Any building is a defiance of gravity. Since earliest times, architects have tackled the challenge of erecting a roof over empty space, setting walls upright, and having the whole stand secure. Their solutions have depended upon the materials they had available, for, as we shall see, certain materials are better suited than others to a particular structural system. There are two basic families of structural systems: the shell system and the skeleton-and-skin system.

Core Concept: Structural Systems
(Your score will be reported to your instructor)

In the shell system one building material provides both structural support and sheathing (outside covering). Buildings made of brick or stone or adobe fall into this category, and so do older (pre-19th-century) wood buildings constructed of heavy timbers, the most obvious example being the log cabin. The structural material comprises the walls and roof, marks the boundary between inside and outside, and is visible as the exterior surface. Shell construction prevailed until the 19th century, when it began to fall out of favor. Today, however, the development of strong cast materials, including many plastics, has brought renewed interest in shell structures.

Page 281

The skeleton-and-skin system might be compared to the human body, which has a rigid bony skeleton to support its basic frame and a more fragile skin for sheathing. We find it in modern skyscrapers, with their steel frames (skeletons) supporting the structure and a sheathing (skin) of glass or some other light material. Also, most houses today—at least in the United States—are built with a skeleton of wood beams nailed together, topped with a sheathing of light wood boards, shingles, aluminum siding, or the like. Skeleton-and-skin construction is largely a product of the Industrial Revolution; not until the mid–19th century could steel for beams or metal nails be manufactured in practical quantities.

Two factors that must be considered in any structural system are weight and tensile strength. Walls must support the weight of the roof, and lower stories must support the weight of upper stories. In other words, all the weight of the building must somehow be carried safely to the ground. You can get a sense of this if you imagine your own body as a structural member. Suppose you are lying flat on your back, your body held rigid. You are going to be lifted high in the air, to become a “roof.” First you are lifted by four people: One supports you under the shoulders, one under the buttocks, one holds your arms extended above your head, another holds your feet. Because your weight is therefore channeled down through four vertical people to the ground, you can hold yourself horizontally with some ease. Next you are lifted by two people, one holding your shoulders, another your feet. A lot of your weight is concentrated in the center of your body, which is unsupported, so eventually you sag in the middle and fall to the floor. Then you are lifted by one person, who holds you at the center of your back. The weight at both ends of your body has nowhere to go, nothing to carry it to the ground, and you sag at both ends.

Core Concept: Mechanics, Force, and Structures
(Your score will be reported to your instructor)

Tensile strength refers to the amount of tensile (stretching) stress a material can withstand before it bends or breaks. As applied to architecture, it especially concerns the ability of a material to span horizontal distances without continuous support from below. Returning to the analogy of the body, imagine you are made not of flesh and blood but of strong plastic or metal. Regardless of how you are held up in the air, you can stay rigid and horizontal, because you have great tensile strength.

If you keep these images in mind, you may find it easier to understand the various structural systems we shall consider below. They are introduced here in roughly the chronological order in which they were developed. As was mentioned earlier, all will be of the shell type until the 19th century.

Load-Bearing Construction

Another term for load-bearing construction is “stacking and piling.” This is the simplest method of making a building, and it is suitable for brick, stone, adobe, ice blocks, and certain modern materials. Essentially, the builder constructs the walls by piling layer upon layer, starting thick at the bottom, getting thinner as the structure rises, and usually tapering inward near the highest point. The whole may then be topped by a lightweight roof, perhaps of thatch or wood. This construction is stable, because its greatest weight is concentrated at the and weight diminishes gradually as the walls grow higher.

Load-bearing structures tend to have few and small openings (if any) in the walls, because the method does not readily allow for support of material above a void, such as a window opening. Yet it would be a mistake to think that such basic methods must produce basic results. The Great Friday Mosque at Djenne, in Mali, is a spectacular example of monumental architecture created from simple techniques and materials (13.1). Constructed of adobe (sun-dried brick) and coated with mud plaster, the imposing walls of this mosque have a plastic, sculptural quality. The photograph shows well the gentle tapering of the walls imposed by the construction technique as well as the small size of the windows that illuminate the covered prayer hall inside. The protruding wooden poles serve to anchor the scaffolding that is erected every few years so that workers can restore the mosque's smooth coating of mud plaster.

Great Friday Mosque, Djenne, Mali. Rebuilt 1907 in the style of a 13th-century original.
Page 282

After stacking and piling, post-and-lintel construction is the most elementary structural method, based on two uprights (the posts) supporting a horizontal crosspiece (the lintel, or beam). This configuration can be continued indefinitely, so that there may be one very long horizontal supported at critical points along the way by vertical posts to carry its weight to the ground. The most common materials for post-and-lintel construction are stone and wood. Since neither has great tensile strength, these materials will yield and cave in when forced to span long distances, so the architect must provide supporting posts at close intervals.

