The importance of interchange
Editor's Note: The following is the keynote address to the Symposium for Teaching Intelligence which was sponsored on 1 and 2 October 1993 by CIA's Center for the Study of Intelligence.
Last summer, a friend of mine was driving on Cape Cod just after a severe storm. A state trooper waved him down. Up ahead a large tree had fallen across the road. He worried that he had missed a "road closed" sign. The trooper bent down, rested his elbows on the window, and said, "If a tree falls when no one is around, does it make a sound?" My friend's comment: "Only in Wellfleet!"
In a sense, we are dealing here with a different version of that old metaphysical conundrum. If scholars and journalists do not know what intelligence agencies have done, can they be said to have done anything? More practically, if scholars and journalists do not tell citizens what intelligence agencies have done for them in the past, why should the citizens expect intelligence agencies to be useful in the future? And the reality is that most scholars and serious journalists do not know enough about the real history of the Intelligence Community to explain to citizens why Congress should drop money into that black box.
To say this is not to disregard the extraordinary accomplishments of those few scholars and journalists who have sought to penetrate what Walter Laqueur called the "world of secrets." A generation ago, the literature on intelligence was at or below the level of literature on business before the arrival of modern business history and business education. Libraries had a few reference works such as R. W. Rowan's Story of Secret Service. These works had solidity and reliability comparable to H. G. Wells's History of the World or Elbert Hubbard's Little Journeys to the Homes of the Great. In most of the literature, fact--if any--was indistinguishable from fiction. Some of the best items actually were works of fiction, as, for example, Somerset Maugham's Ashenden or the British Agent.
During the past 25 years, this condition has changed completely. The amount of serious research on intelligence has been such as to produce--and warrant--several new learned journals. Intelligence and National Security, edited by Christopher Andrew and Michael Handel, is one primarily for historians. The International Journal of Intelligence and CounterIntelligence, edited by F. Reese Brown, is more for political scientists and intelligence professionals. The Joint Military Intelligence College's Defense Intelligence Journal is a new entry. The contents pages of these journals and their often-crowded review sections provide a register of research easily comparable to research in any of the other major subfields of history or political science.
The revolution in intelligence scholarship, however, has been largely self-contained. It has not so far had much effect outside its own inner circle. Writing on intelligence rarely appears in other learned journals--even Diplomatic History and World Politics.
A search through the 1992 citation indexes for social sciences and arts and humanities turns up relatively few entries for titles that, for intelligence specialists, are standard works. For Christopher Andrew's Her Majesty's Secret Service, for example, there are only three citations; for Harry Hinsley's British Intelligence in the Second World War, the same number; for Loch Johnson's America's Secret Power: The CIA in a Democratic Society, four; for John Ranelagh's The Agency, five; for Gregory Treverton's Covert Action, two; for David Kahn's The Codebreakers and Thomas Powers's The Man Who Knew the Secrets, none.
These numbers are all the more disappointing when one notes, as an example, that the three citations of Andrew's Her Majesty's Secret Service are in articles by intelligence specialists--Loch Johnson, Wesley Wark, and Andrew himself, or, as another example, that almost all citations to Johnson's book, Ranelagh's, and Treverton's are in law journal articles concerned with intelligence oversight. As against the three citations of Hinsley's British Intelligence in the Second World War, there are 12 for his 30-year-old book, Power and the Pursuit of Peace. As against three citations for my own collection, Knowing One's Enemies: Intelligence Assessment Before the Two World Wars, there are 10 for my 20-year-old essay collection, "Lessons" of the Past: The Use and Misuse of History in American Foreign Policy.
There is clear need to make research on intelligence better known and better understood outside the company of intelligence specialists. The open question is how to do it. At the risk of starting the conference with an off-key chord instead of a keynote, I will suggest some possible approaches. I will speak first of scholars and second of the Intelligence Community, but it will be my main argument that the precondition for further progress is long-term collaboration between scholars and members of the Intelligence Community, partly along lines suggested by this conference's format.
