The problem of description in Diderot's Salons
Our modern conception of description is largely dependent on the realist novel, in which Gérard Genette distinguishes four genres of narrative: narration, description, dialogue and discourse1. Such a categorization implies a division of the functions of narrative: action is assigned to narration; ideas, the ideology of the story, to discourse; and the speech of the characters to dialogue. Description, for its part, is the poor relation: essentially decorative, it dresses up the landscape and embroiders the action. Its very gratuitousness, according to Roland Barthes, would provide an alibi for authenticity: by producing the detail that will serve no purpose in the unfolding of the narrative, description triggers the real effect of a thing seen, which cannot be invented2.
Faced with Diderot's Salons, it's tempting at first to simply ignore these categories, so different in context and genre from the narrow framework in which they were conceived. However, a diachronic approach could prove fruitful. If we consider that the whole of the Salons is conceived as a vast description, we realize that Diderotian description covers the highlighting of the action of the painting, that this action is in principle always based on an oratorical performance, on the words of the characters, that these words, finally, which contribute nothing to a narrative, biblical or mythological, supposedly known in advance to the spectators, take the place of discourse through the visual play of allegory, moral, religious, or political.
Description is therefore not a genre of narrative: as a total expression, it is a narrative, for which the arrangement of figures in the space of the scene is only a part, a modality, a genre of description. We cannot, therefore, confine description to a logic of the list, to an enumerative system that would oppose the articulations of discourse and the noble structure of narration3, even to demonstrate that with a list we can produce narration all the same. After all, there's nothing to indicate that narration is the end of the story; still less of the Salon.
There is therefore, above all, a sprawling dimension to description which, before the advent of the realist novel, can be defined neither as a modality of the novelistic narrative to be complemented by other modalities, nor as a typified practice of language, based on enumeration and the list, and which would be opposed to another practice, based on the logical sequence of ideas, or the chronological sequence of events. Of this type of conception, which is currently flourishing in all textbooks and dictionaries, we find no trace in the Logique de Port Royal, nor in the article Description of the Encyclopédie, which is inspired by it4.
Description is not, however, defined here according to Aristotelian categories, as ekphrasis, or as epidictic discourse5, even though, as we shall see, this tradition plays a decisive role in Diderot's descriptive practice. For Abbé Mallet, description is above all an "imperfect definition": description is therefore not opposed to narration, but to definition; because it does not reach the essence of the object, but only its attributes, it is experienced as peripheral and worked by lack. Deceptive, it comes only as a supplement to an object that is lacking: it is first of all the absence of the object that the description says.
I. Description as mediation: envelopment, supplement
In the first Salons, Diderot does not name what he does:
"Here's pretty much what you've asked of me. I hope you can make the most of it.
A lot of paintings, my friend, a lot of bad paintings. I like to praise. I'm happy when I admire. I'd like nothing better than to be happy and to admire...
It's a portrait of Marshal d'Estrée" (p. 1936)...
"what you asked me", "en tirer parti", "C'est un portrait": to designate his text, Diderot remains in the indefinite. Very indirectly, he sets a horizon of expectation, praise: "I love to praise. I am happy when I admire." Praise is the genre of ekphrasis: describing a work of art is a piece of oratorical bravura; it's about saying the beautiful in beautiful language; the evocative power of the word will virtually restore the image only on condition of unconditional empathy, of enthusiastic adherence to the painted idea7.
We soon understand that this expectation will be disappointed: "I asked for nothing better than to be happy and admire..." The imperfect tense indicates that one expectation is substituted for another, that the praise demanded by the genre will be turned on its head in a sharp critique as only Diderotian verve is capable of.
Nowhere is there any question of describing: "There's a Annunciation by Restout" (p. 193), "There's a bad Adoration of the Kings by Collin de Vermont" (p. 195), "There's a Retour de chasse by Chardin" (p. 197). The Salon de 1759 is chanted by this litany of presentations, il y a, c'est : objects are designated crudely, detached from any relationship, any subjective articulation, any face-to-face encounter. Praise or blame do not yet explicitly construct an aesthetic relationship.
This relationship begins to be formulated at the start of the Salon de 1761 :
"Here, my friend, are the ideas that went through my head at the sight of the paintings exhibited this year at the Salon. I throw them on paper without bothering either to sort them or to write them down." (P. 201.)
This time, the presentational brutality of the text is consciously exhibited. In 1759, objects were thrown out without mediation, with a il y a, a c'est. In 1761, Diderot says he "throws on paper", not the paintings themselves, or their description, but "the ideas that went through my head at the sight of the paintings". A relationship emerges, and a mediation: between image and text, the ideas are the medium of a description that is not yet said. The brutality of the introductions fades: "The first painting that stopped me was the Portrait of the King." (P. 201.) Here before us is an outline of a stroll, then the pause of a glance, the meeting of the Diderotian eye and the image. "Whatever the dear abbé may say La Madeleine dans le désert is only a very pleasant picture." (P. 202.) In front of this painting, then, there was a lively conversation between Diderot and Galiani, a disagreement in which the image is taken to task and only reaches us inscribed in this circuit of gaze and word. Inscribing the image in order to describe it: description is an escape from the framework of the representation it presentifies only if this framework has first been posited, installed in a relationship, so that a polarity is established; descriptive speech circumscribes its object, but performance undoes all circumscription; enunciation fixes, inscribes the image in a place and in a sociability, but the effect of the statement is metaleptic, it makes us forget the absence of the painting, its nature of representation and artifice: "it's the thing itself"8, out of all relation.
