The following text was drafted in response to the first prompt in AN’s “Post-Pandemic Potentials” series. A previous response, by Mario Carpo, argued that all the changes ushered in by the pandemic are likely to be reversed. Read more about the series here.
As the last of our mid-term reviews were completed in early March, the looming red dots of the COVID-19 tracker were still well east of New Haven—indeed, an ocean away. The design studios on the upper levels of Rudolph Hall were strewn with the typical detritus of the charrette, as our students departed for spring break and a well-earned two-week respite. Within a few days, however, Yale closed our campus out of an abundance of (what is now clearly justified) caution and we had abandoned the building altogether, told our students they would not return to those studios for the rest of term, and were scrambling to heave the entire curriculum into an online mode. Instead of resetting our spaces for the last push of the term, we turned to creating digital collaboration infrastructure, teaching everyone how to use Zoom, and chasing our students around the globe to verify their time zone, home equipment, and connectivity.
Our faculty, left with little option, readied their webcams, learned WhatsApp, and finished the term, exhausted. And while we agreed with everyone that much was lost in the diaspora, the virtual educational experience mirrored the professional, where global architecture firms, having abandoned their studios, were even reporting an uptick in productivity. Already in February I had argued to our students that they were likely to see several economic downturns during their careers; the same is surely true for pandemics, so at bare minimum, this was good practice.
And while both architectural education and practice may have become dependent on Mario’s “technical logic of electronic computation,” it is by no means certain that complete reliance will persist when vaccines and sufficient death and economic destruction convince the citizenry that while this crisis may be over, another is sure to come and we’d best be better prepared. It’s too early to determine the effects of the loss of personal contact, physical models, and actual job site visits on the efficacy of architectural practice, but it seems that the profession has moved—if ever-so-slightly—into the digitized worlds of knowledge work our clients developed years ago and upon which many heavily rely.
But architects are nowhere near Mario’s optimistic but technologically unjustified state of “technical logic of modernity,” a world where the scientific method, descriptive predictions, and models of reality like formulae, equations, and infinitely indexed data mean we no longer have to “know anything” at all because everything is a Google search away. This is certainly true for Mario’s Gmail inbox and a lot on the web, but not the data generated by architects out there. Most of the latter is unformed, generated in wildly differing and disconnect formats, inaccessible, and thereby not particularly useful. In an earlier time before the explosion of data-generating tools this was called the “interoperability problem,” but it’s gone far beyond that. Lacking any cohesive strategy for creating, organizing, or accessing that information, its value goes largely unrealized.
So that steeper hill of technological opportunity—in the language of the prompt—has been out there for a while but architects have avoided or ignored it rather than climb it, or even measure its height. The highest quality data in even an advanced office today is used for shape generation (scripts, Rhino), visualization (VR, VRay), or precision working drawings (Revit). Generative design approaches, artificial intelligence, or behaviorally provocative, predictive models are few and far between.
Nicholas Negroponte’s model of technology evolution, described in detail in 1970’s The Architecture Machine, is particularly instructive. He posited there that new technology first mimics the manual processes it replaces, then augments those same processes, thus paving the way for real innovation. In architecture, we’re stuck at best in the latter stages of that first phase, having ventured a few tentative forays into the second. Extravagant building forms and the occasional energy model notwithstanding, transformation is the peak of another hill not yet seen, much less scaled.
That’s not to say that we’re not technologically adept; the rapid pivot from physical to virtual offices demonstrates a certain grasp of design and virtual collaboration tools. But today’s design studio and construction site are about as far away from the technology that’s given us self-driving cars—detailed predictive models of rapidly changing reality—as the filmstrips I used to watch in elementary school decades ago are to YouTube. This, of course, is not news. As our friends from McKinsey have documented, the building industry is one of the least digitized (and least productive) industries on the planet.
It’s far too early to even pretend to understand our post-pandemic circumstances, making this exercise something like weather prediction: Looking at past conditions and future probabilities and making some informed guesses about the likelihood of certain outcomes. So what are the chances that the virtualized experience of design and construction in the pandemic moves us faster up the path of technological evolution?
The field will likely be pushed forward not by innovative practitioners but rather by clients, who long before the pandemic relied on large data systems, artificial intelligence and machine learning, and predictive simulation to manage their supply chains, workflows, investments, and facilities. Those same clients likely currently occupy buildings controlled by large-scale digital control systems and monitors that generate mountains of useful data about their operations and will demand that architects deploy that data to inform the design of their future facilities.
After working remotely for several months, most of those same clients are realizing that perhaps physical proximity isn’t all it is cracked up to be. They realize that they can select architects from a much bigger pond, and firms will need to focus their competitive edge accordingly; as always, most will compete on price, but a select few will decide that data-informed design is a way to separate from the pack.
And suddenly the architect’s responsibility for the public’s health, safety, and welfare has taken on a new, potentially ominous dimension: epidemiology. Although the tradition of offloading technical responsibility is as old as the professionalization of architecture (at least in the United States), the safe occupancy of space during a pandemic is a quintessentially architectural problem well-suited to technical analysis, simulation, and prediction tempered by a good eye—the watchwords of innovative digital practice—and yet another opportunity that next-generation practices may embrace, and traditional ones vigorously eschew.
Meanwhile, our contractor colleagues promise to push full speed ahead on productivity and labor optimization on their job sites through data and automation and creating demand for technologies emerging from today’s BuildTech venture explosion, and creating increasing demand by builders for design data to drive automation and design-for-manufacturing. Architects can provide it if they can see past hidebound business models and risk approaches. The pandemic has demonstrated that today’s lean, lowest-first-cost supply chains are often abusive, very fragile, and ripe for disruption. Responsibly managing those supply chains means understanding the demand and flows of materials and goods converging on a construction site based on decisions that could originate during design should architects choose. At its root that’s a data modeling problem, and another competitive opportunity for the willing.
Of course, the same innovation headwinds facing architects are blowing even harder during—and likely long after—the pandemic. Tiny margins, low tolerance for risk, and commoditized fee competition leave little financial resources nor the incentive to innovate beyond the margins of productivity. The profession is still squeezing those gains out of its slow and reluctant adoption of BIM toward an uninteresting, and likely less and less efficacious, result: making better construction documents, clearly the path of least resistance. Almost 20 years after architects began to adopt BIM, this is hardly a transformative result. Prioritizing “geometric BIM” to make drawings, rather than “data BIM” to create insight, is a waste of a potential platform that might bring much-needed coherence to the array of architectural data out there, while simultaneously providing new business proposition along the way.
My friend Mario offers two options for designers in a post-pandemic world: “build bunkers or work elsewhere.” Bunkers may well be the uninspired route to virus-proof architecture, but surely we have the energy and tools—many technological—to avoid that grim fate. As for the latter, there actually is no “elsewhere” when, in the modern world, “everywhere” is subject to a virus. And architects working remotely can be anywhere. Perhaps a new generation of practices, invigorated by the possibilities of technology and a newly available global network of talent unencumbered by the vast overhead of office infrastructure, will take a few of the resulting hills.
Phil Bernstein, FAIA, is an associate dean and Professor Adjunct at the Yale School of Architecture and a former vice president at Autodesk. He spent most of his practice career at Pelli Clarke Pelli Architects. His book Architecture Design Data: Practice Competency in the Era of Computation, was published by Birkhauser in 2018.