2
Brunelleschi’s Dome, Verrocchio’s Palla, and Leonardo’s Eye
The palla commission would keep Verrocchio’s bottega busy for almost three years. It came from none other than the Operai dell’Opera del Duomo, the citizens in charge of the construction and adornment of the Florence cathedral. On the face of it, the commission was not too demanding. It did not involve creating statues displaying subtle body motions, expressive faces, or bold drapery folds, nor did it entail the creation of extravagant designs for luxury objects. All Andrea was asked to make was a large metal ball.
But the commission was not routine metallurgic work, either. The palla marked the end of a centuries-long construction process: the new cathedral had been designed around 1296 to replace a crumbling, undersized church that no longer met the needs of a growing population. After approximately one hundred years, the main body of the church was complete. In 1418, the architect Filippo Brunelleschi had won the competition to build a dome covering the area above the altar: it would be a majestic octagonal structure of redbrick and white marble ribs. He also designed efficient lifting machines to convey materials to where they were needed; as a result, the dome was completed in less than three decades, record speed for the time. Leon Battista Alberti, who observed the dome in 1435, right around the time it was finished, marveled at this “enormous construction towering above the skies, vast enough to cover the entire Tuscan population with its shadow.” To this very day the dome defines the city’s skyline.
Brunelleschi’s design called for the dome’s white ribs to converge on an elegant marble “lantern” at the top, which would be crowned by a golden globe surmounted by a cross. By 1461 the lantern had been completed: it was a slender structure with eight tall windows, each about thirty feet high, and massive buttresses with door-like openings that allowed for a continuous passageway around the base of the enormous lantern. This is the walk around the dome visitors still enjoy today. All that remained to complete the cathedral was the golden ball. The Operai left in place the scaffolding and the lifting machine, the castello, so that it could be reused to lift the palla. They expected to make quick work of the job.
But six years passed and nothing happened. The situation was becoming increasingly absurd. The cathedral had been built, but it could not be declared finished because the palla was not installed. To add insult to injury, the beautiful lantern was still invisible, as it remained encased in the massive scaffolding of beams, stairways, and platforms that supported Brunelleschi’s castello, the lifting machine the architect designed to build the lantern.
The failure to install the palla stemmed from the challenge of its size. The ball had to be seen from the street, at a height of over 350 feet, and against an open sky. It was known that the high elevation and the open backdrop would make the orb look smaller than it actually was; to counter this illusion, the palla would have to be enormous. The problem was that such a large ball was too big to pass through the lantern’s narrow windows, which offered the only available passageway to reach the final location. This meant that either the palla had to be too small for proper viewing from below, or it had to be lifted in pieces and assembled on the lantern’s terrace, where some sort of soldering device would have to be built. And since soldering with fire, at a height of 350 feet, on a terrace surrounded by wood scaffolding, was quite dangerous, the building of the palla would be no trivial affair.
Making matters more complex, the Operai wanted a golden palla that would mesmerize viewers as it glittered under the Tuscan sun. They knew it would be too costly to make it of actual gold, and settled for a ball manufactured from a different metal—which metal exactly was up for debate—that would be gilded using melted gold florins. The most effective gilding technique available at the time was mercury gilding, which was done by applying an amalgam of gold and mercury to a metal surface and heating it so that the mercury evaporated and left a thin coating of gold behind. The process involved the use of highly toxic substances (this is the reason it is hardly used today), and often artists had to repeat it several times to achieve the desired coating. It also had to be applied to a finished object, and so the palla would have to be gilded atop the dome as well.
By 1468, all that the patrons had accomplished was to commission two craftsmen to make a bronze base on which the palla would rest. Unable to decide what to do next, they did what Florentines did whenever they faced challenging public works: they called for a public competition. Out of such competitions had emerged great works such as Brunelleschi’s dome itself, as well as Lorenzo Ghiberti’s set of doors for the Baptistery, the octagonal building facing the cathedral where Florentine children were baptized.
The Operai received two proposals. One came from the two craftsmen who had made the base. They proposed to cast the palla in bronze in a single piece, gild it, and attach it directly to the base. The other proposal came from Andrea del Verrocchio. Andrea planned to build a hollow palla to reduce its overall weight, supporting it with a sturdy internal armature that could be firmly anchored to the dome. He proposed to form the exterior of the palla from metal strips hammered into shape on a rock; the strips would then be soldered to one another with burning mirrors.