It wasn't until 1763 that the word description appeared for the first time, to designate precisely this performance that cancels out the object's inscription:
"Having paid this light tribute to the man who instituted the Salon, let's come to the description you're asking me for.
To describe a Salon to my liking and yours, do you know, my friend, what it would take? All kinds of taste, a heart sensitive to all charms, a soul susceptible to an infinite number of different enthusiasms, a variety of style to match the variety of brushes; to be able to be grand or voluptuous with Deshays, simple and true with Chardin, delicate with Vien, pathetic with Greuze, to produce all possible illusions with Vernet. And tell me, where is this Vertumne? You'd have to go all the way to the shores of Lake Geneva to find him, perhaps.
But it's far away, and while one's head rests on one's hands or one's eyes stray into the air, one's mind tires, and one draws nothing but insipid, cold lines. But I'll be even with you if I do my best and tell you my old song again:Si quid nosti rectius istis,
Candidus imperti: si non, his utere9." (P. 237.)
"La description que vous m'en demandez" here clearly designates the entire Salon, and characterizes what remained informal in the 1759 entry, "Voici à peu peu ce que vous m'avez demandé."
The description responds to a request, fills an expectation, a lack; it makes up for the absence of its object, which is not "in front of you", but "far away". It makes up for it for want of anything better: si quid rectius... It's not just a question of responding to Grimm; the description responds to the object it describes. The image, the painting itself, is in demand, since it's a question of deploying "a variety of style that responded to the variety of brushes" : through the play of syllepsis, two tools oppose each other, style on the one hand, the punch, the poet's quill, the brush on the other; two techniques therefore enter into trade, organize an exchange10.
Faced with this demand, one would have to be a Vertumne, declining all the modes of sensibility before all the painter's figures: the Salon description is the exercise in this declension, deploying a range of Diderotian portraits in the same way that, in a single painting, the pictorial scene deploys the range of its figures. In the history painting, the event represented, the scene itself, triggers a range of reactions in the painted characters who participate in or witness it, a range of responses disposing a variety of figures on the canvas; similarly, the overall event of the Salon disposes a variety of paintings in the exhibition space, to which responds the range of descriptive techniques that Diderot is summoned to employ in his Salon.
Description thus constitutes a set of responses in relation to a given event; it unfolds in a range around that event. The event is the heart of description: not only the overall event of the Salon, but, for a given painting, its subject, the action it implements, and therefore the narrative that underpins it11. Description is not a skill, or a discourse functionality that would oppose a narrative functionality: because it responds to the event, it supplants this narrative, it envelops and understands it.
To unfold and achieve this envelopment of the object, description proceeds by empathy12. This is what is meant by the image of Vertumne, found in the exergue of Rameau's Neveu to designate the Nephew's infigurable character, and the multifaceted performance of his pantomimes, but also to characterize the dialogic device: Vertumne's metamorphoses respond to the Master's discourse that is first held by Me, then circulates from Me to Him. But pantomime doesn't respond by opposing one discourse to another, not even one point of view to another: its response is empathetic, miming the discourse of the Other, incorporating it, and rejecting it only through the body that has enveloped it, in spasm and burlesque parody13. Description proceeds in the same way, and for this reason it is at once a mimetic transfer of the object14, a mechanical transposition, so to speak, of the image into the text, and already a critique and judgment, a distancing from the object: in description, incorporation turns into abjection, involution into revolt.
The description "seeks the composition" of the painting: composition is the plastic shaping of the event it represents. Through this shaping, it defines the painting: description is an imperfect definition; bent entirely on the essential restitution of this composition, it captures only its attributes, its dispositions, its order. The order is a mute composition15, a spatial arrangement that doesn't yet make sense. Ordering is the basis of description, from which it points to composition; objective, but insignificant, borderline illegible, ordering kills the object and bears its own disfigurement16.
II. Deceptive dimension of description
Description tends towards an unattainable goal: it is circumstantial and its goal is essential; it envelops its object and, enveloping it, it supplements it but never reaches it. Diderot gives this proliferating logic of the supplement a historical explanation, in the sixth site of the "Promenade Vernet":
"Mores become more polite, barbaric, poetic and picturesque usages cease, and it is incredible the harm this monotonous politeness does to poetry. The philosophical spirit brings with it a dry, sententious style. Abstract expressions containing a large number of phenomena multiply and take the place of figurative expressions. The maxims of Seneca and Tacitus everywhere succeeded the lively descriptions, the tableaux of Titus Livius and Cicero; Fontenelle and La Motte to Bossuet and Fénelon." (P. 620.)
Language and expression began by being "poetic and picturesque", i.e. creating meaning (the poetic) through painting (the picturesque). Figurative expression, which makes sense by showing figures, preceded abstract expression, by which discourse dispenses with seeing and bypasses description. Titus Livius and Cicero express themselves in "tableaux" and "animated descriptions": they describe, as poets. Seneca and Tacitus, favoring the concise formula, the ellipsis, the maxim, proceed by maxims, as philosophers.
