The Ugly Pet

Sustainability should be revolutionary to architectural form. But it hasn’t been, at least not yet.

Oskar von Miller Forum, Munich.
Oskar von Miller Forum, Munich. [David Heymann]

This spring I spent a few days in Munich at the Oskar von Miller Forum, which serves as a locus for interdisciplinary thinking about construction. The Forum building, designed by Thomas Herzog and funded to showcase just how elegantly the Bavarian construction industry can deliver innovative sustainable architecture, feels a bit like the hopeful future of the perfect city. It doesn’t hurt that it’s near the Englischer Garten, as it was summer hot. People were already swimming (and surfing) in the Eisbach, the stream that passes through the park. You can float its long length, then return dripping to your stuff using a regular streetcar line without a ticket (you need a ticket if you’re dressed or dry).

It would be difficult to stay at the Forum without knowing about its environmental agenda and sustainable systems. The building, widely published, is intended as a manifesto — that’s one reason many students and scholars apply to study and work there. Each residence unit is stocked with the beautiful book that describes the project’s “environmentally friendly and emissions-reducing innovations.” The Forum’s director, Werner Lang, shared technical information via email before I arrived, and in Munich he and I walked through the building as he explained it all in depth. So, trying to sort out how the architectural space affects the experience of inhabitation was complicated: it was hard for me to separate my sensations from all the mediation.

The experience of sustainable space is hyper-mediated, each building scripted like some organically and ethically sourced food product.

In general this is a frustrating upshot of the sustainable revolution in architecture. Unlike, say, the development of the steel frame, or the invention of concrete, most sustainable material development does not have any significant, self-identifying formal consequence — sustainably sourced wood looks just like wood that isn’t. And this is true of many building systems as well: how can you know from the architectural form if, for example, the heat source is geothermal? (You can’t, unless you know what to look for). The pure value and capacity of space and form to convey the crucial information falls away, like the disappearance of elegant cursive writing. In order to get traction, the experience of sustainable space has to be hyper-mediated, each and every building scripted like some organically and ethically sourced food product, with an explanation of its goodness provided, somehow. Hence all the words. Frustrating.

Oskar von Miller Forum, Munich.
Oskar von Miller Forum, Munich. [© Oskar von Miller Forum]

At the Forum, then, much non-spatial mediation is needed in order to appreciate the sophisticated systems that have been designed to ensure your comfort in its elegant spaces. For example, the building has in-floor (i.e., hidden) radiant cooling and heating that work in concert with other active and passive temperature-controlling strategies, like mechanical window shades that allow sunlight deep into the building or not, as needed. As a matter of policy, the allowable temperature range in any one space is fairly broad, in part because heating up and cooling down occur slowly by these means, and their control is determined by several interconnected variables. An inhabitant can regulate his or her room temperature by opening or closing a window (or, especially, by opening or closing a second window to help or hinder cross ventilation), but the overall baseline temperature is set centrally. As a result you’ll feel more or less uncomfortable — hot and humid — than you would in similar contemporary buildings. But because you know — because you’ve been told — what the Forum is trying to achieve environmentally, you quickly make your peace with that. You proudly roll up or down your shirtsleeves, put on or take off your shoes (I found the radiant floor to be especially key). In many buildings — in which if you can even open a window not much happens — you’d be frantically searching for the thermostat.

Although each component of the Forum is considered with respect to its effect on all others, each was developed largely in isolation; e.g., what are the issues that need to be resolved for this particular program, for that specific exposure? For this reason the building reads like an assembly of different pieces which fit thoughtfully together but which are not clarified or organized by an over-arching architectural language — as they would be in, for example, a neoclassical building, or a Renaissance palazzo, or a Frank Lloyd Wright building, or anything by SANAA. This pieced-ness is a sensible outcome of one strain of thinking about sustainability: by developing and then monitoring each building component, you can supposedly get an accurate, detailed environmental performance profile to let you know what really works and what doesn’t. The hyper-specificity of the parts is thus theoretically in the service of betterment, and the appearance of the building as an assembly seems self-justifying.