Andrea was probably the most talented bronze caster in Florence—possibly in all of Italy. More important, as Vasari reported, he excelled in “the sciences, particularly geometry.” In fact, he came up with his design for the palla thanks to such expertise. One needed to know spherical geometry in order to cut the outer metal strips in triangles that would perfectly match one another and form a sphere. And one needed knowledge of optics to solder these metal parts with burning mirrors.
Andrea, however, was not awarded the contract for the palla.
The Operai ordered that “it should be made by casting […] in a single piece” and that “under no circumstances should it be made by hammering.” Their deliberation was an explicit rebuttal of Andrea’s proposal. Instead, they contracted the craftsmen who had cast the base to fashion a palla out of a single piece.
These craftsmen set to work. By early August they had cast the ball in an area near the cathedral that the Operai had reserved for them (we must remember that the streets around the cathedral were not paved at that time). They were ready to extract it from its mold. A group of dignitaries gathered to witness the moment. To this day, we do not know what happened, but we do know that something went so terribly wrong that the Operai went back to Verrocchio. A month later, Andrea was at work on his original plan, a hollow ball that would be four Florentine braccia (or about eight feet) in diameter, made of metal strips hammered into shape on a rock and soldered to one another. The plan required the full array of Verrocchio’s inventiveness and technical expertise.
The first step was to find the right material.
Since the Middle Ages, artists had known that if you wanted to beat a metal to a fine edge without it cracking, the best option was unalloyed copper—copper that was not mixed with other metals. They also knew that this unalloyed copper was perfect for gilding, and that the gilding would be more durable if higher-quality copper was used. Cyprus and Germany produced the best copper, and Venice was the major commercial center for its distribution. Years earlier, Lorenzo Ghiberti had taken a trip to the city in search of proper copper for his Gates of Paradise. In June 1469, Andrea made the same trip for the very same reason. When he found copper of the quality he desired, he sent instructions to the Medici bank in Florence to transfer cash to their Venice branch to pay for eight sheets of this fine metal. He returned to Florence and waited for the copper shipments, which arrived in installments between July and October. In the meantime, he designed the specific tools he needed for the palla.
The first tool was rather simple, but it nonetheless required knowledge of spherical geometry if the palla was to be executed to perfection. It was an exact, one-to-one copy of the palla made of stone, which Verrocchio cut into two exact halves: one half was for the Operai, the other for him. He needed these two exact halves for two reasons. One reason was diplomatic. The construction of the Florence cathedral had been marred by bitter disputes between artists and patrons about undelivered work, unsuitable materials, poor building techniques, and even fraud. The best way to ensure that no issue arose with the palla was to provide each party with an identical “model” to assess the work. The Operai kept their half in their offices for centuries, and Andrea saved his own as well: twenty years later, when his workshop was inventoried after his death, it was still there. The other reason for dividing the sphere was of a technical nature. His design required hammering copper sheets into shape, and it was much easier to accomplish this by using a half sphere placed on its flat side than a sphere rolling around.
But what really required Andrea’s technical virtuosity was another aspect of the palla. The eight copper sheets that he had so keenly purchased in Venice had to be cut into triangles and transformed into curved shapes by hammering. The difficulty was to calculate how exactly to cut these flat triangles so that they could eventually be joined together to form a perfect sphere. We do not know how Andrea proceeded, but spherical geometry would have allowed him to create a precise diagram.
Invented by the Greeks, transmitted to the Islamic world, and from there to the Latin West, spherical geometry was concerned with non-Euclidean spaces—that is, spaces that do not have straight and parallel lines but only curved ones. Traditionally, it was the domain of natural philosophers and astronomers, but in the Renaissance, merchants and sailors had become interested in it because it could help them calculate the routes of long-distance voyages around the globe and estimate the time required. Abacus schools had registered this new interest and added spherical geometry to their curricula. Andrea might have picked up some spherical geometry in such a school. But perhaps he also had the chance to discuss spherical geometry with an expert who was one of the most prominent figures of Renaissance Florence and whose name appears in Leonardo’s notes: Paolo dal Pozzo Toscanelli (1397–1482). While Andrea was working on the palla, Toscanelli was building an astronomical instrument inside the Florence cathedral.
