Chapter 1
Architecture as an expressive language
In antiquity, architects relied on the shared language of architecture, embodied in the architectural orders, to convey meaning through their work. The temples of Minerva, Mars and Hercules, for example, were built in the Doric order because its simple, potent form was appropriate to the gravity of these divinities. According to Vitruvius, âBecause of their might, buildings [devoted to these divinities] ought to be erected without embellishments.â Temples to Venus, Flore, Proserpine and the Nymphs required the Corinthian order, which was more appropriate to the gentleness of these goddesses. Temples to Juno, Diana and Bacchus were built in the Ionic order, which best represented these divinities: âThe determinate character of their temples will avoid the severe manner of the Doric and the softer manner of the Corinthian.â In De Architectura (first century BC), Vitruvius described this correctness of expression as decorum, âthe faultless ensemble of a work composed, in accordance with precedent, of approved details.â This fundamental respect for conventions (from the Greek thematismos) was dictated by both custom and nature.1
Vitruvius first wrote of the expressive role of architecture in terms of ornaments and their potential to express historical events. The story of the Caryatids, these statues of women from Caryae holding entablatures, is a good example. The replacement of columns by statues of enslaved women supporting cornices expressed not only their structural role, but the historical event of the defeat of the inhabitants of Caryae by the Greeks. Vitruvius explains:
The Peloponnesian city of Caryae had sided with the enemy, Persia, against Greece. Subsequently, the Greeks, gloriously delivered from war by their victory, by common agreement declared war on the Caryates. And so, when they had captured the town, slaughtered the men, and laid a curse on the inhabitants, they led its noble matrons off into captivity. Nor would they allow these women to put away their stolae and matronly dress; this was done so that they should not simply be exhibited in a single triumphal procession, but should instead be weighted down forever by the burden of shame, forced to pay the price for such grave disloyalty on behalf of their whole city. To this end, the architects active at the time incorporated images of these women in public buildings as weight-bearing structures; thus, in addition, the notorious punishment of the Caryate women would be recalled to future generations.2
For Vitruvius, architecture expressed an order that was both natural and conforming to cultural conventions. Architecture was not only the âart of buildingâ but also included gnomonics and mechanics.3 It spoke of the order of the universe. In 1684, Claude Perrault (1613â88), a member of the AcadĂ©mie Royale des Sciences and best known as an architect for his design of the east colonnade of the Louvre, translated Vitruviusâs treatise, and his extensive notes and commentaries demonstrate some fundamental changes that were transforming the very nature of architecture at the end of the seventeenth century. Whereas Vitruvius argued that architecture consists of ordonnance (a translation of the Greek term taxis), disposition or arrangement (from the Greek diathesis), eurhythmy or proportion, decorum (which Perrault translates as biensĂ©ance), and distribution (from the Greek oeconomia), Perrault insisted that Vitruvius was mistaken and that only ordonnance and disposition should be considered true constituents of architecture. Together, they express the use of each room and the destination of the building:
The ordering (ordonnance) of a building consists in the division of the space we are planning to use. This division is done in such a way that the dimension of every part is appropriate to its use and proportioned to the size of the whole building. . . . The disposition is the placing of all parts according to their quality, that is to say in the order determined by their nature and custom.4
For Vitruvius, however, decorum was also a central part of architecture. It was the aspect that determined âthe correctness of a building,â ensuring that the appearance of a project was âcomposed of approved elements and with authority.â Vitruvius explains that this authority was based on custom and nature. Perrault, however, insists that its origin is in custom (accoustumance) and regards it as the principal authority in architecture. Misreading Vitruvius, or perhaps adding a new emphasis, Perrault writes: âVitruvius seems to imply that custom is the principal authority in architecture.â5 In presuming that decorum was based primarily on custom, Perrault enabled the discipline of architecture to be conceived as an art based on convention. This radical questioning of the natural foundation of architecture introduced an arbitrariness that plunged the whole discipline into a potential crisis of meaning. However, architects did not immediately embrace Perraultâs position; they remained convinced that natural proportions were a fundamental principle, but were obliged to acknowledge the new role of conventions in their theories of architecture.
