A Framework for Reform of the U.S. Intelligence Community: Chapter 5

A Framework for Reform of the
U.S. Intelligence Community

Chapter 5: Reform Ideas to Avoid


While the Intelligence Community needs many reforms, some ostensible reform/reorganizational proposals are poorly conceived and should be rejected. These include some from prominent individuals.

CIA Must Continue to Exist

Reformers in both Congress and the Executive Branch must understand that the national missions that CIA performs are essential to the well-being of the nation and that they must be maintained. There is no substitute for this work. The functions could be performed by another, newly-created organization composed of entirely new individuals, but that step would be time-consuming and expensive. The by now well-known lapses of the Agency do not argue for the CIA's elimination, but rather for a thorough house-cleaning. The misinformation and hysteria on this issue is enormous. Spread by even ostensibly responsible individuals like former SSCI member Senator Daniel Patrick Moynihan, who has introduced legislation calling for the dismemberment of CIA, these extreme views do not contribute to the national debate about how to improve the performance of intelligence.

A further confusion continues to stem from the popular myth that CIA was created to "fight" the Soviet Union in the Cold War and that, with "victory" in that "war" having been declared, there is no more need for CIA. In fact, the National Security Act of 1947, which created CIA, was largely a result of dissatisfaction with the fractured military command, control, and intelligence system of World War II. During the war, the Army and Navy operated with often dysfunctional independence. (Read for example, of the bitter personal battles between Admiral Chester Nimitz and General Douglas MacArthur in the Pacific campaigns for a glimpse of the problem.) There was need for a Defense Department (created in 1949 after the National Military Establishment formed by the 1947 Act proved ineffective) to coordinate defense issues. A stronger foreign policy establishment in the White House was created in the form of the National Security Council. And, the Air Force finally won separate service status.

The emergence of the Soviet Union as a prospective adversary spurred changes, but the United States had a primitive national security structure that was not well suited to serve the needs of a new world power. Other countries long had formal civilian intelligence services. We periodically had military intelligence services. General George Washington had one during the Revolutionary War, which was disbanded after independence. During World War II, despite the fractured Army/Navy relationship, our military signals intelligence service made a major contribution to the war effort by breaking Japanese military codes. The formation of CIA as a national intelligence agency was only our belated joining of international norms.

Keep the DO and DI Separate

DCI Woolsey's major organizational initiative--the physical integration of the CIA's analytic and operations directorates is a terrible idea that should be reversed. Toward the end of his tenure, Woolsey announced plans to create a "partnership" but not, I understand, a full fledged merger between the organizations. Officers from the Directorate of Intelligence and Directorate of Operations were to be physically co-located in some cases--ostensibly to improve cooperation between the components. But the idea has major drawbacks (90). It:

  • Increases the opportunity of a mole, especially in the DI, to gather a broad range of sensitive operational information.

  • Puts together very different cultures, with different missions, that require different managerial styles. In the DO, for example, a more hierarchical military command and control system is appropriate, while the best DI units operate under a system of relatively loose management that guides analytic priorities but which encourages debate and controlled internal controversy.

  • Puts covert operators, who are policy implementers, together with DI officers ostensibly working only to provide information and objective analysis. The mix would taint the perceived objectivity of already suspect analysis. In ethics, the U.S. Government cares about the appearance of impropriety, whether or not actual actions are inapproriate. It should do the same in intelligence.

    Analysts and collectors need to be in regular and close communication, but they work in different ways and do not need to sit next to each other.

    But some distance between operators--acting as either collectors on covert action missions--and analysts does not require separate agencies, as some have proposed. Melvin Goodman, for example, cites British practice--which separates operators (MI6) from analysts (in the Ministry of Defence and the Joint Intelligence Committee)--as a good model (91). While such a separation presumably would reduce somewhat the likelihood of Ames-like counter-intelligence scandals, it has efficiency problems and surely would not prevent another of Goodman's suggested advantages: elimination of advancement of parochial opinions to the White House and State Department. There is no reason why separation would make any difference in this area. As we have seen, DO operators lobby Congress on their own behalf, and a DCI or DDI with a policy bent could try to influence the NSC through force of opinion, judgment, wisdom, or the slant of his analysts' papers. Good senior leadership alone can prevent unwanted intrusion of intelligence into the policy process.

    No Military Control of Covert Action

    The U.S. military should not assume control or conduct of paramilitary covert action, as Robert Gates and some others have proposed (92). I say this for two main reasons: the military cannot handle the job; and, there are major diplomatic and domestic political risks associated with use of uniformed military personnel in such activities. I make these comments as a U.S. Army reservist who has spent most of his military career in intelligence and special operations assignments (93).