Toscanelli was a doctor, an astronomer, and a member of a merchant family. His reputation in spherical geometry was such that he was called on to mediate disputes between the two most famous astronomers of the Renaissance, Johannes Regiomontanus and Nicolas of Cusa, who quarreled bitterly over whether it was possible to transform a circle into a square with the same area, using only simple geometry. (Years later, Leonardo himself became interested in this topic and claimed that he had found a way to solve it; he was wrong, however, and we know today that a circle cannot be squared—indeed, “to square a circle” has become an idiomatic expression for something that is impossible.) Today, Toscanelli is most famous for a map, now lost, that he made in order to demonstrate that the ocean that supposedly lay between Europe and Asia (what would later be called the Atlantic) was considerably smaller than previously believed. Ships, he argued, could sail across this smaller ocean to reach the Indies. In 1474, he wrote a letter to a Portuguese cardinal explaining his hypothesis. Eighteen years later, Christopher Columbus studied the letter and the map before his trip to China, a land he never reached as another continent stood in between.
In the early 1470s, however, Toscanelli was best known for his study of optics and for the astronomical instrument he was designing. The instrument used the Florence cathedral itself as one of its components: when sunlight passed through a hole pierced in the lantern, a beam of light was projected onto the cathedral’s floor, where a mosaic, now lost, could be used to track its movement. The instrument was a “meridian line,” and it was intended to measure how much time it takes the sun to return to the same position in the sky—for instance, the time it takes to move from vernal equinox to vernal equinox, or from summer solstice to summer solstice.
Did Verrocchio and Toscanelli talk about spherical geometry when they met on the lantern’s terrace, or when they shared Brunelleschi’s castello to lift objects they made? We do not know. What we know is that Leonardo recorded Toscanelli’s name in one of his early folios connected to the palla and even owned some of Toscanelli’s notes in Toscanelli’s own hand. Those notes, interestingly enough, dealt with burning mirrors, the tools that were used to assemble the palla.
However Andrea learned spherical geometry, he was a quick study. In less than a year—eleven months to be precise—he built the palla. He cut the copper sheets into triangles. His assistants hammered the copper triangles with wooden mallets over the stone ball and helped lift the pieces from the church’s floor to the platform, where the soldering took place.
As a goldsmith, Andrea was familiar with the soldering tools that were commonly used to assemble sculptures in metal or to make mechanical parts—roundels, screws, pulleys. Considering his bottega’s large output of metal works, he must have owned a few soldering tools, although none were recorded in the inventories of his workshop.
But we do know that Andrea soldered the palla with burning mirrors. And we know this because Leonardo, who was an apprentice in the workshop at the time of the palla construction, wrote about it forty years later.
By then, Leonardo was in his early sixties and a famous artist. He had lost the beauty of his youth but still had an impressive presence; he resembled an ancient philosopher. He was living in Rome, where he designed weapons for the papal army. Legend has it that the Greek mathematician and inventor Archimedes of Syracuse had used a burning mirror to set the Roman fleet on fire at Syracuse in 212 B.C.E. Leonardo filled a small notebook with calculations, mathematical formulas, and notes on the workings of this theoretical weapon. He planned to create a military-grade burning mirror of his own.
In the midst of this frantic work, he jotted a side note on the margin of the page sitting in front of him—folio eighty-four of the notebook he was compiling, which is known today as Manuscript G. He wrote:
Remember the welds that were used to solder the palla of Santa Maria del Fiore. [Make them] of copper hammered over stone as for the triangles of that palla.
The note may seem like a casual recollection of technical details, one of those mental detours the artist took when his mind wandered away from the matter at hand and toward original machines, old philosophical problems, new paintings. But this was much more than that—it was a memory from his youth, a rare peek into his formative years, a glimpse of things that mattered to him when he was an apprentice in Andrea’s workshop. Leonardo did not specify what role he played in making the palla, nor how his master designed the burning mirrors. But the event stuck with him.
