Building Bad: How Architectural Utility is Constrained by Politics and Damaged by Expression
Jonathan Ochshorn

contact | contents | bibliography | illustration credits | ⇦ chapter 5 | chapter 6 | chapter 7 ⇨


book cover

The LEED Reference Guides give buildings points for providing outside air (ventilation) and daylight, in spite of the fact that the relationship between light and air, on the one hand, and sustainability, on the other hand, is hardly self-evident. Let's look at these items separately before examining them together.

The need for bringing outdoor air into buildings arises from the pollution of indoor air, not only from building products and processes (e.g., incomplete combustion from stoves, radioactive radon from the ground, lead from old paints, asbestos from various old finishes and insulation, mold growth associated with excess humidity, and volatile organic compounds—VOCs—from paints, preservatives, aerosol sprays, cleansers, air fresheners, dry-cleaned clothing, certain building materials and furnishings, photocopiers and printers, glues, and permanent markers), but also from people themselves (e.g., increased concentrations of CO2 due to ongoing respiration, as well as all the nasty artifacts associated with scents—perfumes—and deodorants). Assuming that outdoor air is not, itself, polluted, it makes sense to dilute the polluted indoor air with fresh air from outdoors, thereby improving the indoor air quality. Doing so is therefore a function of buildings, and has been considered to be one since the acceptance of "germ theory" in the late 19th century and the subsequent codification of air quality provisions, initially in "hygiene" regulations promulgated by Boards of Health, then in various ad hoc housing laws, and ultimately in building codes.1

Providing fresh air is, in fact, a requirement in modern building codes and a required "prerequisite" in the LEED Reference Guide, but this requirement is fraught with contradictions and difficulties. The most fundamental contradiction is between, on the one hand, the desire for fresh (outdoor) air and, on the other hand, the desire to reduce building energy consumption. This conflict is most acute in air-conditioned buildings during hot, humid weather: compared with recirculating "stale"—but relatively cool and dry—indoor air, dehumidifying and cooling outside air is quite energy-intensive. From this standpoint, the drive to lower energy use is therefore in direct conflict with the desire to improve indoor air quality. This has led to more sophisticated strategies for bringing in fresh air, not on a continuous basis according to flat rates tabulated in building codes, but rather only to the extent that it is actually needed. To determine when a room needs more fresh air, it is theoretically possible to continuously measure levels of likely pollutants in the room, including carbon monoxide, particulate matter, VOCs, and so forth. This being quite costly, the more common approach is to place CO2 sensors in individual rooms, where the level of CO2 is assumed to be an adequate stand-in for the whole array of possible pollutants. Such sensors, if calibrated correctly—something far from certain2—activate mechanical ventilation systems bringing in outdoor air only when a predetermined threshold ("setpoint") is reached; the name of this system, demand-controlled ventilation, reflects this energy-saving strategy.

Legislating increased outdoor air by upwardly revising tabulated ventilation rates (or recommending such increased ventilation rates for LEED points) may also be problematic, paradoxically leading in some cases to more—not less— indoor air quality problems. Joseph Lstiburek makes the case as follows: "Overventilation in hot, humid climates has led to more indoor air problems due to mold resulting from part-load issues than underventilation anywhere else … In my not-so-humble opinion, all of the [Code-based prescriptive ventilation] rates have been just wild guesses without a sound epidemiological basis. But, the resulting mold from overventilation is real and demonstrable."3