    The military--despite the elevation of special operations in stature and funding as part of the Goldwater-Nichols Act of 1986 and growing involvement in non-traditional roles--still thinks conventionally. Conventional force commanders run the services and the Joint Staff, despite the elevation of one special operations officer to four star rank as commander-in-chief of the U.S. Special Operations Command. Even though there are personnel who are sophisticated in "operations other than war"--the current term for what once was called low intensity conflict--there is not an adequate base of personnel experience to develop sophisticated plans or conduct sophisticated operations of the scale a president might ask CIA to perform. There is not the money or the time to prepare such forces. The military personnel management system mandates that people, including intelligence officers, rotate too quickly to develop area expertise. Despite some successes, I also doubt that the military could maintain security as well as CIA does. Moreover, the military's success in covert operations in Southeast Asia in the 1960s and 1970s was poor--for reasons that Defense Department reforms have not ameliorated. This subject warrants significant discussion--about these and other issues including CIA's own very mixed record in conducting covert operations--in closed-door sessions.

    There are several major political reasons to keep the Defense Department out of covert operations:

  • It is important to be able to maintain plausible deniability. Use of American citizen-soldiers, poorly able or untrained in concealing their national heritage, sharply reduces our ability to conceal involvement.

  • Families of soldiers lost in battle are more likely to complain publicly that they sent their loved ones to "defend the country"--not participate in some "dirty" war only tangentially related to America's immediate defense.

  • Unhappy mothers and wives of servicemen lost in covert actions are much more likely to demand "accountability" than the families of intelligence officers who know the rules of the game. The POW/MIA cottage industry of the post-Vietnam War period is more likely to "reinvent" itself to find soldiers lost in covert actions than it is to seek out missing or dead intelligence operatives or agents.

  • It seems inevitable that uniformed military involvement in covert operations would raise questions related to the War Powers Act or related questions about whether the nation is at war or at peace. These would increase the likelihood of yet another oxymoronic public debate about the conduct of "covert" action.

  • Military involvement in covert actions could cause complications with our allies. In particular, it would raise yet more questions about our judgment and concerns about whether treaty terms that require foreign allies to assist each other are in danger of being triggered. Foreign countries may be less willing to ally themselves with us if they sense increased chances of being dragged into a war they do not want.

  • Use of U.S. soldiers in covert actions reduces our moral authority and our diplomatic assertions that negotiations, multi-national organizational fora, and legal proceedings should be the foundations of conflict resolution.

    And, successful covert para-military actions may benefit from or even require the access, sources, and tradecraft of the clandestine service of CIA that the Defense Department cannot match. All of these problems are avoidable through continued separation of the Department of Defense from traditional CIA para-military covert operations.

    No National Imagery Agency

    There is nothing unique about imagery as a source of information that warrants its placement in a separate collection or analytical agency. Indeed, the best imagery analysts are good all-source analysts who rely first on imagery. CIA years ago gave up the idea that "photo interpreters" only look at pictures. It created the Office of Imagery Analysis, which gave full analyst status to imagery analysts. Separation of such analysts into another agency would almost certainly reduce coordination with other analysts, cut overall analytic quality, and lead to yet more organizational turf fighting.

    The range of imagery analysis runs from analysis of the products of national level systems--that is, satellites--to specialized air-breathing platforms' output, to photographs and other images taken by tactical fighters on bombing runs, to ground photographs taken by infantrymen. This work is done at the field level in support of targeteers and tactical commanders who, like General Schwarzkopf in 1991, had to decide if the fire support provided by aircraft has inflicted sufficient damage to allow other tactical initiatives against Iraqi forces. The complexity, type, and time-sensitivity of such work is sharply different than that of the national-level imagery analyst now working at CIA's Office of Imagery Analysis or at DIA--who may be evaluating complex technical issues and conducting the imagery equivalent of basic research. While imagery analysis done as part of bomb damage assessments supports tactical commanders' decisionmaking, Washington work is often long-term and strategic in nature.

    These many types of imagery analysis are not well placed under the same organizational umbrella. Military field commanders will not accept control of assets potentially critical to their tactical decisionmaking to be controlled by out-of-theater organizations--especially civilian ones. And, combatant commanders' concerns may differ sharply from those of the Departments of State or Energy. Still more different are the interests of Agency for International Development officials wanting to know if Landsat can tell them if drought conditions in the Horn of Africa have again worsened to the point where another complex humanitarian emergency looms on the near horizon.

    There is room to streamline the collection of imagery, however. The unusual organizational structure of the NRO as a joint CIA/Air Force outfit is a testament to turf jealousy, not wisdom. If the NRO is maintained, it should come firmly under the contro l of a strengthened DCI, who needs such authority to balance the budgetary demands of the collection and analytic components and agencies against their roles in furthering the greater good of the Intelligence Community, and to ensure that funds roughly go first to meet higher national intelligence priorities, not to the organizations with the best lobbying efforts. The Air Force can continue to perform satellite launch and control functions on a contract basis--following the instructions of its client, the DCI, either directly or through the NRO.

