In recent years the “police brutality” style of thinking has become an increasingly prominent feature of liberal opinion in America. Several years back, when urban crime became big news, liberal politicians and intellectuals often found that their only palatable response was to point out that the police were doing wrong, and denounce them for it. Hence, police brutality. This was the right instinct, but it offered little solace to those who were genuinely panicked about being robbed or murdered. It also failed to ask the harder questions at the root of the problem—such as, what should the police be doing to ease people’s fears, without trampling on others’ rights? While the liberals were intellectually safe with their police brutality position, they left the political field open to those who offered quick answers to the popular outcry, answers like preventive detention and life sentences for drug dealers.

Much the same philosophy is evident in public reaction to the Central Intelligence Agency (CIA). Over the past dozen years, the CIA has provided superabundant evidence for a police brutality approach to its problems. Operations like the Bay of Pigs invasion, the secret subsidies to universities and student groups, and the CIA-financed wars in Laos and the Congo understandably attracted pub- lic attention to what the agency should not be doing. These covert projects, usually referred to as “dirty tricks,” are only a small part of the agency’s official functions, but they have done more to shape America’s image for the rest of the world than the State Department, Pepsi-Cola, Food for Peace, and Henry Kissinger combined.

Even so, a sense of sportsmanship makes it hard for us to passionately denounce these secret activities, since the case against them is so obvious. Perhaps the most persuasive argument against the “clandestine” projects is that they can be discussed in a low-security forum such as this magazine. One can hardly imagine James Bond or the men from KGB watching their exploits exposed in the press. It would be foolish to claim that every rock sheltering a secret operator has been turned over and every plan revealed, but the CIA’s recent security record is not such as to do credit to allegedly undercover agents.

Since 1964, when David Wise and Thomas Ross published The Invisible Government, the trade in CIA exposes has been brisk. In 1966 The New York Times ran a five-part series which, among other offerings, explained how the CIA had poisoned 14,000 sacks of Cuban sugar that were in temporary storage in Puerto Rico. Ramparts told in 1967 about secret subsidies to the National Student Association. And more recent illustrations from Chile, Laos, Cambodia, and Washington, D.C. easily come to mind.

The bungled exploits have had their effect on our relations with both friend and foe. In the early sixties, a CIA agent recruiting local operators in Singapore attracted attention when he plugged in his lie detector and blew out all the lights at his hotel. He was arrested, and the British were infuriated to discover that we didn’t trust their spies to provide us with all the news from the area. Then Washington offered Singapore’s Prime Minister Lee Kwan Yew $3.3 million to keep quiet about the whole affair. Lee held out for 10 times as much, and eventually spilled the whole story to the press.

The Intelligence Community
The CIA may be the most famous of the U. S. spy agencies, but it is hardly the largest or even the most influential. With its 15,000 employees and reported $750-million budget, the CIA is one of the smaller members of what is usually referred to as the “intelligence community.” The largest representation is, not surprisingly, from the Pentagon. In 1961, Robert McNamara created the Defense Intelligence Agency (DIA) as a step toward centralizing the intelligence networks of each of the armed forces. Today, the DIA is thriving, with a $100-million budget and 5,000 employees, but it has made no visible dent in the organizations it was designed to replace. Accord- ing to figures compiled by Senator William Proxmire, Air Force intelligence employs 60,000 people and spends $2.8 billion; the Army has 38,500 intelligence employees and spends $775 million; and the Navy has 10,000 employees again at a cost of $775 million. The State Department has an Intelligence and Research branch (INR) which is relatively small (335 employees, $8-million budget) and concentrates on background research for State’s policies. The National Security Agency, which describes its function as “code-breaking” but is the most secretive of all the organizations, employs 20,000 people and’ spends $1 billion. The “intelligence community” is rounded out by small representations from the Atomic Energy Commission, the FBI, and the Treasury Department. Altogether, the national intelligence systems employ 150,000 people and cost $6.2 billion annually.

