The United States of America was not among the last Roman colonies, but then the idea that it was could appear equally plausible. Across the entire country there is evidence of a rational and powerful society, the background of which could have been Greco-Roman. If the town grids exhibit a lack of closure and an occasional confusion about cardo and decumanus, and if temples to unknown deities are rivaled in their freedom of distortion only by the greatness of their numbers, then all of this might only too easily be explained by remoteness of original models and the relative lateness of colonization. But in spite of insistent visible evidence of grid towns and temples—and even an occasional pantheon—there is no trace, crucially, of forum or agora, thus assuring that the United States is a product of another time and another culture, that its Roman ties are indirect, assumed, or both.

From the absence of a strong tradition of enclosed urban space (that is, of forum or agora), we can conclude that America was principally the product of Neoclassicism, not of Classicism, and that the profile of urban America, with its notably fragile tradition of urban space, is therefore more a result of chronology than or geography or genetics. Both the landscape and its distance from the centers of Western civilization contributed to the basic psychology of the place—to what Vincent Scully describes as "a feeling at once of liberation and of loss"¹—but it was Neoclassicism that provided the urban system for the emerging nation. It was precisely that urban system, applied in the special geographic and psychological context, that turned the ideal of the Enlightenment garden city into reality in the New World.

At the time of its birth in 1776, the United States of America had a total population of approximately 2.25 million. By 1790 this had increased to almost 4 million, but the total urban population was only slightly more than 200,000 people and only two cities were larger than 25,000.² The whole country was regarded as something of a wilderness by foreign visitors, and there were no professionally trained American architects. (Benjamin Latrobe, arriving in 1796, was the first.) Urbanism was barely an ideal, much less a developed tradition or even a necessity. The principal architectural element in the colonies had been the detached house; and even by the time of the Revolution, only the northern cities evinced some pattern of town houses with potentially common walls. The character of the towns was still distinctly medieval. Between this innocent time and the era of aspiring to an expansive future, a second wave of European influence appeared—one that would be of incalculable importance to the urbanization, and subsequent deurbanization, of America. By 1900 this relatively primitive arcadia would be transformed: almost half of the total population of 87,832,000 would be urban, and three cities would be larger than one million.³

The second wave of European influence, which swept America during the 1790s, is analogous to the second wave of Italian influence introduced to medieval France in the 1540s. The principal difference is that this time the imported taste and culture were not Italian but English and French; the imported spatial sensibilities were not those of the Renaissance but of the Enlightenment; and two of the three principal couriers were not foreign, as were Vignola and Serlio, but native Americans: Charles Bulfinch and Thomas Jefferson. Both men were gentlemen-amateur architects, each returned from an extended stay in Europe in the late 1780s, and each tried to introduce an appropriate architecture and urban culture to the developing United States of America. The architectural language of both men was Neoclassicism, but for Bulfinch the tradition was English, whereas for Jefferson it was French—and Roman.⁴

After returning in 1787 from almost two years in Europe, Charles Bulfinch devoted himself for some thirty years to providing his provincial hometown of Boston with an urban architecture. His sources, like his background, were English, primarily Robert Adam and William Chambers. Although his projects for Park Street, Colonnade Row, and the Tontine Crescent no longer exist, they provided the inspiration for many surviving projects that still place Boston closer than most American cities to the European urban tradition. The partially realized Tontine Crescent was especially significant. Planned around 1793, it was the first (and almost the only) American example of a unified enclosed urban space. Although there were earlier squares and greens, such as those of Savannah, Philadelphia, and New Haven, they were neither enclosed nor unified; and although there were subsequent examples of unified enclosed squares, there were very few, and fewer still outside of Boston.

The Tontine Crescent was originally designed as two opposing crescents of regular houses with an oval park in the middle and the Franklin Theater at one end. As finally realized, the northern crescent was replaced by a straight row of houses, but the integrity of the composition was maintained. The Tontine Crescent could have been the prototype for an American tradition of enclosed urban spaces; that this was ultimately not the case may be due as much to the project's English derivation as anything else. Bulfinch's Crescent, along with his other projects, renewed the connection to English architecture and culture that had been broken for thirty years. But while the English tradition may have been highly coveted in Boston, it was predictably less so in the rest of the country. Having just broken away from England through revolution, the rest of America certainly did not want an English architecture as a symbol of their new nation. It may have been more than the Britishness of Bulfinch's Boston, however, that made its acceptance difficult.

