissitudes of moral causality, i.e. the multitude of weak phenomenal connections that blur, disrupt and sometimes radically alter the production of effects. This is not simply a matter of complexifying data within an unchanged logical model; what is at stake is the very principle of syllogistic causality invoked by D'Alembert. D'Alembert then proposes an alternative to the syllogism, analogy, which is scarcely more successful with Diderot: each time, these are simple logical operations, which do not account for the complex causality that governs moral and political affairs.

Diderot calls this complex causality vicissitude: vicissitude is the dynamic expression of time perforated, not essentially by the intermittence of the same voice ceasing and resuming, but by the impossibility of sensitively experiencing all the weak links. You can't hear everything in the distant concert. Time is accessed through this pas-tout of sensitive experience, a pas-tout of jouissance³³ that is first and foremost a synchronic, harmonic, simultaneous pas-tout. Time unfolds from the immediate experience of this vicissitude in the linking of phenomena: the system of links is not clear, is not complete, because its design is blurred by the skein of weak links. The test of time is the test of the intensification of this blurring. The system of links enters into vicissitude, starting from the not-at-all that we experience, that we compensate for by imagination, and from which we derive enjoyment.

What Diderot means here to D'Alembert, at the end of the first of the three interviews in the Rêve, prefigures in some ways what will be Derrida's debate with Ricoeur. At the end of Ricœur's inaugural lecture³⁴ at the Congrès des sociétés de philosophie de langue française held in Montreal in 1971, and following Derrida's paper³⁵, Derrida and Ricoeur had a lively exchange. Ricœur criticized Derrida for not having proposed a theory of discourse, and for having attributed to writing a whole series of "traits" that in reality belong to discourse. Derrida's response is that, while such a theory is necessary, he is upstream of it, at the level of what it presupposes. What are these presuppositions?

"These presuppositions are the ones I drew very schematically this morning, namely that something like the event, for example, was self-evident, that we know what an event was ; Now, a theory of discourse presupposes a theory of the event, a theory of the speech act, a theory of the act as a singular event, and on this concept of the event, for example - but this concept of the event forms a schema [I must have said "chain"] with a whole set of other concepts - I tried to mark out what prevented any supposed event (singular, actual, present, irreplaceable, irrepeatable, etc.) from constituting itself as an event in this sense.) from constituting itself as an event in this philosophical sense, that is, what divided its singularity by the simple fact that this event was a kind of discourse, let's say quite simply a semiological event³⁶."

These exchanges must, of course, be seen in context: if Ricoeur advances the discourse, it's because he has just centered, on that day, his lecture on discourse, while Derrida responds with the event because, on that day too, his paper was about the event. We must also beware of dates: in 1971, Ricoeur had just published Le Conflit des interprétations (1969), but had not yet written either La Métaphore vive (1975) or Temps et récit (1983-1985).

However: here points to what we have identified as the blind spot in Temps and Narrative, the unquestioned evidence of the event, whose dimension of singularity Derrida highlights as irreducible to the generic categories of discourse. Discourse generalizes the event as a semantic structure, while the event erupts upstream of this structure, as non-discursive meaning-making, as a "semiological event" in the sense that semiology accounts for the effects of meaning that cannot be semantized by the categories of discourse. The deconstruction of the event makes it possible to think of the compactness of the event as the irreducible singularity of an advent, and at the same time as that to which we have access only through the lacunar traces of its generic representation in discourse.

Scheme/String. Has there really been a transcription error, or is Derrida playing it up? The event makes schema and makes chain: making schema, it fades into the generic categories of discourse, which it founds; but it is no longer the event; making chain, it remains immersed in a network of concepts that it motivates and even conditions, but which dissolve it as an autonomous philosophical concept, and prevent it from being thought. In a way, this double logical impossibility preserves the event: it instills respect at its threshold. Therein lies the limit of discursive interpretation and, in the gushing polyphony of the philosopher's harpsichord, the post-modern announcement of a grammatology of devices.

Notes