Writing in Foreign Affairs in 1988, then Deputy Director of Central Intelligence Robert Gates described the impact of intelligence-sharing with the Congress in sweeping and--from the standpoint of the executive--problematic terms:
Most specialists writing about the change in recent years in the balance of power between the executive and Congress on national security policy cite Watergate and Vietnam as primary causes. I believe there was a third principal factor: the obtaining by Congress, in the mid-1970s, of access to intelligence information essentially equal to that of the executive branch.
This situation adds extraordinary stress to the relationship between the CIA and policy agencies. Policymakers' suspicions that the CIA uses intelligence to sabotage selected administration policies are often barely concealed. And more than a few Members of Congress are willing to exploit this situation by their own selective use of intelligence that supports their views. The end result is a strengthening of the Congressional hand in policy debates and a great heightening of tensions between the CIA and the rest of the executive branch . . .
The result of these realities is that the CIA today finds itself in a remarkable position, involuntarily poised nearly equidistant between the executive and legislative branches. The administration knows that the CIA is in no position to withhold much information from the Congress and is extremely sensitive to Congressional demands; the Congress has enormous influence and information, yet remains suspicious and mistrustful. Such a central legislative role with respect to an intelligence service is unique in American history and in the world. And policymakers know it."(1)
This chapter seeks to evaluate how intelligence-sharing with Congress has affected key areas of concern: relations between the two branches, the work of Congress itself, the work of the Intelligence Community, and, finally, the relationship between the Intelligence Community and other parts of the executive branch. In doing so, it addresses the issues raised by Gates as well as some he did not raise.
Of those interviewed for this study, few would take issue with Gates' contention that intelligence has made Congress a smarter, more effective critic of the executive branch, often complicating the lives of policy officials. Many note, however, that intelligence analysis provides support for the policies and proposals of an administration as often as it undermines them. Perhaps even more often, it provides ammunition for both sides of a policy debate. Indeed, it is not unusual for Members to draw different conclusions from the same information. Although, as Gates points out, Members of Congress are not above making selective use of intelligence to support their positions on particular issues, many of those interviewed noted that policymakers suffer the same affliction.
Most of those interviewed for this study seemed to believe that intelligence-sharing has, on the whole, improved relations between the two branches. Many pointed out, for example, that, with or without access to intelligence, it is the role of Congress to criticize. "Even if Congress got no intelligence," said one observer, "they would be seen as meddling. And they would be relying in those circumstances upon information provided by policy agencies that might be slanted or incomplete or what have you. Intelligence, on the other hand, is supposed to provide an unbiased, complete version of the facts. If Congress is going to meddle anyway, isn't it better they at least have the facts?"
Giving Congress the facts, this observer went on to say, actually decreases its propensity to meddle: "While it is true that access to intelligence gives Congress something against which to test the President's actions and policies, it usually ends up giving credence to those actions and policies. While the [political opposition] cannot be expected to defer to the President, at least they will have essentially the same information and can understand what is motivating the President. They may disagree, but they start with the same information base."
In a similar vein, one Congressional staffer thought that having access to intelligence at times had actually discouraged leaks of classified information. "If an agreement can be worked out [with the Intelligence Community]," the staffer noted, "with respect to what can be used in public and it gives both sides of an issue enough to go on, I think it actually discourages people [Members and staff] from resorting to leaks."
Several also noted that, because Congress has access to intelligence, it has sometimes managed to avoid irrational legislative responses to world events--responses that would undoubtedly have created serious diplomatic problems for the incumbent administration. As one current Member put it: "Because the leadership has had immediate access to intelligence reporting, they have sometimes been able to stop the panic and craziness up here."
Others noted a salutary impact on the use of intelligence by policymakers. Because officials are aware that Congress has access to intelligence information, they are more likely to take it into account themselves when formulating a particular policy or proposal. If their policy choice should run counter to the intelligence reporting, they realize that one day they may find themselves defending their choice to their Congressional overseers (who have access to the same intelligence).