Even today, while the rest of us live and work as if we had a pretty good idea of what was happening in the world, lights are burning late in the CIA’s Langley, Virginia headquarters, and agents are polishing up plans for the next big operation, the perfect one, the one that will avoid the mistakes made in Singapore and Cuba. Why do they persist? The answer reveals that the CIA is not as different from the Nebraska Department of Highways as both of them would like to think, since each is driven by institutional momentum to continue doing the same things it has always done in the past. In the CIA’s case, the momentum comes from masses of old-time agents. These men were recruited in the days of Stalin, Truman, and Acheson; they were set on their chargers, equipped with lances, pointed in the enemy’s direction, and given a shove. Most of them are still fighting; in fact so many are still with the agency that the then-director, James Schlesinger, had to sheepishly explain to Congress last month that the CIA has one of the government’s worst aging problems.

One can sympathize with these older men, who, like railroad firemen and others whom time has passed by, are not sure what to make of the world they now face. They learned Swahili and Thai in the early days; they attended classes on how to crack a safe or bug an office; they devoted themselves to the struggle to grind out a few extra yards against the other team anywhere in the world. Now they are just trying to keep on doing the only things they know. In many ways, Howard Hunt, one of the convicted Watergate burglars, is the tragic archetype of their class. The tactics, he used against the Democrats and Daniel Ellsberg might have come straight from an instruction manual on “How to Rig the Elections in Chad”; his demeanor since conviction suggests that he would behave the same way if he were captured by the Communist Chinese during a covert mission.

Sympathy for the agents’ problems doesn’t obscure the harmful by-products of their dirty tricks. The drawbacks of secret operations are numerous: they frequently don’t work; they give the President “flexibility” in foreign policy when “restraint” is what we need; they reduce the State Department’s representatives overseas to an undignified status, since the people of Lagos or Lima know what to conclude when the CIA station chief lives in a bigger house and has more agents working for him than the ambassador does; they make American diplomats fearful that the CIA will stage a coup as they are exchanging pleasantries with the soon-to-be-ousted government; they even confuse the CIA’s own analysts, who are never sure whether the political developments they are charting are caused by genuine guerrillas or the CIA.

Even the agency’s official spokesmen seem to be facing these sad facts. James Schlesinger had none of the super-spy aura of many previous CIA directors. Richard Helms, who headed the agency from 1966 until he was pensioned off as ambassador to Iran this spring, was an intelligence “professional”; he was a career spy, and as the CIA’s Deputy Director of Plans, was in charge of covert activities. During the fifties, Director Allen Dulles had also stressed the agency’s secret functions, and at times appeared to be running a clandestine arm of his brother’s State Department.

What of Schlesinger? He is another of the “management men” much in evidence in government these days. Before coming to the CIA, Schlesinger served briefly as head of the Atomic Energy Commission, and before that had spent a short term in the Office of Management and Budget (where he conducted a study of the nation’s intelligence system). Before coming into government, he had been an analyst at the RAND corporation. In leaks to the press between his appointment and his transfer to the Pentagon in early May, Schlesinger suggested that the new CIA would be different from the old in the same way that Schlesinger was different from Helms—less razzle-dazzle, secrecy, and drama; more pipe-smoking and analysis. In his three months at the agency, Schlesinger encouraged more than 1,000 agents, mainly from the “operations” branches, to retire. Of course, reports of scaled-down “operations” cannot be taken at face value, not least because Schlesinger’s successor as Director, William Colby, was promoted from the operations division, where he headed such projects as the “Phoenix” campaign of political terror in Vietnam. But times clearly are changing.

Still, neither criticizing the clandestine activities nor retiring the secret operators takes us far toward figuring out what really should be done with the CIA. If keeping it from doing dirty tricks were the only issue, the simplest solution would be to close down the whole agency. There is something besides secret missions at stake, and though it may seem obvious, the real issue is the importance of the “intelligence” function.