Description is an archaic modality of language. Here, Diderot takes up the reflections he developed in the Lettre sur les sourds following Condillac17, and which led him to define all poetry as "hieroglyphic"18, i.e. as both text and image, as an immediately visual expression and as a discursive journey. Poetry carries within it the history of language and, at its most refined, reconnects with its most primitive origins. A polarity then emerges beyond history, not between narrative and descriptive competence of language, but between poetry and philosophy, between imagination and judgment, between what expression gives to see and what discourse gives to think. With reference to a Petite, très petite ruine by Hubert Robert, Diderot returns to this:
"The more we detail, the more the image we present to the minds of others differs from the one on canvas. For one thing, the extent our imagination gives to objects is always proportionate to the enumeration of the parts. There's a sure way of making the listener mistake an aphid for an elephant. It's simply a matter of pushing to excess the circumstantial anatomy of the living atom. A very natural mechanical habit, especially in good minds, is to seek clarity in their ideas; so they exaggerate, and the point in their mind is a little bigger than the point described, otherwise they wouldn't see it any more inside themselves than outside. Detail in a description produces much the same effect as trituration19. A body fills ten times, a hundred times less space or volume in mass than in molecules. M de Réaumur did not suspect this; but have a few pages of his treatise on insects read to you, and you will detect the same ridiculousness in them as in my descriptions." (P. 704.)
The problem of detail is here a pretext for reviving the Pascalian device of the two infinities. The description makes one mistake "an aphid for an elephant", just as Pascal claims to "paint" in "un ciron" "an infinity of universes"20. Pascal plays on the vertigo that seizes the eye caught between the infinitely great that opens up above it and the infinitely small that steals all landmarks below it, to highlight the precariousness and weakness of man without God. Diderot joins him in criticizing mechanistic materialism21 which, detailing the structure and order of beings, ends up in spite of itself monstrously deforming them and removing them from the visibility it claims to access22. But it's not a question of opposing the vanity of the sensible experience of seeing to the revelation of another, spiritual, visibility of the world. What Diderot is denouncing here is the textualization of description, its mechanistic, or in other words structural, conception as the addition of details, qualities and attributes. Enumerative, description threatens visibility, makes us lose sight of the idea23. The difficulty with description lies in the fact that while it appeals to the eye, to which it provides the geometric data of what is to be seen (such and such an object in such and such a position at such and such a place), it actually addresses the imagination, which scoffs at dispositions and dreams up actions, expressions, sensations. Diderot continues:
"I believe that the eye and the imagination have more or less the same field, or perhaps on the contrary that the field of the imagination is in inverse reason to the field of the eye. Whatever the case, it's impossible for the presbyopic and the myopic, who see so differently in nature, to see in the same way in their heads. Poets, prophets and presbyopes are prone to seeing flies as elephants; myopic philosophers to reducing elephants to flies. Poetry and philosophy are the two ends of the telescope." (Ibid.)
The more the description gives hints for the eye, myopic details, the less latitude the imagination has to dream24: detail kills the image, which the mind can no longer grasp globally. This is why the field of the imagination (whose reverie mentally recreates the image) is in inverse reason to the field of the eye (which mechanically fixes the coordinates of objects in space).
The description thus superimposes two fields of gaze, or two contradictory regimes of visibility: to the geometrical field of the eye it opposes the scopic field of the imagination, to myopic vision, a presbyopic vision. These two visions are distorted views of the referent to be described; they constitute "the two ends of the telescope", a deceptive telescope, which distorts and blinds rather than reveals.
.
Description cannot account for its object. To point out the description in the Salon is to point out the ineffectiveness of the text. Description is only a pale supplement to the picture it describes; it requires its reader to go far beyond what it itself proposes. Of Deshays' Artémise au tombeau de Mausole, Diderot writes:
"But to feel all the effect, all the gloom of this composition, you have to see how these figures are draped, the carelessness, the volume, the disorder that reigns there. It's almost impossible to describe. I have never conceived better how energetic this part, which commonly passes for rather indifferent to art, was, supposing taste, poetry and even genius. Artemisia is dressed in an inconceivable way. This large flap of drapery, pulled back over her head, falling in wide folds over the front, and unfurling over the side of her face turned towards the back, revealing and at the same time emphasizing the whole part of her head exposed to the spectator, is of the greatest manner and produces the most beautiful effect." (P. 335.)
The description fails to account for the composition, which we can only understand if we see the image itself: "you have to see how these figures are draped". "The most beautiful effect", the effect for the eye is indescribable, or at any rate "almost impossible to describe", "inconceivable"25. The play of folds in the drape escapes any textualization.
The technical realization of drapery is an obligatory part of academic teaching. Roger de Piles devotes a long development to it in his Cours de peinture par principes26. The drapery conceals part of Artemisia's afflicted head: it conceals and it let's see, it hides and it exposes, producing for the eye that beat, that pulsation that mimes the request and response in the gaze and produces aesthetic enjoyment. We've seen how description presents itself as a response to a double demand, from its addressee and its referent, the Other of the dialogue and the image to be described. Formulating a response, it incorporates the demand to be satisfied and, through this beating of demand and response, it mimics the scopic play that constitutes every aesthetic relationship, from the gaze to the object it interrogates, from the object to the eye it satisfies, but always otherwise, "in an inconceivable way", surplus, as it were, to the description we could make of it.