Post-and-lintel construction has been, for at least four thousand years, a favorite method of architects for raising a roof and providing for open space underneath. The ruins of a portion of the ancient Egyptian temple of Amon-Mut-Khonsu illustrate the majesty and also the limits of post-and-lintel construction in stone (13.2). Carved as bundles of stems capped by stylized papyrus-flower buds, the stone columns support rows of heavy stone lintels, with each lintel spanning two columns. The lintels would in turn have supported wooden roof beams and roofing. Because stone does not have great tensile strength, the supporting columns must be closely spaced. A large hall erected in post-and-lintel construction was thus a virtual forest of columns inside. We call such spaces hypostyle halls, from the Greek for “beneath columns.” Ancient Egyptians associated hypostyle halls with the primal swamp of creation, where, according to Egyptian belief, the first mound of dry land arose at the dawn of the world. To make that connection clear, they designed their columns as stylized versions of plants that grew in the marshes of the Nile. Surrounded by load-bearing walls pierced high up by small windows, the hypostyle halls of Egyptian temples were dark and mysterious places.

View of the hypostyle from the courtyard temple of Amon-Mut-Khonsu, Luxor. Begun c. 1390 B.C.E. Height of columns 30′.
Photo by Wim Swaan. Library, Getty Research Institute, Los Angeles, Wim Swaan Photograph Collection, 96.P.21.
Page 283

In ancient Greece, the design of post-and-lintel buildings, especially temples, became standardized in certain features. Greek architects developed and codified three major architectural styles, roughly in sequence. We know them as the Greek orders. The most distinctive feature of each was the design of the column (13.3). By the 7th century B.C.E., the Doric style had been introduced. A Doric column has no base, nothing separating it from the floor below; its capital, the topmost part between the shaft of the column and the roof or lintel, is a plain stone slab above a rounded stone. The Ionic style was developed in the 6th century B.C.E. and gradually replaced the Doric. An Ionic column has a stepped base and a carved capital in the form of two graceful spirals known as volutes. The Corinthian style, which appeared in the 4th century B.C.E., is yet more elaborate, having a more detailed base and a capital carved as a stylized bouquet of acanthus leaves.

Column styles of the Greek orders.
Page 284

The most famous and influential work of Greek architecture is certainly the Parthenon, a Doric temple that we will examine in Chapter 14 (see 14.26). Here, we look at the smaller Temple of Athena Nike (13.4), which stands nearby on the hillsite in Athens known as the Acropolis. With their stepped bases and volute capitals, the columns indicate that this is an Ionic temple. The columns support a structure whose remains, reconstructed here in a line drawing (13.5), display other important elements of Greek architecture. The plain, horizontal stone lintels of Egypt are here elaborated into a compound structure called an entablature. The entablature consists of three basic elements. The simple, unadorned band of lintels immediately over the columns is the architrave. The area above the architrave is the frieze, here ornamented with sculpture in relief. The frieze is capped by a shelflike projection called a cornice. The entablature in turn supports a triangular element called a pediment, which is itself crowned by its own cornice. Like the frieze, the pediment would have been ornamented with sculpture in relief. If these elements look familiar to you, it is because they have passed into the vocabulary of Western architecture and form part of the basis of the style we refer to broadly as classical. For centuries, banks, museums, universities, government buildings, and churches have been built using the elements first codified and named by the Greeks, then adapted and modified by the Romans.

Kallikrates. Temple of Athena Nike, from the east, Acropolis, Athens. 427–424 B.C.E. Pentelic marble.
Elevation, Temple of Athena Nike.

Many of the great architectural traditions of the world are based in post-and-lintel construction. The architectural style developed in China provides a good contrast to that of Greece, for while its principles were developed around the same time, the standard material is not stone but wood. We know from terra-cotta models found in tombs that the basic elements of Chinese architecture were in place by the second century B.C.E. During the 6th century C.E., this architectural vocabulary was adopted by Japan along with other elements of Chinese culture. We illustrate it here with a Japanese building, the incomparable Byodo-in (13.6).

Hoodo (Phoenix Hall), Byodo-in Temple, Uji, Kyoto Prefecture, Japan. Heian period, c. 1053.
Bracket system.

