Skip to Content
MIT News feature

The Many Careers of Jay Forrester

Computing pioneer Jay Forrester, SM ’45, developed magnetic-core memory. Then he founded the field of system dynamics. Those are just two of his varied pursuits.

It is a late March day in Massachusetts. The sky is clear, but the air is frigid and the wind fierce. Looking outside, Jay Forrester, SM ’45, turns to glimpse the swaying treetops. He knows all about the power of wind. As a teenager, he harnessed it to bring electricity to his family’s ranch in Nebraska.

Jay Forrester
Jay Forrester, in front of Whirlwind in the MIT Digital Computer Lab, is interviewed by Boston’s Channel 7 in 1957.

A precocious tinkerer, Forrester studied electrical engineering at the University of Nebraska. He arrived at MIT as a graduate student in 1939 and never left, staying on as researcher, professor, and now professor emeritus. He has been at MIT for more than half the time that has passed since the Institute opened its doors.

Yet Forrester’s 76-year MIT tenure is even more notable for its breadth than its duration. He helped develop digital computing. He oversaw the creation of a national air defense system and helped launch MIT’s Lincoln Laboratory. Then he joined what is now the MIT Sloan School of Management and founded the field of system dynamics, which examines complex business, economic, and environmental systems—and the unexpected feedback effects human activity creates within them. “I’ve had several careers,” says Forrester, who turns 97 in July. “Starting with ranch hand.”

When Forrester talks about his life and work—as he did with MIT Technology Review recently at his home outside Boston—it illuminates not just the diversity of the projects he has tackled but the links among those seemingly disparate careers.

Ranch hand

Forrester’s father, Marmaduke Montrose Forrester—“Duke,” for short—was a homesteader who acquired property under Abraham Lincoln’s Homestead Act (which was extended to Nebraska in 1904). When Forrester was born, World War I was still being fought, and he lived in one of the few houses in his area with indoor plumbing (something his mother, Ethel, had insisted upon). Both parents had gone to college and were teachers; by the time Forrester was in the third grade, he was riding a horse to the tiny local school, where his father taught him for two years.

“There were advantages to the one-room schoolhouse,” he says. Being around older students made advanced work seem less intimidating—and gave him an opportunity to explore his interests. The state of Nebraska sent a box of books to each school every year; one year, his school’s box included manuals describing how to make batteries and install burglar alarms. He studied those books and later salvaged old auto parts from a local junkyard to build a wind-driven electric plant for his family’s ranch.

The feat demonstrated Forrester’s willingness to initiate new ventures. But helping with his family’s business would also plant another seed for the eventual blossoming of system dynamics. “A ranch is a cross-roads of economic forces,” Forrester has written, noting that supply and demand, price changes, and agricultural uncertainties are a “dominating part of life” on the homestead. Ranch economics, like larger systems, plainly do not settle at equilibrium—a principal theme in his systems research.

Military engineer

Shortly after Forrester began graduate studies at MIT, he started working in the Institute’s Servomechanisms Laboratory, which was developing controls for radar-based anti-aircraft guns. During World War II, as he recalls it, the Navy requisitioned a prototype radar system from MIT for use on the USS Lexington (the second aircraft carrier bearing that name during the war; the first had been sunk). Forrester designed the hydraulic stabilization system that kept the radar pointing along the horizon. In 1943, the hydraulic controls malfunctioned, and Forrester volunteered to go to Hawaii and fix them. He got more than he bargained for.

While he was still doing his repairs, the Lexington was ordered into battle. He stayed on board while the carrier conducted combat operations near the Gilbert Islands on December 4. “You could see the torpedo bombers coming in,” he says. That night a torpedo hit the Lexington, killing nine. The Lexingtonsurvived and returned to Hawaii with a damaged propeller. “You could tell it just wasn’t right,” says Forrester, who had escaped harm in the control room. The officers remained calm—but his own face may have expressed some anxiety, since one officer quietly asked him to relax.

