Engineering the Harvard Engineer

For most members of Harvard University's Faculty of Arts and Sciences, it was just another monthly meeting, the last of 2006. For Venkatesh Narayanamurti, dean of the Division of Engineering and Applied Sciences, it was one of the most important meetings of his career.

On that December afternoon, the professors gathered, as usual, at the Faculty Room, a spacious chamber in University Hall with sea-green and tan walls, lush Oriental carpets, leather-topped tables and chairs, five crystal chandeliers, and tall arched windows overlooking Harvard Yard. Dozens of oil paintings and marble busts of Harvard's past presidents and other luminaries--William James, Henry Wadsworth Longfellow, and Jean Louis Rodolphe Agassiz, to name a few--add to the aura of gravitas and tradition.

At 4 p.m., after the customary tea was served, Derek Bok, then Harvard's interim president, started the meeting. When it came time to discuss the docket items, Dean Narayanamurti stood up, glanced down at his notes, and then told his colleagues that, following a presentation on the topic he had made early that year, he was now ready to propose that the Division of Engineering and Applied Sciences be renamed the School of Engineering and Applied Sciences.

Despite the name change, the school would remain part of the Faculty of Arts and Sciences, the largest of Harvard's 10 faculty bodies, but it would now have the freedom to grow and establish collaborations across campus. This step, Narayanamurti told his colleagues, would ”help to enable Harvard to meet the changing needs of the times and the challenges posed by the future.”

As he concluded his remarks, he felt a wave of apprehension. The dean had spent the better part of the previous year working out all the details of such a move. Just a few months earlier, though, his well-laid plan appeared to be on the verge of collapse when then-president Lawrence H. Summers, who had come to accept the school upgrade, announced he would step down. Now, as the moment of judgment approached, Narayanamurti still had doubts about how some of his colleagues would respond.

An instant of silence ensued, but no objections were raised; the motion passed by acclamation. The proposal would now go to the university's top governing body, the Harvard Corporation, to be ratified, and then Harvard would finally have the beginnings of what it had notably lacked for its entire 372-year history: a world-class engineering school on a par with the university's other famed dominions, such as business, law, and medicine.

As the meeting moved on to other business matters, Narayanamurti grinned from ear to ear.

Sitting in his corner office in Pierce Hall last fall, Narayanamurti--known to nearly everyone as ”Venky”--recounted that day and the chain of events leading up to it. Engineering at Harvard, he noted, has long been a modest enterprise in comparison with the university's world-famous humanities and social-sciences departments and professional schools. After World War II, when most elite U.S. universities nurtured their engineering schools into full-fledged enterprises, Harvard had lumped its engineering and applied-sciences faculty into one division. The division status didn't command much respect inside or outside Harvard, and even the engineering faculty joked that they spent more time explaining what the term division meant than what they did there.

The transition to school--the first new school at Harvard in 70 years--is designed to fill a glaring gap in Harvard's report card, Narayanamurti says. ”Harvard will always get a grade of incomplete until it has a preeminent engineering school.”

The expansion plan is certainly ambitious. Armed with the Harvard name and a US $999 million endowment, the engineering school plans to increase its faculty from 70 full-time professors to 100, double its graduate student body to 600, enhance undergraduate courses for nonengineering students, and establish more collaborations with other Harvard departments and with industry.

In pushing for a first-rate engineering presence at Harvard, Narayanamurti had the backing of some well-placed academics. Thomas E. Everhart, a former president of Caltech and a Harvard graduate who has presided over Harvard's board of overseers--a body that advises the university on a wide range of issues--says the move was long overdue. ”With engineering and applied sciences becoming much more important in academia and in the economy as a whole, Harvard wasn't doing its share,” he says. Since Narayanamurti became a dean at Harvard nearly 10 years ago, Everhart says, the technology-oriented fields are finally getting much more recognition.

Although Narayanamurti says his plan met with virtually no resistance on campus, people did raise lots of questions, the most obvious being: How would Harvard make a mark in engineering with the mighty Massachusetts Institute of Technology just down the street? And why would Harvard, essentially the world's greatest liberal arts college, want to do engineering anyway?