Built as a palace, Byodo-in was converted to a Buddhist shrine after the death of the original owner in 1052 C.E. Among the works of art it houses is Jocho's sculpture of Amida Buddha, discussed in Chapter 2 (see 2.28). Our first impression is of a weighty and elaborate superstructure of gracefully curved roofs resting—lightly, somehow—on slender wooden columns. The effect is miraculous, for the building seems to float; but how can all of that weight rest on such slender supports? The answer lies in the cluster of interlocking wooden brackets and arms that crowns each column (13.7). Called bracket sets, they distribute the weight of the roof and its large, overhanging eaves evenly onto the wooden columns, allowing each column to bear up to five times the weight it could support directly. Chinese and Japanese architects developed many variations on the bracket set over the centuries, making them larger or smaller, more elaborate or simpler, more prominent or more subtle.

Page 285

The distinctive curving profile of East Asian roofs is made possible by a stepped truss system (13.8). (Western roofs, in contrast, are usually supported by a rigid triangular truss, as in the Greek pediment.) By varying the height of each level of the truss, builders could control the pitch and curve of the roof. Taste in roof styles varied over time and from region to region. Some roofs are steeply pitched and fall in a fancifully exaggerated curve, almost like a ski jump; others are gentler, with a subtle, barely noticeable curve.

Stepped truss roof structure.

The post-and-lintel system, then, offers potential for both structural soundness and grandeur. When applied to wood or stone, however, it leaves one problem unsolved, and that is the spanning of relatively large open spaces. The first attempt at solving this problem was the invention of the round arch.

Page 286
Round Arch and Vault

Although the round arch was used by the ancient peoples of Mesopotamia several centuries before our common era (see 14.9), it was most fully developed by the Romans, who perfected the form in the 2nd century B.C.E. To get a sense of how the arch works, we might go back to the analogy of the body. Imagine that, instead of lying flat on your back, you are bent over forward into a curve, and again you will be lifted into the air. One person will support your hands, another your feet. As long as your body follows the proper arc—that is, your two supporters stand the correct distance apart—you can maintain the pose for some time. If they stand too close together, you start to topple first one way and then the other; if they move too far apart, you have insufficient support in the middle and plunge to the floor. An arch incorporates more complex forces of tension (pulling apart) and compression (pushing together), but the general idea is the same.

The arch has many virtues. In addition to being an attractive form, it enables the architect to open up fairly large spaces in a wall without risking the building's structural soundness. These spaces admit light, reduce the weight of the walls, and decrease the amount of material needed. As utilized by the Romans, the arch is a perfect semicircle, although it may seem elongated if it rests on columns. It is constructed from wedge-shaped pieces of stone that meet at an angle always perpendicular to the curve of the arch. Because of tensions and compressions inherent in the form, the arch is stable only when it is complete, when the topmost stone, the keystone, has been set in place. For that reason, an arch under construction must be supported from below, usually by a wooden framework. In addition, an arch exerts an outward thrust at its base that must be contained (see diagram).

Among the most elegant and enduring of Roman structures based on the arch is the Pont du Gard at Nîmes, France (13.9), built about 15 C.E. when the empire was nearing its farthest expansion (see map, p. 340). At that time Roman industry, commerce, and agriculture were at their peak. Engineering was applied to an ambitious system of public-works projects, not just in Italy, but in the outlying areas as well. The Pont du Gard functioned as an aqueduct, a structure meant to transport water, and its lower level served as a footbridge across the river. That it stands today virtually intact after nearly two thousand years (and is crossed by cyclists on the route of the famous Tour de France bicycle race) testifies to the Romans' brilliant engineering skills. Visually, the Pont du Gard exemplifies the best qualities of arch construction. Solid and heavy, obviously durable, it is shot through with open spaces that make it seem light and its weight-bearing capabilities effortless.

Pont du Gard, Nîmes, France. Early 1st century C.E. Length 902′.
Page 287

When the arch is extended in depth—when it is, in reality, many arches placed flush one behind the other—the result is called a barrel vault. This vault construction makes it possible to create large interior spaces. The Romans made great use of the barrel vault, but for its finest expression we look many hundreds of years later, to the churches of the Middle Ages.

The church of Sainte-Foy (13.10), in the French city of Conques, is an example of the style prevalent throughout western Europe from about 1050 to 1200—a style known as Romanesque. Earlier churches had used the Roman round arch to span the spaces between interior columns that ultimately held up the roof. There were no ceilings, however. Rather, worshipers looked up into a system of wooden trusses and the underside of a pitched roof (see 15.2, 15.3). Imagine looking directly up into the attic of a house and you will get the idea. With the Romanesque style, builders set a stone barrel vault as a ceiling over the nave (the long central area), hiding the roof structure from view. The barrel vault unified the interior visually, providing a soaring, majestic climax to the rhythms announced by the arches below.