I’m interested in how buildings affect the experience of landscape — controlling that experience is one of the foremost responsibilities of an architect.

I mention the outward appearance of the Forum here because I’m interested in how buildings affect the experience of landscape — I think controlling that experience is one of the foremost responsibilities of an architect, and it’s been a hot-button topic since the collapse of the Modern movement. I’m particularly interested in how sustainable buildings might affect the experience of landscape differently — actually better, differently — because, as a human being, I’m hoping for more sustainable architecture, and, as an academic (and as an architect), I’m thinking the consequences should be revolutionary to architecture, as the consequences of every other major technological revolution have been. But they haven’t been, at least not yet. As I argued in an earlier essay in this journal, the architects who are struggling to express a new technology are inevitably feeling their way forward. A building put together from many hyper-specific parts might be the way. But how, then, does such a building affect experience of its setting, of its urban or natural landscape?

Loblolly House, Maryland.
Loblolly House, Taylor Island, Maryland. [Hans Drexler]

In that same earlier essay I discussed the Loblolly House, designed by Kieran Timberlake: a well-known example of contemporary sustainable architecture which exhibits the hyper-specificity of carefully monitored parts I’ve been describing. Its various pieces are even more distinct than those of the Forum: at the time of its construction, in 2006, the building was promoted for the potential replace-ability and recycle-ability of its components, if these were found to under-perform. One’s appreciation of the Loblolly House increases with knowledge — with non-formal mediation — as it does with the Forum, and I compared the building to an ugly pet:

… something that, while ugly by normal aesthetic parameters, becomes beautiful because of the extraordinary amount of care and effort that has gone into seeing the creature to maturity, like nursing a sick cur from the pound to health, and finding you love the damn thing, which has somehow become pleasing to the eye. Here, beauty is conditional rather than universal, and exists in the eye of the beholder, rather than broadly. … The Loblolly House [is] about as homely a well-intentioned construction as is imaginable, something, in terms of aesthetic dimension, that only a mother could love (I say that with admiration).

One conceptual difference between the Oskar von Miller Forum and the Loblolly House is that, at the latter, no concerted effort appears to have been made to ensure that the pieces had something in common before they were put together: this randomness is actually what gives the building, located on a sparsely populated island off Maryland, its counter-intuitive charm. Whereas at the Forum, in Munich, the performance parameters include consideration for historical type-forms and for the urban context, and the pieces go together very elegantly. A limited measure, rhythm, and palette all work to tie the components together, and their differences are further masked by the application of screening devices — like the partly transparent solar array over the central south façade — that impart a larger scale commensurate with the metropolitan setting. The building’s presence in its urban landscape is also predicated on your sense of it as the coordinated sum of the constituent parts. Let’s call the Forum a modified ugly pet — one that’s been trained and groomed. 1

Arthur and Yvonne Boyd Education Centre, West Cambewarra, New South Wales.
Arthur and Yvonne Boyd Education Centre, West Cambewarra, New South Wales. [Lucas Torresi]

Fredericks House, Jamberoo, New South Wales.
Fredericks House, Jamberoo, New South Wales. [Lucas Torresi]

There’s another approach to hyper-specificity in sustainable architecture, one that subsumes the idea of an assembly of disparate components into a harmonious unity. That approach appears most clearly in the work of the great Australian architect Glenn Murcutt. Murcutt’s buildings are famously perfect for their settings; consider, for instance, the Fredericks House and the Arthur and Yvonne Boyd Education Centre. Figuring out why they fit so well is complicated. Murcutt never makes any of the explicit site-, context-, or landscape-derived formal gestures that many architects today use to generate theoretically relevant forms, and his projects are the very opposite of the so-called landform buildings — like Snøhetta’s Norwegian National Opera and Ballet, in Oslo, which is modeled to represent a glacial ice field. Murcutt’s buildings do not represent anything. They seem systematically constructed from standardized kits of parts. And they are certainly absurdly specific: in lectures, Murcutt traces out, in chalk, the (many) exact site, program, and environmental factors at play in his buildings, and as he does so the architectural forms gradually become clear.