From this perspective, Congressional access to intelligence is seen not as a problem for policymakers but rather as a help to them. "Any policymaker worth his salt," said one intelligence official, "should be able to explain to the Congress why he or she is advocating a policy that does not appear supported by the intelligence." Although intelligence may provide ammunition for a particular Senator or Congressman to criticize, the policymaker, having weighed such information in arriving at a chosen policy, should be able to defend his/her position more effectively.
Some policymakers are not so sanguine, however, pointing to instances where they believe intelligence analysis unnecessarily provoked, rather than assuaged, an unruly Congress. They fault analysts for frequently providing intelligence (especially in briefings) that is unduly alarmist because it does not take into account ongoing US actions and/or because it is based on unreliable or incomplete reporting. As a result, Members become needlessly agitated and resort to legislative actions that are unjustified by the circumstances, creating fires that require the involvement of busy policymakers to extinguish.
Many analysts do, indeed, believe they have an obligation to present the "worst case" scenario in their briefings so that Congress will know the outer limits of the downside facing the United States in a given set of circumstances. Members and staff interviewed for this study also expressed a desire to know the "worst case" as the Intelligence Community perceives it (as well as what the Community considers the "most likely" scenario) in order to calibrate their positions on issues.
Whether intelligence-sharing with Congress is seen as ultimately facilitating or impeding relations between the executive and legislative branches, it has, as Gates suggests, clearly complicated them. Still, for all the complications, many see intelligence as having provided, and as continuing to provide, a firmer footing for the dialogue between the branches on national security matters. No one expects a return to the days when Congress deferred to the executive. Given this reality, many say, ways must be found to facilitate collaboration between the branches, rather than allow polarization to grow, so as to ensure that the military and foreign policies of the United States have the support of both branches as well as the American people. They see intelligence-sharing as one means--an important one--for bridging this gap.
Taking the Congress as a whole, however, intelligence analysis (whether in written or verbal form) actually reaches only a small percentage of its Members and bears upon a small proportion of its work. A survey of lawmakers conducted by the CIA's Office of Congressional Affairs in late 1988 not surprisingly found them "overwhelmingly disinterested" in intelligence insofar as the execution of their legislative duties was concerned.(2)
Apart from the intelligence committees, relatively few Congressional staffers have the security clearances needed for access to intelligence, and, for many who hold such clearances, what they see is minimal. As one legislative aide put it, "most staff up here does not have a clue in terms of what is available [in the Intelligence Community]. They see a few documents, but what they see is only the tip of the iceberg. They have no idea what is going on."
Members' lack of interest can be attributed partly to the fact that intelligence does not lend itself to use in a public process. As one SFRC staff member noted: "We [the committee] are part of the public debate. We deal in the realm of the overt--in what actions other governments take and what actions they don't take. While it is still useful to understand what their plans or intent may be, most of what the committee needs to know can be obtained from the New York Times or CNN. And the committee will respond overtly to it."
On occasion, access to intelligence does become important to Members. This occurs, for example, when votes are scheduled on issues that are important to them politically and on which intelligence has a significant bearing--such as, a vote to send US troops into hostilities abroad or a vote to ratify a controversial treaty. Members look to intelligence analysis not only as a source of substantive guidance but also as a way to give them political cover should their vote turn out to be the wrong one.
Intelligence has its greatest impact on Congress, however, through the work of its committees, particularly the committees with national security responsibilities. Most Members and staff of these committees have come to value the analysis provided by intelligence agencies both for its own sake and as a check on the information coming from the policy agencies under their jurisdiction. In this regard, intelligence often provides a "handle" for a committee's oversight activities.
Quite often intelligence will also figure into the consideration of legislation handled by these committees. Such legislative measures include:
Finally, intelligence serves simply to inform Members with respect to world affairs. One Member suggested, in fact, that this was the greatest benefit of intelligence-sharing: "Members do not have to react simplistically [to world events] any longer without the benefit of knowing what the facts are."
