Reinvent the CIA.

The Central Intelligence Agency's current cryptonym for itself in classified message traffic is PNINFINITE. The name is suggestive of the expansive world view that created the agency and sustained it through the Cold War years, but is now chronically out of date.

The CIA of the 1990s is a vast, global bureaucracy in search of a mission. The Company, as it was known in the old days, is now in liquidation. The end of the Cold War has rendered the massive technical programs developed to monitor the Soviet military largely irrelevant. So too the great covert action programs that were established in the 1950s to battle communism around the globe. The problem now is finding a graceful way to pension off the old networks of agents across Europe. And for all the anguish over the recent arrest of alleged Russian "mole" Aldrich Hazen Ames, the old counter-espionage ballet with the KGB is mostly history, too. Russia and America will continue to spy on each other until the end of time, but the stakes aren't the same now. We give Moscow billions of dollars in aid. At least until the Ames case broke, we were also sharing information with the Russians about common problems.

Yet despite all these changes, the old CIA structure largely remains intact. The agency in 1994 still has stations or bases in nearly every nation on the globe. But to what end? In the old days, the CIA case officers could say they were tracking their Soviet counterparts in the great game of counter-espionage. One former CIA officer explained to me once that he had spent much of a tour in Katmandu, Nepal, trying to recruit the local KGB resident there until he realized that the KGB man, all the while, had been trying to recruit him. This sort of shadowboxing kept several generations of CIA officers employed, but it's largely over.

The CIA, probably more than any other agency in the U.S. government, needs to reinvent itself for the post-Cold War world. It doesn't just need tinkering--retargeting its collection programs, smoothing its relations with Congress, looking for new enemies like drug traffickers or economic espionage that will give it a mission for the 1990s. It needs something more fundamental. Senator Daniel Patrick Moynihan has argued that the agency should blow itself up, figuratively speaking, and start over. That may be extreme. But here are a few thoughts about what a reinvented CIA might look like, drawn from conversations with current and former intelligence officials:

* Less Is More. The overriding requirement for a new CIA is that it needs to be smaller--and yes, more elite. There are too many case officers chasing the limited number of real spies the United States should be trying to recruit to our side. Does the agency really need stations in places like Dares Salaam and Kuala Lumpur and Caracas? Certainly the U.S. government will continue to need information from these capitals that will guide policymakers, but that's what the State Department is for. We have ambassadors and political counselors around the world who can meet openly with the local foreign ministry and simply ask the chaps there what's going on (as opposed to recruiting them as spies). Why then do we also need CIA case officers in these Third World capitals?

In the old days, the answer was that we had to go toe-to-toe with the KGB in these outposts to maintain our influence in the non-aligned world. Not any more. Yet the case officers are still out there, trying to recruit agents in the Malaysian foreign ministry, or the Kenyan security service. The real rationale for many of these recruitments, nowadays, is to ensure promotions for case officers, who need to log new recruits the way ad salesmen need to sell ads. Yet this way lies madness. An example of what can happen came several years ago in Ghana when a large network of CIA agents was rolled up by the local security service; several dozen Ghanaian agents and their families had to be relocated to the United States at considerable cost. The question is: Why had they been recruited in the first place?

A smaller clandestine service may be able to recruit the agents the United States needs most to operate in the world of the nineties. The hard targets, these days, are places like North Korea, Iran, China, France, and Japan. Collecting real intelligence about these targets will require a loose matrix of case officers--many of them operating under "nonofficial cover" as business executives abroad. What we have instead is a static structure of case officers posted mostly in embassies and consulates around the world, often with quite flimsy diplomatic cover. They're marching in formation, like the British Redcoats, against an enemy that is hiding behind trees.

* Depoliticize the CIA--Spend Less Time on Capitol Bill. One sure sign that a federal agency is in trouble is when senior officials begin to spend large amounts of time stroking members of Congress. The name of the game becomes protecting the budget, rather than doing the job. That's what happened with the FBI under J. Edgar Hoover, when it boasted of its success in solving car thefts and pretended the Mafia didn't exist. It has become a way of life at the Pentagon. And in recent years, this cozy Congress-o-mania has come to afflict the CIA.

The agency in 1993 gave over 1,000 briefings to members of Congress and their staffs, according to one reliable account. The director of Central Intelligence, James Woolsey, personally visited Capitol Hill over 150 times in his first nine months in the job, and over 80 percent of these meetings are said to have been one-on-one sessions with members, explaining the intelligence community's budget. Some senior CIA career officers have been grumbling that Congress sees more of the director than they do.

As Woolsey's frantic briefing schedule suggests, we now have an intelligence agency that works as much for Congress as it does for the executive branch. The process began in the late 1970s, for understandable reasons. The agency, though never the "rogue elephant" its critics suggested, needed more oversight. Since the congressional takeover of the CIA, two developments have been clear: The cost of producing intelligence has gone up substantially, and the quality of the product (by most accounts) has gone down. Neither trend should surprise readers of The Washington Monthly. Too often, the oversight process is perverted into a form of political logrolling. Rather than encouraging strong management, Congress ends up reinforcing all the agencies' worst tendencies. If you doubt it, consider this story making the rounds at the CIA: Some members of Congress are said to be questioning the agency's modest efforts to close unnecessary stations in Africa, on the theory that a reduced CIA presence there would signal disrespect. Can it be that cutting CIA operations in Africa has become politically incorrect?