Murcutt’s buildings do not represent anything. They seem constructed from standardized kits of parts. They are absurdly specific.

My favorite example of Murcutt’s absurd specificity comes from a story he told me about the design of a detail in the Marika-Alderton house, built in 1994 and located in an Aboriginal community on Australia’s hot, humid northern coast. Like all the architect’s buildings, the house is un-air-conditioned, taking its shape in part from the need to increase or decrease air movement for comfort. The house sits on a spit between the Arafura Sea and an inland lagoon. The migration of dugongs (sea cows) into this lagoon is an important event for the community, a pivotal marker in the natural cycle of the year. It’s the responsibility of young boys to watch for the dugongs, so a crucial programmatic element of the house was a place to sit for this watching.

The height of the building off the ground, Murcutt told me, was set by measuring how high the local species of goanna (a large, omnivorous lizard) could reach and bite: the floor was set just high enough that the boys’ dangling legs would be safe. Because it is easier to see with your eyes in shade, the panel over the viewing space (which is part of an operable system that allows wind to circulate through the space during warm weather but which can be sealed shut during cyclones) was sized to block the angle of the sun during the dugong season. A ledge in the viewing room was calibrated with the average length of the boys’ arms: the boys’ food is set on this ledge to keep it free of dust, but the ledge is low enough so they can easily reach up while waiting and watching.

Marika-Alderton House, Eastern Arnhem Land, Northern Territory, Australia.
Marika-Alderton House, Eastern Arnhem Land, Northern Territory, Australia. [Glenn Murcutt, courtesy of Architecture Foundation Australia]

Marika-Alderton House, Eastern Arnhem Land, Northern Territory, Australia.
Marika-Alderton House, Eastern Arnhem Land, Northern Territory, Australia. [Glenn Murcutt, courtesy Architecture Foundation Australia]

Despite this remarkable attention to specific details — shared in spirit by the Loblolly House and the Forum — the Marika-Alderton house reads as a formally unified entity, discrete, limited and whole; as indeed do all Murcutt buildings. This is a quality not shared by Loblolly or the Forum. Murcutt has said that the magical fit of his designs to their environs arises from this discrete wholeness: the sites accommodate the completeness. I think that is true, but not in the way Murcutt describes it. By way of analogy, he says that siting a building is much like placing a brick — there’s a formally unified entity! — at the edge of the ocean: gradually, through the action of waves, the sand accommodates the brick, surrounding it and softening the impact of its presence. 2 But you just can’t trust architects when they talk! That analogy is misleading; the truth is more nuanced.

Murcutt is notorious for the time he takes on a commission; some clients had to wait five or six years before he completed the design.

Murcutt is notorious for the time he takes to start working on a commission once he’s been hired. He taught at the University of Texas for a brief stint in the early ’90s: back then he told me his clients had to wait five or six years before he completed the design work. But during that period he was busy. He would start with research, gathering data on sun angles, wind speed and direction, temperature and humidity and rainfall averages, subsurface geology, soil and subsoil types, native flora and fauna; then he would progress into the Aboriginal and European cultural histories of a region — he was being commissioned all over Australia — and on and on, all the while trying not to distinguish too definitively between qualitative and quantitative values in data. Also during these years, often unbeknownst to the clients, he’d keep visiting the sites, where, since these were often in the outback, he’d pitch a tent, staying as long as he could, not only observing but also just being there. Then he’d return in another season. And another, and another.

Throughout the process, he’d try to keep his mind open to small perceptions, like the micro-direction of wind across the site; or, especially, how animals moved through the landscape. And he’d wonder, was that a pattern or an anomaly? In the back of his vehicle, he had samples of various materials, like metal siding of all sorts, and he’d leave these out, examining their finish, color, reflectivity, and scale of mark against, for example, the local grasses as they changed from one season to the next. Gradually, he explained, he’d begin to get a picture of what human comfort might mean at a particular place: how much sun and air movement was needed given the humidity, how much heat needed to be conserved or shed, what the views meant, and how they might be used (all of which begins to define the section of a building), what the expected color and clarity of the sky might be, what the relative proportions of the house might be, how important it was to see the building or not (a crucial factor with regard to feeling secure in a remote area), and so on.