Relatively few opportunities exist in the Senate and even fewer in the House of Representatives for Members to educate themselves on international affairs issues. For many Members, service on one of the national security committees is important not because it gives them an opportunity to oversee the operations of the Defense Department or learn the intricacies of the Intelligence Community, but rather because it provides access to information about world affairs they would not otherwise have. In 1992, then Chairman of the Senate Intelligence Committee David L. Boren put it this way:
Some Members request intelligence briefings to educate themselves; others seek out intelligence analysts on a personal basis and establish an ongoing dialogue with them on a topic of interest. The best informed are likely to carry the most weight where international affairs are concerned--with their colleagues, the media, the administration, their constituents, and the public. Their opinions are more apt to be sought and their advice more likely to be heeded. By being informed, they are better able to make a reputation for themselves.
There are, however, pitfalls in all this, even for individual Members. Staking themselves to a position on the basis of intelligence that later proves to be wrong can be embarrassing at best and politically disastrous at worst. (See the discussion concerning the Senate vote on the Persian Gulf resolution in part V below.)
Surprisingly, the intelligence analysts interviewed for this study tended to downplay their influence with Congress. Most seemed to believe that Members usually had their positions staked out and minds made up long before receiving an intelligence briefing. If the briefing lent itself to their views, the Members would take it on board. But few Members were seen by analysts as having changed their positions based on what they heard in an intelligence briefing. Members and staff who were interviewed, on the other hand, generally thought these analysts were selling themselves short.
As a practical matter, it is impossible to quantify the extent to which intelligence information has influenced the oversight or legislative responsibilities of Congressional committees or has affected the actions (or political fortunes) of particular Members. Viewed in the context of the totality of Congress's activities, the information provided by the Intelligence Community could be said to have hardly caused a ripple. But viewed in the context of specific legislative actions--or its influence from time to time on individual Members--intelligence could as easily be seen as having played a key role in determining and shaping Congressional actions and reactions on particular issues.
In 1989 the head of the CIA's Office of Congressional Affairs told Deputy Director Gates that, in his view, Congress had an "unquenchable appetite" for intelligence that, in the long term, could pose serious problems for CIA management.(4)
One of the problems cited was the potential that intelligence support to Congress could overwhelm the available analytical resources to the detriment of consumers in the executive branch. As one executive official noted: "There are not enough resources in the Intelligence Community to provide intelligence support to 535 Members of Congress. They are going to have to draw some lines somewhere."
Most Members of Congress have little appreciation for what these resources are. They ask and they receive. Most perceive a large and faceless bureaucratic machine (funded by taxpayer dollars) with an unlimited capability to churn out analysis and provide briefings on demand.
The reality, of course, is that a relatively small corps of intelligence analysts cover a vast gamut of national security issues for consumers in both the legislature and the executive branch. Together, the analysts comprise a formidable capability. Considering the number of issues they are expected to cover, however, it is clear that this capability actually is spread quite thin. It is not unusual for major areas to be covered by a handful of analysts in a particular agency and for more obscure areas to be covered by only one or two analysts. Although special analytical teams are sometimes put together to deal with looming or ongoing crises, they too usually find themselves stretched to the breaking point, given the demands placed on them in such situations.
For the most part, the analysts interviewed for this study regarded Congress as a legitimate and important recipient of their work. They welcomed the opportunity to support the elected representatives of the people and influence the Congressional role in public policy debates. Indeed, some saw Congress as more open to their influence than policymakers in the executive. Several also noted that, because analysts know their work may someday be scrutinized on the Hill, greater quality control is introduced into it.
Some analysts, however--while they regarded Congress as a legitimate consumer--did not see it as a "serious" one. They thought most Members had neither the time nor the interest to understand or probe what was being briefed to them. They viewed Members' reactions as often shallow and superficial. Some analysts resented having to brief staffers, whom they often saw as "nonplayers." The time and effort required to satisfy Congressional demands took time and effort away from the analysts' first priority: satisfying the pressing needs of decisionmakers in the executive branch.
A number of analysts also pointed to what appeared to be Congress's growing inclination to bring intelligence analysis out into the open, which the analysts saw as imposing additional (and unnecessary) burdens and strains. This occurs either when analysts are required to appear at public hearings of Congressional committees or when intelligence agencies are asked to "sanitize" classified analysis so that it can be made public by a committee or an individual Member (sometimes for a thinly veiled political purpose). Such requests, often couched in such terms as "the American people need to know this," are difficult for intelligence agencies to resist.
