It is difficult to exaggerate how thoroughly the gathering of information on the Soviet Union, and especially its military power, has dominated U.S. intelligence operations since the Cold War began. Today, in light of the diminished Soviet military threat to western Europe, the sharp decline in the Soviet economy and the centrifugal forces pulling the Soviet Union apart, such concentration on that country and its military must inevitably decline. There will be differences over how much of a reduction is necessary, but it is difficult to see how anyone could argue that a substantial adjustment by the intelligence community is not in order.

Any reduction in effort against the Soviet Union will be resisted strongly in the CIA, where the Cold War ethos runs very deep. Similarly, any reduction in effort against the Soviet military threat will be opposed by the various military intelligence agencies when, after all, that threat is their raison d'être. Yet two of those military intelligence agencies collect not only military intelligence but also a major portion of U.S. political and economic data on the Soviet Union: the National Security Agency (NSA), which is responsible for most of our electronic intercepts; and the Satellite Reconnaissance Agency (SRA), which operates all intelligence satellites.1

Not only will new priorities have to be convincingly delineated, but some organizational changes will be needed as well, including a new position of Director of National Intelligence (DNI).


What should the new priorities be? Are there topics of sufficient import to the nation's well-being to replace the Soviets as the focus of U.S. intelligence? There are, and for the simple reason that we live in an information age. Increasingly, having the best information is the key to success in almost any line of endeavor. The tremendous advantage U.S. forces achieved over Iraq in the Gulf War was because Washington knew more about what Baghdad's military was doing than Iraq did about American forces. Or imagine what help it would be in international trade negotiations to know precisely what internal subsidies and other devices the other side is employing. Information always has been power, but today there is more opportunity to obtain good information, and the United States has more capability to do that than any other nation.

This is an advantage not to be forsaken just because the Cold War is over, even though we may not be able to identify today exactly what information will give us the most power a decade or so from now. It will cost very little to maintain or expand U.S. human intelligence efforts. What will be expensive is the building of a robust network of satellites with a variety of sensors. Washington can easily construct a system that will detect any significant activity on the surface of the earth, day or night, under clouds or jungle cover, and with such frequency as to make deliberate evasion difficult. Physicist Edward Teller has estimated that such a system would cost $5 billion to purchase and $1 billion per year to operate. At twice that it would be a bargain because the ability to peer anywhere, anytime, is bound to be of great value in the uncertain new world ahead.

There is much more than this general case for continuing a wide-ranging intelligence effort. The most obvious specific impact of the new world order is that, except for Soviet nuclear weaponry, the preeminent threat to U.S. national security now lies in the economic sphere. The United States has turned from being a major creditor nation to the world's greatest debtor, and there are countless industries where U.S. companies are no longer competitive. We must, then, redefine "national security" by assigning economic strength greater prominence. That means we will need better economic intelligence. The United States does not want to be surprised by such worldwide developments as technological breakthroughs, new mercantilist strategies, sudden shortages of raw materials or unfair or illegal economic practices that disadvantage the country.

If the objective of collecting economic intelligence is to buttress national economic strength, then that requires making our businesses more competitive in the global marketplace. How can we do this with information acquired secretly abroad? Economic intelligence can range from the broad trends that foreign businesses are pursuing, all the way to what individual foreign competitors are bidding against U.S. corporations on specific contracts overseas. Some argue that when it comes to specific data such as competitive bids the government should not become a partner of business and distort the free enterprise system. The United States, however, would have no compunction about stealing military secrets to help it manufacture better weapons.

If economic strength should now be recognized as a vital component of national security, parallel with military power, why should America be concerned about stealing and employing economic secrets? But if the government provides economic intelligence to specific American corporations, how does Washington decide which ones should benefit? What about a corporation that operates in the United States and provides jobs for Americans but is foreign-owned? There are problems galore, but these are for the Commerce Department to handle on a case-by-case basis. One way to help, when there is no other, is to make information public. That may aid U.S. corporations less than some would like, but it also can lessen an advantage foreign corporations have over American firms.