Here we touch on the fundamental paradox of description: prolific, transforming aphids into elephants, it nevertheless always falls short of the image it accounts for: "the best description says so little" (p. 746). Its abundance is a symptom of its powerlessness to realize the program it has set for itself. It always refers back to the initial lack, to the image that is irremediably lacking27. Ideally, one would describe in the presence of the image. The absence is twofold: not only does Diderot only describe the paintings at the Salon when he gets home, hanging on the vagaries of a failing memory, but he describes them for readers who may never see them. Hence his suggestion to Grimm, in the preamble to Salon de 1767:
"Obtain from the opulent people to whom you intend my notebooks, the order or permission to have sketches taken of all the pieces I shall have to talk about; and I answer to you for a brand-new Salon. With the artists of past centuries better known, I would relate the manner and style of a modern to the manner and style of some ancient artist most analogous to his own, and you would immediately have a more precise idea of color, style and chiaroscuro. If there were an order, incidents, a figure, a head, a character, an expression borrowed from Raphael, the Carracci, Titian or another, I would recognize the plagiarism, and I would denounce it to you. A sketch, I don't say done with wit, which would be better though, but a simple sketch would suffice to show you the general disposition, the lights, the shadows, the position of the figures, their action, the masses, the groups, that connecting line that snakes and links the different parts of the composition; you would read my description, and you would have this sketch before your eyes; it would spare me many words, and you would hear more." (P. 517.)
Here appears the idea of an illustrated Salon, which would include not only the "sketch" of the paintings exhibited at the Louvre's Salon carré, but a kind of repertory of the classical works that serve as models for contemporary artists. Such a project brings to mind the drawings with which Gabriel de Saint-Aubin covered the booklets and sales catalogs that passed through his hands. We have an illustrated Salon de 1761, Salon de 1769 and Salon de 177728. But Diderot goes further, suggesting the creation of a veritable iconographic database, enabling borrowings to be spotted, "manner and manner" to be compared, and, as it were, an encyclopedia of techniques and styles to be compiled, in the spirit of the Encyclopédie.
What then becomes of Diderot's description, alongside these images and freed from any technical constraints? It's not a question of textually redoubling an order that the sketch will give from the outset and that the text dissolves as it details it29. The idea is explicitly to save part of the text: the sketch "would save me a lot of words, and you'd hear more". Ideally, the description should convey something beyond itself: it's not the technique, it's the painter's idea that it's aiming at, that it gives us to imagine, this idea expressed not in the order, but in the composition. Fundamentally, the description is not descriptive in the sense of narratology.
It is in this spirit that we must understand the recurring proposal to replace the description of a failed painting with a tale or a digression. In reference to two Servandoni overdoors, Diderot writes:
.
"Allow me to break the monotony of these descriptions a little, and the tedium of these parasitic words, heurté, empâté, vrai, naturel, bien colorié, bien éclairé, chaudement fait, froid, dur, sec, moelleux, that you have heard so much, without what you will hear them again, by some deviation that relaxes us. What is the Hercules of fable?" (P. 350.)
The poet and the philosopher revolt against the technicization of a description reduced to its technical dimension, which is not only the technique of making the painting, but also the technical organization of the space of representation. When it comes to painting, geometrality is merely a prerequisite. To speak of the Hercules of fable, to return to the idea therefore, is not to escape from description, but from its technical monotony.
III. Description as a device
The digression, the tale30 do not break with the poetic logic of the Salon as a whole, which must always be understood as description. On the contrary, they provide its symbolic framework. Continuing the movement of envelopment of the object that characterizes description, they constitute the meta-level from which the imagination can be liberated and the visibility of things restored. Thus, in the preamble to Robert's article in the Salon de 1767 :
"But what's the point, you may ask, of this spread on travelers and voyages? What have these true or false ideas to do with Robert's ruins. As these ruins are numerous, my intention was to set them within a framework that would palliate the monotony of the descriptions, to suppose them existing in some region, in Italy, for example, and to make a supplement to M. l'abbé Richard. For this purpose, it was necessary to read his Voyage d'Italie" (p. 695).
Because here the process of envelopment has failed, the device becomes visible. In the same way that the commentary on Fragonard's Coresus and Callirhoe was embedded in Plato's Lair, that the Vernet of the Salon of 1767 came to be inscribed in the Promenade and conversations with the abbé, Hubert Robert's ruins should have become the stages of an imaginary journey, the sites of a tourist guide. The "écart sur les voyageurs et les voyages" presents itself first and foremost as a departure from descriptive monotony, but it's a departure for a reverse take, through which technical descriptive discourse, threatened by the geometric insignificance of an illegible series of picture ordinances, can hope to make sense and give sight to its object.