    The mechanism for tasking satellite imaging systems also merits review. The long-standing COMIREX structure was cumbersome. It featured too little strategic planning and too much ad hoc targeting based on the bureaucratic power of transient coalitions of consumer agencies fighting for control of limited capacity. A streamlined Defense Department requirements system funnelled through DIA would help. This issue--including an evaluation of recent changes in the requirements management system and the creation of the Central Imagery Office--deserves a thorough examination in a classified context.

    No Separate Open-Source Agency

    The growing body of publicly available information, the Internet, and talk of an "information superhighway" have spawned discussion of a need for separate organizations to handle this "open-source intelligence." Much of this talk is uninformed, however, and even responsible advocates of more use of public information sometimes make extravagant claims of its usefulness.

    No separate agencies are warranted, although better collection and processing of such data seem appropriate. New equipment, operating procedures, internally modified organizations, and attitudes are needed to handle the new informational environment. The Community Open Source Program Office (COSPO), headed by CIA careerist Joe Markowitz, appears to be headed in the right direction. With about one percent of the Intelligence Community's budget, COSPO is developing ways to ensure better access of publicly available information to analysts and thus to polymakers--without compromising the security integrity of classified information systems or intellectual property protected by copyrights (94). A program to put new computer terminals on analysts' desks hardly warrants creation of a new agency.

    It is worth remembering that open-source information has been a staple of intelligence collection for a long, long time. Contrary to now-conventional wisdom, intelligence analysts long have known of and used open source information extensively. Diplomats have collected such "intelligence" information for centuries.

    More recently, CIA has had the Foreign Broadcast Information Service (FBIS) as part of its Directorate of Science and Technology. FBIS conducts open source collection and translation of radio and television broadcasts, translates print media articles, and sometimes releases analytic pieces based on open source reporting. In addition, intelligence analysts regularly read, and for years have read, U.S. and foreign news publications, watch(ed) foreign television broadcasts, and subscribe(d) to a host of specialized publications. Subscription budgets at CIA have been enormous--for both periodicals and analytic services. (As an aside, recall that FBIS translations serve a host of non-government individuals and organizations as well as the intelligence community; CIA thus long has helped promote general American understanding of the world and American commercial interests.)

    Moreover, CIA for years has had a formal program to collect information about foreign affairs, particularly in technically demanding areas, from knowledgeable U.S. citizens. Analysts talk with these non-government specialists in many areas, as well as prepare questions for the full-time collectors to ask. I have done it myself repeatedly. These contacts have been helpful, but rarely in my experience have they provided critical information. There is little qualitatively different about what has been done than that proposed by open source advocates like Robert Steele of Open Source Solutions, Inc. These contacts have been limited, however, and there may be opportunities for expansion of the program.

    Any such expansion must address at least two major concerns of the citizens, however: confidentiality of information; and, the time and other burdens of talking with CIA. In the past, these contacts have been volunteer efforts by loyal citizens. A program to levy greater demands on the business and academic communities should be prepared to offer market rate compensation for professionals' and executives' time, as well as expense reimbursements. If this occurs, however, there is some danger that those individuals might attain unwelcome status as paid "agents"--and thus draw the attention of the counter-intelligence services of the countries in which they do business or have academic exchanges. This attention presumably would limit further the appeal of the program to such people and to their organizations.

    The explosion of electronically transmitted information amounts to an increase in the volume of information made possible through improved technology. There is not, however, any qualitative difference from the open-source collection of the past. Intelligence agencies should continue the process of better gathering the information available and working to improve further ways of using it. There is no need for independent organizations to handle this flow. Open-source information should flow to analytic offices, just as the products of other collection efforts should. This collection should be as integrated as possible with all collection and analysis, not segregated by origin.

    This is not to say that improvements cannot be made. FBIS, in particular, has borne some criticism for slow responses to technology changes. It appears to need substantial modification and enhancement. FBIS's analytic responsibilities might best be shifted to the DI.

    It would be foolish to create analytic organizations that exploit only open-source information. This would cause needless and costly duplication while limiting the capabilities of the open-source-only analyst, and reducing the resources of other analysts presumably to be forbidden to use open source data. Such a proposal is akin to suggesting the replacement of an organization of able-bodied analysts with one having only deaf and blind ones who can use only their sense of touch to acquire information, and another with quadriparaplegic analysts who can only use their eyes to acquire information.

    For insight into the likely effectiveness and reputation of an open-source only analytic group, we should look closely at the analytic clout of FBIS and NSA's current source-specific analytic units. I state these judgments after regular attendance at several dozen monthly "warning meetings" held by National Intelligence Officers in which analysts from throughout the Intelligence Community shared views and forecasts (95).

