Early in the 20th century the Norwegian Roald Amundsen and Englishman Robert Falcon Scott competed to satisfy a pressing cultural desire in the developed nations of the West: to reach the south pole, first. Each explorer famously promoted a different beast to haul their expedition’s burden over the long, arduous icy treks. Amundsen, based on experience, opted for dogs. Scott, on only his second trip to the Antarctic, committed primarily to ponies. These ended up being largely ineffective, and played a role in the chain of events that led to Scott’s calamitous end.
Competitors almost always develop variations in method. Students will actually pursue bad ideas, just to avoid repeating someone else’s strategy.
Though ponies had little track record for the extremes at hand — and Scott was strongly advised to use only dogs by Fridtjof Nansen, another seasoned Norwegian explorer — Scott’s selection was not inherently surprising. It would hardly have made sense for him to use the same method as Amundsen. The expeditions were in competition; moreover, the two men were contemporaries, and aware of each other. If competitors, knowing of each other’s effort, seek to solve the same problem simultaneously, they almost always develop variations in method. Testing alternatives in competition is an effective way to determine if one solution is better, despite the unfortunate (for Scott, in this case) failures it sometimes entails. Anyone who teaches design studio is familiar with this phenomenon: students will actually pursue bad ideas, just to avoid repeating someone else’s strategy.
If you reverse the logic, dogs and ponies in this instance have more in common conceptually — both are being used to try to accomplish the same thing — than their surface differences suggest. Here, in architecture, is a similar situation: There is a powerful conceptual link between historicist buildings, on the one hand, and landform buildings on the other, even though that second hand seems attached to someone else’s arm. The pressing cultural desire (like that need to claim the bottom of the world) binding these buildings’ wildly differing means? Both seek to stabilize — to protect, to uphold, to clarify — the experience of the landscapes in which they are sited.
Stabilize suggests the underlying cultural desire better than any other term. As I’ve argued in earlier essays in this series, both the Post-Modern revolution in architecture and most of the After-Modern work that has followed can be understood as an attempt on the part of architecture to favor the primacy of a host landscape rather than the primacy of an invading building in order to protect place — I hate that word, but we’re stuck with it — against Modern place-less-ness. 1 All of the architects competing to resolve this pressing need stake out a method for how a building must be conceived and constituted so it can best serve its landscape. This is what links the explorations of architects like Robert Stern, Allan Greenberg, or Andres Duany (among the more well-known of what is in fact a substantial portion of the design marketplace) to architects like Steven Holl, Marion Weiss and Michael Manfredi, Antoine Predock, and, of late, Peter Eisenman.
But, as that range of architects suggests, there’s not much agreement about method, only competing strategies occurring across a spectrum from dog to pony (many architects work somewhere in between). The landform strategies are at one extreme, historical simulation — which for years has found its primary expression in the New Urbanists and their offshoots — at the other. Each group gathers information from different aspects of site and context; and each differs in their belief about how architectural form carries landscape-related content. So, which of these approaches is the dog, and which the pony? It’s worth knowing. Scott, after all, froze to death.
The arguments for landform architecture — in which a building is a free physical entity, unburdened by expectations of building-ness, which can be explicitly shaped by mapping the factors and forces that have configured the site (either in its immediate confines, or more largely within its home landscape) — were set out in detail in previous essays. Snøhetta’s Norwegian National Opera and Ballet, modeled to represent a glacial ice field — or perhaps better, since you can clamber all over it, to enact and embody one — is a lovely example, in part because the conceit becomes literal in the winter, when the frozen harbor extends the snowfield (the public plaza) of the building. The logic is self-evident: is there anything better than a glacier? In a time of place-less-ness, a specific landscape (especially a natural landscape) is a source of meaning, value, and stability; the building, so the argument goes, is landscape; therefore the building has meaning and value, and further stabilizes its place.