The congressional takeover of the CIA has produced another, more troubling phenomenon: An agency that resisted politicization even at the height of Watergate (recall Richard Helms' refusal to take the fall for Richard Nixon after the burglary, and his subsequent exile to Iran) is increasingly being drawn into political squabbles. A recent case in point was the flap over the agency's psychological profile of exiled Haitian president Jean Bertrand Aristide. Conservative members of Congress requested the briefing from the conservative national intelligence officer for Latin America, Brian Latell, and requested specifically that he talk about Aristide. Almost immediately, raw intelligence about the Haitian leader's mental health (including a sofar unsubstantiated allegation that he was hospitalized after a nervous breakdown) was leaked. The problem here isn't with the CIA collecting such information or briefing policymakers about it. The country needs an intelligence service that will gather sensitive information about foreign leaders--especially when that information calls U.S. policy into question. The problems arise when these secrets become political footballs, with conservatives punting in one direction, and liberals in another.

* Beware of New Enemies, Like Drug Traffickers. In its eagerness to make itself useful in the post-Cold War world, the CIA (with considerable pushing from Congress) has embraced a new mission combating international drug traffickers. But is this really what the United States wants its intelligence agency to be doing? Last year's flap over CIA penetration of a Venezuelan drug ring illustrates the problems. According to news reports, a CIA/DEA counter-narcotics operation there went awry, and drugs got onto the street in the United States. Why didn't the CIA stop the drugs, critics wonder.

There may be a confusion of roles here. When it comes to the drug war, people understandably expect the CIA to act like the police--catching dealers, seizing drugs, making arrests. They forget that the CIA is an intelligence agency. When a CIA operations officer sees something bad happening, his first reaction isn't to swoop in and break it up, police-style, but to stand in the shadows and watch it for awhile to see where it leads. When such intelligence tactics are directed against drug traffickers, they may ultimately yield valuable information, but they are bound to be corrosive for the CIA. And sometimes, as evidently was the case in Venezuela, the operations will go sour.

The truth is that drugs and the CIA are a bad mix. For 30 years, the CIA has been dogged by allegations that it was involved, through its agents and operatives, in the narcotics trade. The rumors stretch from Laos and Vietnam to Afghanistan to Nicaragua. In some cases, given the agency's need for information from the low-lifes that inhabit the drug world, the rumors are probably true. Given this history, a strong argument can be made that the CIA needs to spend less time around the honey pot of narcotics, not more. Assigning the CIA to stop the narcos may literally be Mission Impossible.

The strange bedfellows problem also applies to another growth area for the ClA--the war against terrorism. The public needs to understand the reality of the CIA's anti-terrorism operations. To get information, it puts its own people inside terrorist cells; and sometimes, when necessary, it strikes deals with terrorists. Some years ago, I wrote a novel describing one such relationship. The book described how, for most of the 1970s, the CIA maintained a secret relationship with Yasser Arafat's chief of intelligence. It was, in many ways, a pact with the devil. But it saved lives. Would today's CIA leadership, busy briefing Congress, be so bold?

* Get Smart about Counter-Intelligence. Behind the recent arrest of Aldrich Ames is a problem more serious, in the long run, than all the machinations of the Russian intelligence service. That problem, according to knowledgeable observers, is the dry rot and bureaucratic incompetence that has come to afflict the agency's spying and covert action arm, the Directorate of Operations.

Though the Ames case got the publicity, it's just one of a series of recent counter-intelligence disasters. During the past dozen years, CIA operations have been compromised in Egypt, East Germany, Cuba, Ghana, and Iran. The underlying mistake in many of these cases was the same: Excessive dependence on the polygraph as a means of establishing the bona-fides of agents being recruited in the field. Ames allegedly passed two polygraph tests after being recruited by Moscow. Double agents from Cuba and East Germany are said to have duped the lie detector, too.

What's worrisome is that these counter-intelligence cock-ups seem illustrative of a larger problem in operations. The agency simply isn't devoting enough brainpower to the task of recruiting and vetling agents. Like the "body count" in McNamara's Vietnam, recruitment has become a numbers game. Insiders complain that case officers are wasting time "developing" agents--i.e., having expensive meals with them--who will never be recruited. Sometimes case officers are claiming to have recruited agents when they have done no such thing. In too many instances, the recruitment mill has drawn in false or useless agents, sometimes with disastrous consequences.

The answer to all these problems, insiders say, is to exercise more care and judgment in targeting operations. The agency in the 1990s should pay less attention to the great god of recruitment and more to the quality of information it is seeking to acquire; less attention to machines like the polygraph, and more to people.

* Rethink Covert Action for the 1990s. In the New World Disorder, the United States often seems powerless to affect events. The options in places like Bosnia, Somalia, and Haiti tend to have an all-or-nothing quality. The post-Vietnam U.S. military wants to fight only popular, winnable wars, where it can overpower the other side and guarantee a quick victory. In practice, that means sending in hundreds of thousands of troops, as in Desert Storm, or doing nothing. It's no surprise, given this limited menu of options, that the United States has generally chosen to do nothing. Where it has deviated from this massive force rule, as in Somalia, the U.S. has regretted it.

Yet this policy of doing nothing to stop the slaughter in places like Bosnia has left many Americans feeling uncomfortable. And it's reasonable to ask whether the CIA could make a useful contribution. Is there a covert option in situations where the overt use of military force is deemed inappropriate? Sometimes covert action would simply mean bringing barbaric practices to light and publicizing them. In other cases, it might mean working more aggressively with the democratic opposition in countries like Serbia, Iraq, and Iran. In a few cases, it might even mean covert paramilitary operations against a country like Serbia, that would raise the price of aggression. Given the dangers associated with these covert options, the wisest course may still be to do nothing. But for the old boys and old girls at PNINFINITE, the questions are at least worth asking.

David Ignatius was an editor of The Washington Monthly from 1975 to 1976. He is an assistant managing editor of The Washington Post and the author of a trilogy of spy novels about the Middle East: Agents of Innocence (1987), SIRO (1991), and the forthcoming The Bank of Fear (June 1994).