Intelligence is usually defined as “evaluated information,” and is sub-classified by the pros into categories ranging from “current intelligence” (up-to-the-minute bulletins) to “national intelligence estimates” (weighty’ analyses of long-term trends). The justification for having an intelligence system at all is that the government needs reliable information on which to base decisions. The CIA’s specific intelligence role is to tell the truth about what is happening overseas.

Just how much truth we need to know, and about what, are the questions we must answer in setting priorities for the intelligence agencies. Before attempting that, it is worth noting that the CIA is unique in its truth-telling role. What sets it apart from other government organizations that send dispatches from abroad (see box) is its agents’ detachment from policy-making. A Foreign Service Officer may be reluctant to point out that his embassy’s policies are failing; an Air Force reconnaissance officer may doctor his evidence to fit his general’s plans; but CIA agents are relatively free of these personal and institutional pressures. “Basically, you need someone who doesn’t give a damn whether a program succeeds or fails,” says Chester Cooper, a former CIA agent and author of The Lost Crusade. “He’s the only one who will tell you where things are going wrong.”

The contrast between the CIA and the Defense Intelligence Agency (DIA) illustrates the virtues of detachment—known in the trade as “having no ax to grind.” The best of the recent books on intelligence problems, Patrick McGarvey’s CIA, The Myth and the Madness, explains why the DIA is congenitally incapable of tell- ing the truth. Until very recently, few of its officers were intelligence specialists; most were career soldiers serving a brief stretch between assignments in Vietnam and Europe. Many were more interested in coming out of their DIA assignment with a good personnel rating than in challenging what looked like dubious intelligence estimates. This quite predictably reduced not only their skill as analysts, but also their objectivity.

The resulting bias was displayed to the public in 1969, during the ABM debate in Congress. Armed with DIA estimates of Russian intentions, Defense Secretary Melvin Laird told his audience some spine-chilling tales. The Russians were planning to build 500 big SS-9 missiles, he said; and “new evidence” indicated that they were also on their way to creating a “first-strike capability” designed to annihilate the U. S. in a nuclear war. The most charitable explanation for these figures is that they came from the DIA’s “worst-case” method of estimation. This line of reasoning, similar to the “what-if-I-die-tomorrow” thoughts that race through the brain at bedtime, starts with basic technical data on Russian capabilities. Then the DIA analysts compute what the enemy could do if he poured all his effort into building missiles—the worst case. This pessimistic approach is healthy for generals contemplating battle or investors entering the stock market, but in predicting Russian missile strength, it leads to dramatic exaggerations. The “errors” are presumably tolerable to the Pentagon, since they support its argument for more missiles to counter the threat.

In the case of the ABM, Laird might well have stampeded the Senate into approving the project were it not for CIA reports. Richard Helms revealed that his agency’s analyses showed that the Russians would end up with 300-odd missiles rather than the 500 DIA predicted (they eventually built 318); that work on ABM sites near Moscow had virtually ceased; and that the “new evidence” of a first-strike capability was at best tenuous.

In fairness, it is worth mentioning that the CIA has not wholly digested the lessons of the ABM debate. The interconnection between its intelligence gathering and its clandestine operations is a classic illustration of tendencies it criticizes in the DIA and other agencies. In planning escapades like the Bay of Pigs, the CIA violates its own fundamental rule about detached analysis, because the agents do have a stake in the information they analyze. Undercover operators are put in the position of generating data to justify the projects they want to carry out. They report that “Yes, the Cubans will rise to join us,” or “No, we won’t get caught if we stage a coup.” Allen Dulles wrote, “Policy-makers tend to become wedded to the policy for which they are responsible. State and Defense employees are no exception to this very human tendency.” Neither are CIA agents.