The criticism of this approach follows circular arguments first posed a century ago by Adolph Loos. A building, (like Amundsen’s tent in the Antarctic) is not landscape; it’s a specific part of landscape — a building, not a glacier — with a specific role to play within the limits of what a building is in landscape. From the point of view of landscape (as opposed to, say, structure or program) most buildings belong to a class of perceived object that almost never enters interpretative consciousness. A building manages this by correctly triangulating landscape-specific expectations regarding constructed form, program, and site. If it does not operate within these limits — which many mapped landform buildings ultimately do not — it disrupts, destabilizing the sense of consistency in the experience of a landscape. (Loos makes an exception to this rule for tombs and monuments; the Opera and Ballet is arguably one of these.)
Here I want to concentrate on the other end of the spectrum, on the argument for design methods that use historical simulacra, one implication of Loos’s arguments. In the past generation these techniques arose from the writings of Robert Venturi, Aldo Rossi, Christopher Alexander, and Christian Norberg-Schulz, among others. Here is their argument, bluntly put. In relating to the world over time — and, in so doing, establishing stable pre-Modern landscapes — buildings have attained their form and appearance through behaving by certain overarching rules (Alexander sought to codify these in A Pattern Language). These rules are modified by local histories of form (as Rossi noted in Architecture of the City). A close study and application of the rules in designing a new building will protect the existing and necessarily valued landscape into which it will soon be placed (those two again, plus Venturi and Charles Moore). And we need to protect that landscape because place is the key to a sense of belonging in an increasingly place-less world (Norberg-Schulz). 2
Delirious New York is, counterintuitively, the most brilliant manifestation of historicist thinking.
In retrospect it certainly sounds plodding. But the mindset casts a profound, complex, and seductive spell. How seductive? Rem Koolhaas’s masterpiece, Delirious New York, roughly contemporary with the books mentioned above, is, counterintuitively, perhaps the most brilliant manifestation of this way of thinking. Look at this magic place, Manhattan; here are the remarkable rules for how buildings behave here; I’m just following those rules. Like Koolhaas — and with only the partial exception of Venturi (both in the remarkably open-minded Complexity and Contradiction, and in the increasingly and intentionally perverse buildings designed in partnership with Denise Scott Brown) — the over-arching argument was not put forth innocently. That era’s manifestos were written to suggest that the general and local rules had to be followed.
That’s because those manifestos were not merely about how to derive form. They arose from pressing (as in: we need to get to the south pole) cultural criticism that, while latent in architecture since Team X, really flourished in the writings of Jane Jacobs, William Whyte, Kevin Lynch, and others. 3 Their criticisms are arguably still valid: in the intervening years not much of a dent has been made in place-less-ness! No wonder the New Urbanists, that founding Post-Modern generation’s most astute progeny, took up the mantle. They’ve made their claim — we should all be making landscapes by means of pre-existing landscape proprieties, since the last time architecture satisfied society was when those were at work — like vengeful librarians seeking to silence the unruly while suggesting all they really want is for you to read for your own betterment. Ha!
Is the task of an architect in creating a building for a landscape like that of a musician playing an instrument in a symphony: to faithfully interpret rather than freely invent?
OK, here’s the thing though. The thesis question that links the New Urbanist goal to its method — if you seek to stabilize the identity of a landscape, and you can isolate a historical prototype for its creation within a context, why not just make buildings today as they once were made, especially if the prototype pre-dates the age of disruption (the full-on advent of the Modern era)? — is actually a fairly good question when you think of it dispassionately. Its extreme answer poses a radical possibility: that an architect’s task in making a landscape is akin to that of a musician playing an instrument in a symphony, faithfully interpreting a score rather than freely creating. The score is the landscape as it was written in the past (inviolate, like a Brahms or a Bach or a Beethoven). In any landscape, buildings may have major roles to play, but these are pre-determined by the score; only rarely is a building asked to interpret, or improvise (or embody, symbolize, enact); mostly it does its small part, ideally through the mechanism of a period instrument.