The digression opens the description to a symbolic field31, just as the eye and the imagination provided, as we have seen, the geometric and scopic fields. Description thus functions as a device, and as an autonomous device, in which all the modalities and genres of narrative are summoned. What distinguishes description from other discursive regimes is not generic: it is its purpose, not its manner, that makes it specific. Description reveals, or at least seeks to make visible, its object. Of course, this visibility presupposes the establishment of a space and the arrangement of objects and figures within that space; but it is only achieved through the composition of an action based on the arranged figures, and the construction of a meaning for that action. The spectacular symbolic frameworks of "L'Antre de Platon" and "Promenade Vernet" should not be seen as virtuoso exceptions to an ordinary practice of description in Salons, where this framework would be absent. There can be no description without judgment, no setting of a scene without the critical overhang of Diderot the poet and philosopher, who will only make this scene a fully visible object if it is also an object of thought. Any judgment opens the text, even if only embryonically, to deviation and digression.
Diderot's presentation of his work in the Preamble to the Salon of 1765 is significant in this respect:
"I will describe the paintings to you, and my description will be such that, with a little imagination and taste, they will be realized in space, and the objects will be placed in them more or less as we have seen them on canvas; and so that one may judge the background that can be made on my censure or on my praise, I will end the Salon with a few reflections on painting, sculpture, engraving and architecture." (P. 293.)
Behind this presentation emerge the three levels of the descriptive device, which also constitute three regimes of visibility. First, there is a kind of objective reality of pictorial space, an arrangement, an order; from this reality, the reader imagines the painting, virtually realizes it in space; finally, the painting is the object of censure or praise, it is judged, evaluated. This device of description applies to the painting as well as to the Salon as a whole: the end of the Salon de 1765, made up of the Essais sur la peinture, is apparently a digression. But this digression provides the principle, the legitimacy of all the judgments made about the works exhibited; it therefore retroactively fixes their symbolic framework.
In this superposition of fields (geometrical, scopic, symbolic), of regimes of visibility (order, imagination, judgment), we can see a polarity taking shape: "with a little imagination / and taste"; "as we have seen them" / "so that we may judge". Imagination, which produces images, is opposed by taste, which evaluates them; what is first given to see by description, is judged at its conclusion. Structural analysis would be content with this polarity to define Diderotian description; it might even oppose descriptive competence and critical competence. Several passages in the Salons seem to point in this direction. At the end of the Preamble to the Salon de 1767, for example, Diderot writes:
"I warned you about my sterility, or rather about the state of exhaustion to which the previous Salons have reduced me. But what you will lose on the side of deviations, views, principles, reflections, I will try to return to you by the accuracy of descriptions and the fairness of judgments." (Pp. 528-9.)
Doesn't he clearly contrast "the accuracy of descriptions and the fairness of judgments" here? Can we not deduce from this that the account of a painting begins with its description, then ends with its judgment, and that Diderot thus alternates description and discourse, two distinct genres of narrative? Or, more subtly, can we define the Diderotian account as a discourse stretched between description and judgment, worked by these two contradictory poetic logics?
There's no doubt that such a polarity exists in the Diderotian practice of description. But we mustn't lose sight of the framework in which it takes place: the point of departure is always an absence, the absence of the tableau or the absence of the Diderotian bravura piece ("ma stérilité", my "exhaustion"). It is to make up for the disappearance of "deviations, views, principles, reflections" that this polarity is posited at the beginning of the Salon de 1767. Description thus functions as a supplement before entering into a polarity; its disappointing monotony itself desperately begs to be made up for by these digressions, of which the Salon de 1767 presents precisely the most brilliant and well-supplied fireworks display. The "deviations", the "principles" are certainly denied here; but this is in order to be able to function fully; they are veiled in order to be revealed, so that the dialectic of demand and the pulsation proper to seeing, founded on surprise and retrospection, can play out: in fact Plato's lair described Coresus and Callirhoe; in fact, the walk with the abbé described the Vernet of the Salon.
The opposition of description and judgment is therefore nothing but an illusion, as these remarks about Loutherbourg's drawings in the Salon de 1767 :
are even more clearly demonstrated.
"Besides, don't forget that I guarantee neither my descriptions, nor my judgment on anything; my descriptions, because there is no memory under the sky that can faithfully win so many diverse compositions; my judgment, because I am neither an artist, nor even an amateur. I'm only telling you what I think, and I'm telling you with all my frankness. If I happen to contradict myself from one moment to the next, it's because from one moment to the next, I've been variously affected; equally impartial, when I praise and when I renounce praise, when I blame and when I renounce criticism. Give a nod of approval to my remarks, when they seem sound to you; and leave the others, for what they are. Everyone has their own way of seeing, thinking and feeling. I won't value mine until it conforms to yours." (p. 749.)
Diderot does seem here to oppose descriptions and judgment. But the former are in the plural, the latter in the singular: the text is made up of descriptions in which Diderot exercises his judgment. What is essential, however, is the claimed contradiction of the descriptions. We find here what was suggested by the figure of Vertumne at the beginning of the Salon of 1763: the description is dialogical; it is both the unequal response of a diversely affected soul and a request made to Grimm, whose soul will or will not be affected in the same way. "I will not value mine until it conforms to yours. A whole game of conformities is set in motion: between the painting, Diderot and Grimm, between several ways "of seeing, thinking, feeling", the three verbs referring to the three levels of the descriptive device.
