Introduction
Robin Middleton
Nicolas Le Camus de Mezieres is usually approached backward, as it were, as the source of some of Etienne-Louis BoulleeÕs more adventurous theories relating to the evocation of architectural character through the effects of light and shade. Boullee greatly enlarged on Le CamusÕs speculations. BoulleeÕs writings were not published until the twentieth century, but they were no doubt known to his contemporaries and entered the architectural memory long before he died. His fame has thus perhaps robbed Le Camus of the recognition that is his due. Le CamusÕs most intriguing work, moreover, Le genu de tarchitecture; ou, LÕanalogie de cet art avec no.) oenoationo (The genius of architecture; or, The analogy of that art with our sensations), 1780, has seemed only partially concerned with bold conceptual matters. His is not the philosophical stance of Boullee. Le CamusÕs study is, indeed, essentially a handbook on the planning of the French hotelÑ the town house, as a rule, of a noble family Ñ though he extended the range of investigation from the merely practical to encompass the whole realm of decorum and the proper manner of stirring ideas and emotions through architectural means. It is a pioneering work; its theory seems at times makeshift and inconclusive.
Le genie de i,architecture has too often been read as an assemblage of tenuously linked parts, enlivened by a handful of brilliant perceptions: an introduction dealing with the way in which feelings are aroused by architectural forms and an exploration of an expressive language to suit; a section on the role of the five orders in providing a traditional language of architectural expression, followed by a reversion to the initial theme; and, almost as an adjunct, a detailed analysis, room by room, of the planning and arrangement of the hotel. This last takes up two-thirds of the book. And half of this, one might note, is devoted to service spaces, servants' quarters, and stables. This proliferation of tiny rooms set aside for specific purposes is not to be equaled in architectural literature until the nineteenth century, and then in England in such works as Mrs. Beeton's Book of Household Management, 1861, or Robert KerrÕs The Gentleman! I/ottoe; or, How to Plan English Reou)ence,i, from the Paroonage to the Palace, 1864. There is in Le CamusÕs work a curious mix of somewhat rarefied theory and ordinary, commonsensical issues. His special concern, as the word geniuo in the title makes clear, was with the nature of individual creation Ñ a particular preoccupation of eighteenth-century architects. By mid-centuiy the phrase "fire and geniusÕÕ (Òda feu et du genieÓ) served as a cant term of praise, though as early as 1652 in Deooeino de pliuieuro pa la Li (Designs for several palaces) Antoine Le Pautre had described his visionary designs as "children of my geniusÕÕ ("enfano de mongenie")d
Le CamusÕs theory, as already indicated, is not always fully worked out. It remains uncertain. But though the parts might not seem at first to relate too well, they do form a whole. His study is quite coherent, as becomes at once evident when the pattern of knowledge and the ideas from which it emerges are considered. Some outline ol these ideas is perhaps necessary before beginning Le CamusÕs book.
The usual means of giving form and expression to a building in classical architecture was through the use of the orders Ñ in Greek architecture the Doric and Ionic as a rule (though the Corinthian Order was also a Greek invention) with the addition in Roman architecture of the Tuscan and the Composite. Variants of all five orders were used throughout the Renaissance. Their proportions and the whole array of elements and moldings that related to them were analyzed and classified by architectural theorists, most notably and usefully by Vignola, whose works were published throughout Europe in many editions.2 A codified system of classical architecture was thus established. All manner of variations were possible, and nothing was, in fact, fixed; but the system was sustained by the notion of an ideal that was thought to have emerged first in Greece and to have attained equal, if not higher, perfection in Rome and in Renaissance Italy. This ideal was, in theoiy, absolute and scarcely to be affected by historical change or cultural differences, although for the French of Louis XIVs reign it seemed that an even higher level of achievement had been attained in France.
During the second half of the seventeenth century this belief in a uni-
versal norm was gradually eroded, in part through an enhanced knowledge of the societies and civilizations of the world and an awakening recognition that the forms of their architecture represented equally valid and coherent systems, but perhaps even more forcefully Ñ certainly more insidiously Ñ through the impact of Cartesian doctrines. For though Cartesians were in no doubt at all that there were universal norms in philosophy and all the arts, they held that these norms were to be established on the basis of reason, not on the basis of precedent. This thesis was, however, of a paradoxical kind, for the standards of good taste were grounded not only in antique precedent but also in custom and usage both in classical times and from the Renaissance onward. Custom and usage represented an expression of the workings of nature, and Nature (la beile nature) in its idealized form was itself the seat ol reason. The tastes of the Cartesians were thus not very different from those of their predecessors. But the system of classical architecture was nonetheless subject to a profound reassessment.
The most articulate and effective reappraisal of classical doctrine was provided by Claude Perrault, a doctor and scientist rather than an architect. This appeared in 1673 in the footnotes of PerraultÕs new translation of De arcbitectura libri decern (The ten books of architecture) of Vitruvius (which, as the only surviving theoretical text from antiquity, had in fact served to establish the authority of the architecture of Greece, the buildings of Greece being to all intents unknown to the architects of Europe until the second half of the eighteenth century). PerraultÕs initial reappraisal was later expanded in his Ordonnance de.Õ cinq espices de colonnes selon la methode des anciens, 1683 (translated into English as A Treatise of the Five Orders of Columns in Architecture in 1708).
As the title alone of this work makes clear, Perrault was in no way intent upon rejecting the authority of classical architecture or the primacy of the orders as the embodiment of the highest standards of beauty and artistic expression. He aimed rather to subject them to new rules of assessment. His ground was the rule of reason. The orders, he held, should not be regarded as sacrosanct; they should not be upheld as expressions of a universal system of harmony discernible, according to Pythagorean theory, in music in the intervals ol the scales, in architecture in the established proportional ratios. Musical harmony was not to be equated with architectural harmony. The ear might indeed register a discord, but the eye was by no means so finely responsive to proportional harmonies. The ear and the eye were instruments of a different kind, and their operations must, on scientific grounds, be considered as distinct. The particular sharpness of this argument was occasioned by the publication in 1679 of Rene Ouvrard s Architecture harmonu/ue; ou, Application de la doctrine des proportions de la musique a Tarchitecture (Harmonic architecture; or, Application of the doc-
trine of musical proportions to architecture).