The Servo Lab work had an important influence on Forrester. “That brought me into the field of feedback systems,” he says. The U.S. military was trying to use radar data to aim its anti-aircraft guns automatically, but that data was imperfect. Understanding the feedback issue—how to correct for the imperfect data and send better information to the artillery—was essential to building accurate systems. In tackling this problem, the engineers developing anti-aircraft weapons recognized that the guns were not just sets of stand-alone parts that could be developed separately and then assembled. They were fully integrated systems: the interactions between parts were critical to their operation.

Forrester says it was only a “gradual realization” that the principles of feedback and systems could apply to society; he wasn’t thinking along those lines in 1943. Still, he notes, the military work was his practical introduction to those concepts.

Computing pioneer

After the war, Forrester began developing an airplane flight simulator. In the fall of 1945, an acquaintance, Perry Crawford ’39, SM ’42, suggested building a digital computer to run the simulations instead of the analog device Forrester was considering. The resulting Navy-funded computer, Whirlwind, used inefficient and highly unreliable vacuum tubes for memory. Then Forrester spotted an ad for a magnetic material that gave him the idea of storing digital data with magnetic fields instead of electrical charges. He led the development of magnetic storage technology that would replace Whirlwind’s vacuum-tube memory—and serve as the industry standard for memory for two decades.

By then, Navy funding for the flight simulator—and thus for the computer—had long since dried up. But after the Russians exploded an atomic bomb in 1949, the Air Force had stepped in to fund the development of Whirlwind, which would become the basis for its SAGE air defense system until the 1980s. Forrester likes to say that computing changed more between 1946 and 1956 than in any decade since. His work on Whirlwind remains the pride of his career in electrical engineering.

But because Whirlwind began as a simulator, it also set the stage for his research in system dynamics, which frequently employs simulations to detect the instabilities that emerge in complex systems. Forrester says now that his experiences with both feedback concepts and simulations were vital in his approach to analyzing social systems.

Lincoln Lab division leader

As director of the Whirlwind/SAGE project, Forrester became head of the largest division of MIT’s new Lincoln Laboratory, where he selected IBM to build the machines and then wrote the contract between IBM and the U.S. Air Force.

That contract stipulated that Forrester’s team had to sign off on every page of the design. A later agreement also required IBM to maintain the equipment, ensuring a high-quality product by giving IBM’s engineers incentive to avoid arduous repair trips to isolated SAGE sites in the dead of winter. “You’ve never seen an organization turn around so fast,” he says with a delighted laugh.

By 1956, Forrester was convinced that many substantial problems in computing had been solved. So some MIT leaders, including the Institute’s president, James Killian, suggested he consider joining MIT’s nascent management school. That year he moved from Lincoln Lab to the faculty of what would become Sloan.

To Forrester, who was already viewed as a computing pioneer, the move was not a drastic shift because, as he would later write, “I was already in management.” Besides, he says today, “my career changes have revolved around happenstances where somebody’s opened the door. And I’ve walked through it to see what’s on the other side.”

Management theorist

At Sloan, Forrester began examining a General Electric refrigerator factory in Kentucky that suffered from boom-and-bust episodes. He could have blamed the larger business cycle and left matters at that. But he began empirically studying the factory’s weekly orders, inventory, production rate, and employees.

Computing data by hand on a single notebook page in what would be the first system dynamics simulation, he found something curious. Management’s method of projecting its future needs, combined with time delays in the manufacturing process, prevented the factory from reacting optimally to changes in demand. So even small shifts in demand produced disproportionate fluctuations in production, along with significant swings in the number of employees being hired and then let go. “The internal structure and policies defined a manufacturing system that tended toward unstable behavior,” he would later write.

In 1961, Forrester published Industrial Dynamics, a management classic that describes this nonlinear nature of business operations. “Everything that changes through time is governed by a feedback system,” he explains. The little changes people make within a system can create big differences. And many interventions, such as ramping up production, often have harmful unintended consequences. To minimize them, you have to understand the dynamics of the whole system. Thus business-­level studies became the foundation of system dynamics—although Forrester-­style conclusions can be a hard sell to executives who think their willpower or charisma will help companies more than a nuts-and-bolts analysis.