In principle, burning mirrors function like any mirror: beams of light that hit their surface are reflected back. But because the surface of burning mirrors is concave, the basic optical rule of reflection acquires an added level of complexity: many of the reflected rays that bounce off the surface do not disperse in the air, as they do in a flat mirror, but instead hit other points of the concave surface, generating additional reflected rays that go off in different directions, which in turn create more reflected rays. Burning mirrors are a special subset of concave mirror: their curvature is designed in such a way that rays that reflect off their surface do not disperse in random directions but rather converge on a single point—the focal point—and so produce intense heat when the source of light is the sun. To use a modern geometrical term, burning mirrors are parabolic mirrors.
We do not know how, exactly, Andrea made his burning mirror for the palla. We do not even know what material he used—metal or glass? We can speculate that the burning mirror was larger than those he used normally in his workshop because of the palla’s size. It would also have to be very accurate. Where, then, would Andrea have acquired the knowledge needed to construct it?
This is not an easy question to answer. When optical manuals were first composed in ancient times, burning mirrors were not discussed. Because burning mirrors create heat, earlier authors placed them in a separate category from the science of vision. This distinction held for centuries until the professor Biagio Pelacani da Parma (1365–1416) arrived on the scene. His book on optics, Questions on Perspective (Questiones [sic] super perspectiva communi), was based on his university lectures and was written in Latin. It was also generously illustrated. But the peculiarity of Pelacani’s book was that it linked university teaching with the work of craftsmen and artists as the author discussed various topics, including burning mirrors and “apparitions,” as he called them, or visual illusions, which he claimed to have experienced personally. Like burning mirrors, these apparitions were the result of the geometry of concave mirrors: they were images of earthly things reflected into the air by clouds that acted like gigantic reflectors. Pelacani’s book and his ideas about applied optics enjoyed considerable success in the following decades, especially in Florence, where copies of his book from the years 1428, 1445, and 1469 survive. It is doubtful that people like Andrea del Verrocchio would have been able to master Pelacani’s Latin. But they may have found certain vernacular books that had been modeled after Pelacani’s book and that made its contents more approachable.
The most important of these vernacular books is now kept in Florence, its author unknown.
The text was a simplified version of Pelacani’s Questions. It was beautifully illustrated. It was addressed to an artist in training and explained the science behind many objects produced in Florentine workshops: reading glasses, gigantic lanterns, fancy optical devices meant as entertainment for dinner parties, and of course burning mirrors, which the author recommended be built in metal. The text also stressed the importance of the air as the medium within which vision occurs—and the medium responsible for many optical illusions. In fact, it offered the example of how towers, partly obscured behind a wall, appear to be farther away than they actually are because of the medium of air—an example Leonardo would later use himself.
We do not know whether Andrea saw this text, or one similar to it, but it is significant that advanced spherical geometry and the optics of burning mirrors could be studied in vernacular texts in Florence during the years of the palla operation. If any of these vernacular texts circulated in Andrea’s workshop, chances are that Leonardo got a glimpse of them as well.
Leonardo, too, was fascinated by burning mirrors, and learned “how to make a concave sphere [spera in cavo] that makes fires.”
Among his early folios are notes and sketches that refer to the workshops of glassmakers who manufactured burning mirrors from glass. These workshops were in the hills around Florence. Their furnaces could be dangerous, reaching temperatures of over one thousand degrees and emitting highly toxic fumes. Florentine law forbade such furnaces from being built within the city’s walls.
One particular folio of about ten by eight inches leaves little doubt that Leonardo had direct knowledge of every stage in the manufacture of burning mirrors made of glass. This folio is carefully executed, each drawing beautifully spaced on the page and combined with carefully worded notes. It looks like a reworked draft of his field notes, and serves as a one-page manual on “how to make a concave sphere that makes fire.” Leonardo may have written it either as a sort of personal pro memoria or as a means of teaching others.