The description/judgment polarity, which only emerges in the Salon of 1767, could be nothing but a red herring32. Rather, what it reveals is the circuit of the aesthetic relationship, which between the viewing subject and the viewed object establishes the pulsation, the interplay of a series of demands and responses. This interplay is part of a non-structural logic of the supplement. It takes us back to the origins of description, to the performance of ekphrasis by which the ancient orator claimed to produce the very image by the sheer force of his verb. To produce the image: this is at once the primitive, archaic force of language, and its most sublime and accomplished manifestation.
.
There's a problem with description in Diderot's Salons. By description, we used to mean a well-established modality of discourse, which we generally contrasted with narration in the novel. But this recent hermeneutical framework makes no sense in the era of the Salons: in the Encyclopédie, description is opposed to definition, not narration. Diderot explicitly presents his Salons as descriptions: but for him there can be no question of simply describing the arrangement of paintings. The description targets the composition, the idea of the painting: the idea is generally narrative, and therefore has nothing to do with what we mean today by description. It is through the idea that description appeals to the reader's imagination; it is the idea that enables praise or criticism to be formulated. But the idea isn't always there. An imperfect definition, aimed at an uncertain idea whose visualization is problematic, description is deceptive. Statutorily doomed to failure, so to speak, it then reveals its enveloping nature, which constitutes it as a device, superimposing three regimes of visibility, for the eye, for the imagination and for judgment. What the Salons reveal about description should lead us to reflect on what we mean today by description, at a time when narrative sovereignty is collapsing and contemporary writing, aiming for the screenplay, claims more than ever to be an installation of visibilities...
Notes
As an interpreter, the describer's gaze defines forms in the image to be described, which he brings back first to "his optical perception", to "his immediate existential experience", then to "a knowledge superadded by culture": he does this visually, not verbally. Verbalization is description. "We must not expect to be able to find a literary source in every case"; the determination of the meaning of the work, which accompanies description, calls rather for "stylistic knowledge". (See Erwin Panofsky, "Le problème de la description d'œuvres", 1931, La Perspective comme forme symbolique, French trans., Minuit, 1975, pp. 238 and 243-4, with the illuminating example of Salomé-Judith.)
G. Genette, "Frontières du récit", Figures II, Seuil, 1969, in which the Homeric-Aristotelian apparatus appears today as a totally anachronistic projection of realist categories. See in particular Points-Seuil, pp. 58-59.
R. Barthes, "L'effet de réel", Littérature et réalité, Seuil, 1982, which clearly posits the corpus in which this effect can be spotted (Flaubert) and recalls the totally unrealistic origin of description, as an ekphrastic performance with no connection to other genres of narrative. See Points-Seuil, p. 84.
On this article, written by Abbé Mallet, see S. Lojkine, L'Œil révolté. Diderot, Salons, éd. J. Chambon, 2007, chap. 1.
References are given in L. Versini's edition, t. IV des Œuvres, Laffont, Bouquins, 1996. For other volumes, the volume number precedes the page.
This hole in the signifying chain clearly marks that between the thing seen and the descriptive text, the mediation, the interpreter, is not language, but the gaze of the describer, who doesn't name things from the outset, even if from the outset he recognizes them and assigns them a cultural, symbolic, signifying status. The symbolic form of the image is not verbal, and B. Vouilloux distorts the perspective, in our opinion, by basing his reflection on description in the Salons on the example of allegorical painting, the most textual, least visual genre in painting, a controversial genre in both Diderot and Abbé Du Bos. (B. Vouilloux, "La description du tableau dans les Salons de Diderot", Poétique, no. 73, Feb. 1988, p. 29.)
About Vien's Saint Germain qui donne une médaille à sainte Geneviève, Diderot writes: "The natures here are neither poetic nor grand; it is the thing itself, without almost any exaggeration. [...] It is the truth of all times and all lands." (P. 213.) In a way, Vien's painting, with its sublimely perfect technique and minimalist ideal, is descriptive painting, employing the means of description. We find the formula again for Durameau's Saint François de Sales agonisant: "It is the very thing, it is the real scene of the moment. [...] For this composition, so true in all its parts, to be the most beautiful there was at the Salon, all it needed was to be painted; for it is not." (P. 768.) Color is the painter's work, the indescribable part of technique. See also, in connection with the Vernet paintings at the 1769 Salon, "the very thing that the artist had imitated on his canvas" (p. 848).
"If you know anything more just than these lines, let me know in all simplicity: failing that, use what I send you." (Horace, Epîtres, I, 6, 67-68. We translate.)
On the function of description in the circulation of the art market's "flows of knowledge and money", see B. Vouilloux, art. cit. p. 27.
Of course such a definition seems more suited to history painting than to still life or landscape, which apparently claim no event to represent. In classical painting, however, the event is almost always the basis of the representation. See our analysis of Vernet's landscapes and Chardin's La Raie (L'Œil révolté. Diderot, Salons, J. Chambon, 2007, p. 389; 452-4. On the other hand, in our view, history intervenes only incidentally as text in the process of description. What intervenes is the event itself, of which both history and painting are, on the same level, representations. No more than history (or discourse), the event is the interpreter of the description: it becomes its referent in the same way as the painting that represents it, through the metaleptic effect of the description. (This is in response to B. Vouilloux, art. cit., p. 37.)