These requests also are often difficult for agencies to accommodate. When giving public testimony, analysts frequently are left to make generalized assertions without being able to explain--for security reasons--the intelligence reporting that led to them. Sources and methods must be talked around. If Members press with questions that call for classified responses, analysts fret about appearing rude or uninformed if they decline or limit an answer. Requests to "sanitize" classified analysis for public release do not increase personal stress, but they do involve time and effort that, from the standpoint of an analyst, is unproductive and a diversion from more important work. Going to this extra effort is especially galling when it is seen only as allowing a Senator or Congressman to score a political point against an opponent.
Several analysts interviewed also deplored the growing number of occasions when Congress insists on having the views of the Intelligence Community before the Community is ready to present them. This has happened several times within the last two years, when portions of draft estimates have been leaked to the press or when Congressional committees have otherwise become aware of their existence. If the Community balks, analysts may be required to present their agency's view or even a personal view. Analysts involved in these episodes say they have been highly disruptive of ongoing work. Either the estimative process has to be drastically accelerated to accommodate the Congressional timetable or individual agencies have to go it alone, forcing them to take positions prematurely without the benefit of the give-and-take of the estimates process.
Whatever frustrations analysts may feel about their relationship with Congress pale in comparison with the frustrations felt by many elsewhere in the executive branch. Policymakers often find their policies and initiatives undermined on the Hill by intelligence briefings. Sometimes the briefings are at odds with what the President or administration spokespeople have said. As a result, policymakers face hostile questioning from the press or from Congress itself. Sometimes they are confronted with intelligence they did not know existed or with analytical conclusions they do not know the basis for. Sometimes other governments are annoyed--diplomatic initiatives are disrupted and negotiations are broken off. "Policymakers should see intelligence agencies," said one intelligence official, "as simply purveyors of information, produced by professionals outside the political arena. Instead, some see us as trying to make trouble for them."
"Policymakers are often frustrated," said one intelligence analyst, "when they have to explain to the Hill why they made a decision on other than the basis indicated by the intelligence. While intelligence is always just one vector in the decisionmaking process, and analysts never advance it as anything more than that, it is not perceived that way on the Hill . . . Policymakers may understand all this themselves, but it doesn't make them any happier when they have to face hostile questioning from some Senator or Congressman."
In principle, intelligence agencies acknowledge an obligation to keep pertinent policymakers apprised of the intelligence analysis being shared with Congress in order to give them time to prepare for and deal with the consequences that are likely to follow. In practice, however, many policymakers find that the performance of intelligence agencies falls woefully short on this score. "The Intelligence Community is so anxious to please its oversight committees," said one former executive official, "that it's hell-bent to get the intelligence up there, regardless of whether it's reliable and regardless of whether they've touched base with the rest of the executive branch."
Another executive official was even more strident: "There is a rush to tell Congress everything, often before it's been notified to us. Whatever they ask for, they get . . . Although they would never put it this way, [intelligence agencies] clearly see themselves as working for the Congress rather than the President."
Intelligence officials also acknowledge a problem. Said one: "There is, in fact, a certain imperative about intelligence. Once it's there, it goes. The emphasis these days is on getting it to the Hill as fast as possible when, in fact, it ought to be on making sure the policymaker is brought in on it before it goes. I know there have been many occasions when intelligence has gone to the Hill without policymakers knowing about it, causing them to ask `who are those guys working for, anyway?' It ought not to happen but it does."
"The real problem that results from this [failing to notify what they plan to brief on the Hill]," said one former executive branch official, "is that it isolates them [the intelligence agencies] from the policymakers who then want to close them out from any involvement in the policy process, to keep them from knowing where policy is headed, and so forth. It becomes a `separate camps' mentality, very destructive of the overall relationship between producers and consumers."
(2) Unpublished draft CIA History Staff study.
(3) Quoted in Smist, p. xvii.
(4) Unpublished draft CIA History Staff study.