Scholars who work on the history, politics, or methodology of intelligence agencies need to address more often and more explicitly questions as to the influence of intelligence on choices made by governments and, more broadly, on currents in international politics and the world economy.
Two examples suggest models. One is John Lewis Gaddis's essay, "Intelligence, Espionage, and Cold War History," originally published in Diplomatic History in 1989, then republished, with some revisions, in his 1992 book, The United States and the End of the Cold War. Drawing largely on work by intelligence history specialists, Gaddis identifies junctures at which clandestinely collected intelligence might have affected choices by the American and Soviet Governments. He concludes that currently available evidence warrants few judgments. He makes the point, however, that the question ought not to be ignored. How did Philby et al influence Moscow's estimates and actions? How did Penkovsky et al influence London's and Washington's?
Apart from October 1962, in the Cuban missile crisis, how did imagery from U-2s and satellites affect Western decisions and positions? Apart from the incident on the eve of the Yom Kippur war described in Henry Kissinger's memoirs, when did raw communications intercepts lead policymakers to conclusions different from those of the experts advising them? (Were there any other occasions when, as a result, the policymaker's conclusion was the right one? The scarcity of other such anecdotes suggests not.) Gaddis's essay identifies questions about intelligence to be posed in all courses about modern international relations. He helps thus to set agenda for scholars doing specialized research on intelligence.
The second example is an essay by Thomas Powers in The New York Review of Books for 13 May 1993. As is typical for The New York Review, the essay makes only token reference to most of the 16 books ostensibly being reviewed. Powers writes instead about the question of whether the United States got its money's worth from the half trillion dollars that he guesses to have been spent on intelligence between 1945 and 1991. He recites story after story reminiscent of those that used to be publicized by Senator William Proxmire as examples of foolish government waste. (He begins by describing a powerful motorboat, manned by anti-Communist Cubans, that cruised Lake Tanganyika during the early 1960s at the expense of the United States. To what purpose--he muses.)
Like Gaddis, Powers deplores the shortage of evidence on particular episodes. His provisional conclusion is, however, at odds with his anecdotes, for he predicts that future historians "are probably going to find that the happy outcome of the cold war depended heavily on the CIA's spies, the NRO's satellites, and the NSA's monitoring of communications." He expresses doubt that this will be so because of the effect of intelligence on particular decisions. "Many small victories and defeats in the cold war have explanations of that sort," Powers writes. "But what American intelligence contributed to the outcome was something quite different--the confidence that we knew what the Soviets were up to, and could afford to contain their forays while waiting for the deep change in attitude which George Kennan had predicted back in 1947.... Intelligence on the grand scale was necessary to the policy of deterrence...; it was the hard-won, detailed knowledge, held by both sides, of what nuclear weapons could do, how many there were, what they were pointed at, and the certainty that they would penetrate any defense."
One can accept Powers's conclusions or question them. What matters is that they address issues necessarily central to survey courses on contemporary history or international relations.
For broad and lasting effect on teaching about international affairs, there will have to be many studies following up or adding to these essays by Gaddis and Powers. In addition to continued scholarly writing on intelligence per se, we need articles and books forcing all serious teachers to turn their students' attention time and time again to the effects and influence of intelligence.
Some historical works already do this for the late 1930s. Building on the work of Wesley Wark, D. C. Watt includes secret intelligence and intelligence analysis as factors in his account of the immediate origins of the European war of 1939, How War Came. In Threshold of War: Franklin D. Roosevelt and American Entry into World War II, Waldo Heinrichs deals with MAGIC and other intelligence as an integral part of the evolution of United States policies in 1941. But few works dealing with later events weave intelligence even into their narratives, let alone into conclusions serviceable for survey courses.