A discussion of mold growth due to increased ventilation rates would not be complete without at least briefly mentioning a 2015 study that attributed ghost sightings in "haunted" houses to toxic mold growth. Professor Shane Rogers of Clarkson University argues that "human experiences reported in many hauntings are similar to mental or neurological symptoms reported by some individuals exposed to toxic molds. It is known that some fungi, such as rye ergot fungus, may cause severe psychosis in humans. … Hauntings," he continues, "are very widely reported phenomena that are not well-researched. They are often reported in older-built structures that may also suffer poor air quality."4 While Sigmund Freud argued in 1919 that "many people experience the feeling [of the uncanny] in the highest degree in relation to death and dead bodies, to the return of the dead, and to spirits and ghosts"5 and Anthony Vidler in 1992 added that "the uncanny has, not unnaturally, found its metaphorical home in architecture: first in the house, haunted or not, that pretends to afford the utmost security,"6 Rogers's research calls into question the notion that "haunting" is a condition of psychological, economic, or political alienation or estrangement. Instead, we can invest the "ghost in the machine"—a phrase introduced by Gilbert Ryle in 1949—with an entirely new meaning: more than 70 years after Ryle, the questionable recommendations of ventilation enthusiasts may well have surrounded the entirely rational fear of global warming and sea-level rise with an extra dose of psychotic paranoia.

Just as fresh air ventilation can be provided either naturally or mechanically, light can be provided either naturally (daylight) or electrically. The International Building Code calls electric light "artificial light," which is a bit strange, since electric light is still light and not a "copy" of light. And just as the provision of fresh air is not necessarily compatible with maximum energy efficiency, the provision of natural daylight may also, in some cases, be in conflict with energy efficiency. This is because glass, through which daylight enters buildings, is not only less thermally resistive than typical opaque walls, but also permits high-frequency radiant energy to pass through, while trapping the lower-frequency energy re-emitted: welcome to the greenhouse effect. While such greenhouse behavior is often idealized as a method to use solar energy passively to heat buildings, the reality is that many commercial buildings are challenged by significant cooling loads, so that the addition of the solar heat gain that inevitably accompanies daylight may be counterproductive from the standpoint of energy efficiency. This potential conflict can be measured by modeling energy loads associated with heating, cooling, and lighting for different window–wall ratios in different climate zones. Making some assumptions about climate, location, orientation, and window–wall ratio, such models have indeed been made, for example, as illustrated in case 1 of Figure 6.1, based on a study of office buildings in Italian cities.7

Graph showing energy demand vs. window-wall ratio.

Figure 6.1. Primary energy demand vs. window-wall ratio in a typical office building comparing two lighting scenarios: a lighting density of 20 W/m2 (case 1) corresponding to fluorescent light fixtures and a lighting density of 5 W/m2 (case 2) with highly efficient LED lighting. Comparing case 1 with case 2, the optimal window-wall ratio goes down from about 25% to about 15%.

What case 1 in this diagram shows schematically is that total energy consumption (demand) in a typical office building using fluorescent lighting—represented by the solid line at the top of the graph—depends on the ratio of window to exterior wall area (WWR) and that, furthermore, there is an optimal WWR at which this total energy demand is at its minimum value (labeled "MIN" in the figure). That this optimal average ratio ranges from 23.5 percent to 32 percent—depending on whether the building walls are relatively uninsulated, or whether the best spectrally selective glazing is used—is the conclusion drawn in this particular study. The long-dashed line in Figure 6.1, representing energy used by lighting, falls dramatically as the WWR increases, corresponding to the need for less electric lighting as daylighting is increased. On the other hand, heating and cooling loads, represented by the shorter dashed lines, increase in a modest way as the WWR increases, since more power for heating and cooling is required when there is more glazing. Adding together these three sources of energy demand—with heating and cooling demand increasing, but lighting decreasing, as WWR increases—creates the low-point in the total energy curve, and therefore the minimum (optimal) value for WWR.

The idea that there is an optimal ratio of window to wall area has stimulated all sorts of research into strategies for using daylighting to reduce overall energy consumption since, at least in the past, electric lighting was a large contributor to overall energy use in buildings. And when energy models assume fluorescent fixtures with a lighting density (the amount of power needed per unit area) of 20W/m2, as is assumed in case 1 of Figure 6.1, the logic of capturing daylight to reduce electric light consumption makes sense when the percentage of window area is relatively modest. The problem is that fluorescent fixtures are rapidly being superseded by LED fixtures that use far less electricity. And when the energy consumed by lighting goes down sufficiently, the total energy used in buildings no longer validates even the relatively modest areas of glass deemed optimal for energy savings with fluorescent lighting. It is now feasible, using LEDs, to design buildings with a lighting density as low as 5.0W/m2, according to California's Codes and Standards Enhancement (CASE) Initiative for 2019.8 And with the lighting density thereby reduced from 20W/m2 to 5.0W/m2 (case 2 in Figure 6.1), the WWR that optimizes the energy costs of heating, cooling, and lighting occurs when the amount of glazing is reduced to about 15 percent of the total exterior wall area. The surprising result of incorporating more and more efficient LED lighting in buildings is that energy demand is actually optimized when almost no glass is placed in exterior walls; in other words, the "MIN" or optimum WWR approaches zero.