In this metaphor, the architect’s task is merely to bring the building in on cue and in tune, to allow the symphony. This requires real skill. There may be range for interpretation, but this sort of work requires suppressing the impulse to differ from the score. Imagine if every classical musician felt like interpreting or improvising or embodying in the middle of a symphony! If a stable landscape has the holistic unity of a beautiful smile, it would be like having had Peter Eisenman as your orthodontist. So, yeah, why not make buildings as they once were is a really good, really radical question. And not only because non-architects so frequently ask it — or, at least, think it!
Now, just to get this out of the way, it’s clear the Post-Modernists and the New Urbanists have not been especially hard on themselves in answering this question architecturally. Answering it well calls for a suppression of invention, a painful mandate given the freedom to experiment with form that architects earned in the Modern era. Perhaps for this reason, a lot of architects developed a have-your-cake-and-eat-it-too design method. Their buildings may begin as literal quotations, but invariably they evolve to allow the architect to comment or embroider on the prototype with the sort of self-referential formal gestures — a famous example is the goofily thin classical order on Venturi and Scott Brown’s New Castle County House — that give Post-Modern buildings, and their successors, the ironic self-awareness (a quality absent from the buildings they seek to emulate) for which they became notorious.
Today — unlike at the start of the Post-Modern — most architects rarely ask the question should we make new buildings look like old ones out loud. But they answer it constantly by default. Based simply on the overwhelming absence of examples in architectural journals and editorial websites — that is, from that part of practice deemed worth sharing — our answer is clear: No, we should not. The same holds true for education. I’m not taking much of a risk by claiming, without having researched this at all, that American architecture schools — Notre Dame is the exception that proves the rule — never ask this question of their students (and the students are too scared to ask).
These aren’t explanations, just evidence of an established point of view so closely held that a convincing justification — No, we should not make buildings as they once were made, and here is why — is remarkably hard to uncover. And because it’s never reconstructed out loud, it just comes out in the vague guilt and fascinated repulsion architects feel when they leaf through, say, Architectural Digest, its pages dripping with new (and often gorgeous) historical excess (plus one contemporary home sitting there patiently, like the designated driver). Actually, this whole essay could be summarized with this simple question: why is Digest, so proudly displayed in most dentist’s waiting rooms, hidden in so many architecture offices?
Architects have no problem replicating Modernism as a period style and bragging about it in magazines like Dwell.
Over the years I’ve only occasionally heard architects offer arguments for why they don’t — and especially why students shouldn’t — make buildings as before. One is: you have to build in the spirit of the times. I invariably discount that reasoning. It’s not an actual argument, just a thinly veiled statement of taste: architects have no problem replicating Modernism as a period style and bragging about it in magazines like Dwell. And, based on the present general and often pleasurable assault on actual reality — the range here is from Las Vegas, to digital representation, to cloning pets, to our acceptance of perpetually youthful celebrities, to Her — I’d say the spirit of the times seems to have no problem admitting full-on replication as part of its fundamental desire!
Here’s a better argument architects sometimes make against historical replication. Occasionally new programs enter the world — actually a better way of phrasing this is that new desires best resolved with buildings continue to arise. These are without historical prototype. What, for example, is the correct form for an Alzheimer’s care facility? It lacks a history of having been. But that dilemma has often been true over the long history of Western civilization, and the traditional solution was to fit the new program to an existing formal prototype, like: new Christian church fits into old Roman basilica. Thinking that there must be a precise fit of form to program is a peculiarly Modern trait, i.e., it’s part of the problematic reasoning that led to the dilemma both landform and replicant buildings seek to resolve. So there goes that line of reasoning.
Actually, the question of fit never really created substantial conceptual problems before the Modern era, or even in that interesting transitional period through the mid-20th century when practices based on historical prototypes continued to flourish. For example, my office at the University of Texas is in Paul Cret’s lovely Goldsmith Hall. Its prototype — a U-shaped mass enclosing a courtyard, the symmetrical legs housing classrooms and offices, the central bar public spaces — has been around for centuries, adapting to or adopted by one use after another. Goldsmith Hall is stunningly well made, and no less so for quoting richly at the level of detail from historic precedents. It has absorbed numerous changes — air-conditioning, more (and more varied) students, multiple curricula, the digital studio — without losing its essential ethic. Though it was completed in 1932, the space it configures in shape and shimmer could have been configured much the same way a century or three earlier.