There were, for Perrault, no absolute standards of beauty in architecture. Beauty was of two sorts: positive beauty, which could be readily recognized in the quality of materials, in precision and neatness of execution, in size, in sheer magnificence, and, as might be expected, in symmetry, and arbitrary beauty, which involved arrangements of forms and shapes, proportion, and the articulation of the whole panoply of elements of classical architecture, which had been established by custom. The latter might be subject to rules, indeed in the interest of uniform standards of taste Perrault was determined that it should be subject to rules, but they could not be regarded as fixed. It was precisely in this area of design that the sensibility and imagination of the true artist might be expressed. For Descartes, of course, imagination was a source of error. Perrault aimed therefore at no more than a modicum of expression controlled by reason.
Yet Perrault's reassessment of the classical tradition on the grounds of reason prompted a significant and creative innovation in architecture. The orders, he argued, had become little more than a decorative system applied to wall surfaces, whereas in classical antiquity the column, which was the essential component of the orders, had served instead as a structural element, which had both constituted the architecture, as it were, and decorated it. And this structural articulation he recognized as a vital aspect not only of true classical architecture but also, altogether surprisingly, of French Gothic architecture. He thus struck a note of cultural relativism. But though Perrault might have referred with approval to the French Gothic cathedrals, he was not setting them forth for imitation. He was observing and analyzing their architectural effects so that they might be subsumed in the classical system. He sought to give expression to these notions in the freestanding colonnade that he and a small committee set up by Jean-Baptiste Colbert Ñ consisting of the architect Louis Le Vau and the painter Charles Le Brun Ñ designed for the east front of the Louvre in 166Z This elevation marked a radical change not only in the handling of the forms of architecture in France but in the grounds of aesthetic judgment. The utilitarian aspect of the orders was now seen to have determined not only the original form of the architecture but to provide a new potential for expression.
Michel de Fremin and the Abbe Jean-Louis de Cordemoy took up Per-raultÕs ideas in the early eighteenth century. They thought likewise primarily in terms of the classical tradition, though, like Perrault, they enlarged their understanding of the rational nature of architecture with reference to the Gothic. Their enthusiasm for Gothic extended even to spatial effects. The classical system they sought to reduce to essentials. De Cordemoy aimed to restrict the use of the orders to the three Greek forms; Fremin was intent to deny even these any overt importance. Both men aimed
at an architecture of common sense (le bon oeno), based on sound building practice in which the shape and form of each element would be explained and tested by reason.
The immediate impact of these ideas was limited, for the science of construction was not greatly advanced in the early years of the eighteenth century; building continued to be a traditional operation not yet susceptible to intellectual analysis. The mathematical basis of structural theory was but slowly formulated during the course of the century. Fremin and CordemoyÕs chief concern, moreover, was with church architecture and in particular with the role of the freestanding column as a supporting element between nave and aisle. Few new churches were erected in France during the first half of the century, though when the greatest church of the age, the church of Sainte-Genevieve (now known as the Pantheon), was initiated in 1757, their ideals were to be given the fullest possible expression. By then they had been reformulated and presented with far greater clarity and coherence by the Abbe Marc-Antoine Laugierm hts Ktoaiour ( architecture, 1753 (translated into English as An Eooay on Architecture in 1755), the most celebrated and mfluential of all eighteenth-century architectural texts. The criteria of architectural excellence were reduced there to those subtended by an idealized prrmttive hut Ñ four upright posts supporting four lintels with a pitched roof above. Architecture became, in essence, no more than a structural framework, albeit a classical one.
Though architecture was to be rigorously tested and purified by reason and reference to origins, architectural understanding was not thereby reduced; rather, it was enlarged. The established classical tradition, rooted in a timeless and fixed ideal, was undermined inexorably, but it would be a mistake to think that it was discredited and cast aside. It was cherished still. And the classical tradition as a whole, not simply that aspect concerned with architecture, could now be explored in terms of the evolving empirical aesthetic. Frenchmen thought in terms of the texts and tropes of antique literature. They still received a classical education. Indeed, when the notion of the absolute authority of the orders was shattered, it was to antique sources that the French turned first to find an alternative form of architectural expression.
The notion of expression in classical aesthetics can perhaps be traced back to Xenophon, who in his Memorabilia recounts a discussion between Socrates and the sculptor Cleiton in which they concur that it is the task of the sculptor to give expression to the individual soul in his figures.3 But in the seventeenth and eighteenth centuries the test of any such ability to give artistic expression to the emotions was usually thought of in terms of a painters ability to represent with conviction the laughing and weeping philosophers, Democritus and Heraclitus. This was long regarded as the sharpest test of a painterÕs powers. It was, however, part of a late
development Ñ a theme taken up in the fifteenth century by the Platonic philosopher Marsilio Ficino from Sidonius, who refers in a letter to paintings of this sort in the gymnasium at Athens.4
The first text of classical antiquity in which the theory of expression was explored at any length was Aristotle's Poetico, which was almost entirely taken up with literary forms, with poetry, tragedy, and comedy, as was HoraceÕs later AroPoetica (Art of poetry). But though these works referred only by implication to the arts of painting and sculpture, and to architecture not at all, they offered so compelling an analysis of the art of arousing and expressing emotions Ñ in particular in the theater Ñ that they became the basic texts on notions of expression in all the arts. Even in classical times the implications of these initial ideas were extended with reference to rhetoric, beginning once again with a fundamental text by Aristotle, the Rhetoric, and subsequently in Cicero's I)e Oratore (On the making of an orator) and QuintilianÕs Irutitutio oratorio. (The training of an orator).