When they play their proper, detached role, CIA analysts are one step nearer objectivity than Foreign Service Officers and military men. And they have other advantages as well. Many of them are semi-academics (one third have Ph.D.s), whose professional advancement depends more on their expertise in the affairs of, say, Ecuador, than on their skill in massaging the ego of the ambassador or general above them. Compared to men in the other agencies, the CIA men have time to reflect on their judgments, to read books, to base their reaction on something other than the latest dispatch from the field.

To gauge the importance of the truth-telling function, one need only imagine what the government’s domestic programs might be like if there were an internal CIA looking them over. The purpose of many programs becomes subverted somewhere in the pipeline between the Washington headquarters and the local office that actually constructs the public housing or operates the Head Start program. But because reports on the problems of these programs have to ascend the very chain of command that caused the problems in the first place, those at the top are generally sheltered from the worst news from the field.

LBJ on Cows and Intelligence

Because of its detachment, the CIA is one of the few groups that can look back on the Vietnam war as a relatively bright spot in its history. Chester Cooper has noted the contrasts between CIA reports on the war and the rosier views coming in from the State Department and the military. As early as 1964, the CIA was questioning the domino theory, saying that “with the possible exception of Cambodia, it is likely that no nation in the area would quickly succumb to Communism as a result of the fall of Laos and Vietnam.” Later the agency doubted that raining bombs onto North Vietnam would reduce either the material or the psychological support for the war, and throughout the sixties the CIA reports on the popularity of South Vietnam’s assorted leaders were more accurate than those of the Pentagon. As further proof that outsiders can get a clearer view of a problem than those causing it, the CIA’s reports on the village- pacification program became sharply more critical in 1965. This, it turns out, was when the Army finally wrested control of the program from CIA operatives.

Watergate burglar James McCord recently alleged that the White House wanted the whole bugging operation blamed on the CIA. He also expressed his more general concern about the way President Nixon was treating the agency:
It appeared to me that the White H9use had for some time been trying to get political control over the CIA
assessments and estimates in order to make them conform to “White House policy.” This could mean that
CIA estimates could then be forced to accord with DOD estimates of future U. S. weapons and hardware
needs…. This smacked of the situation which Hitler’s intelligence chiefs found themselves in, when they
were put in the position of having to tell him what they thought he wanted to hear… instead of what they
really believed…. When linked with what I saw happening to the FBI under Pat Gray… it appeared that the
two government agencies which should be able to prepare their reports … with complete integrity and
honesty … were no longer going to be able to do so.

The role of detached critic does, of course, come at a price, the same price that journalists pay for their privilege of criticizing and complaining. Because bad news is rarely popular with kings, presidents, or administrators, the bearer of bad tidings often loses his influence to those with more reassuring reports. Lyndon Johnson— who, in The Vantage Point, does not mention a single CIA estimate in describing his Vietnam decisions— expressed his appreciation for the agency’s pessimistic reports this way:

Policy-making is like milking a fat cow. You see the milk coming out, you press more, and the milk bubbles and flows. And just as the bucket is full, the cow whips its tail around the bucket and everything is spilled. That’s what the CIA does to policy-making.

In the early days of the Nixon Administration, the truth-tellers in the CIA had reason to fear that they would get an even less-sympathetic hearing than before (see box, page 11).
Henry Kissinger was hardly overwhelmed by the CIA’s insight. After reading a few reports on Vietnam and returning one CIA report with “piece of crap” scrawled across the cover,
Kissinger did not take another CIA estimate seriously in setting his Vietnam policy. But instead of consulting inner oracles or getting reassurance from the generals, as Bundy and
Rostow and Rusk had done, Kissinger created his own intelligence agency within the National Security Council staff. His researchers produced more than a hundred “National Security Study Memoranda” during Nixon’s first term, papers which laid out all the options and arguments in the style Nixon liked. From most accounts, these reports have done the job that CIA estimates once did.