Still, that some type-forms can absorb new programs in a landscape over time does not necessarily mean one should choose to create new buildings in an old manner. The logical causality is backwards. A corollary argument is that you can’t construct those new buildings in the old way, literally. There are new codes — requirements for systems of insulation, types of wiring, etc. — that disallow some old means, and new systems of construction that drive older ones into obsolescence. And yet: many of these changes can be readily hidden, just as Goldsmith Hall’s masonry walls cleverly hide a steel frame. It may be hard to do, but really it’s a question of wanting to do it or not.
A distinct but related argument is that it’s a bad idea to replicate because the desire for environmental sustainability suggests architects should be thinking of future models, rather than maintaining non-sustainable past models. Aside from the observation that most older (very much older) buildings necessarily arose from careful consideration of environmental performance and material accountability, one might easily counter this argument by noting that sustainable performance can be understood as entirely a technical problem, the resolution of which could have little impact on the visible world. It might be possible to make the whole world again in exactly the same form — consider that carbon-sequestering paving is now being tested in Italy — and have it perform tolerably well. Likewise another justification sometimes offered for No it is not OK to imitate the past: it’s harder to reproduce old buildings because of changes in labor and material costs. Well, that is true. Generally speaking, over the past century, adjusted material costs have gone down as a percentage of construction budgets, while adjusted labor costs have gone up. This means you cannot so easily make the handcrafted, finely detailed works most non-architects call up in their imagination when they talk longingly about old buildings.
But harder doesn’t mean wrong. You can design them that way, and you can still construct them. And the opportunities — large enough budgets to pay for the necessary labor — are less rare now than they used to be (thank you, One Percent!). Building budgets have climbed — older architects will still remember the shock of Phillip Johnson’s AT&T building being clad in stone — and the necessary levels of craft and material have become available again. So that’s hardly an argument either. Here’s the telling thing. Today, even when the budget is available to build a code-compliant, carbon-neutral, fully accurate replicant, many architects will still choose not to do what still seems to many non-architects a sensible proposition: why don’t you just make it like one of those great old buildings?
Why not? There must be an answer tied up in the relationship of the method and the goal. Which leaves us back where we began: reversing the logic. Though they do not talk about this much, their actions make clear that many architects must think that the strategy of replication is not the right method to solve the problem. Specifically: many architects must believe it is a bad strategy for creating, through architecture, a larger landscape that is experienced as stable. For them, replication must be the pony, not the dog.
Perhaps the best way to unpack this observation is by means of an interesting example: the Brant House, built in the early 1980s, by the architect Allan Greenberg, and the subject of generous coverage by magazines including Architectural Record and AD. 4 Designed as a partial facsimile of Mt. Vernon, George Washington’s home, and intended to be historically and fully faithful to American estates of the mid 18th century — I mean: not like a cheap suburban knock-off — the house is one of the strangest constructions of the 20th century — as architecturally radical, in its own way, as Mies’s Farnsworth House.
As a formal construct, the Brant House — which is not, like its prototype, on the banks of the Potomac but rather in Connecticut horse country — is not abstracted via analysis of its landscape, at least not explicitly, like, say, the Diamond Ranch High School, by Morphosis, or the City of Culture projects in Santiago de Compostelo, by Eisenman. It appears at first the opposite of a mapping — a process by which the tracings, contours, lines, and projections of various site phenomena are used to define or limit formal possibilities — seeming instead to be the normal by-product of a system of rules that is, or once was, widely understood. In form it is very much in keeping with similar houses constructed, with little to no help from architects, by American gentleman farmers of the colonial and federal eras, many of whom had some familiarity with Palladio, at least as passed down through the English and American gentry.