While there are arguments over how far the U.S. government should go toward providing intelligence data to specific U.S. corporations, there is no question that friendly foreign countries make use of their intelligence services against U.S. businesses. Most everyone would agree that U.S. intelligence agencies should work closely with the American business community to prevent the loss of proprietary information to such foreign-government sponsored espionage. All of this says, then, that in an age of increasing attention to economic strength, there needs to be a more symbiotic relationship between the worlds of intelligence and business.

In addition to a shift toward economic intelligence, the new world order calls for more emphasis on political intelligence in Third World countries. Let us recall that over the past dozen years political events in Iran, Libya, Grenada, Panama, Lebanon and Kuwait have each involved the United States in commitments to the use of military force. The United States did not foresee that it would become involved in any of these situations. We cannot be confident of predicting today where we may choose to intervene tomorrow.

There are national rivalries whose outcomes will be of concern to Washington such as those between the Arab world and Israel, Pakistan and India, North and South Korea, the African National Congress and the South African government, Iraq and Iran. And we know there are global trends toward independence or autonomy for ethnic, religious or national groups and toward democracy in general. These trends will affect countries of interest: Yugoslavia, Romania, India, all of the Arab world and most of Africa. Because of the lessening of Cold War competition in the Third World, Washington may view such unstable situations from a more relaxed posture and be less inclined to become involved. Yet the United States can make good choices only by being well informed.

Finally, there are growing global problems such as terrorism, ecological abuse and drug trafficking that will require the United States to collect a diversity of information. With a global system of data collection, there will be many beneficent uses of information such as helping with the development of remote underdeveloped areas and in disaster-relief operations.


The need for military intelligence in the new world order will diminish but not disappear. So long as the Soviet Union maintains sizable conventional military forces, the United States must account for them, even if with a lesser sense of urgency. And Soviet nuclear forces will demand at least as much attention, perhaps more, if the United States is required to verify increasingly complex arms control agreements. Moreover Washington will want to know about renegade countries that manufacture and store nuclear, chemical and biological weapons. And even those nations' conventional weapons will be of interest, as Washington will want to know in advance of intended military operations, such as the Iraqi invasion of Kuwait. These worldwide checks on weapons and a country's intentions for using them will require a highly intrusive monitoring and verification system. But in a new, more friendly world order, the United States will want that system to be as impersonal-as non-incriminating-as possible.

We will not want to be caught by friends in the act of spying, as that is far more embarrassing than being caught by an adversary who accepts spying as just one more facet of an unpleasant relationship. And it is not only a matter of embarrassment. The effectiveness of the United States in urging order in the new world will be directly related to its ability to bring together coalitions of countries to grapple with international problems. Nevertheless, as we increase emphasis on securing economic intelligence, we will have to spy on the more developed countries-our allies and friends with whom we compete economically-but to whom we turn first for political and military assistance in a crisis. This means that rather than instinctively reaching for human, on-site spying, the United States will want to look to those impersonal technical systems, primarily satellite photography and intercepts.

This is not to say we should eschew human intelligence operations in the new world order, but that human spying must be employed more selectively. That runs exactly counter to what one inevitably hears when there has been an intelligence "failure," such as not predicting Iraq's invasion of Kuwait. The litany is familiar. We should throw more and more human agents against such problems, because the only way to get inside the minds of adversaries and discern intentions is with human agents.

As a general proposition, that is simply not true. Electronic intercepts may be even more useful in discerning intentions. For instance, if a foreign official writes about plans in a message and the United States intercepts it, or if he discusses it and we record it with a listening device, those verbatim intercepts are likely to be more reliable than second-hand reports from an agent. Not only do agents have biases and human fallibilities, there is always the risk that an agent is, after all, working for someone else.

There are a number of recent examples in which technical intelligence systems either uncovered intentions or should have. One was in 1986 when the United States intercepted messages from the Libyan government to its embassy in East Germany, indicating foreknowledge of a terrorist bombing attack in West Berlin. Using that revelation, the U.S. government attempted to thwart the act, but unfortunately the information was not sufficiently specific. Another instance involved the September 1984 terrorist bombing of the U.S. embassy annex in Beirut. Satellite photographs taken before the bombing, but unfortunately overlooked, showed that someone had built a mock-up of the embassy annex and its defenses and had practiced driving trucks through them.