America was born on a cusp of history, a philosophical and architectural turning point between the spatial tradition of the Renaissance and the iconic stirrings of the Enlightenment; and although it would be disingenuous to suggest that the country as a whole was consciously aware of it, there was a general longing for expression. Architects were seeking an American architecture, one that was both an instrument of and an appropriate symbol for a country that was a product of the Enlightenment, not the Renaissance. Of course, the ironies of the gardens of Versailles serving as the model for the capital city of a new democracy have often been noted, but even this ultimate symbol of the ancien régime took on a different meaning when it was reinterpreted by New World Neoclassicism. Washington, D.C. established the importance to America of the French tradition, which, embodied in the French Enlightenment, had its greatest impact on American architecture and urbanism. In this respect, it was Jefferson, and not Bulfinch or Latrobe, who played the role of Serlio to the New World.

Jefferson was an idealist as well as a practical man, and he provided the most poignant architectural and "urban" expression for the new democracy. The expression was so compelling that it still permeates the American psyche and therefore remains problematic to this day, perhaps as much for its symbolic implications as for its anti-urban ones. Early in his career, Jefferson consciously rejected the English architectural tradition for its political associations, and, with the exception of the English landscape garden which he guardedly admired, turned instead to Roman architecture as understood through books, especially those on Palladio. After his stay in Paris, his style became a kind of Franco-Romanism that he considered an appropriate cultural expression.

Jefferson's well-known rejection of the city appeared quite early and was modified only when, late in his life, he grudgingly accepted its necessity. Even as late as 1800, after five years in Paris, he viewed "great cities as pestilential to the morals, the health and the liberties of man."⁵ Indeed, Paris not only failed to change Jefferson's anti-urban convictions, it reinforced them. Paris demonstrated to him that the Neoclassical pavilion in a romantic landscape had the potential to serve as the ideal fabric of a civilized agrarian democracy.

Jefferson replaced Benjamin Franklin as Minister to the Court of France and lived in Paris from August 1784 to September 1789. The experience had a profound effect not only on him, but also on America through the ideas he imported. The influence was almost immediate: after seeing his first and only important Roman temple, the Maison Carrée in Nîmes, Jefferson reinterpreted it, with the help of the French architect Clériseau, as America's first Roman temple, the Virginia State Capitol (1785–89). To point out that Jefferson mistook the Maison Carrée for a republican monument would verge on pettiness about symbolism; what caught Jefferson's eye was the monument's beauty and perfection. In any case, its symbolism is probably close enough.

But "Jefferson's Paris" was not really Roman Paris or even seventeenth-century Paris; rather, it was Paris of the 1780s—frenzied, elegant, avant-garde, Neoclassical Paris—the Paris of Ledoux, of the Palais Royal, and of the pavilions in the new quarters beyond the boulevards.⁶ Paris was a vital, active place during Jefferson's five years there, and thanks to his fanatical record keeping, we are aware of his visits and his interests. Two of the most influential were the newly completed Palais Royal and the new domestic architecture. Both had an enormous impact on Jefferson. As models, they represented extreme opposites: one, the ultimate public gesture, an elaborate urban organism providing both textural continuity and conspicuous spatial focus; the other, the ultimate private icon, prelude to the machine à habiter, the internally responsive, externally expressive solid.

Naturally, it was the latter that had the most pervasive influence on Jefferson's work. During his first year in Paris, he lived in the new Chaussée d'Antin section beyond the boulevards, and since Ledoux's Hôtel de Montmorency and Hôtel Guimard were just around the corner, Jefferson must have known them. He appreciated the sophistication of French planning, and nowhere is this more apparent than in the development of his own house at Monticello. Before Paris, Monticello was an incomplete and slightly awkward combination of a symmetrical English plan with simple rooms and a modified Palladian facade. Afterward, it was not just completed but transformed into an elaborate Franco-Roman villa incorporating all the latest developments from Neoclassical Paris. The outside was modified to read as a one-story pavilion like the Hôtel de Salm, which he had much admired, and the inside was extended and particularized to a degree achieved only in the best French plans. The house has a great variety of rooms, highly developed service areas, and a separation and contrast between the public and private sequences resulting from the ingenious arrangement of the private rooms in two tiers around the double-height public rooms.

