Essay: David Heymann
Landscape Is Our Sex
Tough love in a gray suit. Cary Grant in To Catch a Thief (Paramount Pictures, 1955).
It is not uncommon today for an architect to give a public lecture about a building and gloss over the parameters of its program or the specific needs of its users, speaking instead mostly about the building's site
— its measure, tendencies, desires, structure, mythologies, meaning — as if the problem of architectural design was primarily one of site response. Easily two-thirds of the many talks I've heard architects give over the past 20 years have operated from this basic presumption. While some of these presentations have trafficked in various sorts of disingenuousness, others have been transcendent. Glenn Murcutt gave a lecture at my university some years ago in the latter category. He spoke in front of a large blank chalkboard, and, as he began talking about the site for a construction, he drew out in chalk the various facts (like the hillside slope in section) or the consequence of facts (the arc of a tree that might fall) that were inarguably true. 
Murcutt didn't really refer to anything but the site, and even then primarily as a set of physical parameters — the angle and arc of the summer and winter sun, the direction and speed of cool and warm and strong and weak and moist and dry winds, the height of grasses and brush over the course of the year, the location and shape of outcroppings of rock, the paths that animals took through the site (and the heights
of those animals), the probable limit of a fire line, and so on — each and all of which he sketched or diagrammed. He never actually spoke about or drew the construction
directly. Instead, the building's plan and section gradually became apparent on the chalkboard. After about 40 minutes, these were clear and complete. Murcutt then turned off the lights, and, without saying another word, showed twenty slides of the finished construction — the Simpson-Lee house
. The audience rose in a spontaneous standing ovation. It had been an utterly convincing display of site as source
But not long afterwards I was reminded of the downside of thinking this way — or at least publicly professing to — while visiting the Audrey Jones Beck Building
, an extension to the Museum of Fine Arts in Houston, designed by Rafael Moneo, and completed in 2000. The site for this freestanding construction, which fills a modest city block, is Houston perverse
. At the time of design, its four facing neighbors were a Gothic revival church, a decaying 1970s Yamasaki-wannabe office building, the unclassifiable Warwick (from 1926, once Houston's premier hotel), and the extraordinary museum itself, one half black-steel High Modern (Mies), one half carved limestone Neoclassical (William Ward Watkin). I asked one of the curators about the choice of Moneo, who was awarded the commission over Tadao Ando, Richard Meier, and Robert Venturi and Denise Scott Brown. I presumed the decision hinged on Moneo's skill with difficult contextual parameters. Her answer: "Not really. They all spent a lot of time at the site. Moneo was the only one who actually asked to see the collection
."  Let that
serve as a sobering professional reminder.
Still, I wouldn't recommend relying on Moneo's strategy exclusively. A lot of architects make a living talking — in the messy ways described in a sequence
of earlier essays
published in this journal — about site, landscape, the environment. Architects know this is a shorthand method of tapping into a primary vein of cultural meaningfulness; they rely on this blood link to get ambitious buildings built, to convince
. Among the better convincers is Antoine Predock. In 1999 Predock was hired by the City of Austin to design a new City Hall
. The site is unusually prominent: an entire block along the hard south edge of the city's downtown grid, fronting the empty bluff over the river, bracketed east and west by buildings already symmetrically construed to frame a future City Hall. A major bridge springs southward over the river from the center of the block itself, so the focus of northbound pedestrian and vehicular traffic entering downtown is squarely on the building.