This folio contains numerous sketches. There is a furnace, which he represented from different points of view to make visible its structural components. In a plan view, he visualized the large foundations required to feed a fire that would be intense enough to melt glass. In side views, he showed the cooling system made of buckets of water on a belt and of canals for the passage of air. In a small sketch, he illustrated the furnace in full operation with a burning mirror baking inside. In other sketches, he showed polishing machines used to smooth the mirror’s surface. This is how he described the process of coating the supportive backing of a mirror with glass:
This concave mirror should be made of earth-ware, then it should be baked and returned to the above machine [a polishing machine]; it should then be covered with a glass paste and put in the furnace upside down above ashes.
There is also a diagram with technical instructions for how to design the curvature of a polishing stone. The note reads like a primer on Euclidian geometry, except for a charming aside about Florentine bakeries and their breads:
If you wish to make a concave sphere that makes fires when it is turned toward sunrays […] draw first a pyramid like the one represented here […] And know that the pyramid should be round at its apex, like that of a sugar bread.
Such comparisons between high science and everyday life were also typical of abacus books, which often mixed references to daily routines into their calculations to make it easier for schoolchildren to retain mathematical rules for transactions and measurements. It became a lifelong habit of Leonardo’s to make such comparisons in his own writing.
By the summer of 1470, Andrea and his assistants were done with the soldering.
They moved on to gilding, the final stage of creating the palla.
In August, Andrea received a large number of gold florins from the Operai, who also sent a representative to monitor the gold’s proper use. Under close supervision, Andrea ground the gold, mixed it with mercury, made an amalgam of the two metals, and applied it to the globe. At least three goldsmiths helped him. We do not know how he evaporated the mercury high atop the dome or how he addressed the problem of toxicity. But the gilding was completed successfully and the coated surface rubbed to perfect smoothness.
The result was impressive.
In less than five months, Andrea had transformed a copper assemblage of soldered triangles into a majestic golden ball.
On Monday, May 27, 1471, Andrea oversaw the placement of the palla, with his assistants on hand to help with the maneuver. Some attached the ball to the castello, others pulled the rope, and another group guided the hoist’s revolving arm as it moved the ball from the construction platform to the top of the lantern. Three days later, the same operation was repeated for the cross on top of the globe. On that day, Thursday, May 30, 1471, the cathedral was officially finished.
To mark this important event in the history of the city, “the canons [of the cathedral] and many other people went up [atop the dome], and sang the Te Deum there.” Workers were treated to bread and wine. Grateful Florentines paraded through the city, and the palla came to be seen as one of Andrea’s greatest accomplishments. It even became part of his name. In official documents from 1471 onward, he was no longer called Andrea di Michele di Cione or Andrea del Verrocchio.
He was Andrea della Palla.
Unlike his master, Leonardo did not become “Leonardo della Palla,” but the experience was even more transformative for him than it was for his master. At the very least, he looked over his master’s shoulder and possibly helped hoist the palla atop the dome. An early biographer noted that Leonardo “was most skillful in lifting weight,” which is an odd remark to find in the biography of a genius.
But the notes the sixty-year-old Leonardo made for himself show that he was fascinated with burning mirrors, and perhaps he did start to sketch them around 1470, even though the first notes we have date from seven to nine years later. With each passing year he acquired a deeper and deeper respect for what he later called “the science of art.” Twenty years later, when he began making plans for what he intended to be a major book on painting, he talked clearly about the difference between guessing at how to solve an artistic problem and following rules of science in order to reach the same result every time. “Those who are in love with practice without science [scientia],” he wrote toward the end of his life, “are like the sailor who boards a ship without rudder and compass, who is never certain where he is going.”
Leonardo’s sketches and notes from the years following the triumph of the palla testify to his interest in uniting science and art. While scholars have largely dismissed these notes as primitive relative to his later scientific writings, they are suggestive of the preoccupations of the fledgling artist, and his interest in educating himself.
In a folio on polishing machines for burning mirrors, for instance, he mentioned “chained books [libri incatenati],” which can only be a reference to the books in one of the few public libraries in Florence, which were chained to their desks to prevent theft. In another folio, he sketched Brunelleschi’s castello.
He made blown-up sketches of the mechanical parts Brunelleschi had designed in order to transform a hoist into a more efficient and versatile machine. He was especially interested in the viticcio di lanterna, as he called Brunelleschi’s screw. This was another of Brunelleschi’s fine inventions: a simple screw (vite) that made it possible to raise the castello as the construction of the lantern progressed upward. Leonardo understood clearly how Brunelleschi’s staggering achievements often owed themselves to the design of small technical components.
