THE ARCHITECTURE OF MICHELANGELO
by James S. Ackerman
ę 1966, A. Zwemmer Ltd, London
(all rights reserved)
From the Cover
Michelangelo Buonarroti was principally a sculptor and always claimed that architecture was not his profession; but, with a sculptor's vision, he saw buildings as dynamic organisms - metaphors of the human body - and he designed some of the most impressive architecture in all history. Among his best-known buildings are the Medici Chapel and the Laurentian Library in Florence; the Capitoline Hill, St Peter's and the Porta Pia in Rome. None of his work survives just as he envisaged it; in architecture, as in sculpture, his goals were often so grandiose as to be unattainable. Many of his drawings are preserved and record projects such as those for fortification of Florence and for the San Giovanni de'Fiorentini in Rome, which were never executed, but which are among his most imaginative architectural concepts.
In the early years of the sixteenth century the extraordinary power, wealth and imagination of the Pope, Julius II della Rovere (1503-1513) made Rome the artistic centre of Italy and of Europe and attracted there the most distinguished artists of his age. Chiefly for political reasons, the rise of Rome coincided with the decline of great cities of fifteenth-century Italian culture: Florence, Milan, and Urbino. The new "capital" had no eminent painters, sculptors, or architects of its own, so it had to import them; and they hardly could afford to stay at home. This sudden change in the balance of Italian culture had a revolutionary effect on the arts; while the fifteenth-century courts and city-states had produced "schools" of distinct regional characteristics, the new Rome tended to encourage not so much a Roman as an Italian art. No creative Renaissance artist could fail to be inspired and profoundly affected by the experience of encountering simultaneously the works of ancient architects and sculptors - not only in the ever-present ruins but in dozens of newly founded museums and collections - and those of his greatest contemporaries. Like Paris at the beginning of the present century, Rome provided the uniquely favourable conditions for the evolution of new modes of perception and expression.
I described the results as revolutionary. Since Heinrich W÷lffin's great work on this period,1 the traditional concept of the High Renaissance as the ultimate maturing of the aims of the fifteenth century has been displaced by an awareness that many of the goals of early sixteenth-century artists were formed in vigorous opposition to those of their teachers. What W÷lffin saw in the painting and sculpture was characteristic of architecture, too.
But there is an important difference in the architectural "revolution": it was brought about by one man, Donato Bramante (1444-1514). This reckless but warranted generalization was concocted by a contemporary theorist, twenty-three years after Bramante's death; Sebastin Serlio called him "a man of such gifts in architecture that, with the aid and authority given him by the Pope, one may say that he revived true architecture, which had been buried from the ancients down to that time".2 Bramante, like Raphael, was born in Urbino; he was trained as a painter and ultimately found a position at the court of Milan under Lodovico Sforza. Already in his first architectural work of the late 1470's his interest in spatial volume, three-dimensional massing, and perspective illusions distinguishes him from his contemporaries, though the effect of his innovations was minimized by a conservative and decorative treatment of the wall surfaces. When Milan fell to the French at the end of the century, Bramante moved on to Rome, where the impact of his first introduction to the grandiose complexes of ancient architecture rapidly matured his style. The ruins served to confirm the validity of his earlier goals; they offered a vocabulary far better suited to his monumental aims than the fussy terra-cotta ornament of Lombardy, and they provided countless models in which his ideal of volumetric space and sculptural mass were impressively realized.
Architecture is a costly form of expression, and the encounter of a uniquely creative imagination with a great tradition could not have been of much consequence without the support of an equally distinguished patron. That Julius II sought to emulate the political grandeur of the Caesars just as Bramante learned to restore the physical grandeur of ancient Rome continually delights historians, because the occasion may be ascribed with equal conviction to political, social, or economic determinants, to the chance convergence of great individuals, or to a crisis of style in the arts.
As soon as Bramante had completed small commissions in his early years in Rome (e.g., the cloister of Santa Maria della Pace, 1500; the Tempietto of San Pietro in Montorio, 1502), the Pope saw in his work the echo of his own taste for monumentality and lost interest in Giuliano da Sangallo, the brilliant but more conservative Florentine architect whom he had consistently patronized when a Cardinal. A year after his election to the pontificate, Julius commissioned Bramante to design a new facade for the Vatican Palace and the huge Cortile del Belvedere; in the following year, 1505, he requested plans for the new St Peter's, to replace the decaying fourth-century Basilica. Another commission of unknown date initiated projects for a "Palace of Justice" that would have rivaled the Vatican if it had been finished.