All of these are distinctly French traits, yet one glance reveals that the plan is not French, but something quite different. In the typical Neoclassical French plan, the idiosyncrasies and irregularities are always contained within a rectangular configuration—simple on the outside, complex on the inside—and the central axis of the building is almost always blocked. Jefferson's final plan for Monticello is the opposite. The central Palladian axis is maintained through the sequence of regular public rooms, and the smaller, more specific private rooms are thrown to the outside of the plan. In addition, the perimeter of the plan is loose, articulate, particular; here the center is simple, the perimeter complex. It is as if a French plan had been modified by English influence, rather than the reverse. Later, this kind of modification actually happened when the influence of the English cottage ornée loosened the French perimeter and made the axes of the plan even more casual. Then, in the late nineteenth and early twentieth centuries, the Americans reestablished the axes again while maintaining the English preference for the picturesque perimeter. Thus Jefferson's little house may be seen as a historical shortcut and a magnificent preview of what would be one of America's most poetic urban inventions—the one-family house on Elm Street.

The Palais Royal, Jefferson's other Parisian fascination, had become what Mercier called "the capital of Paris" by the time Jefferson arrived in 1784. Indeed, it was at that time a complex, multifunctional public building, although this had not always been the case. Until 1780, the building proper had been a royal palace facing gardens that were always accessible to the public. The new owner, later to become the Duc d'Orléans, decided to capitalize on the potential of the property and hired Victor Louis to transform the palace complex into a vast multi-use commercial center. The gardens were surrounded and enclosed with arcaded galleries, to the understandable outrage of the public, especially the surrounding property owners. When the building was opened in 1784, however, it was an instant success. Full of shops, restaurants, theaters, and apartments, it became the social center of Paris. The public garden remained, and was even equipped with a subterranean circus. Jefferson visited the Palais Royal frequently and was sufficiently moved to proposed a similar complex for Richmond, Virginia, which would be "a whole square in Richmond improved on some such plan, but accommodated to the circumstances of the place."7

It is tragic that this idea came to naught, however casual its suggestion may have been, for it would have joined the Tontine Crescent in introducing conspicuous urban space to America. The Palais Royal is an invaluable resource both as a fragment and microcosm of the city, and had it been realized, its American counterpart might have been an equally suggestive model for the rapidly growing country. Fortunately, however, this persuasive image informs Jefferson's most important work, the University of Virginia, in a subtle yet profound way.

The University of Virginia sometimes evokes references to Roman fora; it almost always invites comparisons with the original plan (1812) for Union College in Schenectady, New York, by the French émigré architect Jean-Jacques Ramée; and its similarity to the Château de Marly (begun 1679) by J. H. Mansart has occasionally been noted. Yet Jefferson's project is unlike any of these. It is less enclosed and less unified than a Roman forum or Union College; and although it bears a striking resemblance to the arrangement at Marly, its form and its meaning are completely different. The Château de Marly, intended as an escape from the totalitarian rigors of Versailles, is formed of separate but equal pavilions axially related to a similar but larger one for the king, all of which are set in a terraced, formal French garden. At Marly the order is singular and unified, and the part, although expressed, is still controlled by the whole. The University of Virginia, on the other hand, exhibits a more complex order in which the part and the whole exist in a suggestive balance, with many intermediate shades of interpretation. Both the immutability of public truth and the transcendental license of private adjustment are debated at all levels of the organization. It is as if the best qualities of the French Enlightenment and the French Renaissance—of the Neoclassical pavilions and the Palais Royal—had been fused by Jefferson into one spectacular and distinctly American expression.

For political and religious reasons, Jefferson studiously avoided the English university models, and he also avoided the early American campus models because of their ad hoc, inconsequential freedom. Instead, he provided for Virginia a great central space defined by colonnades and individual pavilions. Planted with grass and lined with trees, this space was open on the lower end, and, at Latrobe's suggestion, focused on the library at the upper end. Supporting the space on either side was a rich system of gardens, paths, and an outer ranges of service buildings. Like a democratic version of the Château de Marly, the University of Virginia was at once more subtle and more suggestive. It is no accident that Jefferson referred to his plan as an "academical village," for it is, if anything, a metaphor of society and the city—a Neoclassical ideal "adapted to the circumstances of the place." It certainly is as close as Jefferson came to a coherent idea about the city, and it did provide the United States with a sublime spatial model of the city.