    FBIS analysts cover limited subject matter and receive generally low weights within the analytic community because of the internal bureaucratic status of FBIS and the fact that the analyses are based on incomplete information. FBIS analysts normally are "wall flowers" at warning meetings. This occurs despite the often considerable abilities of FBIS personnel.

    Recall that NSA's analysts are restricted to SIGINT-based analysis. They also have a relatively modest impact within the Intelligence Community's analytic corps because their informational and scope limits are so well known. I know of no significant consumer group that looks primarily to NSA for analysis, although there are certainly some topics that NSA collectors cover better than other collection agencies. This is not to say that NSA has poor analysts. As at FBIS, many are good people. But their mandate is highly restrictive for turf reasons, and the country gets less benefit from their services that it might with unified, all-source analysis. The clear lesson is that source limited analysts have carried less intellectual and bureaucratic clout than those from all-source analytic organizations. I see no reason why this pattern is likely to change.

    We should be equally skeptical of the now-common claims that very high percentages--be they 40 or 65 or 80 or 90 percent--of "intelligence" comes from open sources. There are major methodological problems with such claims. How, for example, are such percentages calculated? Numbers of words? Numbers of paragraphs? Numbers of key points, whatever they are? And, critically, how are the numbers evaluated? There is a qualitative element to intelligence information and analysis that proponents of open-source intelligence rarely mention. All of the individual place names, nouns, verbs, and adjectives of highly classified intelligence reports are "open source"--they too are part of the English language--but their assembly into critical bits of timely data, insightful analysis, and persuasive logic make those collections of words into intelligence products and separate them from dictionaries and atlases. Indeed, there appears to be a loss in the ongoing debate about intelligence reform of the long-known fact that information alone is not intelligence; it becomes intelligence only after proper verification and assessment.

    Beyond the accounting, there are concerns of bias that may be somewhat different than those analysts traditionally have encountered (97). Information enters the public domain because someone thinks others want to read or see it. That means there are selection processes that could reflect commercial or ideological interests and a variety of biases. There is no quality control on the information; it can reflect opinion, rumor, propaganda, or disinformation. Easily found information quickly and voluminously enters the public domain, while tough, complicated issues tend to be short-changed. And so on.

    Intelligence agencies and academics need to do much more research to identify any new types of bias that may creep into the electronic transmission media--and their extent and magnitude--and identify processing mechanisms needed to best use the information. How, for example, is one to assess the usefulness of raw information injected into the Internet when it is phrased in haughty, disparaging language that also is designed to demean another Internet communicator? Are there psychological rules for cleansing the verbiage? Can it be taken at face value, or must it all be rejected as vain frivolity in the absence of corroborating information? How can this all be done when messages are anonymous? And so on.

    Moreover, open-source "intelligence" has an ideological, anti-intelligence aspect that I suspect accounts for a significant part of its popular appeal. As is well-known, U.S. and other Western intelligence services have been assailed for years by some who believe that secret intelligence services are inherently threats to democratic institutions. To these people, the exaggerated claims of open-source intelligence proponents offer a convenient way not only to get more efficiency from government--the ostensible goal--but elimination of institutions they oppose for ideological reasons. If open-source is so good, the reasoning goes, there is no reason for the secret intelligence organs that have such onerous implications for civil liberties.

    Open-source information can partly assume a role that painstakingly gained collection of hard-target intelligence information played 40 years ago--when CIA gathered basic data and published the National Intelligence Survey series of books on countries of the world. The series died in the late 1960s, I understand, as CIA moved away from basic data gathering to become more "policy relevant." But with the prospect growing that U.S. military forces may be deployed with little warning to provide humanitarian assistance or peacekeeping services, there is more need for basic information--maps, demographic data, biographical information, and so on--than traditionally has been collected on the Third World. The NIS books still may be useful for these purposes; I found some on the shelves of the library of Intelligence Center, Pacific during a 1989 Army Reserve tour at Camp Smith, Hawaii.

    Much of the information that military planners need for at least initial planning is publicly available, but has been ignored in recent years. Thus, we continue to experience failures or embarrassments such as invaders of Granada having to rely on oil company maps to get around the island because, I am told, the Defense Mapping Agency did not bother with Granada. More recently, failure to have a basic understanding of the political situation in Somalia in 1993 led to the incompetently humorous spectacle of U.S. Marines in full combat gear "storming" ashore at Mogadishu at night only to be met by cheering Somalis and the bright lights of television cameras. This intelligence embarrassment was better than a hanger filled with full body bags at Dover Air Force Base, but it was a failure nevertheless.

