A Place of My Own

by Michael Pollan

  The July afternoon Joe and I first heard about the engineer—Charlie having accepted Joe’s dare-you invitation to help us cut and nail rafters—the architect came in for a lot of kidding. The day was very much Joe’s. Cutting rafters is a complicated and unforgiving procedure, and Joe had shown up on time and armed with a detailed sketch indicating the precise location and angle of each of the four cuts each rafter needed: the ridge and tail cuts at each end—parallel to one another at a forty-five-degree angle to the rafter’s edge—and the heel and seat cuts that form the “bird’s mouth” where the rafter engages the top of the wall—a rectangular notch that in our case had to have a slightly different depth, or heel cut, on each rafter to account for the fact that our two side walls were not precisely parallel.

  Joe had clearly done his homework, could even spout the formula for determining the length of a rafter: , in which the rise is the height of the gable and the run is the horizontal distance covered by each rafter, or one half the width of the building. For his part, Charlie was feeling somewhat deflated after a punishing meeting with a client, and seemed in no shape to mix it up with a cocky carpenter who’d come equipped with enough geometry to frame a roof single-handed and who didn’t see much point in architects to begin with.

  “Charlie, just explain this to me,” Joe began, gesturing with his big carpenter’s square. “The building is only eight feet by thirteen, correct? It gets a roof that’s framed with four-by-eight rafters and a ridge beam ten inches thick. Plus you’re calling for two collar ties and a pair of king posts, all of which you want us to dowel together. So tell me: How can we possibly have anything to worry about structurally? This building’s been designed for a three-hundred-year storm!”

  Charlie managed a wan smile. He explained, somewhat sheepishly, that he’d needed to check with the engineer on the dimensions and spacing of the straps—the strips of wooden lath that run perpendicularly across the rafters to give us something to nail our shingles to. The fact that our rafters are a full thirty inches apart meant the lath would have to span an unusually great distance. Charlie had wanted to know the minimum dimensions he could safely spec these pieces, since the underside of the roof was to be entirely exposed. If the straps were too heavy, they’d wreck the delicate, rhythmic effect he was aiming for in the ceiling, which he’d told me was going to look something like the inside of a basket or the hull of a wooden boat. Yet if the lath were too light, it was liable to deflect, or bend, under stress.

  While Charlie was working with the engineer to determine the dimension of the straps, he figured it couldn’t hurt to have him run the rest of the calculations on the roof. Charlie explained that any time you have an open, “cathedral” ceiling with no attic, there are special structural problems to solve. As gravity exerts a downward pressure on a roof, the rafters in turn want to push the walls outward, a force that in a traditional structure is countered by the ceiling joists, which tie each pair of rafters together at the bottom, joining them in a taut triangle. But when the living space reaches directly up under the roof, these joists are eliminated, so either the walls have to be sturdy enough to withstand the outward thrust of the rafters or an occasional cross-tie beam must be provided to counteract it. It was Frank Lloyd Wright who pioneered such a ceiling, and it may well have been the novel structural and insulating problems it raised that caused some of his roofs to leak. Following Wright’s example, Charlie wanted to give the interior of my building a pronounced sense of “roofness,” one of those instances in architecture where expressing a structure seems, ironically, to complicate its construction.

  “Anyway, you’ll be happy to hear we’ve got nothing to worry about—the two cross-ties take care of our lateral stresses, and the king posts cut the weight carried by the ridge pole almost in half. And as far as those dowels are concerned, don’t forget that in a storm you have upward forces working on a roof, too.”

  Joe and I both laughed; nothing about my building seemed in danger of blowing away. A few weeks before, as the frame was taking shape, I’d remarked to Charlie about how very heavy it looked. “But it’s not meant to be light,” Charlie had protested. “This is your study, your library—it’s an institution!” When I passed that one on to Joe, a look of concern swept over his face: “Mike, don’t you think there’s another kind of institution we should be talking to Charlie about?”

  But there was nothing funny about the issue as far as Charlie was concerned. Charlie’s nightmares, I knew, featured collapsing roofs and deflecting cantilevers. No doubt the fact that this particular design was being built by a crew consisting 50 percent of me made him even more nervous than usual. Out of the blue, Charlie would phone to reassure himself I was using galvanized nails in the frame; he’d heard about a house on Cape Cod that had simply crumpled to the ground one day, the salt air having rusted its common nails to dust. No doubt such worries disturb the sleep of all architects to one degree or another. When the massive concrete cantilevers of Fallingwater were being poured, Frank Lloyd Wright, delirious with fever at the time, was heard to mumble, “Too heavy! Too heavy!”*

  “You two can laugh,” Charlie said, “but I’m the one who’s ultimately responsible, and it makes me sleep better knowing that an engineer has run all the calculations.” He launched into a story he’d already told me twice before, about an opening-day bridge collapse in Tacoma. Joe chimed in with a few horror stories of his own, the sort of thing I imagine you could get your fill of in the bar at a convention of structural engineers, and by the time he got around to a fatal hotel atrium collapse in Kansas City, we were all feeling pretty good about our roof, about just how beefy it was going to be. While we talked, the three of us were lifting the chunky rafters into place, lining them up over the fin walls as best we could (our frame being, in Joe’s cheerful new formulation, “too hip to be square”) and then toe-nailing them to the ridge beam above and wall plate below with (galvanized) twelve-penny nails almost as fat as pencils. We had all eight rafters securely in place before Charlie had to drive back to Cambridge, and the completed roof frame looked for all the world like a gigantic rib cage, its great fir bones wrapping themselves around a sheltered heart of space. Add to this skeleton a skin of cedar shingles, and you had the very kind of place where a body wouldn’t mind riding out the storm of the century.