The Brant House is one of the strangest buildings of the 20th century — as radical, in its own way, as Mies’s Farnsworth House.
But today the Brant House is a mapping of a certain type. That gentleman farmer tradition is broken, the landed (Loosian) knowledge drained away. Allan Greenberg had to undertake extensive study — measuring colonial era buildings and researching archival documents — to arrive at his systems of disposition, proportion, material, and detail. The house is almost archeological in its desire to re-present a pre-existing condition correctly. Theoretically there is no design invention; just an authorless curation of historical data. In the 18th century authorless meant the actual absence of an architect, today it means an architect working to suppress apparent willfulness by establishing some external controlling logic. Here that controlling logic is: the way buildings had been made before, when the landscape was still perceived (we think) as consistent.
Greenberg optimistically intended for the construction to be a normative part of its landscape — a stable element — representative without representing. He seemed determined to pursue the logical extreme of the radical possibility that the architect does not invent, merely undertakes to perform an allotted portion of the landscape, in support of its symphonic entirety, through the use of a period instrument. In short he sought to favor the whole of a landscape over the precocious possibility of a building expressing itself. That he did so without failure of design nerve — there is no moment when, by creative alteration, he lets you know there is an Architect at work — is why this building is so radical. 5
But of course landscapes are not only construed of sites and constructions. There is also a substantial component of underlying desire at work. To fully understand how this house makes its landscape, it helps to consider the kinds of desires that might turn a very expensive possibility — creating an accurate replica — into a probability. Buildings cost a lot and therefore tend to resolve deep rather than surface desires. So what problem, what desire, would cause a house to enter the world in this form in the latter part of the 20th century? (This is a bit like asking: beyond the culture imperative, what were Scott’s personal, psychological motivations?)
What problem, what desire, would cause a house to enter the world in this form in the latter part of the 20th century?
To answer this question — keeping in mind this is the sort of house many people still want — it will help to imagine you’ve had it built for yourself. Begin by assuming you are a member of the One Percent — what Tom Wolfe memorably called, back in the era when the Brant House was built, a “master of the universe.” You have amassed obscene money in your hedge fund, enough to insure comfort for a growing family, enough to buy hot-market art (the Warhols alone are worth as much as the house). Your particular dilemma — this being New York and its collateral hinterlands — has turned from the acquisition of wealth to its use. Your specific need is to become part of the extended landed society of the City, a society that traces its history back through generations living in extraordinary (and now historic) houses in artfully landscaped East Coast exurbs.
This desire is powerful and problematic, the stuff of tragedy — it’s not for nothing that The Great Gatsby is the essential American novel! You and your growing family do not, as luck would have it, find yourself on a genetic branch of the correct species of tree in the garden in question. So you are going to have to enter by strategy. You’ve purchased land well. Now you need the house. You want something old, redolent of belonging. And what better than Mt. Vernon? Aside from its impeccable historical pedigree, it’s a damn fine house: stately rooms, formal enough to suggest structure, but varied enough to suggest possibilities.
The only problem? Finding an architect to do the work. It’s not just a matter of finding someone who can do the research down to the smallest level of detail. You have to find an architect with no ego — yeah, good luck with that — one whom you do not have to constantly police regarding your literal intent. By which you have meant: absolutely literal. You cannot afford to have the house give anything away. Now, standing in front of it, you cannot perceive any weaknesses. Its every aspect seems convincing, and you are finally about to be at home. The furnishings are in place, the artifacts arranged, the artwork installed. The landscaping is blooming. It’s a freaking stellar day.
Fake Rolexes, knock-off Eames chairs, breast implants: people may be fooled, but you have to sustain the lie.