The new papal buildings confirm the decisive break with early Renaissance architecture already announced in the Tempietto. This building, though one of the smallest in Rome, is the key to High Renaissance architecture because it preserves traditional ideals while establishing the forms of a new age. It is traditional in being a perfect central plan, a composition of two abstract geometrical forms: the cylinder and the hemisphere. But fifteenth-century geometry had never (except in the drawings of Leonardo, which surely influenced Bramante) dealt so successfully with solids: buildings before Bramante, even those with some sense of plasticity, seem to be composed of planes, circles and rectangles rather than of cylinders and cubes, and to be articulated by lines rather than by forms. In the Tempietto the third dimension is fully realized; its geometric solids are made more convincing by deep niches that reveal the mass and density of the wall. Members are designed to mould light and shade so as to convey an impression body. We sense that where the earlier architect drew buildings, Bramante modeled them. Because the Tempietto recites the vocabulary of ancient architecture more scrupulously than its predecessors, it is often misinterpreted as an imitation of a Roman temple. But just the feature that so profoundly influenced the future - the high drum and hemispherical dome - is without precedent in antiquity, a triumph of the imagination.
In the projects for St Peter's (Pl. 51a, Fig. 11a) the new style attains maturity. Here for the first time Bramante manages to coordinate his volumetric control of space and his modeling of mass. The key to this achievement is a new concept of the relationship between void and solid. Space ceases to be a mere absence of mass and becomes a dynamic force that pushes against the solids from all directions, squeezing them into forms never dreamed of by geometricians. The wall, now completely malleable, is an expression of an equilibrium between the equally dynamic demands of space and structural necessity. Nothing remains of the fifteenth-century concept of the wall as a plane, because the goal of the architect is no longer to produce an abstract harmony but rather a sequence of purely visual (as opposed to intellectual) experiences of spatial volumes. It is this accent on the eye rather than on the mind that gives precedence to voids over planes.
Bramante's handling of the wall as a malleable body was inspired by Roman architecture, in particular by the great Baths, but this concept of form could not be revived without the technique that made it possible. The structural basis of the Baths was brick-faced concrete, the most plastic material available to builders. For the Roman architect brick was simply the material that gave rigidity to the concrete, and protected its surface. In the Middle Ages the art of making a strong concrete was virtually forgotten, and bricks, now used as an inexpensive substitute for stone blocks, lost the flexibility afforded by a concrete core. Bramante must have rediscovered the lost art of the Romans. The irrational shapes of the plan of St Peter's (Fig. 11a) - giant slices of toast half eaten by a voracious space - are inconceivable without the cohesiveness of concrete construction, as are the great naves of the Basilica, which could not have been vaulted by early Renaissance structural methods.3 Bramante willed to Michelangelo and his contemporaries an indispensable technical tool for the development of enriched forms.
In the evolution of the design of St Peter's, Bramante left for Michelangelo the realization of an important potential in the malleability of concrete-brick construction; for in spite of his flowing forms, the major spatial volumes of his plan are still isolated from one another. The chapels in the angles of the main cross and, more obviously, the four corner towers, are added to the core rather than fused into it, as may be seen more clearly in elevations (Pl. 51a).
The dynamic characterization of space and mass which was the essence of Bramante's revolution is equally evident in his secular buildings, even when he was concerned primarily with facades. In the fifteenth century it was the nature of a facade to be planar, but Bramante virtually hid the surface by sculptural projections (half-columns, balconies, window pediments, heavy rustications) and spatial recessions (ground floor arcades, and loggias on the upper story, as in the court of the Belvedere and the facade of the Vatican). These innovations are not motivated by mere distaste for the flat forms of the early Renaissance facade but by a positive awareness of the range of expression available in a varied use of light. His projections capture the sun in brilliant high-lights and cast deep shadows; his half-columns softly model the light; his loggias create dark fields that silhouette their columnar supports. In the facades, as in the interior of St Peter's, the purely sensual delights of vision inspire the design. The philosophical impulse of fifteenth-century architecture had become sensual.
Bramante's style rapidly changed the course of Renaissance architecture. This was due not only to its novelty, but to the unprecedented situation created by the great size of his papal projects: for the first time in the Renaissance it became necessary to organize a modern type of architectural firm with a master in charge of a large number of younger architects who were in one sense junior partners, in another sense pupils. Almost every eminent architect of the first half of the sixteenth century, Michelangelo excepted, worked under Bramante in the Vatican "office": Baldassare Peruzzi, Raphael, Antonio da Sangallo, Giulio Romano, and perhaps Jacopo Sansovino. Of these only Peruzzi actually practiced architecture before Bramante's death (e.g. the Villa Farnesina in Rome, 1509); the others learned their profession at the Vatican and later developed Bramante's innovations into individual styles that dominated the second quarter of the century. The effect was felt all over Italy: Peruzzi built in Siena, Raphael in Florence, Sansovino in Venice, Giulio in Mantua, and Sangallo throughout the Papal States. The death of Julius II in 1513 and of Bramante in 1514 simultaneously removed the co-authors of High Renaissance architecture, leaving the monumental Basilica and palaces in such an inchoate state that the next generation found it hard to determine precisely what the original intentions had been. Paradoxically, this was a favourable misfortune, because it liberated the imagination of the younger architects just as they reached maturity. Raphael, Peruzzi and Sangallo, inheriting the leadership of St Peter's and the Vatican, were free to compose variations on the theme of their master, and were actually encouraged to do so by successive popes who wanted distinctive evidence of their own patronage.