The aim of these investigations was to establish the style appropriate to a particular character on a particular occasion in relation to a determined aim. It was a matter of propriety or decorum. Cicero summed up the theoretical tenets most succinctly in a late work, Orator, in which he set forth decorum as the decisive factor in any consideration of style. There are, he wrote, three styles in oratoiy,
the plain style for proof the middle style for pleasure, the vigorous style for persuasion; and in this last is summed up the entire virtue of the orator. Now the man who controls and combines these three varied styles needs rare judgement and great endowment; for he will decide what is needed at any point, and will be able to speak in any way which the case requires. For after all the foundation of eloquence, as of everything else, is wisdom. In an oration, as in life, nothing is harder than to determine what is appropriate. The Greeks call it 7T (tenor; let us call it decorum or "propriety.Ó5 Cicero was quite clear that propriety, or what is correct, is not necessarily concerned with what is right. It was not a question of morality. The rhetorical modes of expression were thus intimately linked to decorum but not to an absolute good. The relationship between the word decorum and decor in architecture need scarcely be stressed. But such notions were to be transposed first to the aesthetics of painting before they were taken up by architectural theorists. Leon Battista Alberti and Leonardo da Vinci were the most famous of the Renaissance exponents of the ideals of the proper representation of character through the depiction of appropriate human forms, gestures, and expressions. But though both were architects and though Alberti certainly used rhetorical theory Ñ CiceroÕs in particularÑto underpin his no-
tions of decorum in architecture, they did not explore these themes as fully as might be expected in their architectural writings. They reserved these notions, on the whole, for painting. They continued to rely for expression in architecture on the characteristics of the orders and their proportional systems.
The key to the transfer of such theories into the realm of architectural aesthetics was provided, not altogether surprisingly, by PerraultÕs contemporary the painter Charles Le Brun. In 1668 Le Brun delivered a lecture on expression to the members of the Academie Royale de Peinture et de Sculpture. He was concerned chiefly with the problems of history painting, with the way in which human emotions might be represented with due decorum so that the event depicted might be convincing. His particular interest was pathognomics, the way in which passing emotions might be expressed in the human body. But the focus of his attention was the human face.
Boldly, Le Brun adapted the scientific aspect of his theory from Descartes sLej paMuino de lame, 1649 (translated into English as The PaMwru of the Soule in 1650). Descartes had conceived of the passions as affectations of the soul. The soul, he thought, was seated in the pineal gland in the center of the brain, and it was from there, through the flow of animal spirits, that the human passions were controlled and given expression in the body. Le Brun argued that as the face was closest to the soul, it was the most expressive part of the body, with the eyes and eyebrows the most revealing indices of expression, as they were almost in direct contact with the soul. He isolated four positions of the eyebrows as indicative of particular emotions, con-cupiscible or irascible, simple or mixed. He sought likewise to demonstrate that the mouth reflected the movement of the heart. The basic modes could, of course, be combined and adjusted, and in a later lecture Le Brun showed how they might be reinforced by the study of physiognomies, the science of the permanent characteristics of the body as affected by the passions.
Le Bruns ideas were circulated in his lifetime in various manuscript versions, but they were not to be published until after his death Ñ Conference our lexpreo-oion, 1698 (translated into English as The Conference upon Expreooton in 1701); many other translations and adaptations followed. Le Brun had aimed at a scientific analysis of the principles governing expression so that painters might work not in imitation of nature but according to its laws, creatively. Instead, his text, and the illustrations accompanying it even more so, came to serve as fixed patterns for expressions, so that particular lines of the eyebrows and mouth were soon recognized throughout Europe as formulae for the representation of such emotions as dejection and joy, astonishment and anger, etc. (figs, laÑe). Le Brun had also indicated in his drawings how human characteristics might be related to the conventional animal passions Ñ the docile
2. Charles Le Brun, human and animal heads expressive of common characteristics, drawing. Paris, Musee du Louvre, no. 28.200. Photo: © R.M.N.
sheep, the brave lion, the cunning fox, etc. (fig. 2) Ñ and these facial types also provided a repertoire of resemblances that were taken up by both painters and connoisseurs. The impact of Le Bruns ideas was immense, for however much his aim might have been distorted and debased, he had demonstrated successfully that character and emotion could be explained and represented in visual terms, that the passions of the soul could be reduced to lines.
Though Le Brun worked on several occasions as an architect, he seems not to have recognized the relevance of such theories to architecture. Indeed, when he attempted to provide expressive forms for each of the twelve pavilions of the Chateau de Marly, he resorted to the established method of applying the orders or their parts to the facades, together with an array of painted conventional and sculptural symbolical motifs indicative of the designation of each Ñ Hercules, Venus, Jupiter, Abundance, etc. (figs. 3-5). Nothing adventurous by way of a linear interpretation of these characteristic qualities is in evidence in his designs. The pavilions have, in any event, been destroyed.
This long exploration of the proper means of expressing character in artistic terms was set firmly within an architectural context only in 1745 in Germain BoffrandÕs Livre dÕarcbitecture (Book of architecture) Ñ the writing of which dates back to 1734. Boffrand was concerned primarily with the presentation of a range of his own designs for hoteli< and chateaux, but his introductory chapters reveal his interest, his obsession one might say with the issues of expression and decorum in architecture. Like most French theorists, he gracefully acknowledged the merits of the architectural forms of other civilizations, other societies, bestowing special praise on the builders of the French Gothic cathedrals for the finesse of their proportional systems, but he continued to uphold the Greek and Roman example as the most fitting for emulation.
Boffrand relied still on the five orders to provide the basic modes of expression. The orders, he said, ranging in character from the rustic to the sublime, must be regarded as the equivalent of the genres in poetry. It is notable that, good classicist that he was, Boffrand set his text in two parallel columns of Latin and French. It is even more noteworthy that in the most finely argued of his chapters, that dealing with character in architecture, he took up the rhetorical device of transposition; he interleaved his text with quotations from HoraceÕs Am Poetiea, a handful of words adjusted or changed in each instance to make explicit the reference to architecture rather than poetry. But though he staunchly upheld the classical tradition, Boffrand aimed to invest architecture with a more complex range of expression borrowed from poetry, the theater, or music so that the purpose of each building should be quite evident in its architecture, both inside and out. The rank of its owner should
3. Painted elevation ofthe Pavilion of Hercules at the Chateau de Marly, designed by Charles Le Brun. Illustrated in Charles Le Brun, Divers desseins de decorations de pavilions (Paris: Ede/inck, circa 1690). Santa Monica, The Getty Centerfor the History of Art and the Humanities.
4? Painted elevation ofthe Pavilion of Venus at the Chateau de Marly, designed by Charles I.e Brun. Illustrated in Charles Te Brun, Divers desseins de decorations de pavilions (Paris: Edelinck, circa 1690). Santa Monica, The Getty Centerfor the History of Art and the Humanities.
likewise be visible, and his intention to evoke a mood of seriousness, gaiety, or joy should at once be effective.
