The 9/11 Commission Report dominates the agenda these days on Capitol Hill and the airwaves in the public debate over the restructuring of the intelligence community (IC). Commission members are now busily arguing for a new National Intelligence Director (NID) coupled with new intelligence fusion centers--for counter-terrorism and proliferation-- which they argue go a long way to fix what ails American intelligence.

Aside from the media splash, it is hard to discern how the commission's recommendations would cut to the heart of our intelligence problems, and not just those associated with the 9/11 tragedy. The portfolio of responsibilities for the proposed NID are little more than a rehash of the responsibilities currently exercised by the Director of Central Intelligence (DCI). And new intelligence centers are not needed because they already exist in the bowels of the CIA. The IC needs wider and deeper horizontal sharing and webbing of intelligence, with a special emphasis on getting the FBI to contribute to the information pool, as befitting the information-technology age--not the creation of a ponderous new layer of bureaucracy reminiscent of the Cold War.

The 9/11 Commission recommends structural, or "hardware", additions when better management and business practices, or "software", in core intelligence functions inside the existing IC--especially at the CIA--are most needed. Although it has been eclipsed by media attention on the commission report, the Senate Select Committee on Intelligence (SSCI) report on intelligence failings in the run up to the Iraq War is in many respects a more perceptive, penetrating and diagnostic study of the CIA's faults.

That the CIA had no spies collecting information on WMD inside Iraq after UN weapons inspectors departed Iraq in 1998 is one of the most damning findings of the SSCI report. And the SSCI, judging from my own 17-year career at the CIA, has precisely diagnosed the root cause of the CIA's failure: "a broken corporate culture and poor management", which "will not be solved by additional funding and personnel."

Acting DCI John McLaughlin, at a press conference defending the Agency's performance, asked, "How do you measure, how do you balance a hundred successes against one failure?" McLaughlin's defense is hardly a credible one in light of the CIA's failures in 9/11 and Iraq--arguably the gravest intelligence debacles in the CIA's fifty-year history--coming in a short span of about two years. These intelligence failures and the SSCI's investigation clearly show the American public that the CIA is broken, no matter how deep into denial the Agency's senior management sinks. And reforms cannot be postponed until after the United States destroys Al-Qaeda and stabilizes the security environments in Iraq and Afghanistan.

The CIA may well score some intelligence successes against lesser adversaries, but it has a history of losing the big games. The CIA failed to recruit spies in the Kremlin during the entire Cold War, in India to warn of New Delhi's 1998 nuclear weapons tests, and early in the evolution of Libya's nuclear weapons program that Tripoli now has surrendered. CIA officials publicly hail the breaking-up of the A.Q. Khan nuclear supply ring as an intelligence coup, but one wonders if the CIA has the human intelligence assets needed to find out whether Khan's operations were officially sanctioned and supported by Pakistan's President Pervez Musharaf and his intelligence services in order to subsidize Islamabad's own nuclear weapons program. In light of the SSCI's assessment that the CIA lacks its own human intelligence assets, one suspects that the CIA is too heavily dependent on its liaison with Pakistani intelligence to find out if Khan was working for, rather than against, Islamabad's strategic interests.

So what were the CIA's case officers in the Directorate of Operations (DO)--charged with recruiting human spies inside regimes and terrorist groups plotting against U.S. interests--doing to redress the lack of human intelligence (HUMINT) in Saddam's Iraq? The answer is, not much. The committee discovered that, "When UN inspectors departed Iraq, the placement of HUMINT agents and the development of unilateral sources inside Iraq were not top priorities for the Intelligence Community." Apparently, the DO was "whistling pass the graveyard" and hoping against the odds that a crisis would not yet again emerge in Iraq to expose the gaping hole in American human intelligence, a hole that had never been patched after the Gulf War.