    Once in Somalia, I am told by participants in the operation, U.S. forces found neither the open sources they had nor national intelligence systems to be very helpful (98). Instead, the military quickly developed its own HUMINT collection system to support operations and to protect United Nations forces from attack. The national Intelligence Community was not well equipped to meet such needs. To make matters worse, there were embarrassments about improper sharing of information with some United Nations forces and simple abandonment of boxes of classified information that an American fortuitously found. Surely the United States can do better than this!

    Certainly we can do better to serve expeditionary forces like Marines and special operations forces, including civil affairs troops, involved in operations other than war. But improving intelligence support for these forces requires little more than more attention to the issue. It does not necessitate a restructuring of the Intelligence Community or a vast increase in resources devoted to open source collection. It can mean, for example, generating short, timely, low-classification situation reports ("sitreps") that can be used by soldiers in the field with little ability to securely store more highly classified material. It should be clear, however, that more highly classified material virtually always is more valuable, and another option is finding ways to get more sensitive material to field users. Low classification material that is 85 percent correct--a figure Robert Steele has repeatedly cited as a good one--is not acceptable. Would you like an enemy who is shooting at you to miss only 85 percent of the time?

    There are, however, some issues that are more fundamentally new about recent changes in information technology that have operational and legal elements. I have no answers to the following questions, and understand that they are not close to being resolved within the Intelligence Community:

  • Can the Internet's public/private encryption characteristics be used to recruit sources? If so, what special security measures are needed? How can bone fides be established?

  • Can U.S. intelligence set up a bulletin board for anonymous tips for information about terrorist activities, for example? If so, how can bone fides be established and pranksters identified? If the tips are anonymous, would CIA be violating U.S. law by unwittingly taking information from or about a U.S. citizen--abroad or in the United States.

  • What are the counter-intelligence dangers of widespread access of U.S. personnel with sensitive government data or proprietary commercial technology to computers which may be offering rewards for treason or commercial espionage? What are the jurisdictional roles of CIA and the FBI when the communications are boundaryless?

  • Efforts are underway to create a secure electronic currency, or "digital cash," to spur international commerce (99). But real growth in this "money supply" would begin to raise questions of central bank control of money supplies, whether it could become another mechanism for "hot money" to contribute to financial instability, trade data reporting, income tax evasion, and so on. To what extent should intelligence be concerned with such issues and how aggressively should it target both the messages (as by NSA, for example) and the human beings and network control organizations (as by the DO, for example) that use them, have paper printouts, and control the hardware/software that accesses the systems?

  • Presuming that drug dealers, money launderers, criminals-for-hire, and terrorists will be active on Internet, what degree of sting activity is appropriate? How should coordination between intelligence and police agencies be handled? What communications law or civil liberties protection provisions might be relevant?

    Treatment of these and undoubtedly other many other information technology-related issues properly resides with the organizations that handle the fundamental issues now. It makes little sense to create an entirely new intelligence community devoted to just one part of the information spectrum.

    No Industrial Espionage

    American entrance into the burgeoning business of industrial espionage is a bad idea that the Intelligence Community properly opposes. I see several reasons not to emulate the former practices of Warsaw Pact states or to join France, Israel, and Japan in this on-going effort:

  • There is little for us to gain. Private sector analysts (and CIA as well) note that the United States continues to be the premier developer of technology in the world, despite the growth of development of good products and techniques throughout the world. We produce the most Nobel science prizes. Any decision to enter the fray in an offensive mode should weigh costs against benefits. We have comparatively few potential benefits against which to balance the risks and costs.

  • Americans cannot keep their mouths shut. Any decision to actively seek foreign secrets would soon become public knowledge. There would be public opposition, and international recriminations. The publicity costs would be considerable.

  • Implicit or explicit American claims to moral superiority, already suspect, would be still weaker. We would look yet more hypocritical.

  • There is no obvious way to distribute the "fruits" of industrial/commercial espionage to American firms. Their people, too, like to talk, and receipt of some information would become public knowledge. This, in turn, would virtually inevitably spark complaints of favoritism in the distribution of such ill-gotten gains. After all, we should be "fair" in distributing our stolen property. Would this not be a wonderful new arena for the trial lawyer community?

  • Asymmetric publicity about U.S. firms' receipt of intelligence information (but not the take of their foreign competitors) would almost certainly provoke some international retribution against American companies, whether they received intelligence data or not, further increasing the costs of such activities.

  • In the short run at least, most HUMINT collectors are poorly equipped to do this kind of work. As mentioned, the typical DO case officer's bag of tradecraft rarely includes the technical expertise needed to spot and recruit good prospects--or to assess the quality of their offerings.

    While industrial espionage is a bad idea for us to adopt, there are plenty of economic missions for intelligence. Both major groupings have been around for a long time. CIA, and to lesser extents DIA and INR, have long examined foreign economic events as part of their assessments of activities abroad. These have ranged from macroeconomic studies of gross trends, to econometric model building, balance of payments assessments, and industrial policies, for example, to sectoral studies of defense industries, labor, agriculture, envronmental issues, and so on. Collection agencies have long supported this analytic work where possible and appropriate. Such work should continue.

