When a celebrated man is 102 years old, the obituaries are mostly ready to go. Appreciations of the architect I.M. Pei, who died in May, flickered across social media. They left the impression that Pei had been important, and old. They featured tasteful images of his shiniest and most photogenic work, as well as of photogenic Pei himself—remembered as a charmer. He smiled for photos. Among architects, who tend to self-seriousness and try to project power by appearing solemn at all times, Pei was almost unique in that respect.
The photogenic work included the 1984–1989 glass entrance pyramid and underground renovation at the Louvre Museum in Paris; its pyramids-in-a-stone-plaza forerunner, the 1968–1978 East Wing of the National Gallery in Washington DC; and the 1963–1979 John F. Kennedy presidential library, a glass and stone cube on Boston Harbor. These monumental cultural projects were Pei’s most publicized buildings, and their images did communicate something of what distinguished Pei as a designer. He was sometimes audacious but never gratuitous. His forms were simple but not simplistic, resolving complicated problems into complex solutions derived from squares, circles, triangles, and all their intersections and permutations. His signature materials were robust and tightly edited: structural cast concrete, stone veneers, and very big windows and skylights supported by steel space-frames (including that infamously allegorical glass ceiling at New York City’s Javits Center, primarily the work of Pei’s onetime associate James Ingo Freed). But all this is only the second half of the story. It may be that the buildings that will eventually matter most are the ones Pei did when he was starting out.
Pei’s stature among architects is hard to convey. However visually entertaining their work, the likely legacies of other American so-called starchitects shrink—some to triviality—beside the decades of modern design that Ieoh Ming Pei produced, from his earliest built work in 1948 to his last project sixty years later. By the time he co-designed the Museum of Islamic Art in Doha in 2008, he was among the last practicing students of a teacher from the Bauhaus. Walter Gropius taught, and later taught beside, Pei at the Harvard Graduate School of Design. Gropius had arrived there in 1937. Pei, the scion of a Suzhou banking family who left Shanghai in 1935 for undergraduate study at Penn and MIT, arrived at Harvard in 1942. His classmates included Paul Rudolph and Philip Johnson.
Modernism in design was not conceived as a style but a refutation of the historicist stylistic revivals that had defined the previous four centuries of architecture in the developing West, forts and factories notwithstanding. Early modern buildings valued user experience over signifying appearance, which enabled them to apply new efficiencies in industrial engineering and standardized manufacturing. A particular modernism was exported by many of the Bauhaus teachers as they became refugees and émigrés—especially in the United States, where it was reassimilated into both high and popular culture. In the doctrinaire postwar period, the modern in design became, for a quarter-century or so, the house style for the midcentury American establishment.
One member of that establishment was New York City real estate developer William Zeckendorf. In an unusual maneuver, Zeckendorf plucked Pei out of Harvard to become an in-house designer at his company Webb and Knapp. Pei would bring along his star student at Harvard, Henry N. Cobb, who became a founding partner in the firm that bore Pei’s name. In varying degrees of association with Zeckendorf, Pei produced an epochal run of apartment buildings and multi-unit residential projects, many of them large-scale and low-cost, in which the sophistication and durability built into modern design practice was applied most directly to the routines and rituals of everyday life. There was 1957–62’s thousand-unit Kips Bay Plaza between 1st and 2nd Avenues in Manhattan; 1957–64’s fifty-six-acre Society Hill development in Philadelphia; 1964’s high-rise Washington Plaza in Pittsburgh; and 1971’s forty-story Harbor Towers affordable housing development in Boston, the work of Cobb. There were 1963’s University Apartments in Chicago’s Hyde Park, two parallel low-lying ten-story superblocks with surrounding townhouses designed with Chicago architect Harry Weese. At a daintier scale and to a higher degree of detailed resolution, there was 1960–1966’s University Village housing project in New York’s Greenwich Village on Houston Street, now known as Silver Towers: three miniature skyscrapers accommodating, at least initially, lower-income Village residents alongside New York University faculty.