To understand the scope of these changes, it is important to consider the debate on the architectural orders that began in the late 1670s and resonated in architectural treatises for over a century.6 In 1683, Claude Perrault published his Ordonnance des cinq espĂšces de colonnes selon la mĂ©thode des anciens, the first architectural treatise to challenge the Vitruvian canon by questioning harmonic proportion as the foundation of architectural orders.7 François Blondel (1618â86), professor at the AcadĂ©mie Royale dâArchitecture and author of the authoritative Cours dâarchitecture, engaged in a debate with Claude Perrault. At issue was whether architectural proportions are based on natural and therefore absolute principles, as argued by Blondel, or whether they result from social conventions and a general consensus among architects, as maintained by Perrault.
In his Ordonnance, Perrault argues that architectural beauty and the meaning of architecture cannot reside in precise proportional relations because there is no rule on which all architects agree. Every architect, he claims, has attempted to perfect the art by adjusting architectural proportions. To support his claim, he compares the writings of various renowned authors and shows the inconsistencies among all previous unified proportional systems. He concludes that the beauty of architecture lies more in âthe grace of its formâ than in âthe exactitude of unvarying proportion.â The different characters attributed to the architectural orders on the basis of their relative proportions âwith little exactitude or precision are the only well-established matters in architecture.â8
Although Perrault was not the first to identify inconsistencies among the proportions that various authors had ascribed to the architectural orders since antiquity, he was the first to reject the explanation of his contemporaries, including François Blondel, who argued that minor discrepancies resulted merely from interpretation problems while the âuniversal idealâ remained unchallenged. To account for the dissimilarities, Perrault rejected the concept of a unified theory of harmony, and instead proposed two kinds of beauty in architecture: positive and arbitrary. âPositive beautyâ was based on what he called âconvincing reasonâ and included the demonstrable quality of craftsmanship. âArbitrary beauty,â on the other hand, was no less important, but was less tangible because it emphasized the composition of the whole and relied on conventions that could vary from one society to another.9 Presuming that the value of architectural proportions is relative, Perrault took the initiative to introduce a new module of his own that slightly adjusted the proportion of each architectural element, so that the pedestals and the heights of columns in the five orders would follow a progression of whole numbers. This âmethod founded on reasonâ is superior to others, Perrault argued, for âit affords memory a greater facility for retaining dimensions.â10 Perrault was indeed a true Cartesian: although he believed in the importance of universal norms for guiding architecture, he also thought that these norms should be based on reason rather than on precedents. Despite what might be expected, François Blondel did not refute Perraultâs simplified method for determining architectural proportions. On the contrary, in his commentary on LâArchitecture françoise des bastimens particuliers (1685) by Louis Savot, Blondel praises the advantages of this method. The work of Perrault, he writes,
contains a method far easier than any other to determine the proportions of the five architectural orders, because their parts follow fixed measurements and are either the same for all the orders, such as the entablatures that measure two diameters in height; or increase by equal intervals, such as the columns that exceed by two thirds of a diameter from one order to the next and the pedestals that increase by only one third of a diameter.
