Reading the Treatises

Most of the world’s buildings were made by people who never read a book about architecture. Their rules lived in the hands and memories of masons and carpenters, passed from master to apprentice, adjusted from one job to the next. But in a few places and periods builders or their patrons wrote the rules down, and once written, the rules changed their nature. They could be argued with, compared, printed, translated and carried to the other side of the world. The architectural treatise is one of the great engines of architectural history, and this part of the book is its library.

The one ancient survivor

Greek architects wrote books about their buildings (Iktinos wrote on the Parthenon, Hermogenes on his Ionic temples), but every one of them is lost. The single ancient treatise to survive complete is the De architectura of Vitruvius, written in the 20s BC. Its survival was a matter of luck: a handful of medieval manuscripts, copied in monasteries without their illustrations. When Renaissance humanists rediscovered it in the fifteenth century it took on an authority out of all proportion to its author’s modest fame, and the whole later tradition of Western architectural theory can be read as a commentary on it.

Vitruvius can be read in full in this library, in the English translation by Morris Hicky Morgan published in 1914, with live figures placed beside the passages they illustrate. Begin with Book III, on symmetry and the classes of temples, and Book IV, on the origins of the orders.

The Renaissance book

The first modern treatise, Leon Battista Alberti’s De re aedificatoria (written by 1452, printed 1485), was a humanist’s book: Latin, unillustrated, addressed to patrons and scholars, and concerned as much with the meaning of architecture as with its practice. Then came printing with woodcuts, and everything changed. Sebastiano Serlio, from 1537, made a book in which the pictures carried the argument, written in Italian for builders. Vignola reduced the orders to thirty-two engraved plates and a few lines of text. Palladio illustrated his own buildings, dimensioned so that readers could build from them. Within a century, anyone who could read a plan could practise classical architecture anywhere in Europe.

Academies, arguments and pattern books

In seventeenth-century France the Royal Academy of Architecture made theory an institution. Its members debated whether proportions were natural or conventional (Claude Perrault against François Blondel) and taught the orders as a discipline. In eighteenth-century Britain and America a different kind of book flourished: the builder’s manual and pattern book, from James Gibbs’s Rules for Drawing to hundreds of cheaper handbooks, which spread the classical vocabulary to provincial carpenters. William Chambers gathered the authorities into a judicious standard textbook.

At the same time, other writers began to question the rules from first principles. Marc-Antoine Laugier’s Essai (1753) imagined the primitive hut of four tree-trunks as the origin and measure of all architecture; J.-N.-L. Durand’s lectures (1802–05) replaced the orders with a method of composition on a grid. In the nineteenth century A. W. N. Pugin and John Ruskin argued for Gothic as the only honest architecture, Viollet-le-Duc read Gothic as rational structure, and Gottfried Semper traced architecture to four primal crafts. Each of these books is summarised in the library.

Beyond Europe

The Western treatise tradition is the best known, but it is not unique. In China the Kaogong ji of the late Zhou period prescribed the plan of the ideal capital, and in 1103 the Song court architect Li Jie completed the Yingzao fashi, a building code of astonishing precision whose module, the cai (a standard section of bracket timber), plays the role of the column diameter in the Western orders. Indian builders followed the vastu and shilpa texts, among them the Mānasāra and the Mayamata, which prescribe the sacred diagram of the site and the proportions of temples and images. Japanese carpenters recorded the kiwari system of proportion in family manuals such as the Shōmei of 1608. These are summarised in the library alongside their European counterparts.

Manifestos

The twentieth century produced fewer treatises and more manifestos: short, polemical texts that proposed not a system of proportions but a new set of principles. Louis Sullivan’s essay on the tall office building (1896) gave modernism its slogan about form and function; Adolf Loos attacked ornament; Le Corbusier reduced a new architecture to five points and later devised a new system of proportion, the Modulor; Robert Venturi answered modernism with a plea for complexity and contradiction; Christopher Alexander tried to write a pattern language that anyone could use. The library summarises these too, in our own words, since they are still under copyright.

How to use the library

Each treatise has its own page with a summary of its argument, a book-by-book synopsis, the rules the atlas takes from it, and live figures illustrating its main ideas. Where a text is in the public domain and a reliable transcription exists, it can be read in full in the reader. The library shelf is arranged by date; the spines marked with a fleuron (❦) are available in full.