    Some of this work also involves studies of developed western countries that are friendly to the United States. This work primarily involves assessment of open source information. CIA does it because other U.S. Government agencies are not equipped to do it; they do not have the resources to do the work. The results are classified because the papers respond to U.S. Government queries which reflect U.S. concerns and therefore are sensitive (100). The fact that the DI assesses the economic prospects of the European Union countries, for example, by no means that CIA is "spying" on them, however. This is an important point. Surely all countries wonder what foreign initiative the American government is going to come up with next--an often difficult task--but their wonderment at and study of American political intricacies does not necessarily mean that they "spy" on us.

    The second major area is counterintelligence. CIA and NSA legitimately defend US economic interests aginst attack from abroad. They can do this by alerting companies to attacks, or breaking up foreign governments' efforts to advance their national interests over American interests through circumvention of the relatively lax rules of international commerce. An example is exposure of a bribe a friendly government paid to a Third World government official to secure a lucrative contract. We do this already, and should continue to do it.

    Put Ames in Perspective

    Americans generally, and critics of CIA in particular, have substantially overeacted to the Aldrich Ames affair. There is need for some perspective. Ames is a traitor who compromised some of CIA's operations and contributed to the deaths of some of CIA's assets. He did much damage. The Agency and Director Woolsey have received much justified criticism for their handling of the Ames affair.

    Ames was a single individual, however. He did not destroy CIA or lead to the destruction of the nation. He was the only sitting CIA officer now known to have worked for Moscow. (Edward Lee Howard apparently passed information to the Soviets only after he left CIA.) Ames is not reason to dismantle the Agency; were this so, the KGB and other Warsaw Pact intelligence services that were riddled with western mole holes would long ago have been shut down. Nor is Ames the reason for the Congressional unhappiness with CIA and the Intelligence Community that led to creation of the Commission. This displeasure had been building for years. For a refresher, read SSCI transcripts of the 1987 and 1991 Gates DCI confirmation hearings.

    I believe reformers should, however, remember that Ames went to work for the Soviets in part because of his disillusionment with CIA and its senior officers. Ames told his sentencing judge that CIA:

    [W]as and is a self-serving sham, carried out by careerist bureaucrats who have managed to deceive several generations of American policymakers and the public about both the necessity and value of their work. (101)

    While Ames was rationalizing, his comments suggest a bureaucratic alienation that many CIA people experienced--I experienced it myself--and which was the direct result of poor management. We should not reject the possibility that there are other moles now in the Intelligence Community--and that bureaucratic alienation was a motive for their treason. Or, perhaps more likely, alienation was one element that led to a reduced defense against more traditional ideological, monetary, or religious pitches. Better leadership can ameliorate though probably not eliminate this security risk.

    Don't Eliminate CIA's Military Analysis

    Robert Gates and others have proposed removal of military analysis from CIA's charter (102). This is an idea that is seriously misguided for several reasons:

  • CIA serves as a valuable source of alternative assessments to those of DIA and the military services. In the many analytic (and bureaucratic) battles between Defense and CIA over Soviet weapons capabilities in previous decades, CIA often (but not always) was correct. Intelligence analysts who are directly subordinate to senior officers with parochial organizational or budgetary interests that spring directly or indirectly from intelligence judgments are not independent. Their judgments have been and properly should be subject to review and independent confirmation. There is a potential for conflict of interest, especially among the military intelligence agencies, that the Intelligence Community organizational structure should take into account.

  • The military services and DIA traditionally have not devoted much effort to areas of the world and issues for which there were not plans, priorities, and troops deployed. Thus, "surprise" deployments to places like Granada, Somalia, and Rwanda meant that there was little information or expertise on hand to guide troops as they deployed. Because we can expect senior commanders to require intelligence support to work on issues they now are concerned with--not possible contingencies or ones their successors in normally short-tenured command positions will face--we can expect continued Defense Department inability to adequately look over the horizon.

  • The Army, I am told, is de-emphasizing expertise among its uniformed intelligence personnel in favor of generalist training. It long has relegated its Foreign Area Officers (FAOs) to dead-end careers--a measure of the service's regard for intelligence expertise. Such steps put more responsibility for expertise on the shoulders of civilian intelligence officers, who must use persuasive judgment, not weight of rank, to influence commanders' decisions. The weakness of the Army's intelligence corps is particularly unfortunate because its personnel are most involved in the complex foreign interventions (such as Haiti, northern Iraq, and Somalia) that require sophisticated understanding of foreign cultures and political institutions. Marine involvements typically are short and relatively unsophisticated--a reflection of the Marine Corps' short logistical tail. It is a good thing, in my judgment, because the Marines lack adequate intelligence and civil affairs training and maintain a mindset that places much more emphasis on destruction than building (103).

