The CIA's reputation, never very high, has significantly declined despite a huge infusion of resources. But its recent roguish behavior is not typical. Indeed, quite the opposite. When the U.S. government embarks on a major effort, its employees, including the CIA, usually salute. Top officials want very much to please their big bosses, and they usually find ways to match the intelligence to the policy proclivities. And when things go wrong, their political masters inevitably turn against the agency. None of this should obscure the general capability of CIA analysts and their dedication to preserving the integrity of the intelligence process.

The CIA's analytical product faces some serious hurdles. First, its importance is inflated in the public eye, even as its reports go mostly unread by top policymakers. And second, if there are intelligence errors on big issues, it casts a pall over the whole intelligence analysis effort. The agency's failure on Iraq's WMD, and its lack of understanding of the potemkin and dysfunctional nature of the Saddam regime, have had a serious impact on American credibility as well as attitudes toward the agency.

It is not the CIA's analytical directorate but the operations directorate--the field-agent effort and the one so important to dealing with terrorism--that has been getting most of the flak, and the one whose career senior leaders have been summarily fired by Mr. Goss. They have been variously accused of delinquencies ranging from gross incompetence to extreme caution--no "risk-takers there." That may be one reason the Defense Department is taking over many of the CIA's activities. This part of the CIA has had trying times, from the Church Committee in the 1970s to their involvement in the Iran-Contra affair to their difficulties in the early Clinton years. They have gone up and down in manpower. In the last few years they have been the beneficiaries of a significant infusion of resources, but if the recent firings are any indication, the enhancements have not yet paid off.

It is difficult for the outsider to draw any conclusion about the effectiveness of the CIA's clandestine efforts. Congressional oversight cannot be given much credence. The agency claims that only its failures become public, that there are many unheralded and unknown achievements, particularly in preventing terrorist incidents. One thing is clear: Greater risk-taking and reorganization of the whole intelligence community does not necessarily produce more intelligence on extremely difficult targets like North Korea. The challenges in acquiring such information are enormous. The Clandestine Service is not likely to be "transformed" by simply changing the organization's wiring diagram and the top people.

The fixes to these "problems" within the professional bureaucracy and between it and the political masters have mainly taken three forms. First, there are times when the political masters have grown so exasperated or so eager to place blame elsewhere that they have fired senior career people. On many occasions, these firings are justified, as with some of the recent moves in the upper reaches of the CIA. But it is hard for senior professionals to swallow righteous firings when those whose mistakes are seen as even more egregious, like several high-level Pentagon officials, retain their positions and are praised for performance. While bloodletting is needed at times to get rid of particular problems, it does not solve systemic ones.

A second tried-and-true formula has been to reorganize the bureaucracy, redirect supporting lines, and move boxes for whole agencies from one place to another. Most times these reorganizations are a waste of time and money. Moving bureaucrats does not necessarily change the culture or the performance. And wherever they are moved, Congress is almost always unmoved. Organizational changes that occur within administrations are often negated by unchanged and parochial congressional relationships--as demonstrated by the utter disregard with which Congress has treated the reforms to its own operations that were recommended by the 9/11 Commission. It is argued that the recent attempts to improve counter-terrorism efforts have proven successful. But a serious evaluation will take several more years.

Third, the final fix resorted to by political masters has been to rail against or ignore the bureaucracy. That is the approach discussed in this article. It has proven very costly to the nation. As we have noted, it is clear in retrospect that U.S. policy in Iraq could have benefited at almost every turn from the advice and information of the professional bureaucracy. From the president on down, political appointees must realize that, yes, bureaucracy is sluggish and out of touch, resistant to change, lacking in imagination, and often wrong, but career professionals can save them from disastrous mistakes. By virtue of having worked seriously on these problems, countries and cultures for years, no one is better than they are at spotting obstacles and landmines. No policy can be successful that fails to anticipate these hurdles. Political masters ignore this expertise at their peril.

Every administration must be zealous in the pursuit of important national security goals. Political appointees, however, must not confuse an absence of candor with loyalty in their career subordinates. Career professionals are being most loyal when they are being candid with their bosses about situations and when they press for a serious examination of policy. They have a sense of American national interests that tries to transcend an individual administration and should be fully and fairly examined before administrations change course.

This will not stop all leaking to the press. But the kind of draconian measures that would be needed to prevent all significant leaks would be bad for the policymaking process and, it could well be argued, worse for the democratic process. Sometimes the people know what's going on or what the choices are only because of these leaks. Ditto for Congress.

Thus, probably the most effective way to address the bureaucracy's weaknesses and take advantage of its strengths is not bloodletting, or reorganization, or ignoring and condemning the bureaucrats. It is having a sensible attitude at the top, starting with presidents and secretaries. Beneath them, the key to making the system work is to have appointees at the assistant secretary level who are responsive to the policy imperatives of the political leadership and willing to engage and draw on the skills of their career experts. The assistant secretary jobs are where the proverbial rubber meets the road.

At this time, the country has a particular need for preserving candor in the departments and a variety of viewpoints from different agencies, particularly in the intelligence world. The War on Terror has given the U.S. government enormous power to do what it thinks is necessary to protect this country. Today, everything done in the name of that effort seems to be acceptable to, even demanded by, the public. The mechanisms for self-examination or self-correction either do not work or are diminishing. Congress provides little serious oversight. The media, particularly on television, are disadvantaged by the secrecy of subject matter and often by their lack of interest. In this post-9/11 environment, the permanent bureaucracy is the last line of defense in possibly subjecting critical policy considerations to the most informed scrutiny. If the administration does not want to consult them seriously, they must themselves persist in trying to make their views known to their leaders in constructive ways that go beyond anonymous leaks to the Washington Post.