1.1 Corinthian order according to Palladio and Scamozzi
Source: R. FrĂ©art de Chambray, ParallĂšle de lâarchitecture antique avec la moderne, 1650
1.2 Proportions of the five architectural orders according to C. Perrault
Source:Ordonnance des cinq especes de colonnes, 1683
Blondel concludes that Perrault succeeded in creating a very effective âidea of the measurements of the architectural orders by taking an average between the biggest and the smallest that one can find in works of Antiquity and in architectsâ books.â11
Blondel did not oppose Perraultâs simplification and rationalization of proportional rules; he opposed the deeper philosophical implications of his theory. Perrault had challenged the premise that architecture is founded on absolute principles. Indeed, the most significant transformation brought about by his theory was the separation between the positive foundations of architecture (commodity, stability, salubrity) and the arbitrary rules based on custom. By promoting arbitrary rules, Perrault enabled architecture to be based on human principles, whereas previously it had always been based on an order that transcended the human condition. In his Cours dâarchitecture (1675â83), Blondel comments on Perraultâs forthcoming book on the architectural orders and questions his challenge to Vitruviusâs canon. Building, Blondel argues, is natural because it is born from necessity. In building, all that has to do with salubrity, stability and commodity is also natural because it is also derived from necessity. Decorum (biensĂ©ance) and decoration are more ambiguous, he says, yet they are also in our nature. Addressing more directly his contention with Perraultâs position, Blondel refutes the reasons invoked for âthe necessity of architectural proportions that are approved only by custom.â12 In effect, Blondel was warning against the long-term consequences of Perraultâs position: if the proportions of architectural orders were nothing more than a shared set of conventions established in antiquity and simply accepted by subsequent generations of architects, there would be no reason to prevent them being substituted by an infinite number of other proportions. The selection of proportions therefore would depend on the taste, experience and intelligence of the architect, thus challenging natural harmony as the basis for meaning in architecture. This was essentially the point of departure for a century-long debate over natural beauty and arbitrary beauty in architecture.
The consequences for practice, however, were not felt immediately. Perraultâs position remained controversial throughout the eighteenth century while Blondelâs teachings endured as the official principles of the AcadĂ©mie Royale dâArchitecture. After Perrault, Vitruvian authority was not replaced immediately by human custom and convention. On the contrary, the proportions of architectural orders retained their former association with nature (through the proportions of the human body), along with their assumed value and their implicit character. Although âcharacterâ was widely sought after by eighteenth-century architects, nature remained the acknowledged source of architectural proportions and its ability to convey meaning in architecture was never really in question. Perraultâs thesis, however, marked the beginning of major transformations in architectural theory that would have profound repercussions in practice. He not only developed his system of proportions into a method that was easier to use but also rejected the need for optical correction, thus giving theory an absolute supremacy over practice and enabling it to become a prescriptive tool.
This complex tension between convention and nature pervaded many spheres of knowledge during the Enlightenment. Although the practice of architecture was unaffected by these new theoretical concerns at the time of the debate between Perrault and Blondel, subsequent generations of architects certainly felt the need to acknowledge the profound ideological change brought about by Perrault. Toward the end of the eighteenth century, the practical implications of Perraultâs theory would become obvious in the work of innovative architects such as Claude-Nicolas Ledoux and Ătienne-Louis BoullĂ©e, who made a radical departure from the traditional âclassicalâ orders of architecture.
Character theory and the language of architecture
Following Perraultâs questioning of the Vitruvian canon, eighteenth-century French theory began to confront the loss of absolute principles in architecture. Although authors of architectural treatises continued to acknowledge Vitruvian sources, they gradually discontinued demonstrations of the natural origin of architectural proportions, thus ending a centuries-long tradition in architectural writing. And although architects still respected the Vitruvian principles, they felt the need to define a new theory of architecture that would acknowledge the growing importance of convention. Throughout the eighteenth century, architectural theory never really lost the desire to reconcile architectural order and cosmological order, but architects realized that they could no longer rely on universal harmony to give meaning to their work. Looking for shared conventions of architecture as an expressive language was an attempt to save architectural meaning. This new interest in the expressive power of architecture would lead to a theory of character.
One of the earliest formulations of this new character theory can be found in the writings of Jacques-François Blondel, author of the Cours dâarchitecture (1771â9) and numerous architectural treatises, and professor at the AcadĂ©mie Royale dâArchitecture from 1756 to 1774 (not to be confused with François Blondel mentioned earlier). It is in one of his early writings, De la distribution des maisons de plaisance (1737â8), an architectural treatise entirely devoted to country houses, that Blondel first declared that the exterior expression of a building should announce its destination. The façades of the main body of a building must be readily identifiable by the richness of their ornamentation and their elevation, Blondel writes, âso that those who only get a view of the exterior can recognize through this sign of distinction the residence of the Master. The other constructions that surround this central building must also express their use, either through sculpture or architectural elements.â13 As the organization and hierarchy of ...