In countering those and other questions, the dean pointed out that engineering is essential to the kinds of interdisciplinary collaborations that drive today's cutting-edge research--collaborations that advance both basic and applied fields. He also argued that a modern liberal arts education needs to encompass technology too.

And as for ”that small school down the road,” as Narayanamurti jokingly refers to MIT, Harvard isn't interested in direct competition. Rather than cover the whole engineering universe, it plans to focus strategically on a few areas--nanotechnology, bioengineering, energy and the environment, computers and society--that, he says, ”can leverage Harvard's strengths.”

At least one MIT luminary backs the Harvard plan. In a recent interview, Charles Vest, a former president of MIT and now president of the National Academy of Engineering, in Washington, D.C., noted that the timing couldn't be better, as the United States is under increasing pressure to train more engineers in the face of competition from Europe and China. ”It's a great symbol to the country that a university that's been built largely around the liberal arts tradition is now saying, ’Look, technology is a big part of the world today, and it needs to be a part of us,' ” Vest says.

If Harvard now appears ready to embrace engineering, it wasn't always so, says Frederick Abernathy, a professor of mechanical engineering and the engineering school's unofficial historian. Over the past century and a half, he says, engineering came close to disappearing from campus on several occasions.

The story begins in 1847, when Harvard started offering a formal technical curriculum under the auspices of the Lawrence Scientific School, named after Abbott Lawrence, a Massachusetts industrialist. The school had a promising start, but it would soon lose its edge to an upstart. In 1861, a geologist named William Barton Rogers, after failing to secure a faculty appointment at Harvard, founded another technical school: MIT.

The Lawrence school soon faced opposition from within Harvard Yard as well. Charles W. Eliot, perhaps Harvard's most influential president, felt that the applied sciences, with ”a practical end constantly in view,” was a poor fit with the university's liberal arts culture. During his tenure, Eliot-- who, ironically, had taught analytic chemistry at MIT before becoming Harvard's president in 1869--did his best to eliminate engineering at Harvard or have it transferred to MIT.

In 1906, he succeeded in dismantling the Lawrence school, whose fragments were absorbed by other parts of Harvard. But his victory was short-lived, thanks to another Massachusetts industrialist, Gordon McKay, who made his fortune developing machinery for the shoe industry. Back in 1891, McKay had designated the Lawrence school as the chief beneficiary of his vast estate. He died in 1903, and six years later Harvard received the first $1 million of his bequest, to be used toward technical education and research.

Finally, in 1918, Harvard--after another failed attempt to combine its engineering efforts with MIT--decided to use the McKay money to reestablish its school of engineering. The school would not get hold of the entire bequest until 1949, as McKay's will provided lifetime payments to one of his two ex-wives, two sons, and several mistresses.

The McKay fortune went on to support 42 professorships, in effect keeping engineering alive at Harvard. But despite its accomplishments [see time line, ”Yes, Harvard Has Engineering”], the overall engineering enterprise never attained world-class status. It never became, well, an MIT.

Things began to change in the mid-1990s, when Harvard president Neil Rudenstine and dean of the Faculty of Arts and Sciences Jeremy Knowles decided that basic scientific research at Harvard needed to be strengthened by more fully embracing engineering and technology. Knowles began consulting experts around the country on how to revitalize the engineering program, including what to look for in a new engineering dean. One of those experts was Narayanamurti.

One weekend in January 1998, Knowles flew to California to meet Narayanamurti, who was then dean of engineering at the University of California, Santa Barbara. ”By the end of that Saturday morning,” Knowles says, ”I realized that I did not just want Venky's advice; I wanted Venky himself!”

Narayanamurti's career began in 1968, when, after receiving a Ph.D. in physics from Cornell University, he joined Bell Laboratories, where he would stay for 19 years. His Harvard office is festooned with memorabilia from the labs, where some of his underlings took to calling him Lord Venky because of his exacting style of management and his eye for spotting promising research.

