Air University Review, September-October 1982

Perspectives on Intelligence

Turning the results of a conference or colloquium into a book is becoming the academic community’s deus ex machina for piling up publishing credits. The work is minimal, the exposure is great, and the low remuneration is more than made up for as the vita lengthens. But for the reader, these instant books are a mixed bag. Often they range widely in quality from one essay to the next, and sometimes they suffer from the effort to pull them all together within a single theme. Nonetheless, they usually contain a gem or two that may make them worth the price of the book.

The two books discussed here tend to tally on the plus side of this equation. Many of the essays are excellent, and a few are truly first-rate. Both books present at least one other side of each issue, either with discussants following each major essay, as in Intelligence Requirements for the 1980’s: Covert Action,* or with several essays on each subject, as in Intelligence Policy and National Security.** Between the two they raise some of the fundamental continuing dilemmas of intelligence and provide even a well-informed intelligence observer with food for thought.

*Ray Godson, editor, Intelligence Requirements for the 1980's: Covert Action (Washington: National Strategy Information Center, Inc., $7.50), 236 pages.

**Robert L. Pfaltzgraff, Jr., Uri Ra'anan, and Warren Milberg, editors, Intelligence Policy and National Security (Hamden, Connecticut: Archon Books, 1981, $32.50), 318 pages.

Ten years ago some of the authors (e.g., Ted Shackley, Hugh Tovar, and Don Purcell) would probably have been distressed to see their names associated with the Central Intelligence Agency (CIA) in print, much less share their firsthand knowledge of covert and clandestine action in rather forthright articles as they do in Covert Action. However, times change and so do attitudes. What does not seem to change are the dilemmas: the ingredients of success and failure in covert actions; the gain or loss from dialogue with the public; the proper balance between traditional human agents and high-technology collection devices; how to improve analysis; centralization versus decentralization of intelligence; the relationship between intelligence and policy—to name a few.

In Covert Action, Hugh Tovar, a thirty-year CIA veteran and Chief of the Covert Action staff, made the following statement about the Bay of Pigs failure:

Was it an intelligence failure? Undoubtedly, and in the grandest sense of the term. It is feckless to argue about guerrilla uprisings or the legion’s survival capabilities. They were ancillary considerations at best. The real questions developed on the idea that Castro was so shallowly rooted in Cuba that he could be shaken by psychological pressures, as Arbenz had been in Guatemala, and then ousted by a comparative handful of troops. It is easy to visualize the sequence. The concept once conceived, probably at senior level, is tested on underlings whose instincts and training guarantee an immediate can-do response. Momentum develops rapidly. Conceptualizing is superseded by planning. Policy emerges in high secrecy and, before anyone realizes it, the project is a living, pulsating, snorting entity with a dynamic all its own. Scrutiny by a disinterested body is all but out of the question under such circumstances. The people at the top get the answers they want. Once astride the tiger, options narrow and will becomes a factor in survival. (p. 198)

Perhaps no single ingredient could alone be fatal: neither the spurious assumptions that buttressed the plan, nor the compartmented secrecy which precluded its objective assessment, nor the unwillingness or inability to stop it once in train. But their cumulative effect spelled failure.

Ted Shackley lays out the ingredients for a successful counterinsurgency operation in his very fine prescriptive paper, "The Uses of Paramilitary Covert Action in the 1980’s," in the same book. His detailed steps to be accomplished in the cadre, incipient, operational, covert war, and conventional war phases provide a valuable systematic road map for making sense out of a kind of warfare that seems on the surface to be amorphous wanderings.

Both Tovar and Shackley, without ever discussing it, make a very persuasive case for sure controls over covert action. Tovar makes a case because so much can go wrong so easily when only a few people are in on the action and they all have something at stake in going forward; Shackley because covert actions can be complex and because they are usually inextricable from overt foreign policy initiatives and have such a high potential for undermining them.

For the past half-dozen years, the intelligence community has carefully nurtured a valuable two-way relationship with the two intelligence overseeing committees of the Congress, the Senate Select Committee on Intelligence and the House Permanent Select Committee on Intelligence. Every covert action was briefed to these committees in detail. Their scrutiny— usually before the fact—guaranteed plans were well-thought-out, persuasive in their logic, and politically acceptable. Their concurrence guaranteed that should something subsequently go wrong, the intelligence community would have knowledgeable friends on the Hill who could ensure that congressional comment would be balanced and fair, and the public could be assured that their representatives supported what had been attempted. Without question this process provided a high yet reasonable standard for covert action. Admiral Stansfield Turner, Director of Central Intelligence during the Carter administration, has stated that this clearance process never prevented the United States from pursuing a covert action that was deemed necessary. The requirement to inform the Congress before the covert action is carried out no longer exists. According to recent comments in the press by committee members, the relationship between the intelligence community and the Congress is beginning to close down, with intelligence officials being less and less forthcoming. While this may make intelligence work easier and may protect against leaks, it also carries the real dangers that Hugh Tovar describes, dangers far more serious in the long run than an occasional leak.

A closed intelligence organization also will affect the quality of analysis turned out. Adda Bozeman in the same book, in an essay titled "Covert Action and Foreign Policy," says, ". . . the most recent U.S. policy and intelligence failures have ensued from these defects in strategic thought and vision." (Here she is describing the misperceptions she sees of the real identities of foreign states and other international actors.) "The responsibility here rests primarily with the academic community, not the Intelligence Community and not with the State Department; for it is after all the former which is traditionally charged with providing educational and professional guidance to the latter." (p. 16)

The article by Ithiel deSola Pool, "Approaches to Intelligence and Social Science" in Intelligence Policy and National Security, suggests that many in society are engaged in the same kind of interpretive work as the intelligence analysts.

