"Professor, this won't take even five minutes of your time."

Thus begins most of the U.S. government background information interviews I have given over the past several years on graduating students who have been accepted for federal employment. Since about half of Georgetown's graduates go into the public sector, federal investigators are numerous and frequent visitors to our campus.

Many of these interviews are worthless. The overworked investigator clearly is anxious to move on to the next case -- and frequently complains about an impossible workload. The questions appear designed only to discover if the student has demonstrated unstable behavior with respect to the use of drugs or alcohol. The questions are asked in a way that invites nonmedical professors to render medical judgments, and this is something that most professors refrain from doing. If a student had the potential to damage U.S. interests the way the Walkers allegedly have, I doubt the current system would uncover it.

Moreover, in my five years as director of Georgetown's School of Foreign Service graduate program, rarely have I been asked to render judgments about subjects on which I am competent. For example, I never have been asked about a student's integrity -- whether he or she has ever been found guilty of academic dishonesty. I never have been asked whether the student gets his or her work done on time. I never have been asked to describe a student's academic strengths and weaknesses.

Most interviews end with what investigators call the catch-all question: "Would you recommend this student for a position of trust and responsibility with the U.S. government?" The question virtually requires an affirmative response unless the student has been found guilty of academic dishonesty or has acknowledged personal problems that would impair his ability to hold a federal government job. In an age when students have the right to see the contents of their security investigation files -- a right that I strongly support -- we professors have an obligation to respond narrowly to such catch-all questions. Whether or not we like a student or whether we have heard via the student grapevine that he or she was drunk at a party should not enter into an answer. Thus, the government's apparent purpose in using the "catch-all question" -- to fish for derogatory information -- is defeated.

Unless this background investigation is changed fundamentally, it will not detect likely problems or warn of potential threats. And by swamping the system with more than a quarter million requests annually for routine background checks of the kind I have described, the current system almost guarantees that investigators won't have the time to ferret out spy rings by conducting enough periodic follow-up investigations of those with access to sensitive national security information.

The recently announced plan by Secretary of the Navy John Lehman to reduce the number of people with access to classified information and to centralize control over the process by which clearances are granted is, in principle. a sensible interim response to a crisis. But how this plan can be executed by a mere 900 naval security investigators with a caseload of more than a million personnel has yet to be explained. And similar plans have not even been announced to cover Army and Air Force personnel or defense industry employees, who in the past have proven to be equally lucrative targets for KGB recruiters.

Perhaps the worst damage the Walker case has wrought is the revelation that the government's security clearance process is outmoded, understaffed and ineffective. But this is not a new story. The system has been found at fault repeatedly by congressional investigations. Just a month before the Walkers were arrested, the Senate held hearings on federal government security clearance programs that revealed, according to Sen. Sam Nunn (D-Ga.), that "the government is already plainly incapable of adequately investigating and reinvestigating all persons seeking security clearances."

Unfortunately, there are no quick, easy or cheap ways to improve the quality and comprehensiveness of this process. So far, the government is going about doing so by administering Band-Aids. Priority should be placed on reducing the number of secrets, for example, as well as the number of people with access to them so the one can be better protected and the other more thoroughly screened. The problem should be attacked on a government-wide basis, not agency by agency; uniform standards and control procedures should be developed and applied stringently to defense industry contractors as well.

What these issues require is the creativity and expertise of a high-level, bipartisan presidential commission. Protecting our secrets in a better way is now too important a job to be left to already overworked -- if not demoralized -- offices of security in a dozen agencies. A presidential commission could ease the pain of changing old ways and help infuse the arcane business of security investigations with new ideas. And since the cost of a new security system likely is to be high, the president will need the Democratic as well as Republican political and business leaders who might serve on such a commission to help persuade Congress to foot the bill.