What he was after was landed knowledge — he wanted to know a place so directly that design decisions would be innate, fully informed, natural.

What he was after was landed knowledge — he wanted to know a place so directly that design decisions would be innate, fully informed, natural (for lack of a better term). The key was the time required to get that knowledge: several visits per year over five years, he argued, was barely enough. Murcutt was then being approached about designing all over the world, but he kept rejecting those offers on the grounds that he couldn’t do justice to locations outside Australia. If Murcutt were merely placing a brick onto a site, he could have taken any of those international commissions. The sites would have accommodated whatever building-brick he put there. That’s why the analogy is nonsense. Brick has nothing to do with it. I think he started using that analogy in the late 1980s to make clear his derision for the then over-hyped idea that a dramatic site-specific formal gesture (think Daniel Libeskind’s Jewish Museum, in Berlin, or Zaha Hadid’s celebrated unbuilt project for The Peak, in Hong Kong) was the means to make a building fit its landscape.

Here’s a better analogy: a Murcutt building is like a turtle or a tree. Viewed panoramically, a natural landscape can seem remarkable for its apparent cohesiveness; but as you get closer you see that the things within it — a tree, a turtle, those flowers, that insect, those rocks, the particular configuration of this burbling stream — are wildly various. One line of reasoning to explain this variety — and I think it is the most apt — is that for any ecosystem the various constituents have evolved to take co-dependent advantage of the same general conditions. So oak tree, for example, is one thing in one ecosystem, and another in another (where it’s typically a new species). The distinction in base variables — sunlight, humidity, available water, soil and subsoil conditions, pollinators, seed dispersers, etc. — means oak will have evolved new characteristics: different form, scale, bark, leaf type, acorn type, maturation pattern, leafing and shedding cycle. Likewise turtle will also be different in different environments. But in all ecosystems in which both are present, turtle and tree will fit with each other without protest. Somehow we sense that fit. I was out in the Hill Country of Texas last week, watching an armadillo rooting under a prickly pear cactus growing out of a limestone boulder in the shade of a cedar elm in which vermillion flycatchers were nesting. All of these things are not like each other in color, scale, form, or any of the means we use to fit buildings into contexts, but they made a more uniform landscape than most groups of buildings.

Simpson-Lee House, Mount Wilson, New South Wales, Australia.
Simpson-Lee House, Mount Wilson, New South Wales, Australia. [Lucas Torresi]

Magney House, Bingie Point, New South Wales, Australia.
Magney House, Bingie Point, New South Wales, Australia. [Anthony Browell, courtesy Architecture Foundation Australia]

For each of the wildly varying sites on which he’s worked around Australia — which, to give you a sense of scale, is about the size of the lower forty-eight — Murcutt has tried to evolve house into a version of itself that will most perfectly survive in its specific ecology. So when you compare, for example, the Magney House, located on the temperate coast of New South Wales, and the Simpson Lee House, in the forested Blue Mountains west of Sydney, you can see the distinct evolution of one thing in two ecosystems. In each case the (again, un-air-conditioned) building reacts directly to environmental parameters, systematically altering its form to ensure a basic level of human comfort: not just enough to survive, but to thrive. This is less mysterious than it sounds. The essence of Murcutt’s design method is to alter a continuous height living space with three variables: the configuration of a screening wall to modulate air circulation and sun angle; the shape of the roof and ceiling to provide shade, generate air turbulence (if desired), and allow a hot air pool (if needed); and the relationship to the ground (the house will be off the ground, for example, if the climate is hot and humid).

Murcutt has tried to evolve house into a version of itself that will most perfectly survive in its specific ecology.

We accept a Murcutt building as a part of its landscape much like the tree or turtle we might find there. This is interesting in part because central to architectural discourse since the collapse of the Modern movement is the question of how to be responsible to place but free of the tyranny of visual context. Obviously turtle or tree, which don’t flourish everywhere, could be replaced by other organisms in the analogy. And you could stretch it with something like crow, which can flourish in many places. Analogies have limits. But I like, in tree — aside from the many kinds of trees, and the outrageous variation that you find in families of trees — the evocation of a building as something rooted, aging slowly and at the mercy of its situation, which it has, to a certain extent, out-designed.

