Seventy percent say they hate modern buildings. Cities keep building them anyway.Do people prefer traditional architecture?

Modernism won the twentieth century outright. Steel and concrete left exposed, ornament stripped away, the complicated curves of traditional facades and street plans scrapped. By the 1950s it dominated global architecture.

The strange part is that even its defenders knew nobody liked it. In 1940 the English architectural historian John Summerson said the public 'can't understand it, never will, and hates it like poison'. In 1977 the critic Wolf von Eckardt announced that modernism was dead and added: 'No one mourned. It was never popular.' One man was sympathetic to modernism, the other hostile, and they said the same thing.

Ask people with photographs and the answer is remarkably consistent

To find out whether that was true, you need data. In 1979 the small town of Metuchen, New Jersey asked Rutgers University to run the first decent version of this: hundreds of local street photographs, rated by residents.

Residents overwhelmingly preferred streets with a small-town feel, detailed facades and a walkable scale, and disliked blank buildings and six-lane roads. What surprised researchers was the level of agreement. Taste was not purely individual.

Since then, around twenty such surveys have been run, covering Britain, the United States, the Netherlands, Portugal and Chile. In every one, more than 60 percent of respondents preferred traditional architecture, and in many cases more than 85 percent. The preference barely moves with age, gender, politics, income or nationality.

Britain is where the evidence is thickest. In 2023 Bath needed a new rugby stadium and got a utilitarian proposal. A small firm offered a classical alternative, Create Streets commissioned polling on the two, and the traditional design won 74 to 26. In 2025 Christ's College Cambridge planned a library by Grafton Architects; Create Streets drew up a Gothic scheme of identical floor area and 71 percent of over 2,000 British adults preferred it.

Most extreme was Marks and Spencer on Oxford Street: a scheme retaining and extending the 1920s facade took 79 percent against 17 percent for demolition and replacement. Even height does not break the pattern. At Shoreditch Works, where the proposed blocks are far taller than the warehouses they replace, 76 to 78 percent still preferred the new scheme.

One group disagrees: architects

In 1987 a Cambridge undergraduate named David Halpern asked students to rate twelve faces and twelve buildings. The faces drew consistent scores. The buildings did not: architecture students diverged sharply from everyone else, often ranking them in the opposite order. The building non-architects rated ugliest was the architecture students' second favourite.

Later research keeps finding the same split. The Portuguese researcher Luis Balula separated experts — architects, planners, developers, academics — from everyone else in Évora. Both groups rated older buildings highly, but they parted ways on modern ones.

This also explains the puzzlement of CABE, a British public body, after surveys in 2002 and 2005. The 2005 report noted regretfully that modernism's emphasis on light, sun and large windows ought to give it a clear advantage over traditional designs, 'but this advantage does not seem to have been understood by home buyers'.

So who is actually deciding

By 2020 the National Civic Art Society put seven pairs of real federal and courthouse buildings in front of Americans. Traditional designs averaged 72 percent, and the figure barely moved with politics: 70 percent of Democrats, 73 percent of Republicans. In 2025 the Dutch group The Aesthetic City found traditional preferences running from 67 to 92 percent, averaging 82 — and it used AI to generate the traditional alternatives, holding weather, angle and building mass constant.

That leaves the real puzzle. Almost every architecture school in the world teaches only modernism, and more than 99 percent of practising architects build only that. Britain has seen hardly a single public commission in a traditional style since the 1950s. What people say they want and what gets built have been separated by a gap seventy years wide.

The evidence is still thin — perhaps only three dozen fully controlled image pairs have ever been polled, mostly in Britain, the Netherlands and the United States. But the direction is clear. Why the gap exists, and when it closes, is for someone else to work out.

In the end, buildings are not a portfolio for architects to admire. They are places for ordinary people to live in, and ordinary people have been voting with photographs for decades without changing their answer.