  It seems difficult if not impossible to avoid figurative language when talking about roofs, they’re so evocative, so much more than the sum of their timbers and shingles and nails. To creatures who depend on them for their survival, it is perhaps inevitable that roofs are symbols of shelter as well as shelters themselves. Seen from afar or in a painting or movie, roofs also symbolize us—our presence in a landscape. Of course people have attached innumerable other meanings to roofs as well, and many of these meanings have changed over time. The traditional gable, for example, meant something very different after modernism than it did before.

  Many of the important battles over style in architectural history can be seen as battles over roof types: the Gothic arch versus the classical pediment, the Greek Revival gable versus the Colonial saltbox, the international style flattop versus all of history’s pitched roofs. In this century, the pitched roof became the most hotly contested symbol in all of architecture. Nothing did more to define modernist architecture than its adoption of the flat roof—and nothing did more to define postmodernism than its resurrection of the gable. Since then, architecture’s avant-garde has sought to explode the very idea of a stable, dependable roof, violently “deconstructing” both the gable and the flattop. But the twentieth-century argument about roofs turns out to be about a lot more than that: it’s really an argument about the very nature of architectural meaning, which seems to have undergone a thorough transformation in the last few years. I’ve come to think this transformation holds a clue to the disappearance of the old idea that architecture was somehow grounded in nature, as well as to the subsequent rise of the kind of literary architecture I had found in the pages of Pro
gressive Architecture. You can get a good view of these developments up on the roof.

  “Starting from zero” was the rallying cry of modern architecture, and for the roof that meant banishing the gable, which the modern movement took as a key symbol of the architectural past—of everything musty and old and sentimental. Arguably the pitched roof (of which the gable is the most basic form) is architecture’s first and most important convention—wasn’t that the point of all those primitive-hut tales?—and under the modernist dispensation all conventions were to be tested against the standard of pure rationalism and function. The demonstrable fact that the pitched roof is supremely functional suggests that modernist rationalism sometimes took a backseat to modernist iconoclasm.

  One of the aims of modern architecture was to rid the sprawling, many-gabled Victorian house of its many ghosts, all the historical encumbrances and psychological baggage that kept us from stepping out into the cleansing light and fresh air of the new century. In this sense modernist architecture was a therapeutic program. “If we eliminate from our hearts and minds all dead concepts in regard to the house,” Le Corbusier wrote, “we shall arrive at the House Machine…healthy (and morally so, too) and beautiful in the same way that the working tools and instruments that accompany our existence are beautiful.” In the modern view, the pitched roof was itself a “dead concept,” but equally unhealthy were all those other dead concepts that got stored underneath the gable, in the attic. For there is where the ghosts of our past reside: the bric-a-brac and mementos that a lifetime collects; the love letters, photographs, and memories that clutter an attic and threaten to bear us back in time.

  Modernism’s program of psychological hygiene sought to rationalize everything about the house, to exorcise its ghosts and render it as unhaunted and transparent as a machine. Glass would supply the transparency, but it was the elimination of the pitched roof and its attic (along with the depths of the basement) that promised to vanquish the dead hand of the past, thereby helping to streamline the house’s occupants for the challenge of the new age. Of course there were some who protested the wholesale housecleaning: Bachelard’s Poetics of Space is an impassioned celebration of attics and basements and all those irrational but nevertheless powerfully symbolic places that modernism had banished from the house. People cannot dream in a “geometric cube,” Bachelard complained. But then, that was the point. The irrational symbolic power of things like roofs and attics is precisely what made them so objectionable.

  It’s hard for us to imagine now just how powerful the taboo against gabled roofs in architecture was until very recently. I say “in architecture,” because of course ordinary home buyers and commercial developers never really surrendered their attachment to pitched roofs, though modernism did manage to diminish the pitches on the vernacular roof, working like some powerful g-force to flatten the steep Victorian gable into the shallow hipped roofs found atop millions of suburban ranches. The architectural historian Vincent Scully writes in The Shingle Style Today that when he set out to build a house for himself in New Haven in 1950, “the model of reality in which I was imprisoned”—he had just completed his dissertation—“made it unthinkable to employ anything other than a flat roof…”

  A dozen years later Robert Venturi single-handedly cracked open this model of reality and freed all the architects who’d been trapped inside it. He built a house for his mother in Chestnut Hill, Pennsylvania, that featured a gigantic, emphatic, in-your-face gable. The Vanna Venturi house, which was completed in 1964, proved to be the opening shot in architecture’s postmodern revolution—“the biggest small building of the second half of the twentieth century,” Scully has called it. Venturi has written that in 1964, even though there were a few single-slope shed roofs creeping back into architecture, the very act of designing a façade “where two slopes met to form a pediment contravened a taboo.” At the time, his big front gable was “both too familiar and too old-fashioned, too rare and too outrageous.”