  • Military decisions have political, economic, social, and other components--areas in which CIA's analysts often are very good. Separation by fiat of these analysts from military intelligence analysts and consumers makes no sense. It may be penny wise but would be pound foolish. Indeed, there should be much better coordination between CIA's non-military analysts and military personnel, particularly those employed in operations other than war (104).

    Moreover, CIA itself in the 1980s and early 1990s devoted relatively little attention to traditional military analysis, with the prominent exception of the Soviet Union and high-interest conflicts like the 1980-88 Iran-Iraq war. In the late 1980s, for example, when the Office of Soviet Analysis decided to purge its files of information on non-Soviet Warsaw Pact countries, it simply dumped them on the literal doorstep of the Office of European Analysis. That office, sure that policymakers had concerns about the role of the Yugoslav military in internal political affairs but seeing no use for traditional military analyses of any East European country including Yugosalvia, directed that the files be destroyed. Most were. EURA's two East European "political-military" analysts became frustrated and left the office. Thus, when Yugoslavia unraveled in the early 1990s, CIA had no base of information or analyst expertise on the Yugoslav military. It therefore had nothing on the armed forces of the newly separate states either.

    EURA took its myopic approach because it had adopted the directives of Robert Gates and his proteges to concentrate on issues that had more immediate pay-offs in terms of papers produced and kudos earned. Because mid-level and office-level officials did not foresee chaos on Yugoslavia, they declined to invest in expertise and data that would have been very handy just a few years later (105). This myopia was and is a sharp departure from the research-oriented work of the 1950s and 1960s, when CIA analysts labored to acquire information and expertise that was quickly deployable in crisis situations (106). Such work was denigrated in the 1980s, and analysts and managers who tried to preserve such work--and even the information--were castigated as relics. The new DI of the 1980s increasingly devoted itself to brief, pithy, short-horizon papers designed to cater to the immediate issues of policymakers' daily calendars. Expertise became bureaucratically much less important than it once was.

    CIA also traditionally did little order-of-battle and other military analysis because it did not view support of the military to be a CIA function. DIA had turf jurisdiction. And, as former Chairman of the National Intelligence Council Christine Williams told me in the late 1980s, the military services were not legitimate customers of CIA because they were policy implementers, not policy makers. This view changed somewhat after General Schwarzkopf's stinging criticism of the performance of intelligence in post-war Congressional testimony--a new military liaison office was created, for example--but it continues to reflect some CIA people's attitudes (107). There should be much closer CIA/military analytic links--not an end to them.

    No New Intelligence Paradigm

    Perhaps no current notion is more pernicious than the idea that the end of the Cold War requires a new way of thinking about intelligence and new organizations to suit. Such beliefs confuse the alteration of the old East-West geo-political paradigm--with the need for accompanying decisions about military force structures and doctrine--with alteration of the intelligence paradigm. In fact, the mission of U.S. intelligence remains exactly the same as it was a decade ago--to provide information and analysis that senior policymakers need to defend and to advance U.S. interests. Reformers may make serious mistakes in mandating new organizations, procedures, or threats if they do not keep this simple fact clearly in mind.

    The point merits repeating. The mission of U.S. intelligence remains the same. Priorities, targets and resources have changed, but the mission has not. There is no need for a new paradigm predicated on the demise of the Soviet Union. Nor is there a need for one because budgets are leaner. Change has been a given in intelligence for a long time. Priorities regularly have changed. Reorganizations regularly have occurred. These are facts. The relatively large change caused by the collapse of communism in Europe is a change in magnitude--not in quality that merits designation as a shift in intelligence paradigm. Confusion about the state of some new world political/economic order among political scientists, politicians, and the public at large does not mean that intelligence officers are confused about their missions.

    The problems of the U.S. Intelligence Community much predate the end of the Soviet Union. They developed massively in the 1980s when the Intelligence Community had plenty of money. They developed because of internal leadership and ethical lapses, not the Cold War or its end. Aldrich Ames says he became a turncoat in April 1985. These problems arose at CIA because senior Agency officers unilaterally, unknown to their policymaking bosses, changed the culture of intelligence from promoting selfless, objective service of national intelligence needs to the satisfying of perceived consumer wants. The ultimate goal, it seems clear to many CIA people, was the bureaucratic advancement of CIA and its senior officials.

