Brunelleschis Dome (2 page)

Yet by 1418 what was by far the grandest building project in Florence had still to be completed. A replacement for the ancient and dilapidated church of Santa Reparata, the new cathedral of Santa Maria del Fiore was intended to be one of the largest in Christendom. Entire forests had been requisitioned to provide timber for it, and huge slabs of marble were being transported along the Arno on flotillas of boats. From the outset its construction had as much to do with civic pride as religious faith: the cathedral was to be built, the Commune of Florence had stipulated, with the greatest lavishness and magnificence possible, and once completed it was to be “a more beautiful and honourable temple than any in any other part of Tuscany.” But it was clear that the builders faced major obstacles, and the closer the cathedral came to completion, the more difficult their task would become.

The way forward should have been clear enough. For the past fifty years the south aisle of the unfinished cathedral had housed a thirty-foot-long scale model of the structure, in effect an artist’s impression of what the cathedral should look like once finished. The problem was that the model included an enormous dome — a dome that, if built, would be the highest and widest vault ever raised. And for fifty years it had been obvious that no one in Florence — or anywhere in Italy, for that matter — had any clear idea how to construct it. The unbuilt dome of Santa Maria del Fiore had therefore become the greatest architectural puzzle of the age. Many experts considered its erection an impossible feat. Even the original planners of the dome had been unable to advise how their project might be completed: they merely expressed a touching faith that at some point in the future God might provide a solution, and architects with a more advanced knowledge would be found.

A section drawing of Santa Maria del Fiore, by Giovanni Battista Nelli.

A ground plan of the cathedral shows the three tribunes, with their chapels.

The foundation stone for the new cathedral had been laid in 1296. The designer and original architect was a master mason named Arnolfo di Cambio, the builder of both the Palazzo Vecchio and the city’s massive new fortifications. Although Arnolfo died soon after construction began, the masons forged on, and over the next few decades a whole section of Florence was razed to make way for the new building. Santa Reparata and another ancient church, San Michele Visdomini, were both demolished, and the inhabitants of the surrounding district were displaced from their homes. Not only the living were evicted: in order to open a piazza in front of the church, the bones of long-dead Florentines were exhumed from their graves surrounding the Baptistery of San Giovanni, which stood a few feet to the west of the building site. In 1339 one of the streets south of the cathedral, the Corso degli Adamari (now the Via dei Calzaiuoli) was lowered so that the cathedral’s height should appear even more impressive to anyone approaching from that direction.

But as Santa Maria del Fiore grew steadily larger, Florence was shrinking. In the autumn of 1347 the Genoese fleet returned to Italy, carrying in its holds not only spices from India but also the Asian black rat, carrier of the Black Death. As much as four-fifths of the population of Florence were to die over the next twelve months, so depopulating the city that Tartar and Circassian slaves were imported to ease the labor shortages. As late as 1355, therefore, nothing existed of the cathedral except for the facade and the walls of the nave. The interior of the church lay open to the elements, like a ruin, and the foundations for the unbuilt east end had been exposed for so long that one of the streets east of the cathedral was known as Lungo di Fondamenti, or “Along the Foundations.”

Over the next decade, however, as the city gradually recovered, work on the cathedral accelerated, and by 1366 the nave had been vaulted and the east end of the church, which included the dome, was ready to be planned. Arnolfo di Cambio had undoubtedly envisioned a dome for the church, but there is no surviving evidence of his original design: sometime in the fourteenth century his model of the cathedral collapsed under its own weight — an ominous sign — and was subsequently lost or demolished. But excavations during the 1970s uncovered the foundations for a dome that was intended to have a span of 62
braccia
, or 119 feet (a Florentine
braccia
being 23 inches, roughly the length of a man’s arm).
¹
With this diameter the cupola of Santa Maria del Fiore would have exceeded by some 12 feet the span of the dome of the world’s most spectacular church, Santa Sophia in Constantinople, which had been built 900 years earlier by the emperor Justinian.

Since the 1330s responsibility for building and funding the cathedral had been in the hands of Florence’s largest, wealthiest and most powerful guild, the Wool Merchants, who administered the Opera del Duomo. None of the wardens running the Opera knew the first thing about building churches: their business was wool, not architecture. It therefore fell to them to appoint someone who did understand the craft, an architect-in-chief, or
capomaestro
, who would create the models and designs for the cathedral and also deal with the masons and other builders involved in the actual construction. In 1366, as planning reached its crucial stage, the
capomaestro
of Santa Maria del Fiore was a man named Giovanni di Lapo Ghini. At the request of the Opera, Giovanni began building a model for the cathedral’s dome. But the wardens also ordered a second model from a group of artists and masons led by another master mason, Neri di Fioravanti.
²
The fate of Santa Maria del Fiore was about to undergo a radical change.

Competition between architects was an old and honored custom. Patrons had been making architects compete against one another for their commissions since at least 448
B.C.
, when the Council of Athens held a public competition for the war memorial it planned to build on the Acropolis. Under these circumstances, it was normal practice for architects to produce models as a means of convincing patrons or panels of judges of the virtues of their particular designs. Made from wood, stone, brick, or even clay or wax, such models allowed the patrons to visualize the dimensions and decorations of the end product much more easily than would a diagram executed on parchment. They were often large and highly detailed — so large, in fact, that in many cases patrons could walk inside to inspect the interior. The brick-and-plaster model for San Petronio in Bologna, for example, built in 1390, was 59 feet long and, therefore, a good deal larger than most houses.