Interior, Sainte-Foy, Conques, France. c. 1050–1120.
Page 288

On the side aisles of Sainte-Foy (not visible in the photograph), the builders employed a series of groin vaults. A groin vault results when two barrel vaults are crossed at right angles to each other, thus directing the weights and stresses down into the four corners. By dividing up a space into square segments known as bays, each of which contains one groin vault, the architects could cover a long span safely and economically. The repetition of bays also creates a satisfying rhythmic pattern.

Pointed Arch and Vault

Although the round arch and the vault of the Romanesque era solved many problems and made many things possible, they nevertheless had certain drawbacks. For one thing, a round arch, to be stable, must be a semicircle; therefore, the height of the arch is limited by its width. Two other difficulties were weight and darkness. Barrel vaults are both literally and visually heavy, calling for huge masses of stone to maintain their structural stability. They exert an outward thrust all along their base, which builders countered by setting them in massive walls that they dared not weaken with light-admitting openings. The Gothic period in Europe, which followed the Romanesque, solved those problems with the pointed arch.

Page 289

The pointed arch, though seemingly not very different from the round one, offers many advantages. Because the sides arc up to a point, weight is channeled down to the ground at a steeper angle, and therefore the arch can be taller. The vault constructed from such an arch also can be much taller than a barrel vault. Architects of the Gothic period found they did not need heavy masses of material throughout the curve of the vault, as long as the major points of intersection were reinforced. These reinforcements, called ribs, are visible in the nave ceiling of Reims Cathedral (13.11).

Nave, Reims Cathedral, France. 1211–c. 1290. Height 125′.

The light captured streaming into the nave of Reims Cathedral in the photograph vividly illustrates another important feature of Gothic church architecture: windows. Whereas Romanesque cathedrals tended to be dark inside, with few and small window openings, Gothic builders strove to open up their walls for large stained glass windows such as the two radiant round windows, called rose windows, visible in the photograph. (Most of the stained glass windows in Reims Cathedral have suffered damage and been replaced with clear glass, which is why the light is so evident in the photograph.) Fearing that the numerous window openings could disastrously weaken walls that were already under pressure from the outward thrust of arches, Gothic builders reinforced their walls from the outside with buttresses, piers, and a new invention, flying buttresses. The principles are easy to understand if you imagine yourself using your own weight to prop up a wall. If you stand next to a wall and press the entire length of your body against it, you are a buttress. If you stand away from the wall and press against it with outstretched arms, your body is a pier, and your arms are flying buttresses. The illustration here (13.12) of the exterior of the Cathedral of Le Mans, in France, shows the Gothic system of buttresses, piers, and flying buttresses, as well as the numerous windows that made them necessary.

Exterior of the Cathedral of Le Mans, France, showing buttresses, piers, and flying buttresses. 1217–54.
Page 290

A dome is an architectural structure generally in the shape of a hemisphere, or half globe. One customary definition of the dome is an arch rotated 360 degrees on its axis, and this is really more accurate, because, for example, the dome based on a pointed arch will be pointed at the top, not perfectly hemispherical. The stresses in a dome are much like those of an arch, except that they are spread in a circle around the dome's perimeter. Unless the dome is buttressed—supported from the outside—from all sides, there is a tendency for it to “explode,” for the stones to pop outward in all directions, causing the dome to collapse.

Like so many other architectural structures, the dome was perfected under the incomparable engineering genius of the Romans, and one of the finest domed buildings ever erected dates from the early 2nd century. It is called the Pantheon (13.13, 13.14, 13.15), which means a temple dedicated to “all the gods”—or, at least, all the gods who were venerated in ancient Rome. As seen from the inside, the Pantheon has a perfect hemispherical dome soaring 142 feet above the floor, resting upon a cylinder almost exactly the same in diameter—140 feet. The ceiling is coffered—ornamented with recessed rectangles, coffers, which lessen its weight. At the very of the dome is an opening 29 feet in diameter called an oculus, or eye, thought to be symbolic of the “eye of Heaven.” This opening provides the sole (and plentiful) illumination for the building. In its conception, then, the Pantheon is amazingly simple, equal in height and width, symmetrical in its structure, round form set upon round form. Yet because of its scale and its satisfying proportions, the effect is overwhelming.

Pantheon, Rome. 118–125 C.E.
Section drawing of the Pantheon.
Interior of the Pantheon.
Page 291

The combined structural possibilities of the dome and the vault enabled the Romans to open up huge spaces such as the Pantheon without interior supports. Another important factor that allowed them to build on such a scale was their use of concrete. Whereas Greek and Egyptian buildings had been made of solid stone, monumental Roman buildings were made of concrete, poured into hollow walls of concrete brick as though into a mold, then faced with stone veneer to look as though they were made of solid stone. An important technological breakthrough, the use of concrete cut costs, sped construction, and enabled building on a grand scale.