Human intelligence is not the sole detector of intentions, but there is an area where it does have a rather exclusive role to play, and it falls in an area where U.S. needs are increasing: forecasting events driven by ground swells in public attitudes, such as public animosities toward an incumbent government, or outbursts of destructive zealotry impelled by ethnic or religious fervor. It may at first seem strange to use the human spying apparatus for public pulse-taking. Human intelligence normally involves recruiting agents close to some important individual or decision-making body, or planting listening devices in the right place or surreptitiously photographing documents. Today, though, the United States faces an increasing number of situations in which the target is not a well-constructed plan that has been hammered out in discussions on which we can eavesdrop, or one which is eventually transcribed and can be copied or intercepted.

For instance, in my time as director of the CIA, our greatest failure was inadequately emphasizing the dangerous waters into which the Shah Reza Pahlavi of Iran was sailing in 1978. What we needed to know was not that Ayatollah Ruhollah Khomeini and his lieutenants were secretly scheming in Paris and employing such unorthodox techniques as sending taped cassettes of propaganda into Iran. If we had obtained that kind of information about Khomeini's plans, we would have questioned whether any of the schemes would have been effective. What we missed was the breadth and intensity of feeling against the shah inside Iran. That was what allowed Khomeini to ignite the country. Those feelings against the shah were shared by disparate groups that came together for the specific and temporary purpose of deposing him.

American intelligence operatives were simply not in touch with the mullahs in the mosques, who were telling the peasants how the shah was "profaning" Islam; nor with the merchants in the bazaars, who were grumbling about the stranglehold of the regime on the economy; or the politically educated, who were exasperated with the shah's unwillingness to share political power. Not only were we not talking with a broad enough cross section of people, we were not sampling attitudes widely enough across the country. In part that was because in the past few decades the State Department, and with it the CIA, have confined their operations only to the capitals and one or two cities in major countries.

Even so U.S. diplomats try to stay abreast of public attitudes, and, logically, it is their task rather than that of the CIA. Unfortunately diplomatic personnel are contributing less and less in this area: in part because the burdens of diplomacy have increased; more so because totalitarian countries, such as the shah's Iran, do not tolerate State Department officers maintaining contact with dissident groups, some of them illegal. CIA officers who were not known by the local government to be employees of the U.S. government could do the job. Or they could recruit local agents to go out and clandestinely sample opinion. The latter would be more effective in any event, as there is a risk that citizens being interviewed would distort their answers if they were giving them to a foreigner. In addition we would want the same individual to do the sampling over a number of years. What is important is not what any single sample says but how it changes; for instance, whether there is increasing or decreasing support or opposition to a government.

One example of the importance of understanding public attitudes is our experience with maintaining peacekeeping forces in Beirut beginning in 1982. U.S. intelligence failed to detect the depth of animosity to this American military presence. When U.S. forces came to be seen as supportive of the Christian-dominated government, rather than as peacekeepers, a wave of terrorist acts from a variety of groups drove both U.S. peacekeeping forces and U.S. diplomats from Lebanon. This transpired not because some Islamic fundamentalist mullahs held secret meetings into which we might possibly have slipped an agent. Rather, the series of bombings, kidnappings, hijackings and murders designed to intimidate Washington came from a number of religious and ethnic groups in Lebanon. It was the cumulative effect of their separate actions that drove us out. Our clandestine intelligence operatives could have been out sampling those attitudes and determining that they were increasing to crisis proportions.

Still another case in which we were unaware of important indigenous attitudes was Nicaragua in the 1980s. Although the CIA was deeply involved in assisting the contras, it was not aware of how strongly the non-contra populace opposed the Sandinista government. Witness the fact that we had no inkling that the anti-Sandinista candidate, Violeta Barrios de Chamorro, would win the February 1990 elections by a landslide. It would have been a challenge to have sampled those attitudes, as many Nicaraguans were intimidated from expressing anti-Sandinista views. But that is precisely the kind of challenge U.S. human intelligence services are trained to meet.

A glaring example of a failure to understand foreign cultures and attitudes was the harebrained idea, spawned in the CIA, that selling arms to Iran in 1985 was the way to support "moderate" Iranians who would undercut the Khomeini government. Yet when the former national security adviser, Robert McFarlane, went secretly to Teheran to negotiate, he found that even someone as high up as Hojatolislam Hashemi Rafsanjani was terrified coming to a meeting, lest he be accused of consorting with the "Great Satan." All of this was seven years after the CIA (in my time) had underestimated the virulence of Islamic fundamentalists toward the United States.