It is these buildings that are the other half of Pei’s story. Kips Bay innovated an alternative to the grim typology that had characterized lower-cost housing developments in New York since the war: small windows punched through brick veneer slapped on a steel frame. Instead, a cast-in-place concrete system turned the facade itself into the structure. This enabled generous picture windows within that cast-concrete grid, freeing up the interiors for flexible floorplans. The complex consisted of two relatively low, very long, and surprisingly thin superblocks, whose attenuated proportions minimized their visual mass and optimized residents’ access to light, air, and views.
In what can be interpreted as at least an aesthetic critique of the so-called urban renewal mass demolitions with which it and several contemporary Pei projects coincided, Society Hill preserved a portion of the acres of low-lying historical urban fabric that it encompassed, and reproduced—with bricks and arches but without pastiche—the human-scaled row houses so characteristic of Philadelphia’s built environment. The fine detailing at Society Hill—built-in benches, platforms, steps, low walls, and other hardscaping that choreographed public and semi-public space—was all of a piece with the project’s monumental design moves, including the tall towers at the edge of the complex.
By moving structure and texture in cast concrete to the facades—as at Kips Bay—and by registering the human scale with the deeply articulated floor-to-ceiling windows at each level, the Society Hill towers were the opposite of the then-prevailing Miesian vision of a skyscraper as a sublimely scaleless volume, shiny in glass skin and steel bones. Whenever Mies had multiple towers to arrange, he would place multiple towers in pinwheel arrangements, giving you the uneasy feeling that they were about to slide past and away from each other. Pei instead composes his towers as if they were gathered people: two of the three towers at Society Hill are parallel and slightly offset, like a couple; the third is turned sideways to address them. At University Village, a long cast-concrete bench-height slab completes the fourth side of a notional courtyard defined by the three slender towers: because it resembles them in scale and material, when you sit on it you feel integrated into the place. Again, the cast concrete facades provide structure, with the deep mullions between inset picture windows providing shade and, at ground level, resolving into multi-level column-like fins that hint at a classical elegance. The towers were positioned on the site so that, optimized for perspectives from heavily-trafficked Houston Street to the south, one tower would always be hiding behind the other two—leaving as much of the sky open as they could.
Many modern architects—epitomized by Pei’s own professed favorite architect Louis Kahn1—tended toward a gnomic, obscurantist discourse, or toward performative mannerisms of ostensible genius—and so toward inimitability and irreproducibility. Pei, on the other hand, tended toward candor—something about his ability to smile, even if disarmingly, converged with his ability to concede uncertainty and to think aloud. That’s as rare, among architects, as smiling. Pei’s professional persona amplified the way in which his projects, in their lucidity and legibility, in their visible method, became for generations of modern designers a kind of open-source code.
The operation of this code within the firm of which Pei and Cobb were founding partners in 1955 (first known as I.M. Pei & Associates and later as Pei Cobb Freed & Partners) served as a kind of microcosm of how its transmission and application later worked in the wider world of the profession. Today, Cobb sometimes gets called, in a normalizing but anachronistic way, a project architect for buildings produced by his and Pei’s firm. But the two design principals and founding partners never collaborated much within the now standard organizational system by which that role is defined. Cobb and Pei instead worked independently, in sometimes wary parallel. “For the most part, starting in the mid-fifties, we worked separately,” Cobb recalled in a 2015 oral history interview. “We shared staff, even though we didn’t engage each other a lot in the work we were doing. . . . I.M. wasn’t into that sort of thing either. Also, we were busy.”
Cobb designed the John Hancock Tower, on Copley Square in Boston, Massachusetts. This office skyscraper, the crowning achievement of his career, epitomized Cobb’s signature concept of a “contingent architecture” that engaged its neighbors not with wan imitation but incisive geometrical reflections, deflections, and proportions. At the Hancock, Cobb designed an astonishingly slender parallelogram volume that (as with Pei’s work at Kips Bay and University Towers, though clad not in cast concrete but tinted reflective glass) maximized the sky visible from critical points in the adjacent square and city. And, for all its sublime scale, Cobb’s design for the Hancock responded with subtlety to the angles of city streets as well as the scale and geometry of nearby historic landmarks like H.H. Richardson’s Trinity Church and McKim Mead & White’s Public Library.