To be sure, the design is only slightly coy about being the center of attention. Though set back from the south edge of the lot to form an open public plaza, the construction is a staggered asymmetrical mass composed of stacked and stepping horizontal concrete and limestone bands that form planted terraces and diverse public spaces. The upper layers, clad in copper, cantilever out variously to provide shaded overhangs. The entire mass is deeply cleft by a wandering central valley
that organizes public circulation and provides natural light. Water (it's air conditioning condensate) drips from a wall along the edge of this deep cut into the dark recesses of the parking garage below. From the central rift a strange copper spike — nicknamed the Armadillo’s Tail
— cantilevers off the north side, pointing directly at the State Capitol (though you cannot perceive this relationship).Austin City Hall, Austin Texas, by Antoine Predock, 2004. [Photos, clockwise from top, by Rafael Marquez, Aaron Plewke and Gary J. Wood]
If the form seems willful and gestural, Predock set about proving, when he first presented the design in public, that it was not. He claimed the form was borrowed directly from the faulted Balcones uplift — the local and deeply revered configuration of fractured limestone layers cut by clear spring-fed streams — as he experienced it along Austin's Bull Creek. Predock showed his own sketches of this landform, and he described the building's condition as if someone (i.e., he himself) had merely (i.e., heroically) lifted out a block of this typological central Texas landform — think of it as an immense stratified wedge of geological lasagna — and placed it neatly on the site. The planted high terraces and the jagged profiles of the upper layers, clad in brown copper, add to the conceit: they (in Predock’s words) "... spill out of the building into the plaza in a way that is analogous to geologic forces in the Hill Country that surround Austin and produce limestone overhangs known as 'balcones.'" 
The City Council, as you can well imagine, swallowed this line whole. Who wouldn't? Landscape is good; building is landscape; therefore building is good. One hears this three-car train of logic constantly in architectural discourse today. It's an enduring staple of public design presentations, magazine articles, student reviews, architects' written descriptions of their own work, and — as noted — public lectures. Nothing sells like landscape. It's our sex
. But is sellability a sound measure of an idea's value? There's that conundrum of a marketplace society! Regardless, for me, at least as an academic, a basic question about perception screams out from this curious form of convincing, like a desperate cry for help. If a building is
a landscape, how does that landscape make
landscape once returned to its landscape? Since — conceptually — the construction rhetorically purports to be the landscape from which it's drawn, shouldn't such an object be invisible
, or at least, hardly visible? After all, the same means — scale and direction of mark, material and color palette, etc. — are used to make a camouflage like MARPAT
. This is a useful analogy, since the same over-arching after-Modern cultural ends — to not disrupt a primary reading of an existing and valued landscape with the radical presence of otherness
— would seem to be desired in architectural design. Isn't that protection somehow implied in these oft-repeated justifications?MARPAT digital camouflage pattern, United States Marine Corps.
Yet many — most — of the buildings conceived by these means over the past two decades are stridently volatile forms, and their landscapes are dramatically altered by their arrival. Consider an early progenitor of the type: Zaha Hadid's fire station for the Vitra complex
in Weil am Rhein, Germany, completed in 1993. The design drawings for this project notoriously (and, in architectural design, familiarly) extended the geometric patterns of the adjacent surroundings — farm boundaries, crop plantings, rail lines, road angles, and so forth — back over the constructible area to provide a matrix of controlling lines in what would seem to be — to the parent landscape — a sympathetic first step. In architectural design this particular act is often (somewhat sloppily) called mapping
; now it is also sloppily referred to as (one type of) parametric design. Here the tracings and projections of various existing though often hidden
site orders are utilized to define or limit formal possibilities. The architectural door prize to this method — abstraction, our Modern birthright — is protected, problematic historical figuration is avoided (whew!), and program is rewritten by the found form in a manner theoretically sympathetic to the existing landscape.
What's then initially surprising about the actual Fire Station — so sympathetically conceived — is how aggressive
it seems to its parent landscape. Ungrateful child!