. . . social scientists, newsmen, diplomats, and intelligence personnel all are, to a large extent, doing the same kind of work. They are all "deciphering" a world in which information is deliberately concealed, in which there are problems of interpretation, in which prediction is difficult, and concerning which they are supposed to be more knowledgeable than the persons to whom they report. (p. 37)

Exchanging ideas, testing assumptions, and engaging in broad-based dialogue are vital to an intellectually vigorous analytic organization, to say nothing of the importance to the analyst of recognition of his work.

More than just permitting dialogue with other academics, the publication of declassified intelligence helps raise the level of public debate by providing unbiased factual data on current issues. Richard Pipes, also in Intelligence Policy, says,

We have superb information in our intelligence community, as can be seen from the publications which the CIA now releases to scholars on such subjects as the Soviet leadership . . . The studies of the Soviet economy produced by the CIA are of high quality as well. (pp. 74-75)

Unfortunately, in the past year a conscious policy decision was made to discontinue publishing unclassified analysis as well as essentially prohibiting analysts from substantive dialogue with anyone outside the intelligence community. This is an extreme reaction to what was believed to be a burdensome incursion on analysts’ time. In fact, it effectively isolates intelligence analysts, encourages a narrowing of their perspectives, and cuts them off from the continually invigorating interaction with their colleagues outside the community, a process through which the quality of analysis is strengthened. As Richard Betts comments in "American Strategic Intelligence: Politics, Priorities, and Direction," perhaps the best paper in Intelligence Policy, "the internal tension between security that protects collection sources and dissemination that improves finished analyses will probably continue." (p. 255) For the moment security is ahead.

Intelligence analysis can also suffer from how analysts are organized and from the character of their relationship with policymakers. William E. Colby, in "Deception and Surprise: Problems of Analysts and Analysis" in Intelligence Policy, says, "intelligence analysis must be organized geographically rather than functionally." (p. 95) Speaking of the same recommendation mode in 1949, he says,

. . . economists, the current political analysts, and the military experts were comfortably settled into separate bureaucratic islands, submitting their analyses to wise generalists to integrate into overall assessments.

The effect was almost uniformly bad. The generalists approached the problem in categories, attaching supplemental economic essays to political estimates and compromising force projections after adversary proceedings between hawks and doves. Emphasis rapidly focused on current political event reporting at the expense of deeper integrative research. (p. 95)

The CIA’s Directorate of Intelligence has just reorganized itself geographically. There are good arguments for retaining its functional organization, the reinforcement and challenge of working with colleagues in the same discipline, for example, but whenever there was a crisis such as Iran or Afghanistan, an analytic task force was formed combining all the diverse disciplines needed to understand and react to the problem. In the long run it seems to make more sense to look at all problems in this interdisciplinary way.

Richard Betts, in the same article referred to earlier, comments on the generally poor communication between analyst and policymaker.

. . . the policymakers who consume intelligence seldom tried seriously to define what they need; indeed many had no idea. "Therefore, intelligence requirements reflect what intelligence managers think the consumers need, and equally important, what they think their organisations can produce." (p. 252)

Unquestionably, most policymakers never learn how to get the most out of intelligence because they either will not or feel they cannot take the time to give meaningful direction to the collection and analysis process. As a result, the most they can hope for is that some prescient intelligence manager guesses right or is quick on his feet.

Another asks how well served is the policy-maker despite his traditional failure to take the helm. In "United States Intelligence Activities: The Role of Congress," Thomas K. Latimer comes to the conclusion that the record is uneven. In citing a study of the analysis of the 1973 oil embargo conducted by the Senate Select Committee on Intelligence Subcommittee on Collection, Production and Quality, he says:

One of the key findings. . . was that certain public sources had done as good or a better job of analysing major issues involved. . . than had the intelligence community. The study also concluded that there had been ample data available to intelligence analysts. They simply failed to analyse adequately that data. (p. 279)

He goes on to cite warning analysis as being a continuing problem but acknowledges the assignment of a senior intelligence officer by the Director of Central Intelligence in 1979 as a "major first step in improving our nation’s warning intelligence." He then offers the lack of adequate warning of the Iranian revolution as a failure of that system and the ample warning of China’s invasion of Vietnam in February 1979 as one of its successes.

However, what the average observer usually fails to concede is that with the diligent search for clues some kinds of events should always be predictable while others will never be. In both cases the warning is only as good as its use by the policymaker. The internal weaknesses of the Shah’s government, for example, were well known to the United States long before his fall from power. Given the constraints on how much spying one can do in a friendly country without rupturing that friendship, we were surprisingly well informed of conditions there. What could not have been predicted then nor could be predicted now in another country is the decision of the ruler not to use the levers of power and control at his disposal when he had always used them in the past under similar circumstances.

Similarly, when political analysts and computers fail to predict the outcome of the U.S. Presidential election even though all the facts are available, it is hardly an intelligence failure to miss the outcome of a foreign election. What is in a person’s head, be he a voter or a shah, can only be guessed at, and that guess will only have a chance of being right when present behavior is consistent with past behavior. The stress created by the kinds of events we are primarily concerned with when we talk of warning does not guarantee high confidence that behavior will be consistent.

Understanding what intelligence agencies can and cannot do is really what these two books are about. The serious observer who wants a balanced look at those capabilities will be well served by reading both.

Captain George E. Thibault, USN (B.S., Tufts University; MA., Boston University; M.S., George Washington University), is Chairman, Department of Military Strategy, National War College. He was formerly Special Assistant to the Director of Central Intelligence, Central Intelligence Agency. Captain Thibault has published articles in Foreign Affairs, New York Times Magazine, Naval War College Review, Washington Post, etc. He is a Distinguished Graduate of the Naval War College.

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

Air & Space Power Home Page | Feedback? Email the Editor