The fact that Michelangelo's career as an architect began in 1516 is directly related to this historical scene. Michelangelo's animosity toward the powerful Bramante kept him out of architecture during Bramante's lifetime. But the election of a Medici, Leo X (1513-1521), as the successor to Julius II, provided opportunities in Florence. Leo, although he chose Bramante's chief disciple, Raphael, to continue the Vatican projects, needed an architect to complete the construction of San Lorenzo, the major Medici monument in Florence. Michelangelo was the obvious choice for this job because he was not only the leading Florentine artist but also a sculptor-painter, ideally equipped to carry out the half-figurative, half-architectural programme envisaged by the Medici family. Besides, the commission served the dual purpose of removing Michelangelo from Rome and of frustrating the completion of the Tomb of Julius II, which would have competed with Medici splendour.
Although Michelangelo's achievements in Florence proved that he was as eminent in architecture as in the other arts, he was excluded from any important Roman commissions so long as any member of Bramante's circle was alive. When Antonio da Sangallo died in 1546, the only member of the circle who survived was Giulio Romano (Raphael d. 1520, Peruzzi . 1536), and it is significant that the Fabbrica of St Peter's called Giulio from Mantua to forestall Michelangelo's appointment as chief architect. But in his death, immediately following Sangallo's, finally left the field open to Michelangelo, now 71 years old.
Yet Michelangelo's personal conflict with Bramante cannot by itself explain why the intrigues that it engendered were so successful in excluding him from architectural commissions in Rome. That the popes of this period - Leo X; another Medici, Clement VII (1523-1534); and Paul III, Farnese (1534-1549) - recognized Michelangelo's pre-eminence is proven by the fact that they tried to monopolize his service as a painter and sculptor. The Medici were even willing to retain his as an architect in Florence after he had fought against them for the independence of the city. The long delay in recognition at Rome must be attributed to the unorthodoxy of his style. It lacked what Vitruvius called decorum: respect for classical traditions. And in first half of the century cultivated Roman taste was attuned to a correct antique vocabulary in classic context. Bramante had formed this taste, and it took a generation to assimilate his innovations.
Raphael was the ideal successor to Bramante. That his concerns as a painter for massive forms and volumetric space in simple compositions of geometric solids were a counterpart of Bramante's architectural goals may be seen in such architectural frescoes as the School of Athens and the Expulsion of Heliodorus. Consequently, when he succeeded to Bramante's post he could pursue his own interests and a the same time design almost as Bramante would have done if he had lived another six years. If Raphael had been less sympathetic to his master, his architecture would certainly be better known. But in major Vatican works, at the Cortile di San Damaso and Belvedere, the two designers are indistinguishable, and uncertainty about the authorship of projects for St Peter's has always worried us. In his work outside the papal circle - Palazzo Vidoni-Caffarelli, Palazzo Branconio d'Aquila, Villa Madama in Rome, and Palazzo Pandolfini Florence - Raphael developed Bramantesque principles and vocabulary into a more individualized expression notable for its greater sophistication, elegance of decoration, and or its success in binding into a unity masses and spaces that Bramante had tended to individualize. The propriety of Raphael's accession to Bramante's throne is further shown by the fact that the very qualities which distinguish him from his predecessor - moderation, respect for continuity, sophistication and elegance, unification of discrete elements - also distinguish his patron, Leo X, from Julius II.
A comparable poetic justice guided the careers of other Bramante followers. Peruzzi, who often worked with linear and planar means of fifteenth-century architecture while concentrating his great ingenuity on exploring new forms and rhythms in plan and elevation (he was the first to exploit the oval plan and curved facade), was employed more in his native Siena than in Rome. That medieval town must have valued him rather for his superficial conservatism than for the extraordinary inventiveness which had too little opportunity for expression, and which now can only be appreciated properly in hundreds of drawings preserved in the Uffizi Gallery.
Giulio Romano, whose three of four small Roman palaces represent a revolt against Bramante's grandeur in the direction of repression, tightness, and an apparently polemic rejection of plasticity and volume, found himself more at home outside Rome, in the court of Mantua, where the tensions induced by the weakness of humanist duchies in a world of power-states could be given expression in a Mannerist architecture of neurotic fantasy (The Ducal Palace, Palazzo del Te).
So the Rome which rejected Michelangelo was equally inhospitable to other non-classic architects. Though Peruzzi, as a Bramante follower, was frequently given a chance to aid in the design of St Peter's and the Vatican and to compete for major commissions (the great hospital of S. Giacomo degli Incurabili, San Giovanni dei Fiorentini), he never was chosen as a chief architect. The victor was always Antonio da Sangallo the Younger, who gave the classic movement its definitive form.