Visitors enter the Pantheon through the rectangular portico, or porch, that is joined somewhat incongruously to it. Here we recognize the characteristic form of the Greek temple as inherited by the Romans: post-and-lintel construction, Corinthian order, entablature, and pediment. In Roman times, an approach to the building was constructed to lead to the portico while obscuring the rest of the temple. Thinking that they were entering a standard post-and-lintel temple, visitors must have been stunned to see the enormous round space open up before their eyes. Tourists today experience the same theatrical surprise.

Page 292

The Pantheon is a rotunda, a round building, and its dome sits naturally on the circular drum of the base. Often, however, architects wish to set a dome over a square building. In that case, a transitional element is required between the circle (at the dome's base) and the square (of the building's top). An elegant solution can be found in Hagia Sophia (the Church of the Holy Wisdom) in Istanbul (13.16, 13.17). Designed by two mathematicians, Anthemius of Tralles and Isidorus of Miletus, Hagia Sophia was built as a church during the 6th century, when Istanbul, then called Constantinople, was the capital of the Byzantine Empire. When the Turks conquered the city in the 15th century, Hagia Sophia was converted for use as a mosque. It was at that time that the four slender towers, minarets, were added. The building is now preserved as a museum. In sheer size and perfection of form, it was the architectural triumph of its time and has seldom been matched since then.

Louis Haghe, after a drawing by Gaspare Fossati. Interior of Hagia Sophia. 1852. Lithograph, sheet size approx. 21 × 15″.
Hagia Sophia, Istanbul. 532–37.

The dome of Hagia Sophia rises 183 feet above the floor, with its weight carried to the ground by heavy stone piers—in this case, squared columns—at the four corners of the immense nave. Around the base of the dome is a row of closely spaced arched windows, which make the heavy dome seem to “float” upward. (The exterior view makes it clear that these windows are situated between buttresses that ring the base of the dome, containing its outward thrust and compensating for any structural weakening caused by the window openings.) Each of the four sides of the building consists of a monumental round arch, and between the arches and the dome are curved triangular sections known as pendentives. It is the function of the pendentives to make a smooth transition between rectangle and dome.

The domes of the Pantheon and the Hagia Sophia serve primarily to open up vast interior spaces. Seen from the outside, their hemispherical form is obscured by the buttressing needed to contain their powerful outward thrust. Yet the dome is such an inherently pleasing form that architects often used it for purely decorative purposes, as an exterior ornament to crown a building. In that case, it is often set high on a drum, a circular base, so that it can be seen from the ground. A famous example of a building crowned by an ornamental dome is the Taj Mahal, in Agra, India (13.18).

Taj Mahal, Agra, India. 1632–53.
Page 293

The Taj Mahal was built in the mid–17th century by the Muslim emperor of India, Shah Jahan, as a tomb for his beloved wife, Arjummand Banu. Although the Taj is nearly as large as Hagia Sophia and possessed of a dome rising some 30 feet higher, it seems comparatively fragile and weightless. Nearly all its exterior lines reach upward, from the graceful pointed arches, to the pointed dome, to the four slender minarets poised at the outside corners. The Taj Mahal, constructed entirely of pure white marble, appears almost as a shimmering mirage that has come to rest for a moment beside the peaceful reflecting pool.

The section drawing (13.19) clarifies how the dome is constructed. Over the underground burial chambers of Shah Jahan and his wife, the large central room of the tomb rises to a domed ceiling. Over this, on the roof of the building, sits a tall drum crowned by a pointed dome. A small entryway gives access to the inside for maintenance purposes, but it is not meant to be visited. The exterior is shaped in a graceful, bulging S-curve silhouette that obscures the actual drum-and-dome structure evident in the cutaway view.

Section, Taj Mahal.
Page 294
Corbelled Arch, Vault, and Dome

Islamic architects knew the use of the arch and the dome because Islam came of age in a part of the world that had belonged first to the Roman and then to the Byzantine Empire. When Islamic rulers settled in India, their architects brought these construction techniques with them, resulting in such buildings as the Taj Mahal. Indigenous Indian architecture, in contrast, does not make use of the arch or the dome but is based on post-and-lintel construction. To create arch, vault, and dome forms, Indian architects used a technique called corbelling. In a corbelled arch, each course (row) of stones extends slightly beyond the one below, until eventually the opening is bridged.

Just as a round Roman arch can be extended in depth to create a vault or rotated to create a dome, so corbelling can create vault forms and, as in the temple interior illustrated here, dome forms (13.20). Ornamented by band upon band of ornate carving and set with figures of the sixteen celestial nymphs, the corbelled dome rests on an octagon of lintels supported by eight columns. Pairs of stone brackets between each column provide additional support. The elaborate, filigreed carving that decorates every available surface testifies to the virtuosity of Indian stoneworkers, in whose skillful hands stone was made to seem as light as lace.