All emotions, Boffrand held, could be expressed in architecture. The added means of expression proposed was that of the line in three basic forms: convex, concave, or straight. He made no mention of Le Brun, but there can be little doubt that he had him in mind when he castigated the architects of the early eighteenth century for designing interiors with lines intertwined to confusion and for evidencing no understanding that such lines were to architecture what tones were to music. By such means the skilled architect might strike chords of sadness or joy, love or hate, grace or terror. The best surviving interiors of the Rococo, it is to be remarked, are those designed by Boltrand himsell for the Hotel de Soubise in Paris, though today they are not often appraised in the terms proposed by Boffrand.
There is a great deal more in Boffrand Õs book that is pertinent to Le genie de larchitecture, in particular his belief that the first principles of architecture rest on the commonsense grounds of health and security and convenience and ease of use and that it is these considerations that should determine the forms and arrangements of hotels and chateaux before the decor appropriate to the station and taste of the client is conceived. But consideration of such aspects of theory may, for the moment, be set aside to explore further the emergence of an empirical aesthetics in the early years of the eighteenth century, for this was the conditioning factor of Le Camus de MezieresÕs novel concept of architecture.
The first clear expression of the ideas that were to shape the empirical aesthetics of the second half of the century is usually thought to be the Abbe Jean-Baptiste DubosÕs Reflexions critiques sur la poesie et sur la peinture, 1719 (translated into English as Critical Reflections on Poetry, Painting, and Music in 1746) in which sentiment is shown to have a more significant role than reason in the formulation of aesthetic judgments. This was, as may be imagined, no more than a relative assessment. The classical tradition and its codes of taste rooted in the notion of Òla belle nature" were not seriously challenged. The work that was to inflect sensationalist thought to real and disruptive effect, however, was the last of the books of the amateur painter and connoisseur Roger de Piles, the Cours de peinture par principes, issued first in 1708 (and again in 1766 and 1791; translated into English as The Principles of Painting in 1743) Ñ a book, one might note, that engrossed Dubos.
De Piles was opposed to much that Le Brun represented: the system of rules established by the Academie (though his own Cours consisted in lectures on a new set of rules that he had put forward to the Academie when he was finally elected in 1699); the emphasis on design and line rather than color in painting; the notion of the genres, etc. He was particularly scathing about Le Bruns reduction of facial ex-
f. Painted elevation of the Pavilion of Jupiter at the Chateau de Marly, designed by Charles Le Brun. Illustrated in Charles Le Brun, Divers desseins de decorations de pavilions (Paris: Edelinck, circa 1690). Santa Monica, The Getty Centerfor the History of Art and the Humanities.
Introduction
pression to a system of lines Ñ a clever theory, he remarked, but one that could not be applied to the nose. De Piles nonetheless ranked Le Brun surprisingly high in the slightly ludicrous grading system that he devised for the evaluation of painters; Le Brun was set below Raphael and Rubens but above Titian.
In his frequent references to Aristotle and Horace, and especially to Quintilian, de Piles is akin to other eighteenth-century theorists. Like them, he continued to analyze painting in the same terms as poetry and oratory, but he was determined also to free painting from the dominance of literary theory. As we have seen, from Alberti to Le Brun, theorists had thought that the expressive range of painting resided in the representation of the gestures, attitudes, and movements of the human body and in particular in the expressions of the face. The subject matter, so often selected from classical mythology or history, was of paramount importance. But de Piles argued that a painting made an instant and purely visual impact on the spectator that had nothing to do with the subject matter and was in no way related to the conventional modes of expression and that this impact derived instead from the effect of the composition and the manner of execution of the whole Ñ the tout ensemble. Furthermore, it was this unity of impact (unite dÕobjet), as opposed to Andre FelibienÕs and Le Bruns unity of subject (unite de ,'ujet), that gave true value and distinction to a painting. The mere sight of a painting should stir a particular passion. In its ability to evoke a spontaneous mood, painting was similar, at least, to music.
This notion was thought by many to be no more than an extension of the principles of decorum, or coneenance, from the figures and parts of a painting to composition as a whole, and this was indeed part of de PilesÕs intent, but he aimed at far more. He thought that the parts should, in fact, be subordinated to the whole and that the power of the work should stem from the arrangement of forms and colors and of light and shade that was determined chiefly by the artistÕs sense of harmony. This sense he termed the fury or enthusiasm of the painter, and he thought that in its noblest form this enthusiasm comprehended the sublime.
The initial sensuous effect of a picture had, of course, to be sustained thereafter by conventional painterly skills in the representation of an illusion, and this was, inevitably, susceptible still to intellectual analysis. But beauty, after the publication of the Couro, could no longer be formulated as a fixed set of rules, ÒBeauty, say they, is nothing real; every one judges it according to his own taste, and tis nothing but what pleases. Ó6
The thrust of de PilesÕs comments was directed at history painting, but his particular interest was landscape painting. He himself collected Dutch and Flemish landscapes, and his propaganda greatly stirred French interest in such works. He liked them, of course, precisely because they did not rely on figures to
establish their character. They could be construed as more purely painterly exercises. De Piles in fact introduced the wordpittoreoque, or picturesque, into the French critical vocabulary to denote a view that was painterly. Only later was the term to be taken up by landscape-gardening theorists.
Landscapes, de Piles explained, could be more totally expressive of a particular theme than other categories of painting, because the elements of which they were composed were imbued with intrinsic qualities. Thus cottages might be introduced into a painting to invoke a rustic theme or clear-cut buildings to evoke a heroic mood, but more essentially affective were the elements of nature itself. A distant view, the sky and the clouds, mountains and other irregularities, trees and rocks, each had the power to conjure up a particular image. "Rocks," he wrote, Òare of themselves gloomy, and only proper for solitudes.Ó7 But it was water in its various states from turbulence to calm that stirred passion most deeply; water was to be regarded as the soul of the landscape.