Sangallo's dictatorship in the style of 1520-1545 can be explained more by his propriety than by his eminence; he was probably the least gifted of Bramante's pupils. The first major Renaissance architect to be trained exclusively in the profession, he began as a carpenter at the Vatican in the early years of the century. His practice never had to be set aside for commissions in the other arts and, being a gifted organizer and entrepreneur, he was able not only to undertake all the important civil and military commissions of the papacy but those of private families, among them the Farnese, as well. Nearly a thousand surviving drawings in the Uffizi are evidence of vast building activity throughout central Italy. He is distinguished less for his innovations than for his capacity to apply the experiments and aesthetic of the High Renaissance to the complete repertory of Renaissance building types. The facade of Santo Spirito in Sassia in Rome is the uninspired source of late sixteenth-century facade design; the Banco di Santo Spirito (Rome) has a two story colossal order over a drafted basement in a context that delighted Baroque architects and had never been entirely abandoned; the Farnese palace (Pl. 39) is a definitive secular structure of the Roman Renaissance, though major components of its design were anticipated by Bramante and Raphael. It is in the plans and models of St Peter's that the symptomatic weakness of Antonio's architecture may be seen (Pl. 51c, Fig. 11). The project is unassailable on the grounds of structure or of Vitruvian decorum, but it is confusing in its multiplicity: infinite numbers of small members compete for attention and negate the grandeur of scale required by the size of the building; the dome is obese, and the ten-storied campanili are Towers of Babel. Antonio's superior technical and archaeological knowledge proved to be no guarantee of ability to achieve coherence or to control fully such raw materials or architecture as space, proportion, light and scale.
Sangallo, as the first architect of the Renaissance trained in his profession, knew more than his contemporaries about the technical aspects of construction. He was frequently called upon to right major faults in Bramante's structures: to fortify the piers of St Peter's and the foundations of the Vatican facade, to rebuild the loggie of the Belvedere, which collapsed in 1536, all of necessity to the detriment of the original design. But technical competence was not a pre-eminent qualification in the eyes of Renaissance critics: Bramante, though called maestro ruinante in allusion to his engineering failures, was universally recognized as the superior architect. Of course, this may be attributed simply to a difference in creative ability, or genius, or whatever one may call it, but it raises an important question for Renaissance architecture, and for Michelangelo in particular: was it possible, in the age of Humanism, for an individual to be fully successful as a specialist? Sangallo, in gaining the advantage of a long apprenticeship in architectural construction, lost the benefits of a generalized body of theoretical knowledge and principles traditionally passed on in the studios of painters and sculptors. Problems of proportion, perspective (the control of space), composition, lighting, etc, as encountered in the figurative arts, were more important in the development of Renaissance architecture than structural concerns, partly because, by contrast to the Gothic period or to the nineteenth century, technology was restricted to a minor role.
In our day, when the concern for technique has threatened to overwhelm all other values in architecture, it is difficult to appreciate the Renaissance view that the sculptors and painters were uniquely qualified as architects by their understanding of universal formal problems. The view was vindicated by the fact that it was the artist who made major technical advances - the technician merely interpreted traditional practices.
The Renaissance architect was forced into a preoccupation with broad principles in one way or another. First of all, he had to find a way to justify a revival of pagan grandeur in a Christian society; this involved, among other dilemmas, a rationalization of the conflicting architectural principles of antiquity and the Middle Ages. Further, as is demonstrated by Sangallo's failure to construct a theory out of devoted study of Vitruvius and Roman monuments, antiquity itself taught no clear and consistent body of principles. To give order to a chaos of inherited concepts, many Renaissance architects - Alberti, Francesco di Giorgio and others in the fifteenth century, Palladio in the sixteenth - developed and published theories of architecture of a metaphysical-mathematical cast. But formalized philosophies were not the sole solution; it is intriguing that nothing was written about architecture (or any other art) in the High Renaissance. This reveals a desire to solve the same problems in a new way; a reaction in all the arts against the abstract principles of the fifteenth century produced a temporary shirt from intellectual-philosophical precepts to visual and psychological ones that could better be expressed in form that in words. This change of emphasis is a key to Michelangelo's achievement, and for this reason I begin the study of his work with some observation on what we know of his architectural ideas.
1. Heinrich W÷lfflin,
Die klassiche Kunst, Munich, 1899.
The Basilica of St Peter
Almost every major architect in sixteenth-century Rome had a hand in designing the Basilica of St Peter; each in succession changed his predecessor's scheme, yet the final product is a cohesive whole, formed more by the genius of the Italian Renaissance than by the imagination of any individual. The evolution of the Basilica shows the degree to which Michelangelo's image of buildings as organisms pervaded the architecture of his time. Although Bramante's successors were inspired by the originality and majesty of his design, each felt free to feed the organism new ideas and to cast off obsolete ones. The oscillation between central and longitudinal plans apparent even in Bramante's drawings continued throughout the century and was halted only with the construction of the nave one hundred years after the foundation. Consistency was assured by the huge scale of the structure; architects were compelled to accept and to accumulate the portions built by their predecessors, and once Bramante had raised the crossing piers, no subsequent innovation could be wholly independent.
Medieval monuments the size of which necessitated comparably long periods of construction were much less cohesive in style. The large French cathedrals grew by the accretion of successive units, each of which reveals the fashion of its time; at Paris and Laon, the bays at the end of the nave differ from the rest, and at Chartres the two fašade towers are entirely dissimilar. Even in the Renaissance, great chateaux such as Blois, Fontainebleau and the Louvre became museums of architectural history in which each wing or court was built as a pure example of the style of its period.
