If he had been born and raised in Japan, Erich Bloch might very well have ended up becoming there what he became here: a tough-minded maverick government bureaucrat championing a unique brand of industrial policy.

That's because Bloch, not unlike the folks at Japan's Ministry of International Trade and Industry, takes an unapologetically realpolitik view of how wealth and knowledge are created.

"I consider that, in order to be a world power, that you have to be in command of your own economy and you have to have an economy that can support your own aspirations," says Bloch, who recently retired as director of the National Science Foundation, the federal government's primary funding arm for the physical sciences. "I don't think you can have that kind of economy without an appropriate base in science and technology."

It's not yet clear that Bloch's successor at the NSF, physicist Walter Massey, will be able to champion the integration of science, technology and economic growth as firmly as Bloch.

In the vast scheme of America's science and technology infrastructures, the NSF is a small agency with a pipsqueak budget that has less political influence than your typical junior senator. Despite its bureaucratic puniness, Bloch adroitly maneuvered the NSF into the turbulent center of national debates about global competitiveness, industrial policy, big science versus little science, university-industry relationships, equipment and education.

While other agency budgets shriveled, Bloch oversaw a near doubling of NSF's total funding to more than $2 billion annually. Where most administration officials performed the Adam Smith hallelujah chorus, Bloch publicly argued that not only did government have a crucial role to play in science and technology, it also should have a bigger role.

By most measures, Bloch ultimately proved to be a ruthlessly successful bureaucratic infighter who managed to garner both attention and resources for his agency. In fact, Bloch's tenure at NSF is a case study of bureaucratic entrepreneurship in the creation of a subtle industrial policy by a man who is very blunt.

The original knock on Bloch -- now in semi-retirement in Washington -- was that the 32-year International Business Machines Corp. engineering veteran couldn't care less about fundamental science research. Critics said he would turn the National Science Foundation into the National Technology Foundation. He would throw over funding for basic scientific research for the filthy lucre of commercialization. What's more, he would even play politics to get what he wanted.

The reality was that Bloch went into NSF with an extremely clever agenda. He would preserve the agency's traditional core scientific constituency but use it to create new alliances with state governments, industry and venture capital. He would use NSF money not just to fund science, but to fund coalitions that would fund science as well. The NSF engineering "centers of excellence," like California Institute of Technology's in biotechnology, at universities around the country to act as a magnet for academic, state, local and corporate funding. Bloch skillfully offered NSF matching grants to get other organizations to pony up funds.

"It's not just what NSF spends," says Bloch, "but how it influences the whole behavior of its communities and the relationships it builds with states and universities; it's a catalyst for things. You've also got to involve industry in science."

The whole idea was to seed structures that could become semi-autonomous and attract NSF-independent political support. In this, Bloch's critics were right on the mark: Bloch was more than willing to play politics to get what he wanted because, in Washington, that's how you get what you want. Bloch explicitly understood that NSF could leverage its funding to broaden its constituencies. More constituencies give you more clout -- and Bloch wanted clout. His bet was that successful innovation can come from clever infrastructures as well as clever individuals.

What's more, the NSF broadened its vocabulary. Bloch argued -- rightly -- that there was a false dichotomy between science and technology.

"There's no dividing line between the two," says Bloch. "The knowledge flows back and forth. The separation between science and technology is a bad separation -- there's no reality behind it."

That, of course, became the rationale for funding more engineering and technology research. For the most part, the NSF directorates were sharp enough to create explicit links between basic science research and the technology missions.

"People have a simplistic view of technology," Bloch asserts. "They think technology is something industry does. There is such a thing as generic technology -- but the cost of doing it is orders of magnitude more even than it was 10 years ago."

Perhaps the single most important shift at the NSF under Bloch was the emphasis on infrastructure development -- whether that infrastructure be science equipment in university labs, the education of undergraduates, supercomputing networks for universities or new industry-academic research ties.

Yes, there was a lot of political squabbling over who should get what funds and resources. Bloch was not above punishing institutions that didn't go along with his vision of infrastructure, and he wasn't shy about injecting himself into decisions made by his subordinates. Bloch certainly wasn't shy about loudly insisting that both the Reagan and Bush administrations step up their commitments to rebuilding the country's science and technology infrastructures.

Erich Bloch's NSF confirms the belief that a bureaucrat with an agenda who knows how to grow new constituencies can have an impact far beyond the original mission of his agency. Bloch has been criticized for building constituencies and new institutions at the expense of funding science. He would argue that these constituencies and institutions are going to be the future of science funding.

To me, the most revealing thing about Bloch's NSF tenure is that it was less about research budgets than the culture of science. Bloch sees science as a complex system of commerce, academe, federal and state participation and multidisciplinary teams. He sees arbitrary distinctions between basic research and applied technology as just that -- arbitrary. He does not see science as a collection of evolving sub-specialties -- little Darwinian finches, each with their own research labs. Most importantly, he doesn't see science as something that's separate from the society that's sponsoring it.

Michael Schrage is a columnist for the Los Angeles Times.