Scientist Extends Arm In Many Areas

Though John A. Armstrong ’56 is a physicist by training, he is knowledgeable in areas as disparate as meteorology and submarines. During his 30-year career at IBM, where he ultimately served as vice president for science and technology, he directed research during an era during which the cutting edge of technology progressed from lasers to microprocessors.

Armstrong settled on physics at a young age, drawn to the exciting field that had recently developed radar and atomic science. At Harvard, he enjoyed his freedom to take courses in history and English, but the impression that science alone was making meaningful progress into the future confirmed his original intention to become a scientist.

“I don’t give a lot of credibility to the notion that the questions social scientists and historians are asking today are better than they were in the 20th or 19th century,” he says.

He traces his lifelong skepticism back to his realization that theories that are not built with scientific rigor are essentially intellectual “fads.”

While in college, Armstrong had what was then a rare opportunity to work at the General Electric (GE) laboratory near his home in Schenectady, N.Y. He describes his position as an internship—before internships were part of educational or corporate culture.

“I knew something about the way science was actually done...which was something I never heard a word about at Harvard College,” Armstrong says of his experience at GE.

In May 1956, Armstrong was about to graduate summa cum laude when his life took an unexpected turn. He had been awarded a National Science Foundation (NSF) fellowship to cover his graduate studies in physics, but his House Master John Conway at Leverett House offered him the chance to travel in Europe for a year on a Sheldon travel fellowship. At the time, Armstrong, who had never ventured further than Washington D.C., decided to take the fellowship.

His physics adviser correctly predicted that he would lose his NSF fellowship, and incorrectly that he would ruin his career in an effort to convince him to refuse the trip and go straight to graduate school.

“That was the worst advice I ever got,” Armstrong recalls.

By disregarding these predictions, Armstrong was left to pursue his doctorate at Harvard, where the physics department already knew his background and the circumstances of the lost fellowship.

As a graduate student and then as a postdoctoral research fellow, Armstrong worked in the areas of nuclear magnetic resonance and nonlinear optics. His post-doctoral adviser was Nicolaas Bloembergen, who won the 1981 Nobel Prize in physics for a paper co-authored by Armstrong on laser spectroscopy and non-linear optics.

However Armstrong, who had no intention of becoming an academic, signed on at IBM after completing his post-doctorate in 1963. There he continued the research he had pursued at Harvard, becoming an expert on the study of lasers.

Colleague John J. Wynne ’64 was hired by Armstrong in 1964 to work in his IBM laboratory. Wynne also studied physics at Harvard under Bloembergen, though not until after Armstrong had left the University.

The team in Armstrong’s lab was known as the Harvard Mafia, according to Wynne. Though Armstrong was Wynne’s supervisor, the two became collaborators and published several papers together, mostly on laser science.

“He didn’t compete with you, he was a team guy,” says Wynne. “But in my case, we had complementary expertise. I was the hands-on experimental guy and he was the theoretical guy.”

By 1981, when he was appointed director of one of IBM’s major research labs at East Fishkill, N.Y., Armstrong no longer conducted original research. Instead, he rose in the administrative ranks, overseeing all of IBM’s research in 1986. At this time, the research department was staffed with 16,000 Ph.D’s and had a budget of $160 million.

Armstrong says that the differences in scale and available resources—rather than the type of work done in each setting—distinguish industrial research from academic research.

“Nature is not organized according to the syllabus of Harvard,” he says. “When you try to actually make new things work, you’re just as likely to stumble across exciting unanswered questions.”

Though working in industry means giving up some freedom over what projects a scientist can choose to pursue, Armstrong points out that academic scientists are also seriously constrained by the realities of funding grants.

This hampers productive work both by subjecting the development of science to the prognostications of the NSF as to what areas are worth pursuing, and by distracting scientists by forcing them to justify the validity of their work, which may not yet be clear.

For example, in 1986 and 1987, at the very beginning of Armstrong’s tenure as director of research, two teams scientists from IBM’s research laboratory in Zurich, Switzterland won the Nobel Prize in physics in two consecutive years.

“During all of their research, they never wrote anything more than two pages long asking for money,” says Armstrong.

Since his retirement in 1993, Armstrong has remained actively involved with the advancement of science in the academic and public sectors. He has served on the National Science Board, the NSF’s board of directors; the National Research Council; the Governing Board of the American Institute of Physics; and the Overseer’s Visiting Committee to Harvard’s Department of Engineering and Applied Science.

He has also served on committees to determine the government’s policies in a wide range of fields. Meteorology and atmospheric science have been a major focus of these projects, including a report he helped produce called Fair Weather, describing the policy issues raised by the tensions between private weather services. He also advised the government to build a new version of the deep-sea submarine, Alvin, that discovered the wreck of the Titanic, and has helped formulate the direction of America’s future in space.

Armstrong attributes his success in working with so many different branches of science and technology partly to the similarities in the kind of thinking and analysis required in every case, and to his rigorous early training.

“If you understand physics, there isn’t anything about the natural sciences that you can’t at the least be a very well-informed layman about,” he says.

—Staff writer Virginia A. Fisher can be reached vafisher@fas.harvard.edu.