“Very often people are just role players within a [company’s] system,” ­Forrester says. “They are not running it; they are acting within it. This has not been a popular idea with people who think they are in charge … but in fact, unless they are knowledgeable in systems, they will fall into a pattern of doing what the system dictates. If they understand the system, they can alter that behavior.”


Forrester wanted an accessible way to explain these management insights to his students. So he devised the Refrigerator Game, soon re-dubbed the Beer Game. In this table game, teams mimic the beer industry’s supply chain, trying to meet consumer demand while minimizing costly inventory. The entire incoming class at Sloan still plays the Beer Game each fall, led by systems expert John Sterman, the Jay W. Forrester Professor of Management.

The Beer Game can be a confounding experience. Even when experienced executives play, their supply chains almost always produce wild spikes in inventory. In this way the game conveys the nonlinear nature of systems and the instability of business operations. Not only do almost all players run into trouble, but when they are asked to explain their teams’ problems after the contest, they usually misdiagnose them.

Jay Forrester
Jay Forrester in 2008.

“The Beer Game is a high-order, nonlinear dynamic system,” Forrester says. “There is nobody who can understand that system just by [casual] observation and thinking about it.”

At its best, the Beer Game also demonstrates the inadequacy of conclusions based on limited information or easy mental models. The game teaches people to question what they believe. It suggests, for example, that firing people while leaving a company’s larger methods intact is “a remarkably low-leverage intervention,” as Sloan professor Nelson Repenning likes to say. Today, the Beer Game is used worldwide—in business schools, management training sessions, and even public schools, where it becomes the Soda Game—to teach systems thinking to students.

Social analyst

Although Forrester earned the National Medal of Technology in 1989 for his contributions to computing, he is probably now best known for the last phase of his career, which applies system dynamics to the broader social world. His books about urban economics (Urban Dynamics, 1969) and global economics (World Dynamics, 1971) attracted a larger audience, and more contention, than he ever expected, he says—especially when some of his followers published The Limits to Growth (1972). This book suggested that economic growth, combined with pollution and constraints on natural resources, among other factors, could lead to a societal collapse in the 21st century. Some environmentalists welcomed the book as an important warning, but many in the (pro-growth) business community resisted the analysis, and some prominent economists were skeptical about the data and methods.

Forrester is guarded about the hubbub over social dynamics, but “I think the books stand all right,” he says. Currently he is working on a new manuscript about economic systems.

The collision between economic activity and environmental sustainability—can we have both?—may become the highest-stakes example of system dynamics. Our actions are not isolated and separate from our surroundings. Those activities—such as industrial growth—can have harmful unintended consequences. Our attempts to correct course may go awry too, unless we thoroughly understand the systems we live in.

“Time after time … you’ll find people are reacting to a problem, they think they know what to do, and they don’t realize that what they’re doing is making a problem,” Forrester says. “This is a vicious [cycle], because as things get worse, there is more incentive to do things, and it gets worse and worse.” And that is why, he adds, “I consider the work we have done in systems far more important than anything we’ve done in computers.” Systems may be complex, but perhaps what we need is the common sense and curiosity of a ranch kid in a Nebraska schoolhouse. 

Keep Reading

Most Popular

Large language models can do jaw-dropping things. But nobody knows exactly why.

And that's a problem. Figuring it out is one of the biggest scientific puzzles of our time and a crucial step towards controlling more powerful future models.

OpenAI teases an amazing new generative video model called Sora

The firm is sharing Sora with a small group of safety testers but the rest of us will have to wait to learn more.

The problem with plug-in hybrids? Their drivers.

Plug-in hybrids are often sold as a transition to EVs, but new data from Europe shows we’re still underestimating the emissions they produce.

Google DeepMind’s new generative model makes Super Mario–like games from scratch

Genie learns how to control games by watching hours and hours of video. It could help train next-gen robots too.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.