With a comparatively puny $850 million budget, an obscure Pentagon agency has managed to set the agenda in the debate on defense high technology:
* The Cruise Missile.
* The Stealth Bomber.
* The proposed Supercomputer project.
* President Reagan's proposal to launch "killer satellites" that would fire laser beams to destroy Russian ICBMs as they left their silos.
All these controversial and, their supporters say, critical defense programs are children of DARPA--the Defense Advanced Research Projects Agency.
DARPA is the Pentagon's high-technology gambler. It invests in the speculative, the chancy and the unconventional, hoping for a high return for the high risk.
The agency is free to ride any learning curve that might lead to a breakthrough technology; so at one extreme, DARPA has explored psychic phenomona; at the other, it roams through high-energy particle physics.
Conceived 25 years ago as the military's response to Sputnik, DARPA has acquired a formidable reputation as the Pentagon's primary innovator. "The cost-effectiveness of DARPA exceeds that of other segments of the Department of Defense," said James Schlesinger, a former secretary of Defense.
The agency has pioneered much of the most important work in computer science--including computer graphics, computer-to-computer communications and artificial intelligence. It also has funded key developments in weapons guidance systems and electronic warfare. In virtually every respect, DARPA research defines the state of the art in defense technology, several Defense analysts said.
But DARPA also is known for its unusually free-wheeling management style. There's a lot of freedom (including the right to fail) and offbeat personalities. It's a model that several companies--including Xerox Corp. and Hewlett-Packard Co.--have looked to for their own innovation efforts. But it's also a style that doesn't conform to the Pentagon's conservative tradition.
"DARPA has always focused on the revolutionary rather than evolutionary," said Dr. George Heilmeier, DARPA's director during the Ford administration and now chief technical officer at Texas Instruments Inc. "It has never really been a part of the status quo."
"We're supposed to be in conflict with the prevailing culture," said a long-time DARPA insider. "We're the counterculture."
Robert S. Cooper, DARPA's current director, isn't fond of that characterization, but he concedes its accuracy. He said that DARPA's difficulty is balancing those counterculture values that make the agency successful with the conflicting demands placed upon it by its main clients--the three services.
"Getting a good idea out of the lab and into the field is one of the most difficult tasks I have," Cooper said.
But he also said that "the DARPA position is the best R&D position in the world."
Cooper, DARPA's director since 1981, is redefining the agency's approach to research as well as its relationship to the services. He is phasing out DARPA's involvement in expensive, big-ticket weapons systems development and switching the emphasis to more basic research.
"We had a disaster in the Teal Ruby project an infrared space satellite detector that's now scheduled for a 1985 launch," Cooper said. "It's passing the $325 million mark and it was always another $100 million to go and three years til launch."
"The budget is now 40 percent dominated by" the big demonstration projects," he said. "Projects of that scale, especially in such exotic technologies as space weaponry, go beyond the cottage-industry management style that DARPA has been most successful at."
That style has enabled DARPA to craft a research and support network that has kept it at technology's cutting edge. Its people are widely esteemed; the organizational structure is skeletal and leans towards the ad hoc. There are no laboratories in DARPA's Arlington offices; the actual research is contracted out to industry and academe. Program managers coordinate and monitor the work. In many ways, DARPA is the Defense Department's venture capital company.
"There are two invariants which weigh heavily in the DARPA success stories," said Dr. Stephen Lukasik, a former DARPA director now with Northrup Corp. "The single most important power was sole-source contracting. You could find the best people to do the work; you didn't have to competitively bid everything.
"The other is size. DARPA was told to be small, so an incredible amount of management hocus-pocus was minimized." DARPA employs approximately 100 people.
Perhaps the most startling aspect of the DARPA style is the rate of turnover, which approaches 25 percent annually. "I would say the turnover issue is a key one," Lukasik said. "It's easier to get a first-rate person for a short time than to get him to throw his lot in with you for life. Also, when they come in, they're full of vinegar and rarin' to go. But after four or five years, you know, you've pretty much heard what they have to say."
