IT APPEARS THAT there is no room for a relationship between Harvard University and the CIA. One prizes candor and freedom, the other secrecy and the strictest control. Recent events at Harvard have made this polarization of interests an institutional problem.

Over the past several months, at least three Harvard scholars have said they have worked for the CIA doing restricted research. Anathema to widely accepted scholarly codes, the restrictions prohibited disclosure of the CIA's sponsorship and forbade publication without agency review. The professors' actions have raised concerns that Harvard's own rules are not sufficient to discourage this kind of activity.

And last week, a top CIA official, Robert M. Gates, tried to salvage his agency's tenuous connections with academia after the Harvard controversy received national attention. But Gates' much-publicized policy speech at the Kennedy School brushed over the sensitive issue of academic freedom, allowing disclosure of funding in some limited cases but holding fast to the CIA's right of review and ultimate control of scholars' work.

Little has changed since the late 1970s when President Derek C. Bok and then-CIA Director Stansfield Turner futilely tried to open a mutually beneficial window of cooperation between Harvard and the CIA. Professors still must submit to unreasonable restrictions if they wish to conduct research for the agency.

Until the CIA realizes that an abundance of classified material, covert funding and censorship have no place in institutions that depend and thrive on openness and freedom, Harvard professors must understand that contracts with the agency are inappropriate. As Bok and Dean of the Faculty of Arts and Sciences A. Michael Spence review the university's outside research regulations, they should make it absolutely clear that no Harvard professor should make research, consulting or other contracts with the CIA or similar agencies when there are strings attached.

Whether personal or institutional, contracts with the CIA must at the very least be acknowledged in all cases. Codes of scholarly conduct should not be suppressed, especially when involving partisan agencies which often raise the wrath and fear of people all over the world.