  What a revealing way to put it! For had Venturi’s gable been only “too familiar and too old-fashioned,” it would not have qualified as modern architecture. Instead of catching the eye of the Vincent Scullys of the world, the Vanna Venturi house would probably have been dismissed as revivalism—as something reactionary and nostalgic—or, worse, simply overlooked as a naïve vernacular building; after all, there had to have been a hundred thousand other pitched roofs erected in 1964. To count as modern architecture, Venturi’s building had to be “rare and outrageous,” too, and that it most certainly was.

  For as anyone with eyes could see, there was something very peculiar about this particular gable. To begin with, it was on the long side of the house, which made it seem way too big—as if it had been exaggerated for effect, which of course it had. Then, right up at the top where the two slopes were supposed to meet, there was this odd space, a kind of gap tooth through which you could make out an oversized chimney rising several feet back from the façade. The gap made it appear as though there were nothing behind the façade; it flattened the gable out and made the whole house look more like a cardboard model of a house than a real, three-dimensional building. Venturi wanted to use a gable (what better ammunition for his assault on modernism?), but not one that could ever be mistaken for an “old-fashioned” gable. So he gave his gable a sharp ironic twist, exaggerating it and hollowing it out until it looked more like a comment on a gable than the thing itself. As Venturi himself puts it, “the pediment used in this fashion becomes a sign, a kind of representation…”

  Venturi’s use of the word “sign” to describe his roof, rather than, say, “symbol,” is significant. Arguably his house in Chestnut Hill invented a whole new voice in which buildings might speak, and the shift from architectural symbols to signs is a key to that transformation. In using the word “sign,” Venturi is drawing on the vocabulary of semiology, which holds that all cultural activities can be profitably read as systems of signs that are structured like languages. Semiologists, and structuralists after them, borrowed their terms from the turn-of-the-century Swiss linguist Ferdinand de Saussure, whose theories have by now reached far beyond linguistics to influence literary studies, the social sciences, art criticism, and even, thanks in no small part to Robert Venturi, modern architecture.

  The relationship of a linguistic sign to the thing it signifies, Saussure maintained, is accidental; signs get their meanings not from the things in the world they refer to, but from the system of signs of which they are a part. That is why a certain combination of letters—ng is an often-cited example—can mean something in one language while remaining completely opaque in another. It follows that the choice of any sign is completely arbitrary, purely a social convention. In Learning from Las Vegas, Venturi’s influential study of architectural meaning, a book that is steeped in semiology, he offers his own example of the “arbitrariness of the signifier”: In the system of Chinese roads signs, green means “stop,” and red means “go.” Venturi encouraged architects to think of gables and columns and arches as signs too, elements as conventional as the letter combination ng or a green stop sign on a Chinese road.

  In the years since Venturi built his mother’s house and published his two seminal manifestos, it has become the conventional wisdom, at least among architecture’s avant-garde, that architecture is a kind of language and that all its various elements—the gables and arches and columns, the axes and patterns of fenestration and materials—are best understood as conventions having less to do with the nature of the world or the human body or even the facts of construction than with the sign system, or language, of architecture itself. This was something radically new. Even modernists had paid pitched roofs and all the other symbols they detested the compliment of taking them seriously, treating them as if they actually had some weight in the world beyond architecture. Also quite new was the divorce Venturi was proposing between the imagery of a building and its underlying structure, a relationship upon which the modernists had sought to grou
nd a whole aesthetic. By redefining a work of architecture as a “decorated shed”—an indifferent structure with signs on it—Venturi had driven a wedge between the meaning and the making of buildings.

  The Vanna Venturi house was the first work of architecture built on the foundation of the new linguistic metaphor. Like the letter combination ng, the various elements of Venturi’s house—its gable and windows, the arch over its entranceway—are meant to be understood chiefly in terms of the language of architecture. In fact Venturi wants to make sure we look no further: he deliberately designed the house to resemble a model so that it would be, in his words, “not real so much as denotative.” The weightless, cardboard look, which has become a hallmark of postmodern architecture, is a way of announcing that the concrete Here of this building is less important than the abstract There of its signification; for Venturi and the countless postmodernists who followed his revolutionary example, the scrim of representation matters more than the reality behind it.

  Thus the thin, abstract gable on the Vanna Venturi house has less to do with the world in which it rains and snows than with the increasingly hermetic world of architecture, which is in fact its true mise-en-scène. The space the building occupies is as much the space of images and information—of “discourse”—as it is the space of experience and place and the weather. Though its roof may well keep the rain off Mrs. Venturi’s head, her son is anxious that we regard it primarily as a communications device, a sign referring us to, and commenting upon, other roofs in architecture—the pediments of the Greek temple; the long, dramatic gable on McKim, Mead, and White’s shingle-style Low house in Bristol, Rhode Island; and, of course, every flattop in the modernist canon.