The increasing efficiency of LED lighting has also rendered the distinction between shallow and deep floor plates moot, at least from the standpoint of energy use, since "advancements in lighting have reduced the sensitivity of the ratio of interior to exterior office spaces and, as a result, limited differences are seen."9

Some disclaimers are in order. First, optimal values for window–wall ratios depend on many variables, including the climate zone, the occupancy type (e.g., office vs. residential), the quality (energy efficiency) of the glazing and the solid portions of the exterior wall, and the building geometry. Second, strategies involving things such as light shelves (horizontal surfaces that bounce daylighting further into deep floor plans) can be deployed to improve the viability of daylighting strategies. Third, there may be other reasons for using glass that have nothing to do with energy efficiency—mainly that humans often like to look outside when they're inside buildings. Nevertheless, simply assuming that daylight saves energy, even with energy and daylight modeling, is no longer self-evident.

So much for fresh air and daylight considered independently. It turns out that reform movements in the late 19th and early 20th centuries—motivated by mid-19th-century studies that "began to indicate a correlation between the lack of light and air, and spread of disease"10 —considered the functions of providing light, from the sun, and fresh air, from the outside, to be inseparable. In tenement housing, with mechanical ventilation non-existent and electric (or coal, oil, natural gas) lighting, where present, providing only a minimal level of illumination, the exterior window seemed like the perfect remedy, since it let in both light (when open or closed) and air (and not only when open; leaky construction practices insured that at least some outside air would enter buildings even with windows closed in the winter months). Yet windows were not always provided for all rooms in low-end housing: the prevailing pattern of speculative tenement construction in places such as New York City took maximum advantage of available property by constructing buildings with up to 90 percent lot coverage. Windows facing the street were effective, but many interior rooms had no connection to the outdoors. Sometimes, but not always, interior rooms would have nominal access to light and air via small air shafts, but in a context where the smells of cooking—on stoves with no mechanical exhaust systems—or of excrement—from outhouses or water closets placed in side or back yards—could be overwhelming, access to fresh air and daylight required legislation that addressed conditions on both the exterior and the interior of buildings.

In terms of the building exterior, minimum dimensions of courts and yards between buildings on adjacent lots in New York City were established, first, by the Tenement House Act of 1879 (creating "dumbbell" or "old law" tenements) and, 22 years later, by the Tenement House Act of 1901 (creating "T-shaped" or "new law" tenements). The latter act mandated larger air shafts and stipulated a maximum lot coverage of 70 percent. To improve conditions on the building interior, operable windows were required in all rooms. Since conventional double-hung windows, with upper and lower operable sashes, provide twice as much daylighting area (transmitted through the glass in both sashes) as ventilation area (since only one sash can be fully opened; or both sashes only half-opened), building codes beginning in the 1940s conveniently stipulated that windows in each habitable room must be at least half openable, perfectly consistent with the prevailing window technology of the time. Minimum window sizes (areas) were established based on some fraction of the room's floor area, this fraction getting smaller and smaller as modern model building codes evolved. In the U.S., the "Uniform Building Code" (UBC) required overall window areas to be at least 12.5 percent of each habitable room's floor area in 1937, with a stipulation that the area be at least 12 square feet 1.1 m2) added in 1946; this was reduced to 10 percent of the room's floor area in 1970, with the minimum window area also reduced from 12 to 10 square feet 0.9 m2) in 1973; and this was further reduced to 8 percent of the room's floor area with the introduction of the International Building Code (IBC) in 2000.