The Hancock Tower “was the center of my life from ’67 to ’76,” Cobb told his interviewer, which was in part because he took up the work under unusual circumstances. “I.M. said, ‘I’m not doing it,’” Cobb recalled, “‘but this is your city. If you want to, you should do it.’” Cobb legendarily drafted the Hancock over a two-week stretch in September of 1967. He was able to work at such speed, I think, because he was processing and perfecting that open-source code. The phrase “corporate architecture” has long and sometimes rightly been a slight, implying the soullessness or calculation that attaches to, say, corporate rock and roll. But because architectural design requires breadth of knowledge and lateral thinking between arcane specialties, singular achievements in practice—like Cobb’s drawing the Hancock in a fortnight—happen precisely within the busy matrix of collective intelligence that a corporate endeavor, whatever else might attach to the word “corporation,” can provide.
Vernacularity has something to do with anonymity—distinct from the industrialized standardizations that were another aspect of modern design. “The modern vernacular can only take you so far,” Pei said to architectural critic Paul Goldberger in a 1979 New York Times profile. “I believe it will still be valid for the sort of buildings that have a degree of anonymity to them—office buildings, for example. The way Mies van der Rohe pointed is still valid. I think my schooling with Webb and Knapp may have set me back philosophically. It made me a pragmatist certainly.” The context for this melancholy observation was the so-called postmodern movement in design, which through formal and rhetorical maneuvers of insistent stylishness, deliberately visible artifice, polemical impracticality, and scholastic affect rendered Pei’s modernism temporarily unfashionable. The modern buildings of Mies pointed toward work that would be so highly regularized by recurring components as to be anonymized, even as their own strenuous idiosyncrasies made their author unmistakable. But more genuinely functional, pragmatic, and commonplace buildings, especially those intentionally designed to be workhorse background buildings and not showpieces—those in a modern vernacular—tend toward a kind of effacement.
The fate of Pei’s finest New York City project offers a cautionary tale of how the modernism of self-effacement or architectural negative capability can get literally erased. Everyone knows about the most glamorous midcentury-modernist terminal at New York City’s John F. Kennedy airport: Eero Saarinen’s 1962 TWA Flight Center, also known as Terminal 5. Saarinen, who died young at 51, was a quicksilver prodigy who could do anything in any style of his era, and did. His terminal was a small but expressive cast concrete shape, as involute as a hothouse flower. It was beautiful and inflexible. After its decommissioning as a terminal in 2001, the Port Authority of New York and New Jersey conserved the structure for potential adaptive reuse, even as its future remained in doubt. Eventually, the same intentionally iconic Jet Age glamor that rendered it not especially adaptable to contemporary requirements by the time JetBlue took it over in 2008 also meant that it couldn’t respectably be demolished—especially by an airline whose own branding emphasized stylishness. Instead, half of the building was eliminated (the flight wings, departure lounges, and portions of the circulation tubes), and a tidied-up version of what remained was encased within two wings of a sleek, tasteful hotel, for which it serves as a kind of lounge, albeit one without the thrill of departure.
The finest building at JFK, however, was the one next door. That was Pei’s 1969 structure for National Airlines, called at the time the “Sundrome” and also known as Terminal 6. It was the equal and opposite of Saarinen’s work. Deliberately restrained as a background building to complement TWA’s exuberance, Pei’s terminal seemed, at first glance, only like a big box. But the more you looked, the more you saw. The overall austerity gave way to meticulous and precise detailing at every scale: the joints and connections were worthy of aerospace; the subtle textures and pale colors all resolved to a visual equivalent of the calm you hope to hear in the voice of your pilot. The capacious building was, in essence, just a vast and deep steel truss roof supported, as if on fingertips, by gracile steel points atop sixteen big cast concrete cylindrical columns inset from its perimeter. Because this constituted the entire structural system, as well as water drainage within the columns themselves, all the walls could be glass, inset in turn just behind the columns to create a shaded and sheltered threshold. In detailing since made ubiquitous in Apple stores but pioneered here, the glass walls had glass structural mullions instead of conventional metal frames. In a fulfillment of a longstanding modern dream, this resulted in a radical flow of the eye—and perhaps of the spirit—between inside and out.