And thankfully so, because today you are hosting the first of what you hope will be many gala parties, to which you have invited not only your neighbors but also those to whose community you aspire. Just now a first car is turning up the drive. What a gorgeous Mercedes! And it is at this exact moment that a trickle of cold sweat down your neck makes you suddenly and keenly realize that your desirable guests will never be fooled. Form: it’s not just about looks! It’s like fake Rolexes, knock-off Eames chairs, breast implants: people may be fooled, but you have to sustain the lie. Authenticity still counts for something! You are now aware of the precise shortcoming. Your house, rather than providing the certainty you desire, busily engenders doubt. It does not stabilize your landscape. Living there will, you suddenly fear, be like standing on quicksand. It has not served you, architecturally.
These ponies? Not a great way to get to that pole!
Unless uncertainty is what you wanted. Which it wasn’t — and still isn’t, in most architecture. Doubt, rather than certainty, is indeed the hallmark of that vast swath of Post-Modern and After-Modern cultural production fascinated by the possibilities of replication. We see it, for example, lovingly taken advantage of in Sherrie Levine’s infamous (and to the Brant house contemporary) photographs of iconic Walker Evans photographs. Here the problem is not the heroic Modern struggle involved in making the artwork; it’s the heroic After-Modern effort required to consume the artwork, which refuses to quit asking the question: what the hell am I?
In fact, at the beginning of the Post-Modern, every field active in aesthetics except for architecture was fascinated with the de-stabilizing strangeness of simulacra, from sampling in early rap to the ambiguous role reversals in the work of Prince (R.I.P.; Prince would actually have been the ideal client for this house, formerly known as Mt. Vernon), from Ash the android in Alien to the Terminator, a mechanical simulacra draped with living flesh. The Brant House — with its flesh of the historical perfectly hiding its modern mechanical systems — is a Terminator, and it fills many architects with dread. Just like the steaming, naked Schwarzenegger freshly arrived from the future, it demands: “Your clothes. Give them to me. Now.”
This strangeness, the uncertainty that arises from simulated historical representation, is one of the reasons abstraction sustains its powerful hold over many architects, particularly those who believe buildings should serve to promote grounded experience, the authentic — which is so often seen as latent or embodied in landscape. At a time when architects perceive a need to protect that power of landscape, the application of historical precedent has turned out to promote not stability but uncertainty, undercutting the very landscapes, the very environments, the strategy seeks to protect. This it is thus deeply unacceptable to many architects, who worry that the world becoming a full scale mock-up of itself protects nothing, and that the people who don’t see that are, frankly, part of the problem.
At the core of the disagreement between landform building and historical replication is a dilemma. Has the rupture of the Modern forever broken any possibility of continuity with the past? Much has been written on this exceptionally complex and unanswerable question — unanswerable because it sets form against knowledge. Consider this example, to cite just one among countless others: The Hopewell Culture National Historic Park in Ohio (formerly known as Mound City) encloses what was once an intact Native American burial ground consisting of a series of constructed hills and embankments along the Ohio River. Most of these archaic Hopewell earthworks were bulldozed to make a level runway and camp — there’s that Modern rupture! — by the U.S. Army during World War I. After the war, in reaction to public outcry, the mounds were re-made by approximation, again by bulldozer. If you don’t know this, be glad: you will experience the site in awe. Because your knowing will, of course, forever color your ability to accept the place as authentic. Your visit will be yet more evidence of that essential After-Modern concept that all experience is mediated by knowledge. (Also, sorry, I should have spoiler alerted you sooner).
Robert Scott, who opted for ponies not dogs, was buried, along with his tent and two of his crew, under a mound near his final camp, a cairn set on a field of snow and ice, at 79°40’S. Curiously, its particular location is unstable. That snowfield is slowly creeping towards the ocean. Here I’ll quote from the Wikipedia entry on Scott — I couldn’t have written this better myself: “A century of storms and snow have covered the cairn and tent, which are now encased in the Ross Ice Shelf as it inches towards the Ross Sea. In 2001 glaciologist Charles R. Bentley estimated that the tent with the bodies was under about 75 feet (23 m) of ice and about 30 miles (48 km) from the point where they died; he speculated that in about 275 years the bodies would reach the Ross Sea, and perhaps float away inside an iceberg.”