CIA DEPUTY DIRECTOR Bobby R. Inman stirred up quite a controversy a few months ago with a warning to scientists that they had better accept a voluntary system of pre-publication censorship. If they did not accept such restrictions, Adm. Inman predicted, scientists would be held responsible by public opinion for the "hemorrhage" of U.S. technology to the Soviet Union, and they would be "wiped away by a tidal wave of public anger."

In a recent congressional appearance, Adm. Inman regretted the tidal wave metaphor, but stood by his prescription. He is trying, he said, to "goad" scientists into taking action before the government is forced to. Restrictions would cover a sweeping range of research from crop projections to "manufacturing procedures"--this despite his acknowledgment that inadvertent disclosure of technological assets through communications, publications and conversations among scientists and engineers accounts for a "very small part of the problem." Seventy percent of scientific and technological losses occur through espionage, Adm. Inman estimated, while legal and illegal industrial transfers account for most of the rest.

Since this country relies heavily for its national security on its technological edge over the Soviets, even relatively small losses would be worth stemming if that could be done at an acceptable cost. The trouble is, it can't. Outside of the present administration, there are few who believe that a sweeping system of government pre-clearance of scientific research could even be imposed. If imposed, it would require legions of highly trained bureaucrats (that is, scientists and engineers who would be much more productively employed doing their own research than reviewing someone else's) to enforce. And if, somehow, such a system were created, the costs, in stifling and delaying U.S. technological advance, would be many times larger than the value of what would be denied to others.

The government already has more means for controlling export losses than it can effectively manage. There are several different export control lists covering weapons and sensitive technologies. These can be invoked against publication of listed technologies. There is a new 800-page "Military Critical Technologies" list under development. There is the Invention and Secrecy Act, which allows the government to impose secrecy on a patent application without justification and with limited opportunity for appeal. And there is the Export Administration Act, whose definition of "export" has been interpreted to cover "oral exchanges of information with foreign nationals in the United States."

The government should focus its attention on narrowing the critical technologies list to usable dimensions and on closing the loopholes and reducing the confusion, delay, overlap and error that surround the administration of the various export control lists. If these things can be done effectively, the case for imposing controls on research will disappear.