Interior of the Jain temple of Dilwara, Vimala temple, Mount Abu, South Rajasthan, India. Completed 1032.

Although to the naked eye a corbelled arch may be indistinguishable from the round arch described previously, it does not function structurally as a round arch does, channeling weight outward and downward, and so does not enable the construction of large, unobstructed interior spaces.

Page 295
Cast-Iron Construction

With the perfection of the post-and-lintel, the arch, and the dome, construction in wood, stone, and brick had gone just about as far as it could go. Not until the introduction of a new building material did the next major breakthrough in structural systems take place. Iron had been known for thousands of years and had been used for tools and objects of all kinds, but only in the 19th century did architects realize that its great strength offered promise for structural support. This principle was demonstrated brilliantly in a project that few contemporary observers took seriously.

In 1851 the city of London was planning a great exhibition, under the sponsorship of Prince Albert, husband of Queen Victoria. The challenge was to house under one roof the “Works of Industry of All Nations,” and the commission for erecting a suitable structure fell to Joseph Paxton, a designer of greenhouses. Paxton raised in Hyde Park a wondrous building framed in cast iron and sheathed in glass—probably the first modern skeleton-and-skin construction ever designed (13.21). The Crystal Palace, as Paxton's creation came to be known, covered more than 17 acres and reached a height of 108 feet. Because of an ingenious system of prefabrication, the whole structure was erected in just sixteen weeks.

Joseph Paxton. Crystal Palace, Hyde Park, London. 1851, destroyed by fire 1936.
Contemporary lithograph by Joseph Nash. Guildhall Library, London.

Visitors to the exhibition considered the Crystal Palace a curiosity—a marvelous one, to be sure, but still an oddity outside the realm of architecture. They could not have foreseen that Paxton's design, solid iron framework clothed in a glass skin, would pave the way for 20th-century architecture. In fact, Paxton had taken a giant step in demonstrating that as long as a building's skeleton held firm, its skin could be light and non-load-bearing. Several intermediary steps would be required before this principle could be translated into today's architecture.

Alexandre Gustave Eiffel. Eiffel Tower, Paris. 1889. Iron, height 934′.
Page 296

Another bold experiment in iron construction came a few decades later just across the English Channel, in France, and involved a plan that many considered to be foolhardy, if not downinsane. Gustave Eiffel, a French engineer, proposed to build in the center of Paris a skeleton iron tower, nearly a thousand feet tall, to act as a centerpiece for the Paris World's Fair of 1889. Nothing of the sort had ever been suggested, much less built. In spite of loud protests, the Eiffel Tower (13.22) was constructed, at a cost of about a million dollars—an unheard-of sum for those times. It rises on four arched columns, which curve inward until they meet in a single tower thrusting up boldly above the cityscape of Paris. (The writer Guy de Maupassant claimed that he lunched in a restaurant on the tower as often as possible, because “it's the only place in Paris where I don't have to see it.”1)

The importance of this singular, remarkable structure for the future of architecture rests on the fact that it was a skeleton that proudly showed itself without benefit of any cosmetic embellishment. No marble, no glass, no tiles, no skin of any kind—just the clean lines drawn in an industrial-age product. Two concepts emerged from this daring construction. First, metal in and of itself can make beautiful architecture. Second, metal can provide a solid framework for a very large structure, self-sustaining and permanent. Today the Eiffel Tower is the ultimate symbol of Paris, and no tourist would pass up a visit. From folly to landmark in a century—such is the course of innovative architecture.

Iron for structural members was not the only breakthrough of the mid–19th century. The Industrial Revolution also introduced a new construction material that was much humbler but equally significant in its implications for architecture: the nail. And for want of that simple little nail, most of the houses we live in today could not have been built.

Balloon-Frame Construction

So far in this chapter, the illustrations have concentrated on grand and public buildings—churches, temples, monuments. These are the glories of architecture, the buildings we admire and travel great distances to see. We should not forget, however, that the overwhelming majority of structures in the world have been houses for people to live in, or domestic architecture.

Until the mid–19th century, houses were of shell construction. They were made of brick or stone (and, in warmer climates, of such materials as reeds and bamboo) with load-bearing construction, or else they were post-and-lintel structures in which heavy timbers were assembled by complicated notching and joinery, sometimes with wooden pegs. Nails, if any, had to be fabricated by hand and were very expensive.