Even though these examples show that the United States has a serious intelligence gap to fill, it will not be easy to induce the government's clandestine services to fill it. CIA case officers are trained in recruiting agents, planting bugs and photographing documents. To ask them to go out and sense public attitudes is almost like asking a fighter pilot to leave his supersonic jet and become a crop duster in a propeller-driven biplane.

This is not to suggest that gauging public attitudes should become the primary role for America's clandestine services. Washington has seen several examples recently when it badly needed their traditional practices of recruiting agents to penetrate governments. One was in determining whether Saddam Hussein was going to attack Kuwait in August 1990. Another was in locating Manuel Antonio Noriega when the United States invaded Panama in December 1989. Unfortunately in neither case were U.S. human intelligence operatives able to uncover the intentions of these two tyrants. Why not?

In Panama the situation was ideal for recruiting agents. The United States has long maintained a large presence there, which should have made it easy to persuade Panamanians that they could benefit from supporting the U.S. government. Yet when American troops moved into Panama it took them an embarrassing four days to track down the dictator. There is no question that Noriega took extreme measures to elude U.S. troops by constantly moving about and having duplicate automobiles and stand-ins. What the United States needed, and obviously did not have, was someone positioned inside Noriega's immediate entourage. The United States also did not have anyone inside Saddam Hussein's inner circle. Iraq was more of a challenge, considering the cultural differences, the difficulties facing any American working in the country at that time and the risks to anyone the CIA might have recruited.

There are several reasons why we were unable to effect penetrations like these and why our human spying will probably never be what we would like. To begin with, the CIA is a relatively young organization and has never developed a sound tradition in this area. In the CIA's early days the inordinate fear of its chief of counter-intelligence, James Angleton, about the possibility of Soviet penetration of the agency inhibited the recruitment of Soviet agents. After Angleton there was such a rush to recruit that many supposed agents were not adequately vetted. The CIA has yet to build a well-balanced clandestine service.

The more immediate problem is that the CIA's case officers operate under a handicap of not being able to maintain good "cover." The U.S. government will not ask American espionage personnel to make the sacrifices that the Soviets ask of their agents. Soviet agents will leave their homes behind, live for long periods in a foreign country, adopt its citizenship and engage in mundane business activities for years before commencing espionage operations. There is no way we could ask Americans to go to the often repressive countries on which we need to spy and live that way. Nor does Washington like to expose American agents to the penalties of being caught spying without holding diplomatic status.

The United States is not even good at maintaining the limited cover it does establish. Most agencies of the U.S. government are reluctant to have CIA clandestine personnel masquerading as their employees. And the CIA's clandestine people themselves are usually unwilling to pay the price of integrating themselves into another department's operations because that requires working full-time at two jobs simultaneously.

In face of these impediments any improvements the government can make to its human intelligence operations are likely to be marginal. Still we should try, as the payoff can be high when successful. All agencies of the U.S. government should be instructed that human spying is a presidentially and congressionally authorized activity, and that they must play a role in it. Some agencies ought even to be required to establish new functions overseas that would provide good cover for the CIA.

The CIA's clandestine service should also do more for itself. First it must appreciate that the cost of cover has increased substantially. At best, between 20-30 percent of a case officer's time will be spent spying. Second the clandestine service must accept the inconvenience of not having a separate CIA office in every embassy. Such offices are too easily identified. One of the most startling facts I came across in my years in the CIA was that the agency had its own officers' club overseas in the 1950s, back in the heyday of human intelligence. That could only have been an invitation to easy identification of the agents' roles.

Third the reward system for espionage must be changed. Presently it rests on numbers of agents recruited, but the focus needs to be on the quality of information obtained, judged over a period of time.

Finally the type of persons brought into American espionage must shift with the change in targets. We will need officers who are conversant with economics and with highly technical issues, such as nuclear proliferation, more than the broad-ranging political scientists who have traditionally constituted the backbone of the clandestine service.