It has been 20 years since silence about ULTRA was broken. One can scarcely count the occasions when someone has since declared that the history of World War II would have to be rewritten. In fact, it has been rewritten only episodically and that mostly via articles in specialized intelligence journals. And it is hard to point to episodes of the Cold War with standard versions that take account of intelligence as Watt and Heinrichs do.
Exceptions that might spring to mind include the Iranian and Guatemalan coups of the 1950s, the "missile gap," the U-2 and Bay of Pigs affairs, and the Iran-Contra imbroglio. In fact, most writings about these episodes illustrate my basic point, for they tell intelligence stories, not stories that illustrate interplay between intelligence and policy. If intelligence is to become an important integral element in teaching about international affairs, scholars need to produce scores of monographic studies detailing this interplay.
But here we come to the Intelligence Community, for other historians cannot do what Watt and Heinrichs have done unless they have similar material with which to work. I am confident that the history of World War II will eventually be rewritten, integrating analysis of the impacts of ULTRA, the Double Cross system, special operations, and the rest, for many documents are open. Together with official histories, they provide a basis for cross-questioning the testimony in autobiographies and trial records. Eventually, the new details will be pieced together with the old, altering the panorama.
What will be the case for the Cold War depends on the extent to which scholars gain access to comparable documentation and guidance. The Director of Central Intelligence is committed to making historical materials accessible. Commitment from the person at the top is not always a guarantee of action by an organization. (My university is a confederation of comparatively autonomous faculties. The faculties do not keep the same calendars. One president of the university made it his goal to get a unified calendar. He failed. It is still often the case that, if I let a Business School student into a spring term course, I have to do so knowing that the student will graduate before I give my final exam.) The almost complete opening of OSS records and the recent release of documents on the Cuban missile crisis and on estimates of the Soviet Union indicate that at least some parts of the Intelligence Community are prepared and equipped to make the Director's wish a reality.
Even if there is no letup in the current momentum, there will always be less release of documentation than scholars desire. If there is any release at all, it will always be more than some intelligence officers think prudent. I myself favor a common rule for all records relating to national security, including all intelligence files--signal intelligence included: namely, a fixed 25- or 30-year rule. On the recent side of the line, the presumption should be nondisclosure with a basis for exceptions much more restrictive and more rarely used than the current Freedom of Information Act. After the fixed date, the presumption should be complete public access, with exceptions being also rare and made only as a result of agency petitions approved by an independent board, appointed by the President with advice and consent of the Senate.
This formula may be unrealistic--as unacceptable to habitual users of the FOIA as to the guardians of secrets. But scholars on the one hand and insiders concerned with accessibility on the other (legislators and legislative staffers included) ought with some sense of urgency to work out regular procedures that can substitute for ad hoc decisions and that can continue over the long term.
Access to documents is only part of what scholars need from the Intelligence Community. An almost equally important need is that specifically addressed through the format of this conference. Scholars need orientation to the world from which the documents emerge so they can understand and evaluate the documents, make informed guesses about the extent to which the essential record is complete or incomplete, and cross-question memoirs and testimony.
Scholars dealing with modern international relations need to try to understand tribes in governments much as anthropologists try to understand tribal communities elsewhere.
Let me give two illustrations of how scholars can go wrong if they do not understand tribal mores. One is at the expense of another scholar. The second is at my own expense. The first has to do with the tribes that wear US military uniforms, the second with those in the US State Department.
Some years ago, an eminent and exceedingly able scholar presented at the Woodrow Wilson Center a paper dealing with postwar planning during World War II by the Joint Chiefs of Staff. The paper made much of some memoranda issuing from a JCS committee composed of very senior officers. Gen. Andrew Goodpaster commented on the paper. Though with characteristic tact, General Goodpaster made the point that those particular senior officers were not ones in whose judgment the chiefs of staff placed great trust. "If you are looking for the memoranda to which General Marshall paid attention," General Goodpaster said, "find those with the initials `GAL' for Col. George A. Lincoln. That was the person Marshall respected."