De Piles's small book had a pronounced effect on attitudes toward painting; it influenced patterns of connoisseurship and collecting, and it penetrated the artistic sensibilities of the age of the Rococo in a host of other ways. Primacy of feeling was then in the ascendant, and de Piles had provided a text that justified the instant response, the instant submission to pleasure or other emotion. The complex and fantastical stage sets that the extravagant Jean-Nicolas Servandoni devised for the Academie Royale de Musique between 1726 and 1742 and created on his own account from 1754 to 1758 (in a series of tableaux in the Salle de Spectacles at the Tuileries) may be seen as one particular response to this demand for the striking illusion, the instantaneous stirring of emotion. And though Denis Diderot was to bring such spectacles into disfavor precisely because they offered no further engagement of the mind, his later demands for the fullest expression of emotion in painting and his acclaim for the works of Jean-Baptiste Greuze are yet other reflections of that realm of feeling in art opened up by de Piles.
Attitudes of this sort were given form in architecture in the intimate luxury of hotel interiors, in particular in the intricate network of rooms, all artfully decorated, that made up the fashionable womanÕs apartment. Although the practicalities of such arrangements were described in some detail by a number of architects in the early years of the eighteenth century, beginning with Sebastien Leblond in 1710, the aesthetics were not much explored beyond the dictates of propriety, or convenance. The handling of space, or rather a sequence of spaces, to determined sensual effect was to be analyzed first by Le Camus de Mezieres. This was his contribution to architectural theory.
The plan of the French hotel was based on the organization of groups of four or five rooms, though the number was not determined, to form apartments of a distinctive character. Each suite of rooms thus set apart was intended for the use of a particular person. The hotel was made up of many different types of apartments, the chief ones being the state apartment, or appartenient deparade, and the private apartment, or ap-partement de commodity. The principal rooms were linked by an enfilade, a sequence of openings on a single axis that provided both access and a visual link between them. The word appartenunt, it is important to stress, was not known in France before the middle of the sixteenth centuiy. The term and the concept were probably introduced from Italy. The first effective embodiment in France of these planning principles is usually considered to be Cardinal Hippolyte dÕEsteÕs hotel, the Grand Ferrare, erected opposite the royal palace of Fontainebleau between 1544 and 1546 by Sebastiano Ser-lio. This was illustrated in variant lorms in SerlioÕs famous sixth book of architecture, devoted to domestic buildings, which was not to be published until the twentieth century. Manuscript versions, however, were certainly known to contemporaries, and the arrangements of the Grand Ferrare were introduced soon enough into Parisian architecture.8
The history of the form is no doubt more complex than this summary suggests. Myra Nan Rosenfeld has argued that the arrangement was in use in France long before Serlio arrived from Italy and has adduced the Hotel de Cluny (variously dated as 1456-1485 or 1485-1498) as an illustration of the type.9 This, however, does not have a symmetrical entrance facade, which, as we shall see, is another essential expression of the order of the French hotel. Rosenfeld has, nonetheless, sought to trace the type beyond the medieval manor to the Roman farmsteads of France and thus to interpret it as an Italian import of even earlier vintage. The matter requires exploration still and needs to be considered, moreover, with reference to the way in which rooms were used. For this is another aspect of French planning that is determining.
The sequence of rooms in an apartment was outlined early Ñ once again by Serlio in the description ol a royal palace in his sixth book of architecture Ñ as a ,ialie, antichambre, chambre, arriere chambre, and cabinet, that is, a vestibule or guard-room, an anteroom for people to wait in before being received, a general living space, followed by a bedroom and a study. By the early eighteenth century the classic sequence for a noble apartment had been established as antichambre, oalle de compagnie, chambre a coucher, cabinet, and arriere cabinet or garderobe, that is, an anteroom (there might well be more than one), followed by a salon, a bedchamber, a study, and a closet.
The bedchamber was not always used as a sleeping chamber, in particular in an apartment of consequence. The bed to be used would be in the room beyond. For the actual use of rooms was from quite early on, and certainly after the move of the court to Versailles, closely patterned on royal practice. The kingÕs bed with its canopy was identified as the seat of power, even more potent perhaps than his canopied throne. The bed was the real focus of royal ritual. The degree of intimacy with the king was at once made apparent by the manner in which one was received in relation to it Ñ whether in the preceding salon or in the bedchamber itself, whether on the dais of the bed or off it. Even when the bedchamber was empty, those passing through the room had to make a bow to the bed. The royal bed had assumed its purely ceremonial role as a lit deparade as early as 1585, when Henri III drew up new regulations for his household management. After disrobing on his bed, the king would retire to sleep with the queen in her bedroom nearby. This arrangement was later in evidence in the royal apartments in the Louvre, though it is to be remarked that the rooms there were still not arranged in enfilade. And thus they remained when Louis XlVoccupied them.
When work began on the Palais des Tuileries in 166-4 to prepare it for royal occupation, the kingÕs and the queenÕs apartments were set on the same floor, as before, but with their individual rooms axially linked. The king was to sleep in the room beyond his state bedchamber. When, in November 1673, the court moved to Versailles after the completion of the first major reconstruction of the chateau, the queen's apartment was in the south wing, the kingÕs in the north. The main rooms of this suite, the Appartement des Planetes, or Appartement du Soleil, formed a magnificent sequence, in enfilade, ending in the king's state bedchamber; his sleeping room was beyond. When Maria Theresa died in 1683, Louis XIV took over the entire first floor of the palace to form his apartment. He then moved his state bedchamber to the center of the palace; inevitably, it could no longer be in enfilade, instead it was on the principal axis of both the extensive garden and the building itself, the focal point of all roads leading to it. The lit de parade seemed, almost, the center of France. The kingÕs sleeping chamber was alongside. Then, in 1701, he determined that his public persona and his private persona should be as one; he decided to sleep in his state bed. This practice was, in turn, reversed by his great-grandson, Louis XV, who was intent to lead something of a private life and to whom many of the more intimate living arrangements of the eighteenth centuiy were to be attributed. But the bed, in accord with royal precedent, nonetheless continued to be the dominant feature in the classic French apartment, the climax, as it were, to the spatial sequence. The more magnificent the apartment the less likely the bed was to be used, other than on a wedding night or for the birth of children.
6.	Ground-floor plan of a town house on a ypx 120ft site with a "salle,Ó "chambre/JaW "cabinet ÕÕmarked. Illustrated in Pierre LeMuet, Maniere de bien bastir pour toutes sortes de personnes, 2nded. (Paris: Francois Langlois, 1647), Õj 7. Middleton Collection.