Giovanni di Lapo Ghini set about building a model that was fairly traditional in style. He planned a typically Gothic structure with thin walls, tall windows, and, to support the dome, external buttresses of the sort adorning so many of the churches built in France during the previous century. Buttresses were one of the prime structural features of Gothic architecture: by accommodating the thrust of the vaults transferred to them from strategic points, they allowed for walls pierced by a multitude of windows to rise to spectacular heights, filling the church with heavenly light — the aspiration of all Gothic builders.

Neri di Fioravanti and his group rejected the external supports proposed by Giovanni di Lapo Ghini, however, and offered a different approach to the structure of the dome. Flying buttresses were rare in Italy, where architects regarded them as ugly and awkward makeshifts.
³
But Neri’s reasons for rejecting them were probably political as much as aesthetic or structural, in that they smacked of the architecture of Florence’s traditional enemies: Germany, France, and Milan. How the German barbarians, the Goths, had covered Europe with their clumsy and disproportionate edifices would later become a popular theme with writers of the Italian Renaissance.

But if no flying buttresses were to be built, how was the dome to be supported? Neri di Fioravanti, the principal master mason in Florence, had extensive experience in vaulting, the most dangerous and difficult architectural maneuver. He was the man responsible for erecting the enormous 60-foot-wide vaults over the great hall of the Bargello as well as the arches of the new Ponte Vecchio after the old bridge was swept away by a flood in 1333. But his plan for the dome of Santa Maria del Fiore was far more ambitious and largely untested: he believed the dome could be prevented from buckling under its own weight not by means of external buttresses but by the incorporation of a series of stone or wooden chains that would run round the circumference, encircling the dome at the points of possible rupture in the same way that an iron hoop contains the staves of a barrel. All of the lines of stress would therefore be absorbed by the structure itself without being channeled to the ground by means of external buttresses. Unlike buttresses, moreover, these circumferential rings, buried in the dome’s masonry, would be invisible. And it was this vision of a massive dome that seemed to rise heavenward without any visible means of support that for the next half century would both inspire and frustrate everyone involved with the project.

The wardens in the Opera del Duomo did not decide between the two models without a good deal of debate. At first Neri and his group seemed to win the day, but Giovanni succeeded in raising questions about the stability of their design. His doubts illustrate a fear that haunted architects in the Middle Ages. Today a patron who hires an engineer takes it for granted that the end product will stand, even through earthquakes and hurricanes. But in the Middle Ages and the Renaissance, before the science of statics was developed, a patron enjoyed no such assurance, and it was not uncommon for buildings to fall down soon after completion, or even during the building process itself. The bell towers in both Pisa and Bologna began to lean while still under construction because of subsidence in the underlying soil, while the vaults in the cathedrals at both Beauvais and Troyes collapsed a relatively short time after being raised. The superstitious attributed these failures to supernatural causes, but to the more knowledgeable the real culprits were the architects and builders who had made fundamental (though imperfectly understood) errors in design.

In the end Giovanni’s concerns led the wardens to stipulate that, although Neri’s model would be adopted, the pillars that supported the dome should be enlarged. But enlarging the pillars would create perhaps even greater problems. Their dimensions were directly related to those of the octagonal tribune, whose perimeter they would form. The foundations for an octagon of 62
braccia
had already been begun: would this groundwork have to be undone? Even more serious, the diameter of the tribune could not be enlarged without a corresponding increase in the span of the cupola. Was it possible to build a dome with a span even larger than 62
braccia
, still without the use of any visible supports?

These questions were addressed at the meeting in August 1367, in which the wardens opted for a dome that would be 10
braccia
wider than the one previously planned. Three months later, in keeping with Florentine democracy — and also, perhaps, with a desire on the part of the wardens to spread the responsibility as widely as possible — the plan was endorsed by a referendum of Florence’s citizens.

The decision to adopt Neri di Fioravanti’s design represents a remarkable leap of faith. No dome approaching this span had been built since antiquity, and with a mean diameter of 143 feet and 6 inches it would exceed that of even the Roman Pantheon, which for over a thousand years had been the world’s largest dome by far. And the cupola of Santa Maria del Fiore would not only be the widest vault ever built: it would also be the highest. The walls of the cathedral were already 140 feet high, above which a tambour (or drum) on which the dome was to rest would rise another 30 feet. The purpose of this tambour was to elevate the dome — to serve, in effect, as a pedestal, raising the dome even higher above the city.
⁴
Vaulting for the cupola would therefore begin at an incredible height of 170 feet, much higher than any of the Gothic vaults built in France during the thirteenth century. Indeed, the highest Gothic vault ever constructed, in the Cathedral of Saint-Pierre at Beauvais, began at just under 126 feet and rose to a maximum height of 157 feet, still a good 13 feet below where the vaulting for the dome of Santa Maria del Fiore was to begin. And the choir at Beauvais spanned only 51 feet, in contrast to the 143 feet proposed for the cupola in Florence. The fact that the main vaults of the choir in Saint-Pierre had collapsed in 1284, little more than a decade after completion, cannot have eased the minds of the skeptics, especially since the architects at Beauvais had made use of both iron tie rods and flying buttresses, the expedients so boldly rejected by the committee of artists and masons.