Even more important than the location of the intelligence system is its focus. How much do we need to know, and about what? Unless we make the priorities clear, we cannot expect the CIA to automatically divine what kind of information will be most important in the coming years. Given the normal laws of bureaucratic behavior, the agency will report in the future the same things it has reported in the past. In many cases the old things are the wrong things to ask now, and new intelligence needs have arisen. In order to set a realistic intelligence policy, there are three. questions to be answered. First, what do we need to know? Second, who should provide the information? Third, how should they get it?

Embarrassing Moments for the CIA

One approach to the first question is to consider some of the widely recognized “intelligence failures” of recent years. A decade ago, the most prominent items on this list might have been the surprise construction of the Berlin Wall, or the delay in detecting the Soviet missiles in Cuba. During the Nixon Administration, there have been seven major occasions on which the intelligence establishment has felt embarrassed by its performance.

  • The overthrow of Prince Sihanouk in Cambodia— although Sihanouk blames this on the CIA (his recent book in called My War With the CIA), the coup startled the Administration because at the Senate’s insistence, American agents had been pulled out of the country.
  • The Son Tay prison raid— the rescuers found that the American POWs they came to save had been moved elsewhere.
  • South Vietnam’s invasion of Laos— intelligence reports predicted a large Communist force in the vicinity, but the military never got the message.
  • The invasion of Cambodia in 1970— in one of the CIA’s worst performances, its agents miscalculated the strategic importance of the port of Sihanoukville.
  • The Allende election in Chile, which analysts had not expected.
  • The 1969 coup in Libya, another surprise.
  • Aerial reconnaissance before the Egyptian-Israeli cease-fire, which misjudged the extent of Soviet rocketry.

The most interesting aspect of this list is what it says about the CIA’s view of crisis-oriented thinking and foreign policy. What caused these “failures”? Either defective military planning (the need for which could be reduced by avoiding similar wars), or an inability to predict the headlines three weeks in advance. There are good reasons for thinking that we shouldn’t try so hard to keep ahead of the breaking news. For one thing, there are limits to how good this kind of intelligence can be; the four illustrations from the Vietnam war suggest the enormous difficulty of installing spies among a people totally different from us in race and culture and overwhelmingly hostile. We are likely to crack the KGB before we get a good agent in Hanoi. Also, current intelligence comes at a sharply rising marginal cost. By reading the news- papers and monitoring the diplomatic reports, we may be 90 percent sure of when a certain government is about to be overthrown. The extra effort required for 95-percent certainty— which usually means spying, bribery, double agents, and informers—is so costly and provocative that it rarely seems worthwhile. But this is just what the CIA feels it has been trained to do. It is also simpler to slip more spies into an area than to think seriously about long-range developments. So the agency often embarks on a mad pursuit of current news bulletins.

If we step away from this fascination with current intelligence for a moment, the incidents of “intelligence failure” seem limited and trivial, compared with some of our failures to perceive more gradual developments. Another list of failures might be headed by our delay in recognizing that the Sino-Soviet split was opening the door to China, or our slow perception of the dollar’s plight over-seas, or the apparently unforeseen diplomatic damage done in Europe and Japan by the Nixon-Connally economic manifestos of 1971. In the coming years, developments like these will affect our national interest far more directly than distant coups or Latin American elections. The CIA can still keep track of the coups, but it should also start directing its attention elsewhere.

W hat follow s is a list of the subjects about which we’ll need reliable intelligence in the next few years. In each situation the questions to answer are what do we need to know, who should provide the information, and how should they get it?

Strategic Intelligence from Russia and China

Until disarmament is achieved, we will need to know about the missiles, nuclear bombs, and submarines of the Russians and Chinese. We may conceivably ignore other world events, but this is one development undeniably related to our national security. Good strategic intelligence should not be important only to the Strangeloves —accurate estimates of the other side’s weaponry is one of the prerequisites to continued peace and stability. Both sides need to know that the other’s deterrent force is large and function- ing. If we care about disarmament, we also need accurate intelligence. Negotiators at the Strategic Arms Limitation Talks found that each side’s satellite intelligence systems—which guaranteed that neither we nor the Russians could cheat on an agreement —were one of the strongest incentives toward serious bargaining. Finally, accurate intelligence reports can help us control the defense budget, as the ABM showdown illustrated.

