Defense officials are proposing to establish a committee to screen scientific papers and prevent them from falling into the hands of the Soviets, according to Steve Bryan, a deputy assistant secretary of defense.

Bryan will ask secretary-level defense officials later this month to name this group, composed mostly of military officials, to screen the work of scientists working under defense contracts.

Stung by what they call a near-leak of data on a highly classified topic at a scientific meeting in August, and after halting the publication or presentation of 100 papers at that meeting, the defense officials hope now to establish a regular panel to review and pre-censor "sensitive" papers from scientific meetings around the country.

Bryan said he hoped such a system would avoid such things as the last-minute barrage of secrecy orders clamped on papers at the August meeting of the Society of Photo-optical Instrumentation Engineers.

Pre-censorship proposals like Bryan's, however, have proved controversial.

Hakime Sakai, a physicist at the University of Massachusetts at Amherst who had two of his papers squelched at the meeting, said "the review process is detrimental to the whole scientific effort." Getting information to colleagues quickly "is vital for scientific progress" and halting the distribution of work "is in direct conflict" with that progress, he said.

Besides, he said, both his papers were in basic atmospheric research, were not sensitive and preliminary results had long before been published in open government publications.

He does not object to government review, he said, but censorship should be extremely limited, only to papers which are clearly sensitive.

There is now no coordinated monitoring of scientific meetings or papers, and Bryan said "we need to do a better job in setting up the standards for what should be censored and working the system right so we don't get surprised."

The proposal was triggered by the incident at the meeting of the photo-optical engineers, at which the Pentagon was surprised to find that even "highly classified information" was being presented in papers to which Soviet scientists would have access.

One of the papers discussed in detail technology for satellite-to-submarine communication. Bryan said he is concerned that there may be many regular scientific meetings and papers which are the source of leaks to the Soviets.

Concerns about such leaks have been raised by government officials with increasing frequency for several years.

According to long-standing rules, defense-funded research is supposed to be reviewed by the contracting agency for sensitivity and the Pentagon notified before publication if sensitive subjects are involved. But in practice the system works only fitfully.

"Clearly there was a better way to do this than the way it was handled in San Diego . The style was not great," Bryan said. But still, he added, "We're absolutely convinced that a good deal of sensitive information was prevented from being leaked. We feel very good about it in that regard."

Richard Wollensak of Itek Inc., president of the photo-optical society, said the group cooperated with the Defense Department because it recognizes a need to protect "sensitive" information. But he said he wants to make sure a last-minute yanking of papers never happens again.

One federal science official said a proposal for such a central monitoring committee to handle sensitive technology is "not unreasonable" provided it stuck to scientific work done under Pentagon contracts and that it monitored only a few meetings and a relatively small number of papers.

"You can't monitor every conference with the word laser in it," the official said, "but it would be extremely important to have a scanning mechanism like this proposed committee to avoid the embarrassment with such heavy-handed methods" as those used in San Diego.