But it makes sense that the constructed object is neither conventionally sympathetic, nor camouflaged. There's nothing about using mapping in design that guarantees
conventional sympathy to landscape. The building may arise from existing lines — as buildings also do for Murcutt and Predock — but the crucial design tasks are the identification and evaluation of these lines on the one hand, and the association of their geometries with building program and form on the other, which are judicious acts associated with intent and desire. This judiciousness is of course entirely the domain of the architect, as the interpretation of confused entrails is the domain of the tribal doctor.Vitra Fire Station, Weil am Rhein, Germany, by Zaha Hadid, 1993. [Top photo by Hélène Binet; aerial photo from Bing Maps; painting by Zaha Hadid Architects]
It's a stunningly subjective domain to be sure, as is evident in the dramatic difference between buildings by Murcutt and Hadid, which all begin with mappings, but differ in the identification of meaningful lines, and the ascription of form and program thereunto. Regarding mappings of the type that Hadid pursues, the mapped lines may be "true" (a tough word, but most designers sense what it means) of a site, but this "truth" is not necessarily made either evident or stronger when transformed into some aspect of a building. A train track's trajectory might, for example, set the angle of a bathroom urinal wall; but — regardless if the original alignment is visible or not — that wall experienced in place won't necessarily bring a Johnny Cash rail song to mind while one is taking a leak. And that's aside from the whole question of whether it's possible to ignore the finger doing the pointing. In any event, the Fire Station returned to its landscape is startling precisely in its difference
, rather than its similarity. In the confusion between ends and means, that odd hangover of the Modern, something is clearly and cleverly lost, though the fog into which it vanishes may be the point of the rhetoric, like using nature to advertise SUVs.
In all fairness, the Vitra Fire Station was designed for an evolving landscape: it's difficult to approximate the difference between the source and its resultant. Perhaps a better and equally well-known prototypical example from roughly this same time is the Jewish Museum in Berlin
, designed by Daniel Libeskind. Libeskind claimed the complex plan geometry of this construction arose by projecting lines from the building's location to sites of Nazi atrocities or institutions elsewhere in the city. This fact is interestingly irrelevant to your unmediated experience of the building as a physical fact, since, like that spike on Austin's City Hall, you cannot reconstruct these linkages directly (by observation, for example). That is, you cannot understand what generates the order by simple inhabitation
. You have to be told
. The geometry is not therefore perceived as inherent to its landscape, except as a pure conceit. It's an addition, not a clarification — more information, not clearer information — and confounding at that. It certainly doesn't settle the landscape, even though it theoretically points out a true order, much as crime scene markings trace the true lines of bullet paths.
Note, however, the desire at work. It is precisely your inability to experience the mapped world
— here aided by entry via underground tunnel — that adds to the disorienting strength of this exceptional building
. That is why this
construction so powerfully embodies its program in its location. The Jewish Museum is not about protecting an existing landscape, so much as upsetting it. And upset it does. Interestingly, at the time of its design you either understood — thinking about the consequence of such a building while examining the drawings — that the building would be dissonant with its landscape; or you understood — listening to the architect’s rhetoric — that it would not
be dissonant with its landscape. Either way the building appeared responsible
. It certainly swayed the competition jury. Given this sort of dramatic success, I certainly don't object to this type of mapping. It has advantages and disadvantages despite the devaluation that the use of the particular and extremely simplified version of mapping in these examples implies intellectually. Here a landscape is construed primarily as a field of consumable vectors, and a building primarily as the bastard child of those vectors.Top: Jewish Museum, Berlin, by Daniel Libeskind, 2001. [Photo by Bitter Bredt]. Middle: Star Plan for the Jewish Museum. [Plan by Studio Daniel Libeskind] Bottom left: Aerial photo of the Jewish Museum. [Photo by Guenter Schneider]. Bottom right: Olympic Sculpture Park, Seattle, Washington, by Weiss/Manfredi, 2011. [Photo by Benjamin Benschneider]
While that reduction bothers me, what I primarily object to (since all design involves reduction) is the unexamined assumption underlying the intellectual stance — if you can call it that — in current architectural discourse: that operating in this manner is a form of site sensitivity
or landscape empathy
, a means of weaving a building into a landscape and so protecting that landscape's essential identity. That does happen some times — the Olympic Sculpture Park
in Seattle, by Weiss/Manfredi, uses similar techniques to equalize presence in landscape — but not often. Like a scribing tool, which can be used to precisely fit one complex surface to another, but also to quite efficiently and irrevocably scratch the hell out of that same surface, mapping has uses of varying consequence in the design of buildings in landscape. But easy acceptance of the new construction by the existing landscape is not a guaranteed consequence of its usage
. Photographs of these powerful buildings make this abundantly clear. Strange rude entities, demanding attention, the Vitra Fire Station and Jewish Museum steadfastly refuse to connect to anything. Like Austin City Hall, they radically alter their landscapes by their construction.
