The Pentagon Wants to Expand University Research Ties. Here’s What It’s Looking For.

Even after the budget-sequestration vote of 2013 put a cap on government agencies’ spending, the Department of Defense was determined to maintain its support for basic research. Now that Congress has temporarily lifted the caps, the department is eager to expand research partnerships at American universities.

Stephen P. Welby, after nearly three decades of technology-related experience in government and industry, has just taken the post of assistant secretary of defense for research and engineering. In that position, he is the chief adviser to the secretary of defense on all matters of science, technology, and research.

Mr. Welby has an undergraduate degree in chemical engineering from the Cooper Union for the Advancement of Science and Art, a master’s degree in business administration from Texas A&M University at Texarkana, and master’s degrees in computer science and applied mathematics from the Johns Hopkins University. In a recent interview at his office in the Pentagon, he described the department’s approach. The following interview has been edited for clarity.

Q. What’s the outlook for Pentagon spending on research now that we may be getting past sequestration?

A. We went through the challenge of sequester and the impacts that had in Fiscal Year 2015; now that we’re in Fiscal Year 2016 we’re operating under the budget agreement that’s relieved us of those caps, at least in the near term.

But the budget is pressurized by an enormous demand signal for modernization across the Army, Navy, Air Force, and Marine Corps; the growth in personnel costs across the force; the demands of the war. Despite all those other pressures, we have preserved the basic research budget, the total science-and-technology investment across the department.

We’re not investing in basic research as an endpoint in itself. We are not the National Science Foundation. But we do see that the investments that we make in very basic work — in core material science, in basic physics — have long tails that are relevant to downstream applied work, which has military relevance.

Q. As for projections of a flat line for research in the future, are you expecting that, or expecting Congress to raise above that?

A. I can’t anticipate congressional action. But quite frankly, in today’s budget environment, flat is the new up. We’ve been able to protect ourselves from a decline, and that’s really all we can ask for, given the pressure on the budget.

Q. Does the Pentagon’s focus on applied research drive a lot of its priorities in basic research?

A. There is this kind of pipeline model that suggests there is: basic, applied, and then the directed, more mature kind of work — and we start to work our way up the food chain. I don’t think that model is realistic here in the 21st century.

In many of the areas we’re interested in, interesting things pop out of basic research and quickly become applied. Some things have to stew for a longer time, with more fundamental discovery, and then rapidly move into application. Some things have to dwell in the application space for a while because they’re real hard problems to be able to solve.

So I think that model is more complex than the budget categories reflect, and it really is about the quality of work, the nature of the relationships that we have with smart folks who are thinking about the future.

Q. What’s the temperature that you see right now on university campuses toward this kind of work? Do you feel hamstrung in any way by the reception that you get?

A. Folks recognize and are very attracted to the model by which the department funds basic research. We are often the funder of choice for folks with interesting ideas, due to the nature of the grants we offer, the long-term relationships we tend to develop.

As you start to come up the scale to more applied topics, you do get more complicated kinds of issues — whether folks want that kind of work occurring in their spaces or in adjacent spaces, whether that’s part of the university or not. I’m encouraged by folks kind of working their way through those kinds of issues with us.

We are very sensitive to even the perceptions that we might execute any kind of prior control on publication, for example. All our basic research is of course freely publishable and open and available.

Q. Among universities seeking economic spinoffs, where would you suggest they focus in terms of unmet needs?

A. Our goal is the work, not the spinoff. One of the things the department is very deeply engaged in is the MURI activities — the Multidisciplinary University Research Initiative — which are multiyear programs that focus across disciplines. Universities around the country compete around those topic areas. We establish centers on campuses, often centers across campuses, that allow folks to build a concentration of work over time in an area of interest.

We have today 319 of those centers located across the country — this is not a small-scale program — and over the last decade, since the program was initiated, those efforts have generated almost 25,000 papers, a half-million citations. It’s been an enormously productive program.

Q. What are some especially valuable areas of expertise?

A. One is the entire area of software-enabled capabilities — people throw the artificial-intelligence word around, but, really, systems that can track their environment in different ways, systems that can help demonstrate autonomous robotic capabilities, folks who are thinking about novel planning systems, systems that do classification. The whole big-data problem is one that we have in spades.

There’s a strong push in the department to think about the future of communications, and in particular thinking about ways to be much more agile in our use of the electromagnetic spectrum.

There’s a growing interest in robotics technology in general. That includes not just things walking around but things that fly, things that operate in the ocean, things that operate under the surface of the ocean, things that operate in space.

One thing that we don’t quite know what to make of yet is the whole topic of synthetic biology, which all of us are kind of watching as an enormously productive area at the moment that has no obvious military application.

We have a really rich domain. The department operates 63 laboratories and three central laboratories — 66 total laboratories with 39,000 scientists and engineers who work directly for the department, all of whom I think are here because of mission and because of the really interesting problems we have in the department. We’re certainly not recruiting or retaining them by how much we pay, which is something that troubles me, but they’re here because of the depth of the problems.

Q. As for cybersecurity, are universities getting it right? Do they need to become more creative rather than formulaic?

A. Much of what we’re doing today is, if you will, like putting your fingers in the dike. A lot of it is a patchwork of trying to understand how we operate with the fundamentally flawed systems that were developed in an era when security wasn’t really part of the design of much of the core infrastructure which our networked world has built.

I think we’re really in a space where in the future we’re going to have to start thinking about how we design security in fundamentally different ways, rather than this kind of “I’m going to hack you until you break and then repair that.” That break-test-fix kind of story is not scalable.

Q. To what degree do you see the university community thinking that way?

A. There’s an enormous demand signal today for the current state of cybersecurity, and I worry a little bit that that gets in the way of thinking long term. That’s important today, to make sure that we’re locking the door. But longer term, we’ve got to think about entirely new ways to build systems that eliminate this problem entirely rather than just working through it.

There’s a mix across universities and across researchers in terms of what they’re doing. Right now there’s not an obvious research strategy for how I might reinvent our network-enabled world for an era where there are bad guys on the net trying to do things. I worry about things like the Internet of Things, and the emerging capabilities that we see there, just increasing that exposure.

But I’m encouraged to see groups that are thinking more deeply all the way down to “How do I redesign, from the processor on out, how I start thinking about trust in systems?” That’s going to be a robust area of work, not just here at the department but across the community. This is not just a Defense Department problem; this is kind of a global problem.

Paul Basken covers university research and its intersection with government policy. He can be found on Twitter @pbasken, or reached by email at paul.basken@chronicle.com.