In sum, the new world order will yield technical systems that will serve as a sword, the broad cutting edge of intelligence collection, and human spying operations that will serve as the rapier, to be applied judiciously to very specific requirements. Above all, America will have the ability to integrate both the human and the technical efforts that combine for successful collection. Each system has its strengths and its weaknesses. We must make them play to each other.


Ensuring such coordination is the responsibility of the Director of Central Intelligence. In theory he has the authority to carry it out, but in practice the mechanisms available to him are not sufficient. What is required is a relationship between the DCI and the collection agencies similar to that which military field commanders have to the army, navy and air force, and which was one of the keys to General Norman Schwarzkopf's success in the Gulf War. Schwarzkopf, as a unified commander, had full operational control over all forces assigned to him. The departments of army, navy and air force were responsible for providing support to those forces, that is, for obtaining and training the necessary people, equipment and supplies, but not for directing their operations in the field.

The DCI is in an analogous position to a unified commander. He must operate the intelligence-collection assets of the country hour-by-hour every day. It happens, though, that his two most important collection assets, the NSA and SRA, have always been under the military command structure. It is convenient to have military units send up satellites and man listening posts on aircraft, ships and military bases; and the military has a vital need for the kinds of intelligence the technical systems of NSA and SRA collect. But if we are to shift priorities to economics and politics, we need some mechanism for ensuring that military priorities do not always prevail.

Today the DCI can order up whatever collection operations he wants, but the directors of these agencies know their bread is buttered in the Pentagon. To get their attention the DCI, rather than the secretary of defense, should have the authority to appoint and fire these directors. They are, after all, two of his most key subordinates in the entire intelligence community. The DCI also needs a staff with the capability and the authority to issue operating orders to all agencies collecting intelligence, human or technical. In short the intelligence requirements of the new world order make it increasingly important that a single authority balance the economic, political and military intelligence requirements of the nation. That someone cannot be in the military command. It has to be the DCI.

The DCI also needs authority to control the dissemination of the information collected. One major reason the Congress created the position in 1947 was to avoid the mistakes made just prior to the Japanese attack on Pearl Harbor, when the army, navy and State Department intelligence bureaus were not fully sharing the clues each had about Japanese intentions. Today the collection agencies are almost as autonomous, with regard to excluding other agencies from what they have learned. Their rationale for this compartmentalization is to protect the "sources and methods" by which the information was gathered. That is a legitimate enough concern, but there is such excessive withholding that, in my opinion, the United States is just as vulnerable to a Pearl Harbor now as in 1941.

Another problem runs in the opposite direction. The collecting agencies have tended to expand their assignment: from collecting the data to interpreting it. This is dangerous because there is no assurance that any single collection agency has all of the relevant information or skills for such interpretation. In 1979 the NSA produced a report that determined that the Soviets had introduced a new "combat brigade" into Cuba, a conclusion that went beyond reporting what the NSA had actually observed. NSA had not checked with the CIA or State Department on the history of this brigade, and hence its conclusion was wrong. This episode had a damaging impact on the ongoing SALT II negotiations.


All told there will be considerable challenges in reorienting the collection agencies to a new set of priorities and in improving their total performance. Relatively speaking, though, there is an even greater need for improvement on the analytic side of the intelligence community. For one thing the U.S. government will rely less on spying to obtain information. In a more democratic and open new world order, there will be more information available from public debate, from the growing number of arrangements for on-site inspections for arms control purposes and from increasing international trade and travel.

Beyond that the quality of intelligence analysis has never met our expectations. Senator Daniel P. Moynihan (D-NY) has recommended that the CIA be placed under the secretary of state. He contended: "For a quarter of a century, the CIA has been repeatedly wrong about the major political and economic questions entrusted to its analysis."2 Significantly, the senator cited "political and economic" analysis, not military. With military intelligence it is vital to know your opponent's military capabilities so that you can prepare to meet them. Yet gauging military capabilities is a much more straightforward, factual exercise than, say, forecasting a political revolution such as the fall of the shah or an economic explosion such as the transformation of South Korea over the past two decades.