How do we get this information? At one time we had to rely on secret agents who tried to sneak into Soviet defense installations. With the U-2 incident in 1960, the public was first informed of the shift from spies to aerial reconnaissance. Since then, in a rare example of technology’s being the friend of man, advances in reconnaissance techniques have not only made the reports far more accurate, but also made our means of obtaining them far less obtrusive. The electronic and photographic satellites that survey Russia and China can photograph objects 18 inches long and detect the changing of tubes at a Russian radar station. The satellite techniques are expensive—they account for the huge Air Force intelligence budget—but they are certainly the least provoc- ative and probably the most effective way of finding out what we need to know.

Once we have the satellites, do we need anything else? A former CIA agent who helped prepare estimates of Russia’s strategic strength in the sixties and who worked on Helms’ ABM report says that he and his colleagues relied on two main sources: technical evidence, collected by the satellites and deductive evidence, such as estimates from American rocket experts of how long it would take the Russians to build systems similar to ours. The agent said, “You start with the press reports and the information from normal diplomatic channels; that makes up about 50 per cent of the information you use. Another 45 per cent is from the technical reports and the experts you consult here. About two or three per cent is your own analysis. The rest—the spies and the secret hocus-pocus—is usually worthless.”

Managing the technical intelligence gives the CIA, the DIA, and the Air Force something to do; the Air Force and the DIA maintain the satellites and reconnaissance planes and the CIA analyzes the information. That leaves us with the Army and Navy intelligence systems and very little to assign them to. These intelligence systems have one justifiable function, which is to prepare “order of battle” reports on enemy troops. These are rather straightforward accounts of how many men and tanks are lined up where. The Army and Navy are not always content with this prosaic task and often try to expand their intelligence role. The capture of the Pueblo off the coast of North Korea was one of the consequences of this restlessness. The Pueblo’s mission was to locate North Korean radar stations— something the satellites could have done with no risk of provocation. As a general rule, when it’s possible to find out by satellite, the military should stay away. One means of enforcing. this policy would be to disband the Army, Navy, and Air Force intelligence systems and concentrate their few essential functions in a smaller DIA.

Political Intelligence from Russia and China

The Pentagon’s performance in the ABM debate is an example of what technical intelligence can do when it is divorced from political common sense. To realistically interpret photographs of missile sites, we need to know whether the Kremlin is going to stress weapons or refrigerators in next year’s budget. A shortage of good political intelligence was also largely responsible for the general fear of China which prevailed here until so recently. After the Korean War, in which the Chinese made what we considered an irrational attack, our government could not shake the conviction that Mao’s disciplined hordes would take other unpredictable steps. Who could tell whether the Cultural Revolution would lead to an invasion of Russia or an attack on Taiwan or Japan? The current, almost cloying, wave of Sinophilia illustrates the relief many of us feel in substituting panda bears for armed Red Guards as a symbol of China.

The Chinese example also tells a lot about how we should collect political information. China was diplomatically and culturally closed to us during the fifties and sixties, but the CIA was theoretically free to slip in as many spies as it could. Not many got through, and the lesson may be that diplomatic contact is more important than it often appears to be. In an age of hot lines and world-wide TV, it is easy, and fairly accurate, to conclude that the Foreign Service is a decorative appendage; but one of the main reasons for the diplomats’ decline is that their main function—taking the political pulse of a country—has been usurped by the CIA. There is scant evidence that the CIA’s political re- porting is always superior to that of the Foreign Service—the memoirs of ambassadors such as George Kennan and Charles Bohlen show that they were among the most sensitive to shifts within the Kremlin. When the CIA tries to horn in, it not only creates unnecessary suspicion, but also further undermines an already demoralized Foreign Service.