The basic point is one that any academic should appreciate. We all know that seniority, eminence, and such are not necessarily good indicators of who carries weight in department meetings. To thread one's way through the immense volume of papers in any modern government agency, scholars need the kind of guidance that General Goodpaster offered--about whose initials mattered to whom. That information has to come from people who were there.
The second illustration--at my expense--concerns an exceedingly able graduate student who went through the records from our Embassy in Tehran that had been reconstructed from shredded fragments by those who occupied it during the Carter administration hostage crisis. The student reported with some excitement having found numbers of State Department reports belying conventional wisdom on American blindness to the weaknesses of the Shah. Long before the Shah's fall, these reports were calling attention to weaknesses in his regime and to the growing power of the mullahs.
I disappointed the student by questioning whether he really had a basis for challenging the conventional belief. His new documents were mostly airgrams, not cables, and were of relatively low classification. From my own experiences as a consultant, chiefly in the Defense Department, I had arrived at an anthropological rule of thumb, namely that documents aimed for notice at the policy level would nearly always be made to appear very urgent and very restricted in circulation. (This applied equally to State Department and intelligence traffic for Southeast Asia which I had seen in the 1960s. And I remember being told by a very experienced American Ambassador that an airgram classified as Confidential was not expected to be read by anyone. It was something for the file, just in case the Embassy's backside needed someday to be covered.)
Very recently, however, a Harvard colleague with Bush administration experience in the State Department and the White House told me that I was too categorical in what I said to the student. Because of the cumbersomeness of precautions against hostile communications interception, this colleague said, it has become customary for some State Department posts to use airgrams rather than cables for their most sensitive reports. Also, this colleague noted that the use of classification or of marks such as "EXDIS" or "NODIS" as means of getting attention has come to vary from bureau to bureau.
These two examples illustrate the need for someone doing research on foreign policy or international relations to use the approaches not only of the anthropologist but also of the historically oriented anthropologist. We have to pay attention to organizational cultures and to changes in those cultures over time.
This is not easy to do in any circumstances. It is particularly hard for organizational cultures in the Intelligence Community. For the Pentagon or the State Department or their counterparts abroad, the inquisitive scholar can at least start with some help from organization manuals, appropriations hearings or their equivalents, and public records giving some indication of fast- and slow-track career paths. For CIA, the DIA, the NRO, or NSA, all such materials are classified.
Until the extensive hearings on Robert Gates's confirmation as DCI, nothing in the public record gave outside scholars any insight into the cultures producing intelligence estimates. And, though we now have some other examples of direct testimony, an astonishingly large quantity of indirect testimony fed through reporters, and a number of memoirs, we have to use such evidence with extraordinary caution. Among other things, we have to remember that Intelligence Community compartmentation means not only that the left hand may not know what the right hand has done but also that the left thumb may not know what has been done by the left forefinger. We also have to remember that many intelligence officers are trained to be skillful liars.
If intelligence is to become a major standard component in teaching about international relations, the Intelligence Community will need both to make available documentary records and to encourage the direct personal interchange that will enable scholars to understand and use those records. This conference offers a model of how to promote such interchange. The participating scholars will benefit, and the benefits will spread to other scholars and to students. I hope that members of the Intelligence Community will perceive the long-term usefulness of making the scholarly community better able to understand what intelligence officers do and how what they do relates to what others in government do.
For it is in the interest of the Intelligence Community to have its work dealt with as part of the warp and woof of international relations even if, as is sure to be the case, the descriptions are sometimes unflattering or critical or worse. In England, it is said that students who graduate from the better public schools do not necessarily leave knowing Greek and Latin, but they leave profoundly convinced of the existence of these languages. It will be an advance if Americans who study international relations emerge convinced of the existence of their intelligence agencies. And, as one last exhortation--of sorts, let me quote the wise observation of Mark Twain: "History never comes out right. Historians exist to remedy that defect."