A Murcutt building is not an ugly pet. It is very much a whole creature.

But I especially like turtle, since turtle has a greater capacity than tree to adjust its location — to situate itself — beyond the circumstance of where a seed falls. And the opening and closing of a building is somehow closer to what a living turtle and its carapace can accomplish in reaction to stimuli. A tree is responsive to a limited extent, by curling its leaves, which is like opening single windows in the Forum. But a turtle can fully retract into its own shade, which somehow suggests the active component of dwelling within. And then I perversely like turtle because it is a discrete entity, and people have kept turtles as pets simply because they are fascinating, entire, and beautiful. A Murcutt building is not an ugly pet. It is very much a whole creature, an organism beyond the sum of its parts. As mysterious as it sounds, it is understood to require that entirety in order to belong to its place.

Office Building 2226, Lustenau, Austria.
Office Building 2226, Lustenau, Austria. [Eduard Hueber]

One last thing. In the Loblolly House, the Oskar von Miller Forum, and any Murcutt house — all remarkable environmentally sensitive buildings — we see a hyper-specificity (though differently applied) that may turn out to be a characteristic of the new architecture of the sustainable. But that may turn out to be a false lead. Turtle is always turtle. But a building will not always remain a turtle. It may need to become a tree, or a cow, or a thistle, or a church, or a yoga studio, or a data storage facility. Central to sustainability is a separate concept that complicates the seeming self-justification of these architects’ varying but related means. Program — that notorious driver of Modern form — will have to fit loosely into future buildings, the carbon footprint of which trends downward as a function of longer lifespan. Perhaps the correct analogy is turtle shell, one that somehow accommodates a far greater range of creatures within its perfection.

Program — that notorious driver of Modern form — will have to fit loosely into future buildings.

To date that possibility is best exemplified by the small office building designed a few years ago by the Austrian firm Baumschlager Eberle. Located in Lustenau, on the German border, “2226” functions without a cooling or heating system: the name refers to the target range for the interior temperature in Celsius. Like the Forum, it’s a constructed manifesto. The interior spaces of the six-story masonry building are warmed primarily with the heat put off by the inhabitants’ bodies and their computers, secondarily by passive solar gain through the windows. They’re cooled by flushing out the air at night, which brings down the base temperature of its substantial thermal mass. Here hyper-specificity occurs only in the conception of the generic section. 2226 shares with many recent projects in Europe something at once relatively new, but also very old, predating the Modern: the reduction of a building to a simple, regular, systematic box, generally devoid of eccentric formal gestures, in which the formal recognition of particular use or location is entirely suppressed. It is — despite the little rotation of the mass to mark an entry — interestingly monumental, but perhaps its performance in its landscape is worthy of such exalted status at this time.

  1. From the Renaissance through the 19th century, the design of an important institutional building in the generally continuous context of most European cities might have been begun by determining from the adjacent context what the scale (both overall, and of parts) of the primary façade was going to be, then figuring out how the program was going to be most sensibly disposed within the limited flexibility granted by the order of the exterior. The Forum does this in reverse: the site parameters (like the scale of adjacent buildings) are used to tune the exterior. If it is not as lucidly related to its context, that points to an interesting question about what the presence of a sustainable building in an urban condition should be: to what degree should it cooperate? Here, in any event, you see sustainable technologies subsumed into the larger post-Modern discourse of landscape stability (respect for site, context, etc.) that preceded the rise of environmental concern.
  2. This analogy, often quoted or paraphrased, is put in context nicely in E.M Farrelly’s introductory essay in Three Houses: Glenn Murcutt (New York: Phaidon Press, 1993).
David Heymann, “The Ugly Pet,” Places Journal, September 2017. Accessed 28 Sep 2023.

If you would like to comment on this article, or anything else on Places Journal, visit our Facebook page or send us a message on Twitter.