The describer's identification with the painter is sometimes explicit: "If all it took to be an artist was to feel keenly the beauties of nature and art, to carry in one's bosom a tender heart, to have received a mobile soul with the lightest breath, to have been born one whom the sight or reading of a beautiful thing intoxicates, transports, makes sovereignly happy, I would exclaim as I embraced you, throwing my arms around the neck of Loutherbourg or Greuze: My friends, son pittor anch'io. " (P. 268.) "Et moi aussi je suis peintre" would be the formula pronounced by Correggio in front of a painting by Raphael; it was famous in the XVIIIe century; it concluded the preface to L'Esprit des lois and is repeated in the Éloge de Montesquieu in Volume V of the Encyclopédie.
See M. Hobson, "Pantomime, spasm and parataxis: Le Neveu de Rameau", Revue de métaphysique et de morale, no. 89 (1984), A. Colin, pp. 197-213, and S. Lojkine, "Discours du maître, image du bouffon, dispositif du dialogue : Le Neveu de Rameau", Discours, image, dispositif, dir. Ph. Ortel, L'Harmattan, 2008.
About Falconet's U Figure de femme assise, Diderot writes: "I reread my description, and I find it modeled on the figure." (P. 448.)
The article Composition in the Encyclopédie, which is by Diderot, essentially defines composition as choosing the moment in history to represent. This is why, from the point of view of description, composition and event can be identified.
For example, regarding Hector reproaching Pâris for his cowardice of Challe: "Will you soon be finished?" you say.... Wait, wait; you're not there. [...] Did you follow me? Did it all work out in your head? [...] In truth, I must have a very indulgent imagination, to have taken on all this. And perhaps you expect me to give you a detailed critique of this world. You want me to finish, and we'll never finish. By the way, you can count on this description being accurate to within a hair's breadth; it's a tour de force, on my part that is." (P. 343.)
Étienne Bonnot de Condillac, Essai sur l'origine des connaissances humaines, 1746, Alive, 1998. The fundamental notion of this Essai is the notion of liaison: imagination and memory bind perceptions and sensations. The liaison is not a juxtaposition (the structure of the enumeration, the list), but a superposition (which presupposes a device).
"What is this [poetic] spirit? I have sometimes felt its presence; but all I know of it is that it is what makes things said and represented all at once; that at the same time as the understanding grasps them, the soul is moved by them, the imagination sees them, and the ear hears them ; and that discourse is no longer merely a sequence of energetic terms that expound thought with force and nobility, but that it is a fabric of hieroglyphs piled one on top of the other that paint it. I could say in this sense that all poetry is emblematic." (P. 34.)
"Trituration, in Pharmacy, is the action of reducing a solid body to a subtle powder. It is also called lévigation, pulverization, &c. [...] Trituration, is also said, in Medicine, of the action of the stomach on foodstuffs, which renders them suitable for nutrition. See Stomach, &c. Some physicians claim that digestion is done by trituration, & not by fermentation; otherwise that the stomach does nothing but grind & attenuate the alimens to render them fit for nutrition." (Encyclopédie, tome XVI.) Trituration was the mechanistic, Cartesian explanation of digestion; by the eighteenth century, this explanation had become obsolete.
Pascal, Pensées, no. 230, "Disproportion de l'homme", ed. Ph. Sellier, Mercure de France, 1976, p. 127.
The mechanistic position is represented here by the trituration model and by Réaumur (see La Promenade du sceptique, §43-44, I, 117-8). Diderot had a conflicted relationship with him. Réaumur had forbidden him to attend the cataract operation mentioned at the start of the Lettre sur les aveugles (I, 139). He had then accused him of plagiarizing for the Encyclopédie the plates he was preparing for the Academy of Sciences (1756).
The description highlights the confusion, the disjointedness of the composition. Thus of Leprince's Russian scenes: "And to bring out the disjointedness of all these objects, I'm going to describe this painting here, as if it were a Chardin" (p. 409), i.e. this scene as if it were a still life.
Speaking of Doyen's Miracle des Ardents, the order of which he suspected had been borrowed from an engraving after Rubens, Diderot wrote: "It is difficult to execute a painting from a given and detailed description; it is perhaps even more difficult to execute it from a print. Hence the missed understanding of chiaroscuro. Nothing that moves away, closer, unites, separates, moves forward, backward, binds, flees; no more harmony, no more sharpness, no more effect, no more magic. Hence figures pushed too far forward will be too large, and others pushed too far back will be too small; or more commonly, all crowded together, no more expanse, no more air, no more field, no depth, confusion of cut-out objects artfully stuck one on top of the other; twenty different scenes happening as if between two boards, between two panellings separated only by the thickness of the canvas and the border. Add that, while the lack of air and perspective carries the figures from the front to the back and from the back to the front; by a second curse, they will still seem driven from left to right and from right to left, or held as if by force within the enclosure of the canvas; so that this obstacle removed, one would fear that everything would escape and go scattering in the surrounding space." (P. 657.) The description, like the engraving, disseminates the scene and shatters what Roger de Piles calls "the whole together" (op. cit., p. 65).