Hadid and Libeskind are smart; I doubt they actually believe that site mappings of this type result in site sensitive buildings, in the conventional sense of sensitive: having a sensibility that is supportive without being disruptive, like the sympathetic attending a funeral
. But, if conventional sensitivity isn't what Hadid and Libeskind are after, the implication is still somehow in the air. Worse still is the suggestion that so operating is somehow related to sustainability. A good current example of this particular perversion — aside from the geological landscape formation formal design tactic Index
of Stan Allen and Marc McQuade’s Landform Building
— is the script of the Audio Guide that accompanies the traveling exhibit of Hadid's recent work (itself in a Hadid-designed portable gallery).  The relationship of Hadid's parametrics to natural systems is a constant sub-text in the project descriptions, complete with the regular and insistent implication that the buildings must therefore be doing the right thing environmentally. The high point of this rhetorical absurdity comes in exonerating the cladding of Hadid's (interesting) design for the new Opera Hall in Abu Dhabi
. The system's geometry, we are told in reverent voice over, derives from the limbs, branches and leaves of deciduous trees. What could possibly be wrong with deciduous trees? The building may be sited in the goddamn desert
, but, I mean, aside from that
? Top: Performing Arts Center, Saadiyat Island, Abu Dhabi, by Zaha Hadid, 2007. [Rendering by Zaha Hadid Architects] Bottom: Detail of “Index,” from Landform Building, edited by Stan Allen and Marc McQuade (Lars Müller, 2011). Click image to enlarge.
No, let's please ignore the whole ridiculous sustainability tie-in of this sort of mapping and just focus on the formal construction of consistent landscape
. Returning to that funeral mourner analogy for a moment, what many mapped designs propose, while rhetorically claiming to be conventionally sympathetic, is formally closer to that truly shocking scene in Hitchcock's To Catch A Thief
, in which Cary Grant's character, among the supportive in grief, slaps, brutally, the face of the distraught daughter of the deceased, who has just accused (duplicitously, we later learn) Grant's character of complicity in the death. Yeah, that sort of sympathy — tough love in a gray suit
— is hard to sell these days as an architect, even if principled. Better to stick with pretending the slap is a gesture insisted upon by the site
, rather than a conscious act on your part.
Clearly what it means for a building to be "sensitive" to its site is more complicated than I have initially set out here; I merely want to note the essential arbitrariness
of the definition of site that is at work in these projects, and, by extension, in the work of the many architects (and landscape architects) who, to varying degrees, use the same means. Here site is only that limited field of physical facts or factors, no matter how far flung, and at the expense of all others, that the designer needs to justify the form of composition that he or she wants to make, regardless of the landscape
. This is of course Rem Koolhaas's notorious Paranoid-Critical Method at work, the gist of which, as defined in Koolhaas's great Delirious New York
, being that the willfully predetermined design solution — or at least the willful predetermination of what will be formally acceptable in a design solution — is masked in innocence by seemingly neutral methods of production relying on apparently external parameters.
 We know why
all buildings by Hadid or Libeskind — actually of many architects — end up looking more or less the same, regardless of their sites. Though the exact form may not be known in advance, the formal characteristics are. That is why the orders mapped sometimes turn out to be hidden, discovered
, conveniently, by the architect.