The extreme differentiation is the manifestation of a peculiarly French logic, but it is found in Italy to a lesser degree. At the Ducal Palaces in Venice and Urbino, Gothic portions were retained and completed in their original form, while new construction was initiated in Renaissance style. The Certosa of Pavia remained consistent until, in the 1490's, a fašade of an entirely different design was added to complete the church; and at the Cathedral of Florence, Brunelleschi retained the basic scheme of the fourteenth-century dome project, but added a lantern and aediculas inspired by ancient architecture. As long as Renaissance architects were forced ton continue medieval structures, inconsistencies were inevitable. Only buildings started in the Quattrocento could be entirely harmonious in style, but they posed another problem so vexing that, whenever their construction extended over a long period, they often remained, like the palaces and churches of Brunelleschi and Alberti, unfinished. The mathematicalprinciples of Quattrocento design established an interdependence among elements in the plan and elevation that encouraged consistency but discouraged flexibility. The design of a structure begun in accordance with a modular system of proportions could not be changed much, and the architects who succeeded Brunelleschi at San Lorenzo and at Santo Spirito had to adhere anonymously to his style. This became more difficult as time passed and as the style became old-fashioned, so that when Michelangelo was called to design the New Sacristy and fašade of San Lorenzo he could not avoid innovations that differed radically in character from Brunelleschi's forms.
The style of the early sixteenth century was less restricting to the extent that it was less geometrical; moreover, a new attitude was encouraged by professional and technological changes. While most Florentine Quattrocento buildings were small in scale and could be designed and supervised by one architect, the grandiose schemes of the following century turned the fabbrica into a community in which elder architects were partners and younger ones students. Because Raphael, Peruzzi, and Sangallo had worked with Bramante at St Peter's and the Vatican Palace, and because Sangallo assisted Raphael at the Villa Madama, there was no break in continuity when the masters died. Patrons awarded commissions on the basis of competitions and sometimes - as in the project for the San Lorenzo fašade - attempted to enforce collaboration. By the mid-century it was possible for Julius III to assign the relatively modest programme for the Villa Giulia to a tem of three architects: Vignola, Ammanati and Vasari, with Michelangelo as a consultant. In architecture as in the political structure of the Renaissance state, size promoted collaboration, centralization and continuity, and kept designers as well as princes from disrupting the orderly evolution of the institutions they directed.
Structural factors, above all, secured the organic growth of St Peter's. Bramante, in visualizing the Basilica as an expansion of spatial volumes and masses about a vast central area, made the crossing the heart of a cellular structure (Fig. 11a). Every element in his design depended for its stability upon the four central piers, and the dome, in turn, depended on the buttressing powers of the four arms. So the construction had to proceed uniformly outward from the core toward the periphery. This radical evolution differed radically from the chain-like process demanded by the bay-system of Gothic structures, in which spatial frames, each depending on the neighbouring frames for stability, had to be raised in sequences beginning at the apse, a the fašade, or any terminal point in the plan.
Though the Gothic system survived into the Renaissance, the autonomy of the single bay often gave way to what might be called a box system, in which cubic or cylindrical volumes were applied to a core; even the central-plan buildings of the Quattrocento give the impression of having been built up the addition of autonomous units. The uniqueness of Bramante's St Peter's project - visible in the plan (Fig. 11a) - was in the interdependence of the core and its arms. A study of the malleable wall masses of ancient Roman architecture must have helped Bramante to break down the confines of the Quattrocento box, but it was the Byzantines, not the Romans, who had found techniques for integrating domed and longitudinal volumes.
Consciously or not, Bramante revived the structural principles of Hagia Sofia in Constantinople, where all spaces had been generated outward from a domed core. Surviving drawings from Bramante's workshop indicate that the four crossing piers were raised before the final form of the arms had been determined, and for decades after his death each of his successors in turn was free to clothe his skeleton in a new skin. Sixteenth-century views of the Basilica (Pls. 52a-b, 53a) show how its radial evolution gave Michelangelo a maximum of freedom in designing the exterior facades.
The interior volumes, however, were firmly fixed at the time of Sangallo's death in 1546: one arm had been completed entirely, another partially, so that the remaining arms could not be changed; the vaults that form aisles around the crossing, between the outer buttressing piers and the crossing piers, had been built, too. Even when Michelangelo got leave to lop off the outer rings of the hemicycles that terminated all but the fašade arms, he was constrained to keep the inner ring, and could reform only its exterior plan (Fig. 12). The limitations here were greater even than those imposed on the design of the Medici Chapel: the interior could be influenced only by the design of the central dome, the four domed areas at the corners, and the hemispherical vaulting at the ends of the arms. Michelangelo was left in undisputed command solely of the lighting, since these restrictions did not limit the formation of the exterior surfaces. But after his death in 1564, most of his plans for the interior were altered: della Porta redesigned the central dome and those of the four corner chapels, so that all we can see of Michelangelo on the interior of St Peter's is the main drum and the vaulting of the terminal hemicycles; but the original character of both is entirely changed by an overlay of seventeenth-century ornament and veneers.