This is not intended to portray Thomas Jefferson as the sole source of American architecture and urbanism, nor to slight Benjamin Latrobe. Rather, it is simply to argue that at a critical time in its history—at its birth as a nation—the United States inherited an architectural language that was fundamentally not urban, but suburban; that the principal source was not English, Spanish, or American Indian, but French; and that the most persuasive purveyor of that language was Thomas Jefferson.

Fortunately, this inherited language was not absolutely pure; like most products of revolution, it contained traces of the previous system. Thus, although Jefferson's work and French Neoclassicism demonstrated Enlightenment preference for the evocative solid, there was still a trace of the Renaissance in the faint structure of urban space. For example, in late-eighteenth-century French engravings, such as those by Krafft and Ransonnette,8 or Ledoux,9 domestic pavilions were rendered in an apparently continuous romantic landscape, whereas the real buildings that they depict—the pavilions of suburban Paris—were always contained by walls and the streets were defined by gates and service buildings. It can be argued that in France it was necessity that compromised the new ideal—necessities of space, economics, and tradition—and to a certain extent this may be true. The United States, however, was hardly encumbered by these circumstances, which makes it all the more remarkable that Jefferson considered both the symbolic power of the buildings and the sublime presence of space to be, to some degree, necessary to the formulation of the public realm. This may in fact be Jefferson's most important legacy: the continued insistence on both the ideal and the circumstantial, even in the absence of obvious necessity. For Jefferson, circumstance seems to validate the ideal and, furthermore, the ideal only seems to achieve meaning within that union. From the Neoclassical propensities of the American constitution to the anatomy of the small American town, this debate is inextricably present as principle and as fact. Though in the United States the balance may be tipped in favor of the private and the circumstantial over the public and the ideal, there was, for at least a century and a half, a dialogue.

Jefferson's interpretation of the French tradition in his two principal works, the University of Virginia and Monticello, relates directly to what can only be described as two uniquely American contributions to urbanism: the American college campus and the essence of the small American town, Elm Street.

From the earliest colleges of the colonial era to the land-grant universities of the frontier, the American campus has been a simulated city that, with few exceptions, is distinctly unlike European models. Loose arrangements of freestanding buildings meld with the landscape to suggest an almost urban space. Not all campuses possess clearly defined quadrangles, but the ones that do seem to be more focused and provocative. They also represent the most tangible tradition of enclosed urban space in the United States, and that is directly attributable to Thomas Jefferson. The image of Jefferson's Virginia lawn is imprinted on countless campuses in America, from Bowdoin to Cornell to Columbia to Minnesota. The order is usually less rigorous than the original, but the stable presence of public space remains, enriched and tirelessly supportive of peripheral freedoms. On campus one has a veiled sense of being in some kind of primitive urban laboratory, where urbanism is being dissolved into landscape or perhaps reconstituted in spite of it. And is not the campus almost always more substantial, more urban than itshost body, and a continual commentary upon it? As a model of the city, the American campus may well be more suggestive than the real thing; certainly it is one of America's truly original contributions to urbanism.

For all their vitality, American cities are still derivative of the European urban tradition and invite frequently unfavorable comparison. But the American small town is dream town, not to be found in Europe and therefore not comparable. It is the Enlightenment garden city rendered in infinite variations. If the Town Green is not always sublime, Church Street is usually redeeming, though Main Street may in desperation steer very close to toy town. But Elm Street—Elm Street elevates it all, occasionally to the level of poetry. With Elm Street, European hegemony in the development of urban streets is challenged, and in the process Serlio's "Satyric Scene," long ignored by Europe, is resurrected, civilized, and brought center stage. Elm Street is the quintessential tree-lined residential street. It is the Neoclassical street that the French could never or would never build, the penultimate stage in a long process of formal and social inversion. That process of inversion may be traced by examining a series of street types.10

In a typical street in the Marais section of Paris, one such as the rue des Franc-Bourgeois, the space of the street is emphatically defined by a continuous mass of urban hôtels. Even when the forecourt of the hôtel is screened from the street by only a wall and a gate, the ends of the service wings are of sufficient mass to positively define the street. The private world is completely screened from the public world, and clearly defined forecourts serve as transitional spaces.