Many good architects — I think, to varying degrees, very many (but not all) — operate this way today.  Consider, by way of example, the (beautiful) Diamond Ranch High School
by Morphosis. Its form of mapping is less obvious or extreme than that used by Hadid and Libeskind, but it is therefore perhaps more representative of how this way of working has infiltrated design thinking. Here the design (in theory) arises as a series of overlapping diagrams that relate aspects of the underlying site conditions (topography, geology, etc.) to things like mass, circulation, materiality. In each instance, it is possible for the architect to point away from personal desire and register a decision in relation to a tangible aspect of the site. The net consequence is to diffuse apparent willfulness, even though this design is as willful as any other.Top and middle: Diamond Ranch High School, Pomona, California, by Morphosis, 2000. [Photo by Timothy Hursley] Bottom: Box canyon near Mecca, California. [Photo by Doug Carlson]
Of course, there is the whole question of means and ends. The Diamond Ranch High School is a useful case, as its ends represent a consistent set of desired results present in works of this type. At Diamond Ranch — as with many of these constructions — it is not difficult to associate the constructions with natural landform (in this instance with eroded box canyons of the West). Here, whatever the means used, landform is abstracted to a point where it becomes constructible (or vice versa: various construction techniques are acted upon until they begin to resemble landform). In addition to representing landform abstractly, the forms generated are typically the most powerful in their locations. This surprising dominance — the contentious issue at the heart of this essay — is not obtained by traditional means: oppressive axiality, radical symmetry, heroic scale. On the contrary, the buildings achieve their fore-grounded presence by monumental picturesque-ness.
 In this regard they are conventionally sculptural
: three-dimensionally muscular, tending towards complex superimpositions, and extended, often slightly bent or folded forms that rely heavily on the reading of extended, horizontally emphatic planes. This particular definition of sculptural
has little to do with sculpture as an evolving cultural activity, so much as with a sort of frozen idea based on early Modern geometric compositional strategies. Sculpture has long been off doing way more interesting things
Design justification is a slippery slope. No matter how ardently architects claim a design arises from a landscape, if they weren't happy with the formal consequences, they wouldn't make the argument. In my experience with architects and students, justification as often comes after the design as before. This is the heart of the Paranoid-Critical Method. I realize that one hallmark of after-Modern theory is the problem of influence
— can one really say what comes first in a creative process? — but there is a logical limit to relativism, and my bullshit meter says that explaining the ends of these designs by the means of their parent landscapes is a bit like a psychopath blaming the mess made on his or her parents: true in part, but not really the important truth, which lies with the victims — direct and collateral — of the action.
Despite the rise over the past two decades of Paranoid Critical site mapping, there is a stunning lack of rigor to its means-justify-ends argument. The range of objects-in-landscape this process currently produces falls between the two poles of Predock, who maps indirectly, by simile if you will, and Hadid, who maps directly, metaphorically (Diamond Ranch High School and its many kin sit nicely between the two). To reiterate: mapping is justified as a form of respect to landscape that is essentially public in spirit. The design seems, after all, a form of diagnosis
, and the architect can say: the landscape made this, not me
But it is clear the form of mapping used in these examples generates constructed experiential landscapes in which it is impossible to perceive any linkage to the original order from which the map arose. These constructions may have begun as diagnosis, but, just as a diagnosis is rarely the cure
, so a map itself is not directly the landscape it maps. Landscapes do not make maps, humans do. As human constructs, mappings do not automatically retain the (seeming) innocence of maps. Their presence as constructions — rather than as abstractions — is not necessarily stabilizing. In these powerful examples, as I have suggested, the opposite is often true. The sites, having transferred content to the constructions, are reduced in experience from source to setting in the perception of landscape
. It may be possible to map a landscape, but the affect of these constructions is to render the old maps obsolete. You are, no doubt, aware of the irony. Or, maybe — like many of my colleagues — you’re just repressing that awareness!