The extent to which Michelangelo was able to impose his personal style upon St Peter's without essentially altering the interior is astonishing. We can see in comparing his plan to Sangallo's (Fig. 11, 12) that a few strokes of the pen were sufficient to change a complex and confused form into a simple and cohesively organized unit. Sangallo, in taking from Bramante the scheme of a major cross echoed in four lesser crosses at the corners, had expanded the latter to constitute isolated pockets of space no longer knit into the fabric of the crossing; similarly, his semicircular ambulatories became independent corridors - superfluous successions of volumes and Orders which forced him into absurd devices for lighting the main arms. (Pl. 51c, far right). Michelangelo, by merely walling off the entrances to each of Sangallo's disconnected spaces, made one church out of many; he surpassed the clarity that he admired in Bramante's plan in substituting for the concept of major and minor crosses a more unified one of an integrated cross-and-square, so that all circulation within the Basilica should bring the visitor back to its core. The solution was strikingly simple, and far more economical than any proposed before: it even seems obvious, once it is familiar; but in a generation distinguished for great architects, it took one trained as a sculptor to discover a form that would express the organic unity of the structure.
Unity was Michelangelo's contribution to St Peter's; he transformed the interior into a continuum of space, the exterior into a cohesive body. In the exterior massing he was restricted less by earlier construction, since his predecessors had not arrived at the outer periphery. Here again, the problem was to find a form which would integrate two autonomous motifs in the plan - the cross and the square - and again it was solved with the simplest and most economical means (Fig. 11). With a minimum of construction the secondary buttressing piers were transformed to serve entirely new practical and expressive functions. Inside, the passages which Sangallo had cut through the piers were ingeniously converted into stairwells; outside, the diagonal faces of the piers bound the hemicycles of the cross to the angles of the square in such a way that the two shapes were fused without losing their distinctness. The solution was technically impeccable; it changed the form of the piers without affecting their structural function and it efficiently solved the problem of lighting the stairwells. Aesthetically, it was an inspired breach of classical dogma. In plan, the piers were formed essentially as mirror-images of the crossing-piers. But unlike the crossing piers, their diagonal outer faces do not form a forty-five degree angle; they were drawn on the principle that a straight line is the shortest distance between two points, without regard for the angle of incidence, and in violation of Renaissance laws of geometry and proportion. Michelangelo interpreted these diagonals as building elements- as muscles, not the limits of a regular polygon. Simple as the form seems to a modern eye, it represents - even more than the oval and trapezoid of the Campidoglio - a bold and difficult revolt against the immemorial sovereignty of rational geometric figures in architecture.
Comparison with Sangallo's plan reveals the skill with which Michelangelo resolved the continuing conflict between the centralized and longitudinal schemes (Fig. 11). Sangallo had artificially appended a nave and fašade onto one arm, forming, in effect, another church. Michelangelo differentiated the fašade arm just enough to give the Basilica a major axis without prejudicing the centrality of the interior. The Pantheon-like columnar porch emphasized the entrance axis, yet permitted the pilaster system of the side and rear elevations to continue across the fašade without interruption. Moreover, the pediment carried over the forward row of columns was low enough to leave an unimpeded view of the dome from the piazza (a virtue lacking both in Sangallo's and in Maderno's designs); its triangular form would have directed the eye toward the dome, while its proportions and forward projection would have announced the scale and significance of the nave beyond.
The fašade was to be a screen before the undulating mass of the Basilica; it is astonishing how much Michelangelo managed to alter Bramante's formulation of the character of this mass (Pls. 50, 51a, 60). Bramante saw the exterior as a society of distinct geometrical forms bound together by proportion, Michelangelo as a single body so cohesively organized that the differing functions and structural features of the interior plan barely can be discerned. The structural technique - a revival of the heavy, plastic wall-masses of Roman and Byzantine architecture - permitted Michelangelo to treat the body of the Basilica as a sculptural block, and left him free in the choice of surface articulation; the exterior Orders were to be exclusively expressive. Perhaps this is why the colossal pilasters and the strips behind them were distinguished so clearly from the wall surfaces (Pls. 63, 64): they carry a projecting segment of the entablature so that the whole decorative apparatus appears as a detachable overlay (at the Capitol, where similar pilasters have an essential structural function, they support an unbroken entablature). Fenestration was the sole limiting factor: it dictated a tripartite division of the hemicycle elevations and inspired the rhythmical sequence of broad and narrow bays separated by pilasters. The dynamic vertical accents of the pilasters, reinforced by the strips behind them, by the projections in the entablatures, and by the multiplication of shadows that results from compressing two pilasters into one the bends around each angle, entirely overwhelms the discontinuous horizontals of the window and niche frames. The dominance of verticals makes the Basilica appear to grow upward rather than to weigh ponderously on the ground; it suggests an aspiration comparable only to the effects of Gothic architecture, and anticipates a climax in the equally Gothic buttresses and ribs of the dome.