"The average duration of a DARPA project is three and a half years, and the technological employe at DARPA works here an average of three and a half years," Cooper said. Every level of DARPA is affected. Cooper is DARPA's 10th director.
That kind of turnover makes it more difficult for very-long-range projects to be sustained, Cooper conceded. For one, the services simply can wait out programs that they don't like; for another, people go into the job knowing their days are numbered.
Still, the personnel churn assures DARPA a continuous flow of new ideas and research opportunities. Consequently, personality as much as technological opportunity alters DARPA's research efforts. As DARPA's directors changed, so did the agency.
The earliest DARPA projects back in 1958 included work in photovoltaic cells, satellite communications and anti-gravity (which never really got off the ground). "Just a year after they were announced, DARPA had a program in lasers," recalled Herbert York, DARPA's first chief scientist and now serves on the Defense Sciences Board.
DARPA's research scope is immense; it embraces computers, materials sciences, biology and virtually every conceivable military application of sophisticated electronics. Many of the nation's top scientists and engineers have been DARPA employes or contractors at one time or another.
In the last decade, DARPA began budgeting for large demonstration projects such as Teal Ruby and the Stealth Bomber. Cooper and his predecessor have been trying to change that. Ironically, the very changes that keep the agency vital force a constant reevaluation of its mission.
DARPA is positioning itself for what it sees as the next great leap in weapons support technology: the Supercomputer. Cooper said that DARPA will fund research to create computers "with reasoning ability and natural language" understanding. The agency is exploring both hardware and software elements of supercomputing and eventually expects to budget more than half a billion dollars to the various projects associated with it.
DARPA is also responsible for overseeing President Reagan's proposal to intercept Soviet missiles in space before they reach the United States by using satellite weaponry. Cooper said that the proposal "represents a radical departure from the policies of the past. Can I see a time in the future when this is possible? Certainly."
However, Cooper declined to speculate when that time may be.
Cooper seems particularly optimistic about the agency's foray into new materials that wou the iron, steel and chrome that go into tanks and planes. "Materials science will go wild," he said. "We're now building new materials out sed polymers that can be twisted and braided into new materials with new properties. There will be a whole neww materials with different properties of ductility, weight and toughness. The military and commercial possibilities arefound."
These materials eventually will be the basis for new weaponry. That potential underscores a percept defense: that DARPA technology plays a critical role in defining defense policy. Technological opportunities cy options.
"There are policy implications of the new technology," said Cooper. "You have to encourage the military to do thel thinking about technologies in their tactics and doctrine. More and more, the technology crosses service boundaries; it doesn't fit doctrin the organizations."
Cooper likens DARPA in that respect to the research and development arms of major cores, top management doesn't know how to incorporate innovation into its strategic plans and aspirations, he saiobservers said that DARPA's major flaw is the difficulty it has in "marketing" its programs to the services. CDARPA doing road shows for top brass and has lobbied the services hard, but there is still the impression thaty operates more in isolation than cooperation.
"From the point of view of the taxpayer, DARPA's output has been extremely high. From the point of n the battlefield, it's been extremely poor," asserted Tony Battista, a senior staffer on the House Armed Servee.
The problem, ironically enough, is that well-worn Army phrase, Catch-22. The very structure that gives the freedom to innovate is often the major obstacle to the successful transfer of its programs. The services s--revolutions aren't compatible with existing systems; services don't like dealing with personnel turnover anddiction to an outsider agency. But if DARPA tailors its programs to suit the services' desires, it loses that hat lets it stab for a breakthrough. Cooper points out that this is a problem that many major corporations face in their own research and development efforts.
The tension and co probably are unresolvable, according to several formal DARPA directors. They contend that it is important foro maintain its own sense of mission.
Said former DARPA director Lukasik, "If everybody is zigging, you need a zag--and that's DARPA."