Architects are not required to be intellectually rigorous. They only have to convincingly sell ideas about how value is embedded in form in order to build. It is precisely this need that makes architectural intellectualism so suspect, and, just to reiterate, nothing sells today like landscape, however shoddily conflated in theory. The conceptual problems I have been describing are thus not a big deal professionally (though they are frankly unforgiveable within academic architectural discourse). Still, given that architects are, in my experience, a fairly conscientious group, it's interesting to speculate why so many continue to go out on this particularly logic-challenged limb, despite its evident flaws as theory. Igualada Cemetery, Igualada, Spain, by Enric Miralles and Carme Pinos, 1994. [Top image by Marc Teer; two bottom images by Flickr user Velcro]
I can think of three probable causes. Only the first is hopeful. It is possible that a new generation of spatial desire exists, one that seeks to frame public space through the concatenation of the sorts of private experience these formal moves promote (see note 7). This possibility first appears (appropriately) in the uncertain geometry of the Vietnam War Memorial
, but it is to date most precisely confirmed by the Igualada Cemetery
, the masterpiece of Enric Miralles, which points to this possibility with startling conciseness. Here the form of the new public landscape arises as a concretization of an existing man-made landform — an excavated quarry — the form of which is entirely without precedent in public experience. There is no shared toolkit to decipher this space. Here public-ness cannot be that I trust that you are feeling what I am — as I do when I am with you strolling the axis at Versailles. It can only be that, later, we are surprised to discover that we felt the same.
The impossibility of common interpretation is a hallmark of after-Modern theory.  How to convincingly give this idea public form is a pressing impulse that shows up in all sorts of forms of cultural production — multiplicity of interpretation shows up as the blatant subject in, for example, the work of the sculptor Tony Cragg
— though perhaps least coherently in architecture. . But I doubt this essentially new and urgent desire is driving most of these architects. Here's why: I've heard far too many architects describing their design process as tuned to the particulars of their specific sites — and these sites have been as varied as you can possibly imagine — and yet the many buildings designed for all these crazily varied landscapes fall within an extraordinarily narrow formal range. See if you can differentiate the landscape-mapped designs based on site difference as you work your way through the Phaidon Atlas
. These surprisingly similar buildings — from around the globe! — constitute an identifiable oeuvre, the consistency of which only serves to further flatten
the distinction of their parent landscapes, exposing the lie beneath the claim.
No, the reason is probably shallower. I think many architects today use landscape in the way I have been describing to mask their desire to work by the means and in the generous and luxurious and deeply desirable language of abstraction
, that (by the public largely underappreciated but within architecture still correctly heroic) birthright of the Modern, rather than with the thin forms of representation of the Post-Modern, brought to you by the short-sighted idiots who sold out our birthright for magical sheetrock details. By equating abstraction with landform (as with the Austin City Hall), or with the processes that generate landform (as with the Diamond Valley High School), or with the palimpsestic layers of information built up in the human formation of landscape (as with the Vitra Fire Station), or with hidden landscape orders that must be brought to light (as with the Jewish Museum), these architects are cloaking the abstract with a mantle of landscape valor, while sidestepping the more contentious issue of language. Inadvertently or intentionally — it doesn’t matter — many architects today equate landscape and architectural form to keep alive the tradition of a building’s right to abstract sculptural presence. Landscape is valued, while architecture is not
. The rhetorical use of landscape in these instances is a form of deceit, a means-versus-ends intellectual slight of hand. These are timid times culturally. So maybe the gap between the pitch and the product is OK. Or not. Increasingly my experience is that maybe the gap doesn’t exist in architects’ minds, and we’ve actually come to convince ourselves of — to believe — the strange twisted content of our own misleading sales pitch. It happens; it's happened before. One thing though. The sculptural dynamics of all of these buildings, their startling hallmark, actually point darkly towards a third probable cause. Why does the architect trot out this specious logic? Because "the architect
...” — these now the words of Adolf Loos, writing on precisely this same subject  just over a hundred years ago — "...has no culture