About 1833, in Chicago, the technique of balloon-frame construction was introduced. Balloon-frame construction is a true skeleton-and-skin method. It developed from two innovations: improved methods for milling lumber and mass-produced nails. In this system, the builder first erects a framework or skeleton by nailing together sturdy but lightweight boards (the familiar 2-by-4 “stud”), then adds a roof and sheathes the walls in clapboard, shingles, stucco, or whatever the homeowner wishes. Glass for windows can be used lavishly, as long as it does not interrupt the underlying wood structure, since the sheathing plays little part in holding the building together.

When houses of this type were introduced, the term “balloon framing” was meant to be sarcastic. Skeptics thought the buildings would soon fall down, or burst just like balloons. But some of the earliest balloon-frame houses stand firm today, and this method is still the most popular for new house construction in Western countries.

The balloon frame, of course, has its limitations. Wood beams 2 by 4 inches thick cannot support a skyscraper ten or fifty stories high, and that was the very sort of building architects had begun to dream of late in the 19th century. For such soaring ambitions, a new material was needed, and it was found. The material was steel.

Page 297
Steel-Frame Construction

Although multistory buildings have been with us since the Roman Empire, the development of the skyscraper, as we know it, required two late-19th-century innovations: the elevator and steel-frame construction. Steel-frame construction, like balloon framing, is a true skeleton-and-skin arrangement. Rather than piling floor upon floor, with each of the lower stories supporting those above it, the builders first erect a steel “cage” that is capable of sustaining the entire weight of the building; then they apply a skin of some other material. But people could hardly be expected to walk all the way to the of a ten-story building, to say nothing of a skyscraper. Hence, another invention made its appearance, the elevator.

What many consider to be the first genuinely modern building was designed by Louis Sullivan and built between 1890 and 1891 in St. Louis. Known as the Wainwright Building (13.23), it employed a steel framework sheathed in masonry. Other architects had experimented with steel support but had carefully covered their structures in heavy stone so as to reflect traditional architectural forms and make the construction seem reliably sturdy. Centuries of precedent had prepared the public to expect bigness to go hand in hand with heaviness. Sullivan broke new ground by making his sheathing light, letting the skin of his building echo, even celebrate, the steel framing underneath. Regular bays of windows on the seven office floors are separated by strong vertical lines, and the four corners of the building are emphasized by vertical piers. The Wainwright Building's message is subtle, but we cannot mistake it: The nation had stopped growing outward and started growing up.

Louis Sullivan. Wainwright Building, St. Louis, 1890–91.
Page 298

Sullivan's design looks forward to the 20th century, but it nevertheless clings to certain architectural details rooted in classical history, most notably the heavy cornice (the projecting roof ornament) that terminates upward movement at the of the building. In a very few decades, even those backward glances into the architectural past would become rare.

Toward the middle of the 20th century, skyscrapers began to take over the downtown areas of major cities, and city planners had to grapple with unprecedented problems. How high is too high? How much airspace should a building consume? What provision, if any, should be made to prevent tall buildings from completely blocking out the sunlight from the streets below? In New York and certain other cities, ordinances were passed that resulted in a number of look-alike and architecturally undistinguished buildings. The laws required that if a building filled the ground space of a city block right up to the sidewalk, it could rise for only a certain number of feet or stories before being “stepped back,” or narrowed; then it could rise for only a specified number of additional feet before being stepped back again. The resultant structures came to be known as “wedding-cake” buildings. A few architects, however, found more creative ways of meeting the airspace requirement. Those working in the International style designed some of the most admired American skyscrapers during the 1950s and 1960s. International style architecture emphasized clean lines, geometric (usually rectilinear) form, and an avoidance of superficial decoration. The “bones” of a building were supposed to show and to be the only ornament necessary. A classic example of this pure style is Lever House.

Page 299

Lever House in New York (13.24), designed by the architectural firm of Skidmore, Owings, and Merrill and built in 1952, was heralded as a breath of fresh air in the smog of look-alike structures. Its sleek understated form was widely copied but never equaled. Lever House might be compared to two shimmering glass dominoes, one resting horizontally on freestanding supports, the other balanced upand off-center on the first. At a time when most architects of office buildings strove to fill every square inch of airspace to which they were entitled—both vertically and horizontally—the elegant Lever House drew back and raised its slender rectangle aloof from its neighbors, surrounded by free space. Even its base does not rest on the ground but rides on thin supports to allow for open plazas and passageways beneath the building. Practically no other system of construction except steel frame could have made possible this graceful form.

Gordon Bunshaft of Skidmore, Owings, and Merrill. Lever House, New York. 1952.

Also made feasible by steel, suspension is the structural method we associate primarily with bridges, although it has been employed for some buildings as well. The concept of suspension was developed for bridges late in the 19th century. In essence, the weight of the structure is suspended from steel cables supported on vertical pylons, driven into the ground. A long bridge, such as the Golden Gate Bridge in San Francisco (13.25), may have only two sets of pylons planted in the riverbed, but the steel cables suspended under tension from their towers are strong enough to support a span between them almost four-fifths of a mile long. With their long sweeping curves and slender lines, suspension structures are among the most graceful in architecture.