The French manner of setting up tables for eating, almost at will, but usually in one of the anterooms of an apartment, can likewise be traced back to royal custom. As early as 1501 when the Chateau de Blois was being prepared for an exceptional reception of the Archduke Philippe Le Beau, one of the preliminary rooms in the suite was designated in the description of the arrangements as "the room in which the king eatsÓ ("At .mile oil mangeoit le Roy")- Evidence of such specialization of use, however, appears only in 1637 in the Hotel dArsenal in Paris and not for several years more does it become at all common to set aside a room specially for eating. Even in the eighteenth century, the more flexible arrangement maintained, though it had become more general by then to use the second anteroom for eating. When a separate ralle a manger was included in a plan, it usually opened off the entrance vestibule, isolated from the enfilade of an apartment, for it was thought of as a room of disruptive activity, noisy and smelly.
The placing of furniture in French hotel), the ranging of fixed sofas and chairs around the walls of reception rooms with no more than a handful of movable chairs and stools in the center, can likewise be related to the etiquette of the court. Though far too little is yet known of these patterns of use to judge of their effect on architecture. The earliest surviving decor of a room, the Marechale de La Meil-lerayeÕs rooms in the Arsenal, Paris, dates from 1637.
The literature of French planning begins, perhaps, with books of manners, some of which date back to the fourteenth century; though the first of consequence were Baldassare CastiglioneÕs II lihro delcortegia.no, 1528 (translated into French by 1537 and into English as The Covrtyer in 1561) and Desiderius ErasmusÕs equally famous De civilitate niorum puerilium (Good manners for children), 1530 (appearing in French for the first time in 1537 and again and again thereafter). The latter book, as the title makes plain, was directed to the upbringing of children, but this was no more, it would seem, than a tactful device. Other such books followed: another Italian import, Giovanni della CasaÕs II Galateo, 1558 (translated into French as Le Gala-thee; ou, La manure et farron comme le gentilhomme re doitgouiurner en toute compagnie in 1562 and into English as Galateo . . . ; or rather, A Treatise on the Alanerr and Behauiourr, It Beboueth a Man to Vre and Erchewe in 1576); Claude Hours de CalviacÕs La civile hon-ertetepour lej enfantd (Good manners for children), 1559, modeled in part on ErasmusÕs work; Nicolas FaretÕs LÕhonnete homme; ou, L'art de plaire a cour, 1630 (translated into English as The llonerl Alan; or, The Art to Pleare in Court in 1632). Not until the appearance of Antoine de CourtinÕs Nouveau traitede la civilite quire pratique en France parmi lev honnetev gene, 1671 (translated into English as The Ruler of Civility the same year), however, is the relation of manners to architecture tentatively engaged. Behavior in the antiehambre is defined, as is that in la chambre oil ert le lit', in neither room is one to
wear a hat, in the bedchamber one is not to sit on the bed, etc. The concept of propriety of behavior, or bienneance, is linked directly, more than once in the book, to this same concept in architecture. Later books, Claude FleuryÕs Leo devoirs dej maitrej et dej domejtiquej (The duties of masters and servants), 1688, or AudigerÕs La mauon reqleeet dart de diriger In million (The well-run house and the art of running a house), 1692, in which the duties of masters and servants are defined in detail, serve even more effectively to indicate the differentiations of usage, and thus of architectural space, that were gradually emerging in the seventeenth century. But once again these and their successors Ñ and there was a spate of such publications in the eighteenth century Ñ have not yet received sufficient careful study with reference to surviving inventories and plans to provide any real understanding.
The first book on architecture that offers more Ñ though only marginally more Ñ than illustrations of the areas enclosed by walls is Pierre Le MuetÕs Man'iere de bien bantirpour touted jortej deperjonnej, 1623 (issued in enlarged form in 1647; translated into English as The Art of Fair Building in 1670). There, the position of the great bed (as opposed to the ubiquitous folding beds that were set up anywhere) is indicated in the main room of each apartment. Labeling of the rooms reveals that the term jalle is used for a room of general use, a chambre is a room for sleeping (fig. 6), while on the additional plates of 1647, newly fashionable terms such as petite jalle d manger, anti-chambre, and alcove are included (fig. 7). This last is sometimes thought to have been an import from Spain, though one of the first recorded in France was that introduced by the Marquise de Rambouillet into her house in the rue Saint-Thomas du Louvre between 1619 and 1620. The marquise was born in Rome, the daughter of the Prin-cipessa Savelli. She is credited with a great deal more in terms of the refinement of comfort in architecture: the separation of spaces of representation from those used for private enjoyment, the proper control of heating, the lowering of window sills to enlarge the view to the garden, etc. But these were all incorporated into a hotel of somewhat eccentric organization.
Other books illustrating architectural plans followed Le Muet s, but they offer no more than the briefest commentaiy at best. The first full text on the principles of planning appeared only in 1691, when Augustin-Charles Daviler included a chapter, ÒDe la distribution des plansÓ ("On planning arrangements Õ) in his Courj dÕarchitecture (Course of architecture). There, he illustrated and described the arrangement of a large town house of his own design (figs. 8, 9). But the form was by then firmly fixed. It consisted of an entrance court flanked by stables and service wings, dominated by the main residential block, or corpj de loqin, with a private garden beyond. DavilerÕs corpj de login is two rooms deep, an arrangement that is thought to have been adopted first in 1639 in Louis Le VauÕs Flotel Tambonneau, to be taken up a
7.	Ground-floorplan ofa town house on a site f 2ft wide with a 'petitesalle a manger ÕÕmarked. Illustrated in Pierre Le Muet, Augmentations de nouveaux bastimens faits en France (Paris: Francois Langlois, 1647), 2. Middleton Collection.
8.	Ground-floorplan of a town house. Illustrated in Augustin- Charles Daviler, Cours dÕarchitecturt (Paris:Jean Mariette, 1738), 199,pi. 61.
Santa Monica, The Getty Centerfor the History of Art and the Humanities.
9.	First-floorplan of a town house. Illustrated in Augustin- Charles Daviler, Cours dÕarchiteeture (Paris:Jean Mariette, 1738), 203,pi. 62.