While Saarinen’s midcentury modern building directed your attention toward itself, Pei’s vernacular modern building directed your attention—so rightly for an airport—to the sky all around it2. The round columns and big roof evoked—just—something of a temple, something holy if you needed it to be. But rationally, these features enabled the interior to be flexible and adaptively reused over time—as it was, by successive airlines including JetBlue itself between 2000 and 2008. Complex circulation sequences could be choreographed under that broad ceiling, some thirty or forty feet overhead, that made the whole terminal feel like one big room, open on all sides to air and light. Terminal 6 pioneered a particular arrangement of arrivals and departures levels and locations that has since become standard practice. The structure’s rational layout and profile suggested that the building could be straightforwardly extended along its long axis, conceivably doubling airside gates.
In 2011, JetBlue caused Terminal 6 to be demolished. This can be lamented as a failure of the same stewardship shown to Saarinen’s building. It was also an ecological folly: 80 percent of the systematic energy of any given building is generally concerned with its original material extraction, production, transportation, and construction; and only 20 percent with its subsequent lifetime operations. The airline constructed its own new terminal, at greater cost in carbon and everything else, a triangular structure with metal cladding and lots of drywall, cramped between the tarmac and the remainder of Saarinen’s building.
I don’t know why they did that. Maybe it was some basic distaste for the old. Or some prideful disinclination to inherit some older airline’s hand-me-down clothes, however exquisitely suited to retailoring. Maybe it was some overstimulated impatience with the seeming dullness of the design restraint typical of Cobb and Pei. Or some obdurate dereliction of the duty of adaption during the climate catastrophe that demands so much of this quality from us. Maybe there were technical intricacies that made demolition seem irresistibly cheaper—or more satisfying, in the way of a boxer’s knockout punch. The labor required to abate the asbestos and lead paint sometimes found in such vintage buildings can also factor into these decisions—whether or not that was so for Pei’s Terminal 6, those materials were successfully removed during the semi-preservation of Saarinen’s Terminal 5.
The current Jet Blue terminal has about double the number of gates as the old Terminal 6. But the airline plans to extend its new terminal sideways, back over onto the former site of Pei’s building, now mostly a parking lot. Why didn’t they just keep Terminal 6? Back in 2010, David Barger, the airline’s president, addressed Cobb, who had questioned the building’s destruction. “While I share your passion for classic terminal designs,” Barger wrote in reply to Cobb’s dispassionate case for continued conservation and use of a sound and serviceable structure, “I have concluded the time has passed for the . . . building to serve any functional purpose.” Barger promised a “permanent display of . . . photographs and other architectural artifacts so future generations can continue to appreciate the beauty of Terminal 6.” Cobb’s case had not been about beauty as some kind of superfluity, like an ornament. “Spaciousness. Generosity. Dignity. Calm. Order. Precision. Restraint,” Cobb had said, in an echo of Baudelaire’s talismanic modern poem, “L’Invitation au Voyage,” “these are qualities that people who are having to deal with the stresses of travel appreciate.” Substitute life for travel and you have a retroactive manifesto for all of vernacular modernism—and all of Cobb’s and Pei’s work.
I don’t know how the same culture that brought forth a building like Terminal 6 in 1968 became the one that so very casually demolished it only forty years later. Although it’s noteworthy that Pei lived for over a century, it matters more that he sustained a modern vernacular for fifty years. When the word “modern” first came into ordinary English usage, it meant ordinary, commonplace, and everyday: “They say miracles are past,” goes the line in All’s Well that Ends Well, “and we have our philosophical persons to make modern and familiar, things supernatural and causeless.” Because Pei’s early work was modern in just this sense, it should become his most enduring and influential legacy. When Jackie Kennedy interviewed Pei for the presidential library job in 1963, she visited him in a small office still smelling of new white paint. When she praised the fresh flowers on display and asked if the office always had them, he famously confessed that he’d just put them there for her that morning. “We had nothing to show her,” he would recall decades later. “My work had been unglamorous.”