Our concern for the Foreign Service shouldn’t be pushed too far. As the Vietnam war record indicates, there is value in having competing information sources reporting from the field. Checking the quality of FSO’s reports would be an ideal assignment for the CIA. But in most cases, the agency could do the job with annual visits to the diplomatic outposts, rather than setting up over- sized CIA stations all over the globe.

Political Intelligence from the Rest of the World

When the focus moves away from Russia and China, the question of what we need to know becomes more and more tangled with the question of what our world role should be. One way to separate the issues may be to examine each area of the globe and ask what kinds of developments are clearly of concern to us.

We start with Western Europe and Japan. Here the political intelligence from diplomatic, academic, and journalistic sources is overwhelming. As for military intelligence, if we cannot trust these countries to tell us what they are planning, it hardly makes sense to share our nuclear secrets with them or join them in NATO or other defense pacts.

That leaves Africa, Asia, Latin America, and the other disparate regions usually classified as the Third World. These continents, and especially South America, have traditionally suffered the most from close CIA scrutiny. The harvest from these activities is not only the agency’s affiliation with many repressive regimes, but also the astounding paranoia about the CIA which prevails abroad. Within the last 18 months, headlines have appeared in Iraq, Egypt, India, and four Latin American countries attributing domestic unrest to the CIA.

The two standard justifications for Third-World intelligence are, first, the threat of communist subversion, either military or political; and second, the danger of wars, coups, and other violent outbursts. Detecting the first threat, if it exists, does not require large contingents of secret agents. It is hard to believe that no one among our diplomats, satellites, and even tourists would notice the arrival of foreign troops in a small country. If the “subversion” is political, then it is probably wise to know no more about it than the diplomats can observe; if we know when the coup is going to happen, we may be tempted to prevent it.

The CIA’s efforts have backfired in another way. During the 1960s, some perceptive Southerners complained that the only thing keeping the Ku Klux Klan alive was the FBI informers. Similarly, by trying to infiltrate every left-wing movement in Latin America, the CIA has inadvertently nurtured threats that otherwise would have withered on their own.

The second justification—the danger of war—should be taken more seriously. Even if we do not plan to rush into regional wars with quite the alacrity we have displayed in the past, disturbances of the peace anywhere in the world concern us. There is always the danger that a battle in the Middle East, or between India and Pakistan, will somehow draw other powers into the fray. More basically, if we want to encourage a more peaceful world, we should understand why people fight. A lack of this cultural understanding is the weakness of much of our Third-World reporting. So much stress is placed on palace gossip that the larger picture is often obscured.

During the 1971 India-Pakistan war, the CIA made a splash in the National Security Council with a dispatch showing that. India planned to take over the whole of Pakistan. The information (which later proved distorted) came from an informer in the Indian cabinet. It was a triumph of the old-style espionage, but wrong. What the CIA should have been looking for were the long-range causes of the fighting. What had set the Hindus and Moslems at each others’ throats? Were they likely to fight again? Most of the recent warfare in the Third World has risen not from narrowly political or ideological causes, but from religious, cultural, and tribal tensions: consider Biafra, Uganda, Pakistan, the Middle East, and the “demographic war” between Honduras and El Salvador. So trying to understand the violence by sifting through embassy rumors is doomed from the start. Looking for the roots of violence would not require the contingent of spies and saboteurs we now deploy in the Third World; instead, the raw material for such reflection would be academic and diplomatic observation, and the conclusions would come from thoughtful analysis rather than spying.

In discussing this point, Adam Yarmolinsky, one of Robert McNamara’s advisers at the Pentagon, said to me, “My gut feeling is that if I have a chance to know, I want to know. If knowing involves risks, I’ll face them later. That is preferable to the risk of not knowing.” Perhaps, but why not minimize the risk by concentrating on the issues that matter and cutting back the purely political espionage?