Golden Gate Bridge, San Francisco. 1937. Joseph B. Strauss, chief engineer; O. H. Ammann, Charles Derleth, Jr., and Leon S. Moisseiff, consulting engineers; Irving F. Morrow, consulting architect.
Page 300
Reinforced Concrete

Concrete is an old material that was known and used by the Romans. A mixture of cement, gravel, and water, concrete can be poured, will assume the shape of any mold, and then will set to hardness. Its major problem is that it tends to be brittle and has low tensile strength. This problem is often observed in the thin concrete slabs used for sidewalks and patios, which may crack and split apart as a result of weight and weather. Late in the 19th century, however, a method was developed for reinforcing concrete forms by imbedding iron rods inside the concrete before it hardened. The iron contributes tensile strength, while the concrete provides shape and surface. Reinforced concrete, also known as ferroconcrete, has been used in a wide variety of structures, often in those with free-form, organic shapes. Although it may seem at first to be a skeleton-and-skin construction, ferroconcrete actually works more like a shell, because the iron rods (or sometimes a steel mesh) and concrete are bonded permanently and can form structures that are self-sustaining, even when very thin.

A special kind of ferroconcrete construction—precast reinforced sections—was used to create the soaring shell-like forms of the Sydney Opera House in Australia (13.26). The Opera House, which is really an all-round entertainment complex, is almost as famous for its construction difficulties as it is for its extraordinary design. So daring was its concept that the necessary technology virtually had to be invented as the project went along. Planned as a symbol of the great port city in whose harbor it stands, the Opera House gives the impression of a wonderful clipper ship at full sail. Three sets of pointed shells, oriented in different directions, turn the building into a giant sculpture in which walls and roof are one. Reinforced concrete is the sort of material that allows the builder to experiment and try new techniques, that allows the architect to dream impossible dreams.

Joern Utzon. Sydney Opera House, Australia. 1959–72. Reinforced concrete, height of highest shell 200′.
Page 301
Geodesic Domes

Of all the structural systems, probably the only one that can be attributed to a single individual is the geodesic dome, which was developed by the American architectural engineer R. Buckminster Fuller. Fuller's dome is essentially a bubble, formed by a network of metal rods arranged in triangles and further organized into tetrahedrons. (A tetrahedron is a three-dimensional geometric figure having four faces.) This metal framework can be sheathed in any of several lightweight materials, including wood, glass, and plastic.

The geodesic dome offers a combination of advantages never before available in architecture. Although very light in weight in relation to size, it is amazingly strong, because its structure rests on a mathematically sophisticated use of the triangle. Because it requires no interior support, all the space encompassed by the dome can be used with total freedom. A geodesic dome can be built in any size. In theory, at least, a structurally sound geodesic dome 2 miles across could be built, although nothing of that scope has ever been attempted. Perhaps most important for modern building techniques, Fuller's dome is based on a modular system of construction. Individual segments—modules—can be prefabricated to allow for extremely quick assembly of even a large dome. And finally, because of the flexibility in choice of sheathing materials, there are virtually endless options for climate and light control.

Fuller patented the geodesic dome in 1947, but it was not until twenty years later, when his design served as the U.S. Pavilion at the Montreal World's Fair, that the public's attention was awakened to its possibilities. The dome at Expo 67 (13.27) astonished the architectural world and fairgoers alike. It was 250 feet in diameter (about the size of a football field rounded off) and, being sheathed in translucent material, lighted up the sky at night like a giant spaceship set down on earth.

R. Buckminster Fuller. U.S. Pavilion, Expo 67, Montreal. 1967. Geodesic dome, diameter 250′.
Page 302

After Expo 67, some people predicted that before long all houses and public buildings would be geodesic domes. That dream has faded considerably, but Fuller's dome has proved exceptionally well suited for government and scientific operations in arctic climates. To build a habitable structure in the freezing wastes of Antarctica, for example, requires a lightweight material that can be shipped and assembled easily, great strength to withstand below-freezing temperatures and high winds, and control of the interior environment. The geodesic dome meets all those requirements.

In this brief survey of the major structural systems, we have seen that the form of architecture and its method of construction are determined largely by the materials available. Wood readily lends itself to post-and-lintel construction and balloon-frame construction; stone works for post-and-lintel also, as well as for the arch and the dome; metal allows for steel-frame construction, suspension, or reinforced concrete; and so on. But there is another factor—often a more important one—that affects the shape of architecture, and that is the purpose a building will serve.