In 1987, Narayanamurti left Bell Labs and became vice president of research and exploratory technology at Sandia National Laboratories, in Albuquerque. Five years later, he moved on to Santa Barbara. There he expanded the faculty, fostered interdisciplinary programs, and established ties to the local high-tech business community. But the challenge of revamping Harvard engineering appealed to Narayanamurti, and in mid-1998 he moved to Cambridge.

Over the next several years, he hired a dozen new faculty members--he stole at least three from MIT--to strengthen such areas as artificial intelligence, materials science, and biomechanics. The academic stars he recruited include Lene Vestergaard Hau, an applied physicist who devised experiments to slow and stop light pulses and then reconstitute them, and Federico Capasso, a former Bell Labs researcher and coinventor of the quantum cascade laser.

The expanded faculty, in turn, helped reinvigorate education in the division, especially at the graduate level. Under Narayanamurti's tenure, the graduate student body doubled to 350, with the number of applicants jumping nearly threefold, to more than 1200.

Narayanamurti's own profile accompanied that growth. On campus, he is said to know everybody and be known by everybody. His nickname is now so ubiquitous that he is quoted in the Harvard Crimson simply as Dean Venky. Ask about his personality and people say things like ”builder,” a ”dynamo,” a ”bubbling enthusiast.” ”You walk into his office and you feel bombarded with ideas, and you come out feeling like you can take on the world,” says Greg Morrisett, a professor of computer science. He said he came to Harvard in 2004 because he liked the direction Narayanamurti was taking. ”I saw that this place was going to take off.”

Despite all of Narayanamurti's efforts, though, the Harvard engineering program remains small compared with those of the top tech schools [see table, ”Elite Engineering”]. Its faculty is about half the size of those of Princeton and Caltech, schools that Narayanamurti considers peers, and its research spending is a fraction of what big schools like MIT, Stanford, or Berkeley spend. In the latest U.S. News & World Report ranking of top U.S. engineering graduate programs, Harvard comes in at a disappointing 23.

After several years in the job, Narayanamurti says, he came to the conclusion that the division would become preeminent only if it drastically raised its profile. In particular, the idea of creating a first-rate engineering school had been around for a while. Narayanamurti's bosses, Rudenstine and Knowles, supported the idea, and so did a subcommittee of Harvard's board of overseers, which recommended the upgrade to school as early as 2002.

But when Narayanamurti began discussing the idea with Harvard president Larry Summers, Rudenstine's successor, the dean discovered that Summers was skeptical at first. The two men held several long, often heated conversations, and finally, by mid-2004, Summers came around to the idea.

With the support of the overseers and the president, it seemed as though Narayanamurti's plan was falling into place. He even found time to contemplate his retirement, and on 31 May 2005, he surprised his faculty with an e-mail announcing that he would step down the following year.

Then the Summers scandal broke.

After making some controversial remarks on women in science, Summers became engulfed in criticism. A ”lack of confidence” vote by the powerful Faculty of Arts and Sciences led to his eventual resignation in February 2006. For a while the venerable institution appeared leaderless, and nothing much seemed to get done.

Given the circumstances, Narayanamurti agreed to stay as dean until a new president was hired--and until the new engineering school was established. (Some at Harvard remember events slightly differently. One professor describes Narayanamurti's attitude as more like ”Give me the school or else I won't stay.”)

With Harvard's top echelon still in flux, Narayanamurti bolstered his efforts to rally his own faculty around the idea of an engineering school. ”You'd pass Venky in the hall and he'd tell you about it,” says Vahid Tarokh, a professor of electrical engineering. ”Person by person, Venky--a one-man army--got everybody on board.”

But convincing the rest of the Faculty of Arts and Sciences, which had the power to make or break the deal, proved a different matter. Some people, Narayanamurti says, feared that engineering could rapidly swell and begin siphoning money away from other disciplines. So Narayanamurti took to the paths of the Harvard campus and spent six months persuading colleagues in various departments and schools.

Theda Skocpol, a professor of government and sociology and former dean of the Graduate School of Arts and Sciences, says she found Narayanamurti's arguments credible. ”We received assurances that, as in the past, [the engineering school's] faculty and undergraduate and graduate activities would remain closely tied with other parts of the university,” Skocpol says. ”This was understood as a welcome evolution rather than any kind of upheaval.”