    This was a true paradigm shift. The de facto mission of the organization changed radically. Reformers of CIA and the Intelligence Community need to remake the now culturally entrenched paradigm promulgated in the early- and mid-1980s. They should not mistake the changed political/diplomatic environment of the post-Cold War period as a requirement for a systemic change in intelligence. The demise of the Soviet regime is irrelevant to the internal workings of CIA and the conduct of intelligence--though it clearly is important at the levels of targeting, priorities, and funding. Without fundamental internal ethical and cultural reform, all the reconfiguration of organization charts and banter about new targets, new threats, and new agendas will fail to address the key causes of consumers' and Congress' dissatisfaction with the performance of U.S. intelligence.


    90. See Frank Gaffney, Jr., "Apres Woolsey... le deluge?" The Washington Times, January 3, 1995, p. A12, and Melvin A. Goodman, "We Need Two C.I.A.'s," The New York Times, July 21, 1994.

    91. Melvin A. Goodman, "We Need Two C.I.A.'s," The New York Times, July 21, 1994.

    92. Robert M. Gates, "A Leaner, Keener CIA," The Washington Post, January 30, 1995, p. A15.

    93. I graduated from the Special Forces Officer Course and served my active duty tour as executive officer with the 1st Special Forces Group in Okinawa. I spent two months serving in a mildly covert operational training capacity in s outheast Asia in 1972. I did reserve tours with the U.S. Special Operations Command at MacDill Air Force Base in Florida and the Army's 1st Special Operations Command at Fort Bragg. I now am Assistant G-2 with the 352d Civil Affairs Command, Riverdale, Maryland. For over ten years, I served in intelligence assignments.

    94. For a summary of Markowitz's comments at luncheon in February 1995, see Open Source Solutions, OSS Notices, Volume 3, Issue 2, February 28, 1995, pp. 5-6.

    95. In the period 1987-89, I was senior analyst on the staff of the National Intelligence Officer for Warning and was responsible for, among other things, compiling a monthly summary of the results of the warning meetings the regional and some of the functional NIOs held roughly monthly. To do this, I usually attended two or three of the meetings myself and summarized the notes of the others, which usually were complied by the assistant NIOs. These notes had agency attributions for comments and were similar to my own experiences. The dominant analysts came from the all-source agencies--CIA, DIA, and INR.

    96. See Open Source Solutions, OSS Notices, Volume 3, Issue 2, February 28, 1995, p. 2, for example.

    97. For a good treatment, see John W. Williams, "Intelligence Analysis and the Mass Media in the Age of CNN," paper presented at a conference on intelligence analysis sponsored by the Canadian Association for Security and Intelligence Studies, Ottawa, October 28-30, 1994.

    98. This conclusion has attained some currency. See, for example, John Jandora, "Threat Parameters for Operations Other than War," Parameters, Vol. XXV, No. 1, Spring 1995, especially p. 57. Defense Department components are looking intensively at how to improve shortcomings in this arena.

    99. See Timothy C. May, "Crypto Anarchy and Virtual Communities," at tcmay@netcom.com.

    100. Other reasons are that CIA is sensitive about sharing its thoughts on many subjects, and is reluctant to release information on the topics it studies. In these cases, it places low-level classifications on its products, even th ough the topics and and many source materials can be found in good newspapers and magazines.

    101. See Walter Pincus, "White House Labors to Redefine Role of Intelligence Community," The Washington Post, June 13, 1994, p. A8.

    102. Robert M. Gates, "A Leaner, Keener CIA," The Washington Post, January 30, 1995, p. A15.

    103. In Haiti in 1994, for example, Marine and Army units landed at the about the same time, but the Marines withdrew shortly thereafter, turning the job over to the Army. Before the withdrawal, a Marine patrol opened fire on a grou p of yelling Haitians, killing some 10 of them. Better trained Army units, through this writing, did not lose their composure to the point of shooting any Haitians.

    104. Traditionally, CIA has provided little direct support to military commands. CIA representatives to the major commands have tended to be senior personnel, sometimes on retirement tours, who were diplomats as much as conduits of a series of published works. NSA, by contrast, employed more, comparatively junior personnel who worked in offices directly with military analysts. They provide insights, information, and advice on ways to acquire information from the world-wide NSA str ucture. There was also a much more user-friendly telephone connection to NSA, which facilitated communication.

    Moreover, from the CIA perspective, for many years there was little incentive to serve military consumers because: the military were policy implementers, not makers, and thus unimportant; and, DIA for turf reasons stood in the way of better direct CIA/mil itary ties.

    105. I noted this deficiency to EURA managers Ron Miller and John McLaughlin in 1987. They were unreceptive to my concerns. The Office of the Inspector General was similarly unimpressed.

    106. During this period, DI analysts compiled basic data into standard format books, called National Intelligence Surveys, that were readily available for contingencies and contained data on harbor water depth and a host of mundane i nformation of little use to senior policymakers but of critical importance to military planners.

    107. See Molly Moore, "Schwarzkopf: War Intelligence Flawed," The Washington Post, June 13, 1991, p. A1.

    Back to Table of Contents