One of the cardinal weaknesses of American intelligence analysis has been getting large bureaucracies to go out on a limb, that is, to stake their reputations on forecasting. For instance, as late as 1980 I was briefing the Joint Economic Committee of Congress in open session that the rate of growth of the Soviet Union's gross national product was declining, but that the rate of Soviet spending on defense was climbing. Neither I nor the CIA's analysts reached the conclusion that eventually something had to give: that there would be a political and economic crisis. Yet it was only a little more than four years after I left the CIA that Mikhail Gorbachev entered the scene and found the situation so desperate that he moved quickly into the highly risky waters of perestroika and glasnost.

We should not gloss over the enormity of this failure to forecast the magnitude of the Soviet crisis. We know now that there were many Soviet academics, economists and political thinkers, other than those officially presented to us by the Soviet government, who understood long before 1980 that the Soviet economic system was broken and that it was only a matter of time before someone had to try to repair it, as had Khrushchev. Yet I never heard a suggestion from the CIA, or the intelligence arms of the departments of defense or state, that numerous Soviets recognized a growing, systemic economic problem.

Today we hear some revisionist rumblings that the CIA did in fact see the Soviet collapse emerging after all. If some individual CIA analysts were more prescient than the corporate view, their ideas were filtered out in the bureaucratic process; and it is the corporate view that counts because that is what reaches the president and his advisers. On this one, the corporate view missed by a mile.

Why were so many of us so insensitive to the inevitable? One reason, I believe, is that there are too few incentives in the intelligence community for forecasting. Our analytic bureaus, especially those dealing with the U.S.S.R., have become such large bureaucracies that, in the process of moving a forecast up the line from the individual analyst to the head of the agency, iconoclastic ideas get watered down in favor of what has been the accepted wisdom. For instance, one reason the government failed to anticipate the Soviet economic crisis adequately was that it assumed the old Stalinist rules about repressing dissent still prevailed, that public attitudes did not count in the Soviet Union. Yet Gorbachev made his radical moves not just because economic productivity was declining, but because the attitudes of the Soviet public toward their government, their work and their communist party were so negative. There was no longer any hope that fear and repression could drive the Soviet people into working harder.

Another reason for many of the analytic shortcomings is that our analytical agencies do not have an adequate grasp of the cultures of many countries with which we must deal. This of course goes back in part to our failure to feel the pulse in other countries, but it is an even larger problem of education and appreciation.

Still another explanation of our analytic shortcomings is the clash between fresh intelligence and established policy. Predicting the shah's fall, for example, would have challenged a basic U.S. foreign policy in the Persian Gulf area. Escaping the confines of policy will always be a problem for intelligence analysts. It should be less for CIA analysts because the CIA is not involved in making policy, as are other agencies of the intelligence community such as Defense, State and Treasury. The single greatest disadvantage of Senator Moynihan's proposal would be its subordination of the CIA to a policymaking organization. Most DCIS have not taken public positions on policy issues. Still, on major issues of foreign policy, any analyst knows the thrust the president is taking and, thus, must feel some pressure to support his ultimate boss. We need then to find ways to strengthen the sense of independence of analysts throughout the community.

If we do not accept Senator Moynihan's suggestion, how else can we promote better analysis? Giving the DCI more authority over the collecting agencies is a partial solution to improving collection, but it is not the answer to the problem of analysis. The last thing we need is forced conformity of analysis. Let me point out, by way of contrast, that conformity of intelligence to political ideology has always been one of the great weaknesses of Soviet intelligence. I once had the opportunity to ask a senior Soviet defector whether the top Soviet leadership received reasonably accurate assessments of why the United States was doing what it was. He replied that even very senior people like then-Ambassador Anatoly Dobrynin, who certainly understood the United States well, did not dare report with full frankness if their views ran contrary to established doctrine.

Analysts need to be given incentives to be more rigorous, to take into account the perspectives of other nations and to present well-reasoned ideas that run counter to accepted wisdom or policy. The government could help by establishing awards, perhaps with monetary compensation, for forecasts that proved to be accurate, even if overlooked by the system. Analysts would be encouraged to keep documentary evidence of positions they had taken. The president could help by authorizing the DCI to shift individual analysts between agencies on temporary assignments. This could help break ingrained thinking.

