Skip to Content

Things We Don't Know We Don't Know

We all rely on shortcuts: rules of thumb, accepted convention, common knowledge. What if they're wrong?

"We must have a body." —Gilles Deleuze
"We do not even know what a body can do." — Baruch Spinoza

Architects must have buildings, but we do not even know what buildings can do. Like the at-once utter familiarity and utter strangeness of our own bodies, we know little of how buildings actually perform and behave. This is perhaps best illustrated by too many recent LEED Platinum–certified buildings that are documented to perform worse than baseline code buildings.

The recent renewed focus on building science and building performance in the discipline and practice of architecture is therefore very welcome (especially after a period defined by the theatrics of quarreling styles). In the context of increasing demand for diminishing resources, this turn in attention promises fundamental transformations for architecture in the coming decades. Given the paucity of knowledge about the actual performance of buildings, it may seem that the value of embracing building-science research would be the introduction of more certitude into design practice. But the greater value, and greater need, is the introduction of doubt. Currently, there may be no more efficacious way to increase both the rigor and the vigor of the profession.

The more one learns about building science, the more one begins to question central assumptions — assumptions that are as widely taught as they are pervasively practiced. The logics of air conditioning, multilayered wall assemblies, and R-values, for example, become more dubious the more one thinks deeply and systemically about bodies, building performance, and global resources. Study of the assumptions at the beginnings of air-conditioning science reveals compelling insights and oversights: Willis Havilland Carrier brilliantly ignored the role of radiant heat transfer in his calculations (even though that is the body's primary mode of heat transfer) because he was solving a problem for hygroscopic machines and, later, designing a new industry; he was not designing for the comfort of humans or the performance of buildings. Likewise, the flawed concept of R-values also ignores multiple forms of heat transfer such as the important but neglected role of thermal diffusivity for more massive materials or the role of convective flows in insulating materials such as batt insulation, which undermines their capacity to resist heat flow. In short, the more an architect thinks about the relationship between our physiology and the science of heat transfer, to consider only one example, the more apparent the fact that many platitudes of contemporary construction are based not on sound science but instead on a broad network of sometimes unwarranted suppositions and habits. If the discipline of architecture is to become more deliberate, more effective, and therefore more respected, architects will need to develop a habit of questioning assumptions: They cannot acquiesce to conventions and customs any more than they can capitulate to the rhetorical escalations of "new" techniques and technologies.

Exposing the flaws

Questioning assumptions is key to any effort to advance the integration of building science with building design. Too often, cutting-edge, if not glib, techniques are seen as driving revolutions in architecture — a worn-out trope borrowed from early Modernism, when new materials and technologies promised new architectures. However, as German philosopher Peter Sloterdijk has suggested, history has not been a process of revolutionary modernization. Instead, as Sloterdijk has written, it has been about a process of "explication," which he defines as "the revealing inclusion of latencies and background data in manifest operations." In other words, real progress is most often made by reconsidering what we think we know and reexamining the layers of presumed fact, supposed truth, and accepted practice. Any substantive, meaningful shifts in the practice of architecture in the coming decades will likely emerge from a similar process: the overt explication of prior practices.

What might such shifts look like? I can offer a couple of suggestive examples from my own recent work. Based on research that questions the assumptions and practices of 20th-century architecture, my work has focused on three topics that suggest three new modes for building design. First is the role of thermally active surfaces in architecture as an alternative to air conditioning; this paradigm finally activates the corpus of the body and the building in the same thermodynamic space. Second is an examination of lower-technology, higher-performance design rather than the planned obsolescence of higher-technology, lower-performance approaches of recent decades; this is a mongrel paradigm of durability, adaptability, and resilience that leverages the intelligence of both archaic and contemporary techniques. The third is based on an overt recognition of Einstein's observation that matter is but captured energy. Energy and material systems have been taught, designed, and engineered as disparate entities — a thoroughly false division that chronically handicaps architects. An alternative, integrated paradigm conflates energy and matter, thereby drawing attention to new hybrid approaches to building materials and systems.

Knowing what we don't know

"There are things we don't know we don't know ." — Donald Rumsfeld

Even now, as we are beginning to better understand building performance and behaviors, buildings are becoming increasingly obscure. For example, we know slightly more, perhaps, about the performance of certain building assemblies, but the geography and ecology of the materials that constitute those assemblies are largely as indeterminate as ever. This is a disconcerting split in knowledge.

Moreover, the reliability of new research is not guaranteed. New knowledge about building performance is often based on energy modeling and simulation, which give the appearance of accuracy and objectivity. But nearly all numerical models of reality are as incomplete as they are inaccurate. Further, they typically serve to answer only small questions in architecture. Numerical models are equally remarkable for what they tell us and what they conceal about building performance. (This is especially true for weak simulation programs such as Ecotect that are attractive because of their easy, graphic interface.) Numerical models do not verify the actual performance of a building or building assembly; instead, they verify performance measured against the assumptions embedded in the parameters of the model itself. What current digital models ignore is as important as what they analyze; their output cannot be any better than their inputs. The words of philosopher George Grant apply to the current interest in building simulation: "We can hold in our minds the enormous benefits of a technological society, but we cannot so easily hold the ways it may have deprived us, because technique is ourselves."

Admittedly, recognizing the failings of this technological and computational determinism can lead to despair: How can we make any progress in the face of so many unknowns? The answer lies in a perhaps unexpected realm: judgment. Judgment itself is a profoundly sophisticated and integrating algorithm, a robust method of modeling and simulation. In a period of increased interest in science and technology in design, judgment prevents us from descending into pure technique, from becoming pure technique ourselves. As such, science and technology in architecture should always remain a subset of judgment. Despite the availability of increasingly sophisticated simulation techniques, the most refined processor and algorithm in architecture remains the integrating capacity of the mind, and the most subtle thermodynamic and physiological processor remains the body. This is what can make architecture so rich, and maintaining this hierarchy is necessary if we are to integrate design with life rather than subjugate life to mere technique.

The most consequential aspects of any science experiment are, first, the examination of the assumptions that condition the experiment and, second, the evaluation of the results of the experiment — in short, the application of explication and judgment. The practice of architecture — that which will determine the future of the profession — must always be in this sense an experimental practice, one that applies extensive explication and robust judgment to the integration of science and design. Only then we will begin to know those things we don't know we don't know.