JIM MELCHER seems the sort of man who is as comfortable with a soldering iron as with his personal computer. Taut and wiry, his fringe of hair cropped short, the electrical engineering professor at MIT bikes to work each day, and helps utility companies generate electricity more efficiently. ``The systems that keep the lights lit are rotting,'' he worries. ``Transformers that were supposed to work for 20 years are now supposed to work 40.'' Dr. Melcher considers himself a political conservative. But unlike most conservatives, who see ``the government'' as one thing and the Pentagon as quite another, to Melcher they are one and the same. And he thinks the Pentagon is skewing the American economy in ways that are far from beneficial. While the Japanese are building the next generation of electrical transformers, he says, General Electric has converted its own transformer plant in Pittsfield, Mass. to a torpedo factory.
``Defense is a much more certain way to make a dollar and a much better rate of return,'' he says.
It is from this standpoint that Melcher views the Strategic Defense Initiative. To be sure, SDI has been the target of campus criticism almost from the beginning. Program officials infuriated MIT and other institutions, for example, by suggesting to Congress that they had endorsed the program. Over 6,800 university scientists and engineers from all over the United States have signed a petition pledging not to accept SDI funds, including 57 percent of the members of the top 20 physics departments.
But this opposition has been based largely on the view that SDI probably wouldn't work and that it would make the world less safe. As it happens, the SDI Office has had little problem finding researchers willing to take its money.
And as these funds begin to seep through university research labs, they are prompting concerns on the economic front. How will this diversion of talent and endeavor affect America's ability to compete economically?
``Everyone is trying to emulate us but us,'' Melcher says, speaking of the way even countries like China are striving to free productive enterprise from the kind of government intervention that he thinks SDI represents.
Not all in the scientific community share these concerns. Air Force Lt. Gen. James A. Abrahamson, who heads the Strategic Defense Initiative Office, has pledged to apply the results of SDI research ``across all facets of our economy and society.'' Doubts persist nevertheless, and not just in the predictable liberal camp. ``At a time when competition in semiconductors, computers, and communications is so intense, it is disconcerting to see billions of research dollars pumped into `star wars' when these funds could be better invested to assure our leadership in these major global markets,'' wrote Ray Stata of the Massachusetts High Tech Council, a politically conservative group that was a major supporter of the state's tax-cutting Prop. 2,
Some think SDI is just another step down an unfortunate path. America devotes as large a share of its national economy to R&D as both West Germany and Japan. But where these countries put almost all their resources into the civilian economy, at least 40 percent of America's research is geared toward the military. ``The situation is so bad, `star wars' can't do much more damage,'' says Arnold Kerr, an engineering professor at the University of Delaware.
But Vera Kistiakowsky, a physics professor at MIT, speaks for many when she calls SDI ``particularly pernicious.'' In the past, the Defense Department has often funded basic research, which it then sought to apply to weapons. In fact, this was the case with much research now included in the SDI program, before President Reagan's speech declaring a single, focused mission. Now, the Pentagon is setting its sights on a ``narrow band'' of technology, Kistiakowsky says. ``It will skew national research.... It will change what students get educated in.''
``It's an insidious sort of thing,'' adds John Bardeen of the University of Illinois, who has won two Nobel Prizes in physics. ``At universities, students will be trained in problems in [the SDI] area. When they get a job, they will get a job in that area.'' One result is a drain on a limited supply of researchers. Peter Zimmerman, a nuclear physicist currently with the Carnegie Endowment for Peace, says less-prestigious schools ``are finding it impossible to get good post-doctoral researchers in fields gobbled up by SDI.''
The competition for trained personnel means American firms have to pay more to attract good engineers, which leads to higher production costs, at least in the short term, says Frank Lichtenberg of the Columbia Graduate School of Business. And once researchers have gone to work on military projects, they find it extremely difficult to transfer back to the civilian arena, say people in the field.
``Some of our brightest people are seduced into defense systems,'' says Bruno Weinschel, a high-tech entrepreneur and president of the Institute of Electrical and Electronics Engineers, who spent 10 years in defense work in the '40s and '50s. ``I had the finest lab facilities. State of the art. It's fascinating - much more interesting than trying to save a couple of pennies in manufacturing a radio. But that's what the balance of payments depends on.''
In the past, there was less concern over such diversion of research talent because military technologies could be applied fairly readily to civilian uses. But today, SDI snd other defense technologies are ``so esoteric,'' Mr. Weinschel says, that the ``spin-off is minimal.''
In dollar amounts, SDI funding for colleges and universities is relatively modest so far, somewhere in the $100 million to $200 million a year range. Defense Department money generally ranks behind medical and National Science Foundation grants as a source of funding for university research.
But the total SDI research enterprise rivals that of the Manhattan Project during World War II. And there is great concern over how the university share will be targeted. At MIT, for example, the Pentagon provides only about 18 percent of federal funding, but more than 50 percent in departments such as Electrical Engineering and Computer Science - precisely the ones crucial to civilian industry at this time. This pattern, moreover, holds nationwide. ``These are the people who could make the best argument for private-sector sustenance,'' Melcher says.
``I wish they would come in with all that money and say, `The key to SDI is having a good manufacturing base,''' says H.Kent Bowen, an MIT professor who is working to revive the manufacturing arts there.