“You need to know what they’re doing to each other under the table,” Samuel Adams, an outspoken CIA agent said, in justifying the need for clandestine operations. “In a country like Uganda, if you just read the papers you won’t have a clue as to what is going on.” If Amin and Uganda were a perfect analogue to Hitler and Germany, the point would be valid. But we must ask two questions about cases like Amin’s. First, is what’s going on under the table of concern to us? In Uganda, the immediate answer is no. In Israel, or Egypt, or South Africa, or Pakistan, the answer may be yes; these are countries with the potential to involve many other countries in their fighting. This list will expand if nuclear weap- ons are more widely distributed. Second, are spies the only way to find out about the sub rosa activities? Our diplomats in Uganda won’t just be reading the papers, and their idea of what’s really going on should be accurate enough to meet our needs.


Counter-espionage means protecting your country’s secrets from the other country’s spies. It usually involves such ploys as infiltrating the opposition’s intelligence system. From all reports, the CIA has not done well in penetrating the KGB. But as long as there is any data we should withhold from the Russians, the effort is worthwhile. There is at least one piece of information that fits this category: the technical details and deployment plans for our Polaris submarine fleet. The Polaris is the heart of our deterrent system, and it is too easy to imagine that in some Pentagon office there is a board showing where the subs are at any given moment. Prudence suggests that we should at least keep that room off-limits for KGB agents and have a way of knowing when they have seen the plans.

Another kind of counter-espionage is less necessary. CIA agents are constantly guarding against communist subversion at such vital links in ‘our military defenses as the Kagnew military base in Ethiopia. It is altogether too easy to imagine that somewhere there is an American military base of no strategic importance, whose only function is to employ CIA agents to prevent it from being spied on. A simpler solution might be to take the base away.


If there is one effort in which the world’s intelligence systems can cooperate rather than compete, it is in preventing terrorism. Even those who sympathize with many of the grievances of the Irish Republican Army or the Black September group find it hard to stomach their violent excesses. In combatting terrorists, the CIA has a rare chance to put its secret tactics and dirty tricks to good use. Already the agency has begun work in this area. After the Black September killings at the Munich Olympic Games, a “Cabinet Committee to Combat Terrorism” was created in the State Department. The CIA feeds reports to the Committee, which in turn dispatches warnings to U. S. diplomats abroad.

Economic Intelligence

If we look for the international forces that are actually going to affect the way we live in the next decade and forget for the moment about an abstract threat from Uganda, we must confront the dramatic changes in the world economy which have become increasingly apparent in the last few years. This does not mean expanding the data on Russian beet and steel production that the CIA has been collecting for years. Instead, it means realizing that we—like the British since World Wars II—will soon be in the position of living fat or lean as international economic laws dictate. Changes now occurring in Japan and Western Europe may affect our un- employment level more than anything done by the Council of Economic Advisers. Resource and environmental questions are involved as well. How will competing for the same flow of Arabian oil affect our already tepid relations with Japan? What will we do if the Arabs decide they don’t need to sell us any more oil? The Chileans may claim all the fish 500 miles out to sea as theirs; or the African countries may affiliate more tightly with the Common Market; these and a hundred similar developments will affect us in an immediate, material fashion.

Just because economic changes are not secret, there is a tendency not to take them as seriously as missiles or subversive movements. But they are the most difficult challenge the CIA’s analysts will face in the coming years. In the last two decades the agency has had to shift focus. For example, when communist activity died down in Africa, agents put their techniques to use in Southeast Asia. In general, it has been a question since 1947 of applying similar tactics to different parts of the globe. Now it is a question of applying entirely new theories and ideas. The information itself is easy to come by; it is available in economic reports and academic studies. Changing the outlook is harder, but it is what the CIA should be doing.

James Fallows

James Fallows began his magazine career at the Washington Monthly in the 1970s, later serving as editor of U.S. News & World Report and as a White House speechwriter. He has written 12 books, the most recent being Our Towns, coauthored with his wife, Deborah Fallows, which was a national best-seller and the basis of a 2021 HBO documentary.