Through these incremental revisions, buildings codes permitted the area of windows required for light and air to be reduced, yet an even more radical vision began to be implemented that implicitly questioned the functional status of windows as elements necessary to promote human health through the provision of light and air. Certainly, the development of air-conditioning systems and mechanical ventilation in sealed commercial buildings provided evidence that the function of providing light and air was not intrinsically dependent on having operable windows that provided "natural" light and ventilation. Natural ventilation was the first casualty. Beginning in 1970, the UBC no longer required windows to be openable in habitable rooms as long as some fresh air was delivered by other means: "In lieu of openable windows for natural ventilation, a mechanical ventilation system may be provided. Such system shall be capable of providing two air changes per hour in all guest rooms, dormitories, habitable rooms, and in public corridors. One-fifth of the air supply shall be taken from the outside."11 Ironically, the option of mechanical ventilation is now—for most new residences—a requirement, since it is no longer rational to assume that outside air will infiltrate through leaky windows and doors and, in doing so, provide sufficient fresh air to dilute indoor pollutants when those windows and doors are closed. And in parts of the world where outdoor air is too polluted to provide any consistent relief from indoor air pollution, it makes some sense to completely invert the assumptions made by those late 19th-century reformers—that the function of windows was to introduce natural light and air into buildings—and instead prohibit operable windows in such locations, relying instead on mechanical ventilation with effective filters.

Perhaps more surprising than the negation of natural ventilation is the gradual elimination of requirements for windows to provide natural light. Beginning with the 1970 UBC, skylights were allowed to provide natural light instead of windows; in the 1994 UBC, artificial (i.e., electric) light was permitted to replace daylight in kitchens only, with natural light still required in other habitable rooms; and finally, with the inauguration of the 2000 IBC, all spaces were given the option of using artificial light instead of natural light, with a minimum illumination requirement of only 10 footcandles (108 lux). Thus, 99 years after New York City's Tenement House Act of 1901 required operable windows in all rooms, a point was reached in the U.S. where—at least outside of New York City12—neither natural lighting nor natural ventilation was deemed necessary in any occupied room.

A discussion of light and air as functions of buildings would not be complete without placing such functionality in a political context. While modern building codes may have eliminated requirements for natural ventilation and daylight, requirements for the provision of some sort of light and air are unlikely to be challenged, since they can easily be justified on the basis of health and safety considerations. Yet such considerations are not absolute, and the boundary between health and safety, on the one hand (e.g., through the provision of light and air), and the freedom to use one's property as one sees fit, on the other hand, has often been contested.

This can be seen in the "Matter of Application of Jacobs," where the New York State Court of Appeals ruled in 1885 that an 1884 New York State Law entitled "An act to improve the public health by prohibiting the manufacture of cigars and preparation of tobacco in any form in tenement-houses in certain cases, and regulating the use of tenement-houses in certain cases" was unconstitutional.13 What is most interesting about this case is not that the court attacked the "well-established health-related rationale for exercise of a state's police powers,"14 but rather that it used rather extreme language in making a case for property and freedom. In other words, the opinion did not actually deny the right of governments to pass legislation designed to protect public health, but insisted on a clear and consistent rationale:

When a health law is challenged in the courts as unconstitutional on the ground that it arbitrarily interferes with personal liberty and private property without due process of law, the courts must be able to see that it has at least in fact some relation to the public health, that the public health is the end actually aimed at, and that it is appropriate and adapted to that end. This we have not been able to see in this law, and we must, therefore, pronounce it unconstitutional and void.15

Yet the court's underlying bias in favor of property rights clearly shows up when it argues that any legislation curtailing property rights potentially leads us down a slippery slope towards autocratic government, thereby threatening our birthright of freedom:

Such legislation may invade one class of rights to-day and another to-morrow, and if it can be sanctioned under the Constitution, while far removed in time we will not be far away in practical statesmanship from those ages when governmental prefects supervised the building of houses, the rearing of cattle, the sowing of seed and the reaping of grain, and governmental ordinances regulated the movements and labor of artisans, the rate of wages, the price of food, the diet and clothing of the people, and a large range of other affairs long since in all civilized lands regarded as outside of governmental functions. Such governmental interferences disturb the normal adjustments of the social fabric, and usually derange the delicate and complicated machinery of industry and cause a score of ills while attempting the removal of one.16

This notion of a slippery slope—with any governmental intervention posing a threat to property rights and freedom—is hardly an anachronism of 19th-century jurisprudence, but is alive and well in the modern Supreme Court, supported by the Fifth and Fourteenth Amendments to the U.S. Constitution, the former preventing governmental "taking" of private property except for the public good and when compensated; and the latter prohibiting the state from depriving "any person of life, liberty, or property, without due process of law." For example, Supreme Court Justice Antonin Scalia, writing in Lucas v. South Carolina Coastal Council in 1992, was more than willing to cite Justice Oliver Wendell Holmes's opinion in Pennsylvania Coal Co. v. Mahon et al. (1922) which reprised the "slippery slope" argument made in the 1885 case. Georgetown Law Professor J. Peter Byrne, quoting Scalia quoting Holmes, concludes that Scalia's "assessment of property use regulations was warped by his fear that if 'the uses of private property were subject to unbridled, uncompensated qualification under the police power, "the natural tendency of human nature [would be] to extend the qualification more and more until at last private property disappear[ed]."'"17

Thus, the apparently self-evident functional requirements for light and air in buildings are regulated by governments through building codes that prescribe minimum window areas (or minimum levels of "artificial light" and/or mechanical ventilation) and, more indirectly, through zoning ordinances that regulate the massing or bulk of buildings (in part to ensure that light and air reach buildings and public rights-of-way). Yet the functional provision of minimum standards for public health and safety is constantly threatened by claims for the freedom to dispose of one's property without constraint. And while the trajectory of court opinions in the U.S.—at least since the 1926 validation of zoning in "Euclid v. Ambler"—seems to point in the direction of increasing governmental intervention on behalf of public health and safety, the advocates for freedom and the unbridled exercise of property rights have hardly conceded defeat: "Liberals might root against the government … But they should be careful what they wish for. The conservative majority can, and most likely will, rule against the government using broad theories that would also eat away at the constitutional foundations of the New Deal system, which is essential for protecting health and safety, the environment and much else."18


1 Plunz, History of Housing in New York City.

2 Fisk, Faulkner, and Sullivan, "Accuracy of CO2 Sensors."

3 Lstiburek, "Energy Flow Across Enclosures," 64 (footnote).

4 "Can 'Ghosts' Cause Bad Air? Poor Indoor Air Quality and 'Sightings,'" ScienceDaily, March 31, 2015, here.

5 Freud, "The 'Uncanny,'" 241.

6 Vidler, The Architectural Uncanny, 11.

7 Marino, Nucara, and Pietrafesa, "Window-to-Wall Ratio."

8 Uraine et al., "Indoor Lighting Power Densities," 151 (see Table 75).

9 ASHRAE, Advanced Energy Design Guide.

10 Plunz, History of Housing in New York City, xxxii.

11 International Conference of Building Officials, Uniform Building Code Vol. I (Pasadena: ICBO, 1970), 81.

12 As of this writing, habitable rooms in dwellings are still required to have both natural lighting and ventilation in New York City, which has a municipal building code that is different, in this respect, from the IBC, from which it is derived, as well as the New York State Building Code.

13 Matter of Application of Jacobs, 98 N.Y. 98.

14 Furner, "Defining the Public Good," 244.

15 Matter of Application of Jacobs, 98 N.Y. 98.

16 Matter of Application of Jacobs, 98 N.Y. 98.

17 Byrne, "A Hobbesian Bundle," 735–36.

18 Eric Posner, "The Far-Reaching Threats of a Conservative Court," New York Times, October 2, 2018, here.

contact | contents | bibliography | illustration credits | ⇦ chapter 5 | chapter 6 | chapter 7 ⇨