Moreover Congress could enact legislation to provide intelligence analysts a better opportunity to attend academic institutions, participate in professional conferences, travel and live abroad, acquire language skills and thus become true experts in their areas. Congress could also authorize an expansion of the DCI's personal analytic staff, presently known as the National Intelligence Council. An enlarged NIC would still be small enough to prevent redundant tinkering with analysts' work as it goes up the organizational chain. An enlarged NIC also could be a partial solution to the problem of the collecting agencies' excessive restriction of dissemination. This would be a core group of analysts who would be cleared for all, or almost all, information-our insurance against another failure like Pearl Harbor.


Congress should amend its 1947 legislation, which assigns the DCI the dual roles of coordinator of the intelligence community and head of the CIA. Then, there were no satellites to be operated on a national basis; the NSA had yet to be established, and electronic intercept work was being done by the military; the CIA was just being formed; and the entire intelligence community was much smaller, with military espionage divided between two modest components, army and navy intelligence. Today the intelligence budget runs to billions; we have a multitude of national satellites with a variety of sensors; the CIA is reported to have many thousands of people; the NSA has many more personnel, as well as some of the world's most powerful computers; and there is a Defense Intelligence Agency, an Air Force Intelligence Agency, plus a whole host of military laboratories dedicated to individual subjects, such as aerodynamics, underwater acoustics and tank armor.

The intelligence community has grown in size, complexity and importance to the point where it now requires a full-time leader, a Director of National Intelligence (DNI). That said, there has long been a conflict of interest in having one person direct both the community and one of its key elements, the CIA. For instance with spending on national security set to be cut substantially, the United States will need an impartial arbiter to determine how to eliminate the very considerable duplication within the intelligence community. The real clincher is that no matter how small or how large the adjustments in intelligence on the Soviet Union and on military matters need to be, they will be resisted. That resistance will flow from long-established convictions that have meaningful roots beneath them. The most meaningful is that the United States must not endanger its military capabilities, even if our leaders think that those capabilities will be needed less in the future.

Over the past decade, the military has striven for even greater control over intelligence operations, including a determined effort to duplicate the CIA's human-intelligence capabilities. In the process it has managed to shift intelligence funds to the point that over one-third of the total expenditure is under the Pentagon's control, not the DCI's; and another half jointly controlled. The military's concerns behind this shift are not unfounded. The CIA has never paid sufficient attention to, nor been very good at, military intelligence other than strategic nuclear matters.

We have a challenge then to reconcile this growing, overall influence of the military and the declining importance of the military segment of intelligence. Again, the answer lies in an impartial DNI who does not have to defend the CIA, as does the DCI, but who can devote full time to adjudicating as impartially as possible with the intelligence community. Still it would be both wise and politic to provide in law for military representation in the new DNI's organization. Provision should be made that either the DNI or his deputy would be either a retired or active military officer; and that the DNI would have three statutory deputies, at least one of whom would be a three or four-star military officer.

This officer could serve as the deputy for budgets. Such a position already exists within the intelligence community staff. The only change needed is that Congress would send the funds for intelligence directly to the DNI, and this deputy would disburse them to the various agencies of the community. Or, the military officer could be the deputy for analysis, again a position that already exists as head of the NIC. An enlarged NIC is essential to the concept of a DNI because without a sufficiently large analytic staff to keep abreast of developments, the DNI could be too dependent on others.

Finally, the military is more likely to fit well into the position of deputy for collection. This function exists today in the limited form of committees for the coordination of human, electronic and photographic collection. What is needed is more manpower to direct these operations. Day-to-day support would remain with the parent departments, primarily the Department of Defense.

Hopefully the creation of a DNI can be done by shifting people around rather than increasing the total number of people in intelligence assignments. Larger staffs and more central control are not always good solutions to problems. In this case centralization has been progressing almost from the day the position of DCI was created. Each successive director has found it necessary to play a more active role with the community, and presidents have augmented his authority progressively. Now the dramatic world events in 1989 and thereafter should tell us that we must implement dramatic changes if our intelligence capabilities are to keep abreast of these times.

1 For reasons that are difficult to comprehend, the true name of this agency is classified. "Satellite Reconnaissance Agency" (SRA) is a fictional title that will be used for convenience in this article.

2 The New York Times, May 19, 1991.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • Stansfield Turner was Director of Central Intelligence, 1977-81. He is author, most recently, of Terrorism and Democracy.
  • More By Stansfield Turner