Turning again to Bramante's elevation (Pl. 51a) we find an entirely opposing effect; horizontals dominate in spite of high campanili, and the weight of the structure is expressed by the accumulation of masses toward the earth, beginning with the low ribless dome and its stepped base, which seems to settle into the drum. Bramante, who developed the plan from the crossing outward, must have designed the elevation from the dome downward. For him, the great central volume was the cause of the design; for Michelangelo it was the result. Such a distinction is warranted by the peculiar chronology of Michelangelo's studies for the construction; the design was not wholly fixed at the start, but grew as the builders advanced upwards from the foundations. At the beginning, only the lower portions were determined definitively: probably the model of 1546/1547 had a bare attic, and no fašade, roof or domes. When the existing attic was built in 1557, it was left without an exterior facing, with the intention of adjusting its design to future decisions on the dome. In the same year the drum was begun, before the construction of a dome model in 1558-1561 (Pl. 57b). Between 1561 and Michelangelo's death in 1564, the dome was again revised, the attic was designed, and the fašade project, which was dependent on the definition of the attic, was tentatively sketched in plan. This does not mean that Michelangelo ignored the dome until the end: his earliest studies for it (Pls. 54, 55a) pre-date the model of the lower portions. But these studies constantly evolved as Michelangelo watched the walls rise and saw the effects of his vigorous verticals in full scale. We can imagine that the definitive design of 1546 for the paired colossal pilasters was accompanied by a decision to use external ribs on the dome and paired columns on the buttresses. If Michelangelo ever considered retaining Bramante's smooth, stepped hemisphere, he would have abandoned the thought before generating a dynamic upward thrust in the lower part of the building. But only the ribs and buttresses survived to the end; the design of the drum and the lantern changed, and above all, the profile of the dome, which developed from the elevated curve of Plate 54 to the hemisphere of Plates 60 and 61.
The progressive lowering of the dome is a key to understanding of Michelangelo's purpose, yet modern critics were at first reluctant to accept it as a fact. A progression from the spherical dome of the engravings (Pls. 60 and 61) to the raised profile of Pl. 54 (now recognized as an early study) to the dome executed by della Porta (Pl. 63) seemed natural; moreover, it is admirably suited to the popular Wolfflinian theory of a somehow preordained and systematic evolution from classic Renaissance to dynamic Baroque forms. The irrelevance of these presuppositions is sufficiently proven by the elevated profile proposed by the most "classic" of early Cinquecento architects, Antonio Sangallo (Pl. 51c) and by the low dome of Michelangelo's San Giovanni de' Fiorentini (Pls. 70b, 71a-b), which is contemporary to the St Peter's dome model.
Shortly after determining the insistent verticals of his elevation, Michelangelo wrote to Florence for measurements of the Cathedral lantern. The Florentine cupola had exerted a strong influence upon him from the start; he took from it the double-shell construction, the raised profile and octagonal lantern of Pl. 54, the rib construction and the drum oculi of Pl. 55a. The Cathedral cupola was the only available prototype of scale comparable to St Peter's, and its medieval rib construction gave a secure and sufficiently calculable means of controlling great loads. The Gothic profile was congenial to the vertical thrust of the colossal Order, which could not have been resolved in a sunken or smooth dome of Bramante's type. Bramante's dome (Pl. 51a), with its solid mass of masonry, and without external buttressing, would have been excessively difficult - perhaps impossible - to build over such a span. In his last work, the Torre Borgia cupola of 1513 (Pl. 51b), Bramante embraced the structural and expressive potentialities of the Gothic rib: this design probably suggested to Michelangelo the advantages of increasing the eight ribs of the Florentine dome to sixteen, as a means of avoiding an over-emphasis on planes.
Bramante's dramatic Torre Borgia lantern had the effect of resolving the forces of the converging ribs; Michelangelo also found the lantern to be the key to his design: before and after writing to Florence in 1547 he was preoccupied with its form and proportion more than with the dome profile (Pls. 54 - with five lantern-elevations and two plans - 55a-b); on completion of the dome model in 1561 he was still uncertain of the lantern scheme, and later even Vasari was confused about the final design. Since the accents of the paired colossal pilasters on the body of the Basilica were to be channeled into the paired drum columns and from there into the dome ribs, the lantern became the climaz and resolution of the dynamics of the entire composition. The letter and the drawings show that Michelangelo cared less about the dome profile than about the ratio in height between an elevated dome with a low lantern and a hemispherical dome with a high lantern.