A distinct difference may be noted in a typical Rococo street, such as the rue de Grenelle in the Faubourg St.-Germain. Although the plans of the hôtels appear similar to those of the Marais, the forecourts are defined by low service wings, and the main living blocks tend to assert themselves as pavilions between the forecourt and the garden. As a result, the street is defined primarily by gates and screen walls. Here the private realm has become more emphatic, the public realm has become less so. From the Rococo street to the American suburb is a surprisingly short hump, but there is one important intermediate step.

In the Neoclassical streets north of the boulevards, such as the rue Poissonnière, a condition of detached, independent houses is finally achieved. The service wings have disappeared, and the houses generally sit free. They are still separated by garden walls, however, and the street is still defined by rather substantial service buildings and gates. Formally and socially, these Neoclassical hôtels are the inverse of those of the Marais: void has become solid, yet the positive armature of the public realm is insistently maintained.

On Elm Street, however, there is no need for even the gates and party walls of the rue Poissonnière: the forecourt has become the front yard, the garden the backyard. As in idealized French Neoclassical renderings, the house sits free in continuous Arcadia, discreetly varying but always addressing the great colonnade of trees that structure the public realm. Here public and private are both accommodated in a respectful dialogue; and adherence to the conventions of street, sidewalks, front yard, porch, and public rooms is precisely what allows invention and private variation.

Unfortunately, the very characteristics that give the small American town its positive qualities are also those that make it extremely vulnerable. Endless nuance and interpretation are possible as long as the delicate balance of solids and voids is maintained. But the Neoclassical system is inherently biased in favor of the private icon, so balance is doubly dependent on maintenance of the public realm, the underpinnings of which are still, however faint, those of the classical structure of space—of street and square. What is not defined by the buildings must be completed by the trees, and slight weakening of either element can result in serious erosion of the system. The devastation caused by Dutch Elm disease and its professional equivalent, urban renewal, just after mid-century was rivaled in the extent of damage only by modern architecture.

Growing slowly, quietly maturing, modern architecture in America was like a time bomb planted during the Enlightenment, armed during the 1920s, and set to explode after World War II. And explode it did. Indeed, in some ways it is difficult to believe that America was not the theater of World War II. Even the damage caused by Dutch Elm disease seems minor when compared to that caused by architects and city planners who, in a spirit of postwar optimism, began to dismantle American towns and cities with reckless abandon. Urban renewal prepared the way by removing significant portions of existing urban fabric, but modern architecture delivered the coup de grâce with its insistence on the mute object and its rejection of the convention of street and square. Neoclassicism may have carried with it traces of the Renaissance in the faint structure of its urban space, but with modern architecture those traces—the public component of urbanism—disappeared as both idea and fact.11

The triumph of modern architecture in postwar America thus marks the last stage of a long transformation from a kind of tyranny of the public realm in the heyday of Louis XIV at Versailles to an equally demanding tyranny of the private realm in postwar Houston, or Los Angeles, or St. Louis. And if one cherishes the advance of individual liberty after the ancien régime, then, equally, one just as readily laments the passage of civic responsibility and the urban forms that express and promote it. Ideally (as in the ville radieuse), the city of modern architecture was to be totally public. As commandeered by the Americans, however, it became totally private. Even wealth, which before had carried civic responsibility, became private indulgence at all levels of society. Everything now appears to have been cheapened and made more superficial in the head optimism following World War II: the plan voisin before the war begat Boston's Prudential Center after the war; Radburn before the war became Levittown after the war. And to compare the serpentine streets and ranch houses of postwar suburbia to the eloquence of Elm Street would be unbearable if it did not again illustrate that it was in America in the third quarter of the twentieth century that the public realm finally collapsed—formally, socially, psychologically.  From Versailles to Berkeley took approximately three hundred years; small wonder there is confusion in the ranks of the architectural profession.