Santa Monica, The Getty Centerfor the History of Art and the Humanities.
few years later, in 1648, by Frangois Mansart for the Hotel de Jars, which served thereafter as the model; but though the double range of rooms allowed for far more flexibility of planning than the single range, the latter continued to be used until well i nto the eighteenth century.
Daviler s hotel is composed, as might be expected, of apartments on both the ground and the first floors Ñ the state apartment, orappartement de parade, being set on the first floor, still following the Italian usage, which derives from the imperial palaces of antiquity. The dining room, or oalle a manger, is set off the entrance vestibule, separated from the main enfiladed. Daviler offered much detailed advice as to the size and finish of the various rooms and technical information of much specificity, indicating that the science of planning was by then an established affair.
When Sebastien Leblond issued a second edition of DavilerÕs treatise in 1710, however, he thought it necessary to add yet another chapter, "De la nouvelle maniere de distribuer les plansÓ (ÒOn the new manner of planningÓ) to explain more fully the latest refinements. For in Louis XIVs declining years, a society enriched as a result of his endless campaigns and increasingly independent had established itself in Paris, rather than on the fringe of the court of Versailles, and was there exploring the delights of privacy and intimacy in architecture with an increasing degree of sophistication. Pierre Cailleteau, known as Lassurance, assistant to the king's architect, Jules Hardouin-Mansart, had built a series of hotel' that had initiated the new trend, starting with the Hotel Rothelin in 1700 (figs. 10, 11). Yet, as LeblondÕs text reveals, the principles of planning had not much changed. The general and even the detailed arrangement of the hotel remained much as described by Daviler. The kitchen and service rooms were now no longer in the basement of the corpe de logit but in the service wings flanking the entrance court. And the state reception rooms were now on the ground floor, opening directly onto the garden. LeblondÕs real interest, however, is directed rather to the provision of service stairs and the introduction of numerous small rooms for the storage of clothes or commodes (chaitee pc recce) or, preferably, the latest flushing toilets (fig. 12).
This indicates the pattern of change in the planning of the French hotel (and also the French chateau, for they were ordered on the same principles) during the first half of the eighteenth century. These changes were described at length and acclaimed by a succession of theorists Ñ by Jean Courtonne in 1725, by Gilles Tierce-let in 1728, by Jacques-Frangois Blondel in 1737 (and later in 1752 and in 1771), by Charles-Etienne Briseux in 1743, and, as we have seen, by Germain Boffrand in 1745. They confidently asserted that the new art of planning was the greatest of French contributions to architecture; and even that proud Scot Robert Adam recognized the primacy of the French in this respect in his Worke in Architecture, 1773 onward. Their
i o. Ground-floorplan of the HotelRothelin, rue de Grenelle, Paris, designed byLassurance and erected circa 1700. Illustrated inJacques-Frangois Blondel, Architecture franchise (Paris: Charles-Antoinejombert, 1752),
2: pi. 44 (facingp. 232). Santa Monica, The Getty Center for the History of Art and the Humanities.
11.	Entrance elevation ofthe Hotel Rothelin, rue de Grenelle, Paris, designed byLassurance and erected circa 1700. Illustrated inJacques-Frangois Blondel, Architecture frangoise (Paris: Charles-Antoinejombert, 1752),
2: pi. 45 (followingp. 232). Santa Monica, The Getty Center for the History of Art and the Humanities.
great achievement was to perfect the apartment as a sequence of spaces of ever-increasing comfort and intimacy, culminating in a warren of small service rooms with bathrooms and toilets supplied with hot and cold water and something of a drainage system. The whole could now be serviced with the utmost discretion.
Even in the grandest of seventeenth-century mansions, rooms had been used indiscriminately by many members of a family and by passing servants. There was an easy promiscuity. But in the eighteenth century, with the opening up of the realm of feeling and especially individual sentiment, privacy took on a new value. Rooms were more often set aside for personal use, which meant, of course, that intimate relationships could be more successfully explored. Masters were increasingly separated from servants. Attitudes toward servants changed noticeably in these years; even personal servants were treated more often as simple employees, less and less as members of an extended family or household.
Not only were service stairs introduced to provide separate access to rooms but whole networks of corridors as well. These had never been part of the planning arrangements on the main floors. The difficulties encountered by architects in integrating these service spaces and corridors into their plans are still in evidence in some of Jacques-Francois BlondelÕs designs in De la distribution des maisons deplai-sance (On the planning of country houses), 1737, and even more so in Charles-Etienne BriseuxÕs LÕart de batir des maisons de campagne (The art of building country houses), 1743 (fig. 13), where they emerge as the crudest of manipulations. Their coherent assimilation was rendered even more difficult by the fact that the new rationalism and the growing concern for expression Ñ clearly articulated by both Jean Cour-tonne and Blondel Ñ dictated that the interior form of a building should be closely related to its external form. There was to be a neat fit. The outside must not only provide a clear indication of the inside arrangement but must also serve to establish the character of the whole, which was to be sustained throughout a building.
These adjustments in patterns of living and in attitudes led to an increasing specialization in the use of rooms; more were added and the purpose of each determined from the start so that the decor and finishes could more fittingly reflect character and use. Rational criteria, however, were sometimes at odds with one another. The adjustment of rooms to altogether appropriate sizes and the introduction of a wide variety of shapes Ñ hexagons, circles, and ovals (the proliferation of which both Courtonne and Blondel thought somewhat vulgar) Ñ meant that the walls of an apartment on one level might not correspond to those on another level. This led to much manipulation of structure, often rendering the whole quite unstable Ñ as Pierre Patte balefully remarked in Monuments eriges a la gloire de Louis XV (Monuments erected to the glory of Louis XV), 1765 Ñand thus served as a curb to the sheer virtuosity
12.	Plans and sections illustrating aflushing toilet. Illustrated inJacques-Frangois Blonde!, De la distribution des maisons de plaisance (Paris: Charles-Antoine Jombert: 1738), 2:139, pi. 86, no. 3. SantaMonica, The Getty Center for the History of Art and the Humanities.
13.	Ground- and first-floor plan of a country house. Illustratedin Charles-Etienne Briseux, LÕart de batir des maisons de campagne (Paris: J. B. Gihert, 1761), i: 83, pi. 43. Boston, Courtesy of the Trustees of the Boston Public Library.