The final solution recorded by Duperac (Pls. 60, 61) shows the lantern raised on a high podium to compensate for the lowering of the dome, so that the overall height of the Basilica would not have been much less than in the early designs. Moreover the diminution in the width of the dome ribs toward the top would have preserved by perspective illusion the original effect of the elevated profile. The hemispherical profile represents not so much a rejection of Gothic in favour of classic prototypes as an internal crisis in Michelangelo's style. In the space of twelve-to-fourteen years between the design of the lower order and the construction of the dome model, he had turned from the active tensions of the Campidoglio project and the frescoes of the Cappella Paolina to the subjective gravity of San Giovanni dei Fiorentini and the late Passion drawings. The state of mind that produced the reserve and calm horizontality of the San Giovanni model (1559-1560) cannot have been wholly congenial the uninhibited verticality of the initial St Peter's designs (1546): the hemispherical dome (1558-1561) approaches the mood of San Giovanni without denying the forces generated in the body of the Basilica; the steps in the ribs and the rings of dormer windows reinforce the new sedative element. We know from Dosio's drawings that Michelangelo thought at one point of combining the low dome and the low lantern, but he must have found through experiments on the model that this would over-emphasize the shift in style; the early spirit had to be resolved in the lantern.
After completing the dome model, Michelangelo in his last years turned to the design of the attic, where he achieved the same balance of force and repose. He may have made no decisive designs before this time since the model was built without an attic facing (Pls. 58a, 59a), but the construction of the apertures implies that he had intended to give the window frames a vertical axis. In the revised design, the apertures were covered by horizontal frames which help to inhibit the vertical surge. The new accent was to have been reinforced by a continuous balustrade (Pl. 58b).
The restraint of vertical forces in the final project did not result in the kind of tensions found at the Campidoglio, but in equilibrium gained without loss of vigour. The co-existence of static and dynamic forms - a product of the profound introspection of Michelangelo's late years - was too subtle to be understood by contemporaries. In executing the existing dome, della Porta could not rise to the challenge of Michelangelo's testament; in his details he greatly reduced its rigour by eliminating the distinctions between horizontal-circumferential accents and vertical-radical ones. His rich decoration obscured and softened the clarity of Michelangelo's transitions, and disconnected the bones of the structure. By thinning the ribs and their supports, and eliminating their perspective diminution, by elevating the dome profiles and lowering the lantern, della Porta summoned the more familiar image of the Florentine dome. But the aspiring effect of della Porta's dome would have been more powerfully achieved in Michelangelo's final solution, where the climax at the lantern is amplified by the contrasting calm of the dome. We cannot tell how Michelangelo's minor domes would have influenced the final solution: though he probably planned them, he apparently left no designs; those on the engravings seem to be by Vignola, while the existing ones were build from della Porta's design.
Michelangelo's dome fused the forms of antiquity and the Middle Ages in a way incomprehensible to della Porta, who had to return to the more consistent solution of the early studies, and to many modern critics, who failed to see the logic behind the evolution of the design. But for all its deficiencies, della Porta's dome preserved the essential potency of the original concept, and gave the architects of the Baroque one of their most compelling sources of inspiration.
While Michelangelo absorbed certain medieval forms into the predominantly Roman character of Bramante's Basilica, the final design was so thoroughly transformed by the individuality of his own style that it no longer symbolized its traditional roots. It was a statement so unique and so powerful that it became itself a symbol for future centuries. The form of the dome was to become the receptacle for the expression of civic as well as religious ideals; even in Protestant countries where its association with the center of Catholicism might have discouraged emulation, the functions of local and national government are carried on under the cover of replicas of Michelangelo's dome.
Twenty-five years after Michelangelo's death, his design for St Peter's as emended by della Porta was represented on a fresco in the Vatican library (Pl. 58b). The Basilica appears in the center of a huge square surrounded by porticoes designed in the style of Serlio, the construction of which would have required the removal of the Vatican Palace. None of the architects of St Peter's could have hoped to demolish the palace, but the fresco represents more than a painter's fantasy; it demonstrates a great sensitivity to the spirit of the design. The artist returned to a fifteenth-century formula typified in Perugino's Delivery of the Keys in the Sistine Chapel, in which a monumental central-plan structure appears in the center and to the rear of a vast piazza with a pavement marked off into squares. Perugino and his latter-day heirs illustrate the principles of Leone Battista Alberti, who demanded that the principal "temple" of the city should be centralized in plan, that it should be isolated in the center of an ample square, and that it should be raised on a podium to elevate it from worldly things. Alberti would have approved of Michelangelo's pedimented entrance-porch which, in fact, he had used himself in his Mantuan churches.
In all of the centralized projects for St Peter's the impact of the form would have been severely compromised by the congestion of the surroundings. The observer would have been frustrated by the fact that while the form of the Basilica invited him to circulate freely around it, the buildings on either side and the slope of the Vatican hill barred the way. Circulation was invited much more by Michelangelo's design (Pl. 50) than by Bramante's, where block-like forms established finite, self-sufficient planes. Michelangelo, constrained by the portions already built to retain the ideals of the Quattrocento, but unwilling to compromise the kinetic force of his own style, brought into focus the paradox between the early Renaissance aesthetic of stability and centrality and the late Renaissance aesthetic - in the foundation of which he played a dominant role - of movement, axis, and climax. No wholly successful solution to this paradox was possible; one alternative is represented in the fresco; another - prompted by the symbolism and liturgy as well as by the taste of the Counter-Reformation - in the existing Basilica, where centrality was destroyed and the effect of the dome obscured by the extension of the nave.