Policymakers and military commanders during the Gulf War--from Secretary of State James Baker, to Deputy National Security Advisor Robert Gates, to the commander of coalition forces Norman Schwarzkopf--have lamented in their memoirs that the CIA had failed to produce human intelligence of sufficient quality to illuminate Saddam's perception and decision-making process as he prepared to invade Kuwait and to wage war against coalition forces. The DO also failed to penetrate the mammoth nuclear and biological weapons programs that Saddam was then hiding. Given this legacy of intelligence weakness, it is no mystery why the Bush Administration was worried that the CIA in 2002 was again grossly underestimating Saddam's WMD programs.

The DO's failure in Iraq stems in no small measure from its heavy institutional and cultural bias toward recruiting and running agents in place. Case officers are trained and rewarded for spotting, assessing and recruiting agents, even if their agents produce third-rate intelligence of marginal relevance to U.S. national security interests. do officers are loath to admit it, but spies with top quality intelligence, by and large, are not seduced by case officers to commit treason. A close reading of memoirs by former and distinguished CIA case officers of all generations and who worked in a variety of regions shows that the best spies are not seduced, but volunteer their services to the United States.

Looking ahead, the DO needs to concentrate on vetting and debriefing "walk-ins" who volunteer intelligence to CIA case officers overseas while encouraging and facilitating defections from the intelligence targets that matter most to American interests. The CIA today, for example, should be sparing no expense or effort to encourage defections from Iran, whether from members of the Revolutionary Guard or technicians and scientists from Iran's suspected nuclear weapons program. Although defections offer a one-time snapshot of clandestine activities, one snapshot is better than none. CIA analysts, if given a critical mass of reporting from dozens of defectors from Iran, could paste together a clearer picture of Iran's nuclear weapons program than we have today.

Many commentators and observers take as a "lesson learned" from CIA's Iraq intelligence failure that defectors cannot be trusted. But neither can paid or "controlled" human assets on the CIA's payroll. They too have vested interests in telling the United States what they think Washington wants to hear, if only to keep those CIA paychecks coming. What CIA analysts need--much like investigative reporters--is a wide variety of reporting sources that can be cross-checked against other sources to fathom ground truth.

The CIA needs to break its umbilical cord to U.S. diplomatic facilities overseas as its primarily infrastructure for human intelligence collection. The SSCI found that human operations

"against a closed society like Iraqi prior to Operation Iraqi Freedom were hobbled by the Intelligence Community's dependence on having an official U.S. presence in-country to mount clandestine HUMINT collection efforts."

To be sure, case officers based under diplomatic cover in these facilities will be critical to receiving and debriefing walk-ins and defectors, as well as for liaison with host intelligence services. The United States, however, needs more robust human intelligence collection means against hard targets such as North Korea, China, Saudi Arabia and Syria, where America has no official diplomatic presence or where the counterintelligence environment is too tight for case officers under official cover to operate effectively.

The CIA needs to substantially bolster the use of non-official cover officers (NOCs) who have no connections to the American diplomatic infrastructure. Such officers can melt into areas rich in hard-target HUMINT collection opportunities such as the Muslim expatriate communities in Europe that are hotbeds for Al-Qaeda recruitment, indoctrination and logistics, and, for instance, in the Chinese expatriate communities of Asia. CIA NOCs in the past have suffered from neglect at the hands of old-school DO managers, who dominate the CIA's bureaucratic power structure. As intelligence expert James Bamford has observed, NOCs are forced to operate under the authority of the DO's regional offices, which traditionally look down upon their NOC counterparts.

One of the first orders of business for the next DCI should be to give NOC officers their own separate chain of command that runs directly into the DCI's suite. The infusion of competition between a rejuvenated and independent NOC program could be constructively managed by the DCI to put a "fire" under the backsides of the recalcitrant, risk averse and old-school DO officers. As it stands today, the old-school DO is a monopoly that needs to be deregulated under the DCI's authority.

The CIA's NOC program also has suffered in the past due to salaries for its officers. NOCs might be earning six-figure salaries, for example, running legitimate businesses overseas and be required to surrender those earnings for substantially smaller government salaries. The next DCI needs to push for legislation on Capitol Hill to allow NOCs to keep both salaries in the future. NOCs, after all, are earning the money by working two jobs and are taking greater risks than traditional DO case officers because they are not protected by diplomatic immunity.