14.	Ground-floorplan of the abbots house at Villers- Cotterets, designed by Jacques-Frangois Blondel in 176 f, showing how symmetries are maintained in individual rooms and even in their relationship, despite the vagaries of site conditions. Illustrated in Jacques-Frangois Blondel, Cours dÕarchitecture (Paris: La Veuve Desaint, 177}), 4:pi 46. Santa Monica, The Getty Center for the History of Art and the Humanities.
of Rococo planning. Similarly, the requirements of internal planning, however skillfully they might be resolved, were not always allowed to determine the external forms of buildings. Both Daviler and Blondel were in no doubt that in the noblest of hotel) the magnificence and symmetry of the architecture as presented to the world was of far greater importance than ease of circulation or comfort. These might be sacrificed without qualm when rank and decorum decreed.
Symmetry was of overriding import in establishing the image of the French hotel, and once again, this might often be achieved only through the neglect of comfort or even common sense. When only one service wing flanked an entrance court, the elevation of the corpo de logut might not lie on the same axis as the garden front, a displacement that was readily enough masked by avoiding a cross axis in the plan. This expedient accorded with the wish to enter each apartment at its extremity, through the first room of the enfilade Ñ once again to lay emphasis on the extent of architectural magnificence opened up to the visitor. These enfiladeo were ranged, almost invariably, through doors set alongside the garden front, which meant that if an appearance of symmetry was to be maintained within rooms, false doors or equivalent elements must be provided at the rear of the rooms. The chimneypiece, with a looking glass over, had always to be placed centrally in a wall Ñ usually in the wall facing the door of the enfilade so that it should be seen in all its splendor on first entry Ñ and this was balanced on the opposite side of the room with an equivalent projecting console with its looking glass over. Courtonne and Blondel, once again, though they liked reflections, were chary of any too lavtsh use of the looking glass. The main items of furniture, as one might expect, had to be symmetrically disposed, the bed in particular. Blondel was of the opinion that a bed might be set in a corner position only in a bourgeois establishment.
This deeply felt need for an appearance of symmetry was a spur to the most extraordinarily ingenious feats of planning. Architects learned to dispose even the most awkwardly shaped sites so that the buildings upon them and the rooms within presented an air of the finest order and regularity. Such contrrvance, like the provision of false doors, was regarded as no more than a proper expression of rationalist ideals. For though the extreme refinement of hotel design in the early years of the eighteenth century is clearly a reflection of the expansion of individual sentiment, an essential part of the new empiricist grasp of the world, it is equally an outcome of that earlier affirmation of the rule of reason, an expression of those Cartesian concerns for sound practice and utility (figs. 14,15).
This triumph was to be given a further twist of grace in the second halt of the century. A succession of architects had learned to look at the architecture of classical antiquity anew and, far from becoming mere imitators of its forms, had ab-
sorbed its lessons with a marked perception and expanded them in fashioning an architecture of a new sharpness and precision of geometry, which was at the same time responsive to the fullest range of human feelings. Julien-David Leroy, for example, the first architect to provide a convincing record of the buildings of classical Greece in his magnificent folio Leo rubies des plus beaux monuments de la Grece (The ruins of the finest monuments of Greece), 1758, was sufficiently dispassionate in his appreciation of the actual forms and details of those newly revealed monuments to suggest six years later in his short Hut tour de la disposition et des formes differentes que les ebretiens out donnees a leurs temples (History of the arrangement and different forms given to their churches by Christians) that the sensations experienced when walking through or within even the noblest classical portico might be aroused equally successfully by an avenue of trees.
This gives a clue to the aesthetics explored soon after by Le Camus de Mezieres. His seminal ideas can be traced to Leroy. Leroy Õs stance is that of a lover of nature, a lover perhaps of formal gardens Ñ though this is not certain. For that aesthetics based on a recourse to feeling and individual experience, which Roger de Piles had revealed, gave a new value at first to landscape painting and then to nature itself, not so much in its unrefined state as in the contrived form of the landscape garden. It was the cultivation of the picturesque vision that enabled Leroy and, later, Le Camus de Mezieres to see architecture in a new way. But the process of change was slow, for in France symmetiy was a quality of almost divine import, and it was not readily to be abandoned. A taste for the overgrown formal garden was cultivated in theyears of the regence and after, no doubt influenced by the Dutch and Flemish landscape paintings that were then being collected by the more discerning connoisseurs. This was another response to de Piles. For the pleasure indulged in these gardens was of the painterly kind. They were seen in picturesque terms, as scenes from the Ruisdaels, Philips Wouwermans, or Jean-Antoine Watteau.
Another French painter who responded to the wanton neglect of the formal garden in these years was Jean-Baptiste Oudiy, who led outings on Sundays and public holidays to the Prince de GuiseÕs overgrown park at Arceuil, just south of Paris, where painters such as Charles-Joseph Natoire, Frangois Boucher, Andre Portail, Jean-Georges Wille, and others sketched from nature. One of this group was Claude-Henri Watelet, a rich amateur, who is usually credited with the making of the first picturesque garden in France at Moulin-Joli, an estate consisting of an island and the banks of the Seine near Bezons, to the west of Paris. He purchased the site around 1755 and slowly transformed it into a countiy retreat, aided by Boucher and others. The layout of the paths on the island was, perhaps not surprisingly geometric, though the effect was assiduously wild and unkempt. The fashion for gardens
i $. Detailplan ofthe bedroom and related rooms in the abbots house at Villers-Cotterets, designed by Jacques-Frangois Blondel in 1765, illustrating the complexity and finesse of geometrical composition. Illustrated inJacques-Francois Blondel\ Cours dÕarchitecture (Paris: La Veuve Desaint, 1773), 4:pi 48. Santa Monica, The Getty Centerfor the History of Art and the Humanities.
of a sustained picturesque composition, modeled on English examples, did not take hold in France until about 1770, and then tentatively and awkwardly. It was at about this time that Watelet wrote his Essaisur les jardins (Essay on gardens), 1774, intended as the first part of a more general study of taste. This was the first French work on the aims and methods of picturesque gardening. Le Camus de Mezieres was to dedicate his Le genie de / architecture to Watelet and to refer more than once to his example.