This early folio is revelatory for another reason. Right below Brunelleschi’s viticcio di lanterna, Leonardo attempted a crude picture of an eye, a first effort that has gone completely unnoticed until now. He drew the eye using black ink, and the ink has faded to a pale brown such that it is hard to see today. But the writing below it is still legible. It includes questions such as “Why does the eye perceive big things that reach its surface [superfice] as small?” And phrases such as “the pupil is a convex mirror [popilla é specchio cavo],” and what looks like “a glass ball filled of water [palla di vetro piena d’acqua].” The passage is short and convoluted, but those key phrases point directly to some of the most complex theories about the anatomy of the eye, which many authors discussed in the opening chapters of their books on optics.
In another folio about a polishing machine, he twice mentioned “Giovanni d’Amerigo Benci and company [Giovanni d’Americho Benci et chompare].” Giovanni d’Amerigo Benci was a member of a wealthy Florentine family, a patron of the arts, and the owner of a library of books on philosophy and science. His family palace was just a few blocks from Andrea’s workshop. In the mid-1470s, Leonardo painted a portrait of his sister Ginevra. Did he also talk science with Giovanni?
In yet another folio, this one pertaining to water clocks, Leonardo jotted down a list of people who perhaps defined the contours of his intellectual world: a painter, an architect, an astronomer, a physician, a notary, an abacus teacher, and a humanist. On the list appears the name of Paolo dal Pozzo Toscanelli. Toscanelli was fifty-five years older than Leonardo, but the two had many interests in common. Among Leonardo’s papers are notes in Toscanelli’s handwriting on burning mirrors, which were copies of notes by the famous German astronomer Regiomontanus, who worked for the pope in Rome, and who in turn had taken them from the eleventh-century Arab philosopher, Ibn al-Haytham.
On that same list appears the name of a Greek scholar, John Argyropoulos, who lectured on Aristotelian philosophy in Florence until 1471 and who translated, at the request of Cosimo de’ Medici, Aristotle’s Physics (Physica) and On the Heavens (De coelo et mundo), two fundamental texts on optics, reflectivity, and the atmosphere. Argyropoulos also translated Aristotle’s On the Soul (De anima). It discussed how the immaterial soul exists only within a physical body, and was used as a university textbook in the Renaissance.
But the most revealing indication of Leonardo’s early interest in optics can be found on the back of two beautiful drawings representing Saint Sebastian that have been dated to the years 1478–80: one has been in a German museum for decades, but the other has surfaced only recently. He sketched shadow drawings, the oldest that survive, and jotted down some notes.
Unfortunately, these notes are truncated. In one drawing he started to write “That part of the” but breaks off the sentence. He started again in the other drawing: “That part of the […] will be more […] that by larger luminous angle,” and left the sentence incomplete again. Those are early drafts, but they are consequential ones. The aim of these shadow drawings is clear: to understand the science behind penumbras. Carmen Bambach argues that these sketchy notes are “drafts for passages that the artist would later rewrite for cleaner redactions.” When, about a decade later, Leonardo returned to these shadow drawings in another folio, he refined the notes and wrote in full sentences, although in convoluted prose (he never mastered the beauty of written language): “the wall will be darker or more luminous [depending on whether it] will be darkened or illuminated by light [rays] or shadow [rays] that have a larger [or smaller] angle [of reflection].”
He also added a full geometrical demonstration that was missing in the early sketch, although it is certainly possible that the intricacy of that geometrical demonstration was known to him from his early training and it simply did not survive among his papers.
All of these clues make it highly plausible that at some point in his early training Leonardo read books that were seminal works on the science of optics. We know copies of these books were in the city’s public libraries as well as in the private libraries of wealthy Florentines whom Leonardo befriended. Among his many other talents, the young Leonardo was genuinely sociable. At this point he showed no trace of the mercurial temperament he later developed.
What Leonardo must have taken from these books would prove highly consequential for his art, and perhaps the most consequential of all was the Book of Optics, written by the Arab philosopher Ibn al-Haytham.