When Colin Ellard's team walked people past buildings in London and Toronto wearing skin conductance sensors, they measured what anyone who walks through a modern city already suspects. Blank, repetitive facades produced measurable physiological stress. The body responded before the mind had formed an opinion.
Skin conductance doesn't have taste. It measures arousal — threat response, cognitive load. The finding was straightforward: exposure to low-complexity architecture registers in the nervous system as a form of environmental hostility.
Ellard's own interpretation is blunt: "All the connections are there to make the argument that if you're chronically exposed to a low-complexity streetscape, over time it's going to produce a measurable health effect."
The question is why we built this way at all.
What the Brain Was Built to See
The human visual system evolved with fractals. The branching self-similar patterns of trees, coastlines, river deltas, mountain ridges — every scale reveals more structure. The eye moves fluidly, always finding something new at the next level of magnification.
Nikos Salingaros, a mathematician who has worked extensively on architectural mathematics, quantified this in built terms. Traditional buildings across every culture — European Gothic, Islamic geometric, Japanese timber, Mughal carved stone — show coherent detail at six or more hierarchical levels of scale. A cathedral facade reads differently from across a square, from across a street, from arm's length, from inches away. There is information at every distance.
A glass curtain wall and a blank concrete plane offer roughly two levels: the building as a whole, and the mullion grid or panel joints. Between those scales, nothing. The visual system, tuned by millions of years of evolution to process richly structured environments, finds nothing to process. What registers physiologically is not mere boredom. It is closer to sensory deprivation — a mismatch between what the processing apparatus expects and what the environment provides.
Neuroscientist Adam Coburn and his colleagues demonstrated in 2017 that aesthetically unpleasant architecture activates the anterior midcingulate cortex, a region that receives direct input from the amygdala. Fear circuitry. Not metaphorically ugly. Neurologically aversive.
When Ornament Became a Crime
Architecture did not drift gradually toward minimalism. It was driven there at several specific historical moments, by people who understood they were making a break with everything that came before.
On January 21, 1910, Adolf Loos delivered a lecture in Vienna that would shape the next century of building. The title would later become infamous: "Ornament and Crime." His argument was explicit — "the evolution of culture marches with the elimination of ornament from useful objects." Ornament, he argued, was wasteful, degenerate, a marker of arrested development.
What's almost never quoted is what Loos said fourteen years later: "I never meant that decoration should be ruthlessly and systematically done away with. Only when time has made it disappear, can it never be applied again." His followers had already gone much further than he intended. They always do.
Le Corbusier's 1923 manifesto completed the argument. "A house is a machine for living in." Buildings should serve function as machines serve function. Ornament was not merely wasteful — it was philosophically wrong, a category error, evidence that the designer had not understood what a building actually was.
These were not aesthetic preferences. They were explicit moral positions. Ornament became coded as backward, bourgeois, unmodern. Aesthetics had been transformed into ethics. When you make ugliness a moral obligation, you make beauty a form of cowardice.
By 1933, CIAM had turned these ideas into planning doctrine. Cities would be rationalized into strict functional zones — living, working, recreation, circulation. Housing would be solved with tower blocks at wide spacing, freeing the ground plane for light and air. The charter was adopted across postwar Europe, where bombed-out cities needed fast housing. The ideology got implemented at enormous scale before anyone had gathered data on whether it worked.
Pruitt-Igoe in St. Louis was completed in 1956 and demolition began in 1972. Sixteen years. An entire neighborhood had been cleared to build it. The rubble was left where it fell. Charles Jencks called that first implosion "the death of modern architecture." What he meant was that the theory had met its test case — and failed visibly enough that the cameras showed up.
The System That Selects for Ugliness
Christopher Alexander's diagnosis was careful on this point. He did not primarily blame individual architects. "The architects who fully accepted the modern machine have hardly been more than pawns in the game which is much larger than they are."
The game is a system. Value engineering removes beauty from project budgets as a line item. The phrase heard in planning meetings — "let's save some room in the budget for a little architecture in the atrium" — tells you everything. Architecture has been reclassified as optional decoration. Strip it and you save money directly.
Approval committees compound this. Design proposals go through rounds of review by committees with strong aesthetic training in one direction. Anything that deviates from the established consensus takes more time, more explanation, more risk to the project. Risk-averse clients avoid all of it. The design regresses to the mean.
Taylorist construction finishes the job. When the craftsperson who designed a building also built it or inhabited it, every decision was made with feedback — errors became apparent and were corrected. Industrial construction separates design, drawing, fabrication, and installation into distinct labor categories with minimal information flow between them. The drawing is final. The building follows the drawing. Nobody on site has authority to adapt the design to real conditions. What results is not craftsmanship. It is execution.
The Same Pattern Everywhere
This sequence is not unique to architecture. Every domain that industrialized went through it: specialization, separation of design from production, optimization for metrics that could be measured while sacrificing qualities that couldn't.
Industrial agriculture eliminated soil complexity in exchange for short-term yield. The medical system eliminated patient relationships in exchange for diagnostic throughput. Software eliminated human judgment in exchange for process compliance. In each case, the system became highly efficient at producing the measurable output and increasingly poor at producing the thing the system existed for.
The ugliness of modern buildings is not primarily an aesthetic failure. It is an output of a system optimized for everything except what the built environment is for. Construction speed. Cost per square meter. Regulatory approval. Investor return. None of these metrics measure what happens to a person who walks past the building every day for twenty years.
The fractal complexity of traditional architecture was not a stylistic choice. It was the output of a different system — one in which makers adapted buildings over time, in which detail accumulated meaning through generations of use and modification, in which the person who built something lived with the consequences. That system produced environments that are neurologically compatible with how human beings evolved to experience the world.
The skin conductance sensor doesn't lie. The cortisol response is the signal that something is wrong. The question is whether a system optimized for other things is capable of receiving that signal — or whether it is too busy measuring something else.