So when Narayanamurti put the motion before the Faculty of Arts and Sciences on 12 December 2006, he had done his homework. And after the unanimous vote passed, he was ecstatic. It also helps that Harvard's new president, Drew Gilpin Faust, is on board with the plan. In her speech at the school's launch last September, she credited Narayanamurti with maneuvering Harvard toward a more tech-friendly outlook, calling him ”the North Star of our engineering galaxy.”

After the Harvard Corporation ratified the engineering plan in February 2007, Narayanamurti created several committees to handle the transition from division to school, dealing with faculty growth, financial issues, collaboration with industry, and even the design for the school's seal. He continues to push for stronger cross-disciplinary collaborations, which he says are made easier by the fact that the school isn't divided into departments.

His vision of a school with few divisional barriers has helped attract new faculty. Capasso, who joined Harvard in 2003, recalls that he had entertained a number of offers from universities, but he ”would not at first glance take Harvard very seriously. What changed my mind was that they were putting their money where their mouth was.” In his case, the main attraction was the creation of the Center for Nanoscale Systems, which brought together several Harvard groups and is now based at the new Laboratory for Integrated Science and Engineering [see photos, ”Nanoscience, Megabuilding” and ”Underground Research”].

Susan Graham, a professor of computer science at the University of California, Berkeley, and a former president of the Harvard Board of Overseers who backed the upgrade to school status, notes that the open structure seems to prevent the usual departmental squabbling over resources. ”The question now,” she says, ”is whether this will scale as they grow.” As Narayanamurti continues to expand the faculty, he says he has been contemplating Graham's question. He says that once the faculty has about 100 full-time professors, the school will reach a ”critical mass” to do significant things while remaining a manageable group.

He also wants to extend the school's interdisciplinary culture to the classroom. He asked his faculty to create new technology courses for students who are interested in, say, medicine or economics. At the same time, the engineering school is reviewing its curriculum to incorporate more hands-on learning.

Lynn Andrea Stein approves of the move. On sabbatical at Harvard, she is a professor of computer science and engineering at the Franklin W. Olin College of Engineering, in Needham, Mass., a new school that has gained attention for its innovative curriculum [see ”The Olin Experiment,” IEEE Spectrum, May 2006]. She adds, though, that Harvard shouldn't simply try to replicate Olin's ideas. ”A lot of the questions that motivated the creation of Olin are also motivating changes at Harvard, and Harvard will have its own unique response,” she says.

Narayanamurti sees the ideal graduate from his school as a renaissance engineer, one with a grounding in both technology and societal issues. ”Those who want to be pure technologists should go to MIT,” he says. ”At Harvard we want to create people who know how things work but also how the world works.”

Farther down Massachusetts Avenue, at MIT, people aren't completely oblivious to Harvard's expansion plans. It's just what MIT needs, one professor commented, ”a sharp kick in the butt.” Not so, predicted another: after a few years Harvard will just settle for being a very good but very small school of applied sciences, like those found at a lot of other liberal arts colleges.

Nobody at Harvard seems ready to concede just yet, and in fact, the school is already thinking about its long-term prospects. In January 2007, the university created a committee to advise Harvard's president and governing bodies on how to boost collaborative efforts in science and engineering. The committee, of which Narayanamurti is a member, should shape major expansion plans, including the Allston Initiative.

Having exhausted its land on the Cambridge side of the Charles, Harvard now intends to build a second campus in the Allston neighborhood of Boston. At press time, the university was set to begin construction on a $1 billion science complex. It's been suggested that the engineering school, in whole or in part, could move to Allston one day, but no concrete plans have been made. It's clear, however, that engineering and applied sciences will be a central part of the new development.

As for Narayanamurti, what's next? In mid-February, he announced he would step down as dean and return to teaching and research. And this time he means it. But he plans to stick around to see the changes he helped initiate. ”It's going to be one of those things that you don't notice day to day,” he says, ”but when you come back in five years or so, it will be a tremendous change.”

With additional reporting by William Sweet in Cambridge, Mass.