On Casanove's Une Bataille, Diderot similarly writes: "The more you detail; each little detail always having something vague and indeterminate; the more you complicate the problem for the imagination. The same is true of a battle, a landscape, as well as the portrait of an absent woman; the more of her features you give the artist, the more you perplex him. [...] [The head] is all the more at ease, the less it knows about doing and ordering." (Pp. 666-7.) And in the digression from Renou's Jésus-Christ: "I suppose that in beginning the long and minute description of his figure, the poet has the whole of it in his head; how will he make me pass this whole? If he tells me about the hair, I see it; if he tells me about the forehead, I see it, but it doesn't go with the hair I've seen. If he tells me about eyebrows, nose, mouth, cheeks, chin, neck, throat, I see them; but each of these parts, which are successively indicated to me, no longer fits with the whole of the previous ones, so he forces me either to have in my imagination only an incorrect figure, or to retouch my figure with each new feature he announces to me. One stroke alone, one great stroke, leave the rest to my imagination; that is true taste, that is great taste." (Pp. 779-780.)
Similarly about Vien's La Marchande d'amours: "Et puis c'est une élégance dans les attitudes, dans les corps, dans les physionomies, dans les vêtements; une tranquillité dans la composition; une finesse; tant de charmes partout, qu'il est impossible de les décrire." (P. 203.)
Roger De Piles, Cours de peinture par principes, 1707, Gallimard, 1989, Tel, pp. 54-65 ("des draperies", "de l'ordre des plis", "de la diverse nature des étoffes", "de la variété des couleurs dans les étoffes", "des draperies en abrégé").
Thus, about Casanove's Une Petite Battle: "I judge these subjects, without describing them. You can't describe a battle. You have to see it." (P. 663.)
They are kept in the Cabinet des Estampes of the Bibliothèque nationale de France. For 1761, Rés. YD2-1132-8°; for 1769, Rés. YD2-1133-8°; for 1777, Rés. YD2-1134-8°.
Similarly about a sketch of Ruines by Hubert Robert: "Here is a very simple description, a composition that is no less so, and of which it is nevertheless very difficult to form a just idea, without having seen it. In spite of the care taken not to pronounce anything, to be short and vague; according to what I have said, twenty artists would make twenty paintings in which the objects I have indicated would be found, and more or less in the places I have marked for them, without resembling each other or Robert's sketch. Let's give it a try! and agree on the necessity of a sketch. The most shapeless will tell better and faster, at least about the general order, than the most rigorous and careful description." (P. 716.)
About a Joseph sold by his brothers by Amand: "Opt, my friend: do you want the description of this painting, or do you prefer a tale? " (P. 422.) Similarly, in front of Leprince's Berceau: "I'm bored with making and you apparently with reading descriptions of paintings. Out of pity for you and for me, listen to a tale." (P. 681.)
This is why we cannot subscribe to the idea that "description completes the closure of the system of representation" (B. Vouilloux, art. cit., p. 45).
More generally, "the simple primary description of a work of art [...] is already in truth an interpretation having to do with the history of forms [...]. This description implicitly includes this interpretation" (E. Panofsky, op. cit., p. 243).
Référence de l'article
Stéphane Lojkine, « Le problème de la description dans les Salons de Diderot », Diderot studies, XXX, 2008, p. 53-72.
Diderot
Archive mise à jour depuis 2006
Diderot
Les Salons
L'institution des Salons
Peindre la scène : Diderot au Salon (année 2022)
Les Salons de Diderot, de l’ekphrasis au journal
Décrire l’image : Genèse de la critique d’art dans les Salons de Diderot
Le problème de la description dans les Salons de Diderot
La Russie de Leprince vue par Diderot
La jambe d’Hersé
De la figure à l’image
Les Essais sur la peinture
Atteinte et révolte : l'Antre de Platon
Les Salons de Diderot, ou la rhétorique détournée
Le technique contre l’idéal
Le prédicateur et le cadavre
Le commerce de la peinture dans les Salons de Diderot
Le modèle contre l'allégorie
Diderot, le goût de l’art
Peindre en philosophe
« Dans le moment qui précède l'explosion… »
Le goût de Diderot : une expérience du seuil
L'Œil révolté - La relation esthétique
S'agit-il d'une scène ? La Chaste Suzanne de Vanloo
Quand Diderot fait l'histoire d'une scène de genre
Diderot philosophe
Diderot, les premières années
Diderot, une pensée par l’image
Beauté aveugle et monstruosité sensible
La Lettre sur les sourds aux origines de la pensée
L’Encyclopédie, édition et subversion
Le décentrement matérialiste du champ des connaissances dans l’Encyclopédie
Le matérialisme biologique du Rêve de D'Alembert
Matérialisme et modélisation scientifique dans Le Rêve de D’Alembert
Incompréhensible et brutalité dans Le Rêve de D’Alembert
Discours du maître, image du bouffon, dispositif du dialogue
Du détachement à la révolte
Imagination chimique et poétique de l’après-texte
« Et l'auteur anonyme n'est pas un lâche… »
Histoire, procédure, vicissitude
Le temps comme refus de la refiguration
Sauver l'événement : Diderot, Ricœur, Derrida
Théâtre, roman, contes
La scène au salon : Le Fils naturel
Dispositif du Paradoxe
Dépréciation de la décoration : De la Poésie dramatique (1758)
Le Fils naturel, de la tragédie de l’inceste à l’imaginaire du continu
Parole, jouissance, révolte
La scène absente
Suzanne refuse de prononcer ses vœux
Gessner avec Diderot : les trois similitudes