Will intelligence reform help spies see terror threats before it's too late?
In early July 2001, an FBI field agent in Arizona, alarmed that al Qaeda operatives were training at U.S. flight schools, penned a memo recommending a national program to track suspicious flight school attendants. The single-line synopsis of the memo, with portions redacted by black marker, appeared on several Web sites, including The Memory Hole. It read: "UBL [Usama bin Laden]... supporters attending civil aviation universities/colleges in state of Arizona."
This was but one of the sobering facts brought to light first in joint hearings held by the House and Senate intelligence committees and then by the 9-11 Commission. Using 20-20 hindsight, these facts have been viewed as intelligence failures.
Whether that's a fair assessment or not, it was the premise that drove the passage last December of the National Intelligence Reform Act. The act creates a director of national intelligence (DNI) to oversee the 15 government agencies involved in intelligence and a National Counterterrorism Center (NCTC), among other measures. (Several sections of the bill are unrelated to intelligence collection and are, therefore, not part of this analysis.) The question now is whether the reorganization will help the intelligence community get any closer to the tall order the nation sets as its goal: 20-20 foresight.
First step. While the bill authorizes resources to increase the number of federal analysts and collection agents and calls on the DNI to set standards for collection and analysis, it is largely a top-down reorganization that may not achieve the level of change required, say critics like David Rothkopf, senior associate with the Carnegie Endowment for Peace. The government must "find a way to start dealing with the core issues within these agencies, changing the culture, putting a high emphasis on finding objective truths and giving policy makers the information when they need it," he says.
The bill is a "good first step," nonetheless, Rothkopf says. As one congressman said during the House floor debate over the bill, it advances the ball down the field toward the goal of homeland security, even if it doesn't score a touchdown.
That's a view widely shared by intelligence professionals, even though they see flaws in the new law. Former Director of the CIA, Admiral Stansfield Turner, for example, agrees that the intelligence reform bill was on the right track when it moved from having the head of the CIA also be the director of all the intelligence agencies within the government. The newly created DNI "will not be running any agency and will, therefore, hopefully, listen to all of them," says Turner.
"Pre-9-11 the Director [of Central Intelligence] wasn't listening to the diverse views because he was concentrating on running the CIA.... Whereas this new DNI will not have an agency of his own to listen to and give preference to," he says.
Further, the new law gives the DNI "authority to be sure that there is adequate exchange of information--as there wasn't before 9-11--between the various agencies," says Turner. But the most difficult problem the new DNI will face is "to get these 15 agencies to set aside their own parochial turf considerations and do what is best for intelligence," he notes.
To encourage the development of a truly national strategy on intelligence, the bill also establishes a Joint Intelligence Community Council chaired by the DNI and composed of the secretaries of State, Treasury, Defense, Energy, and Homeland Security, plus the Attorney General and others the President may designate.
But the DNI's powers to get the intelligence agencies to work together toward any unified objectives set by the council has its limits. For example, the Joint Military Intelligence Program is not part of the newly consolidated National Intelligence Program that the DNI oversees, although the DNI "shall participate in" and consult with Defense with regard to military intelligence budgets and use of resources.
The military's remaining autonomy is seen as a weakness by some observers. Turner says that under the bill "the various collection systems like satellite photographs and electronic listening are too much under the control of the Defense Department."
This criticism is significant given that during his tenure at the CIA, Turner led the intelligence community in adapting to a new era of real-time photographic satellites, forming the basis for the current system. "In the Cold War," he says, "there may have been a perverse logic to the Defense Department controlling 80 percent of the intelligence assets because the threat was a military threat. In the war on terrorism, there's not an excuse for the military being that dominant in our intelligence because they're only part of the solution, and somebody with a national perspective, not a military perspective...should be in charge of that 80 percent."
Another former director of the CIA, James Woolsey, disagrees with Turner's assessment, however. "The military was not the problem pre 9-11," he says.
Regardless of who controls the military intelligence gathering, the government needs to think more strategically than operationally about intelligence, says Ernest May, Charles Warren Professor of American history at Harvard. The reorganization does move intelligence in that direction with the creation of the NCTC, which presents "a model for the way in which the government might organize itself to deal with a variety of problems," he says.
The bill extends that model to weapons of mass destruction by also establishing a National Counter Proliferation Center that will be the primary organization in the government for analyzing and integrating all intelligence pertaining to the proliferation of weapons of mass destruction. The bill also gives the DNI the authority to establish other intelligence centers focused on new threats as the need arises.
Budget. While the bill does establish a strong DNI with a role in the budget process, it does not endow the position with the clear and absolute budget authority the 9-11 Commission proposed. In fact, because the issue was contentious, the bill "doesn't really settle the question of how the budget would be developed, nor does it really settle the question of what are the ultimate managerial responsibilities of the new director," says May.
Turner agrees that the bill is not "firm enough in regard to budgetary authority." However, he says that the president has the power to imbue the DNI with the requisite authority to ensure effectiveness, regardless of who determines the budget for, say, spy satellites.
Others are skeptical. "Any time we create csars who are essentially bureaucratically orphaned from any control of any large organization, in the past there was not a real strong reason to think they [would] become powerful in Washington budget wars," says Michael O'Hanlon, senior fellow at the Brookings Institute. Although O'Hanlon acknowledges that the new DNI will have substantial authority, he says "it's very hard for these people who are outside of agencies to make a huge difference operationally and in the details."
Career clout. Though O'Hanlon sees the DNI as potentially orphaned, with no agency as its own, the position is intended to be viewed by all agencies as the ultimate boss. In that context, an important aspect to the new DNI is that "now there is one person who will have something to do with the careers of everyone in intelligence," says May.
The career issue is key, says Rothkopf, adding that the government should work toward "jointness" where "people who work for different agencies are trained in other agencies and can't advance in their careers unless they have worked in different agencies."
With that goal in mind, the new law calls on the DNI, in consultation with the heads of the other agencies, to "make service in more than one element of the intelligence community a condition of promotion to such positions within the intelligence community as the Director shall specify."
Similarly, another provision addresses the need for integrating the training of agents across agencies. It calls for cross-disciplinary education and joint training exercises.
At the same time, the act directs the FBI to continue to follow a dual criminal justice/national intelligence mission and to build its domestic intelligence collection capabilities by developing agents, analysts, linguists, and surveillance specialists.
The act addresses analysis and integration of intelligence through its creation of the NCTC, but it does not make similar structural changes to the collection process. It simply directs the DNI to establish objectives, priorities, and requirements for the timely collection, processing, analysis, and dissemination of national intelligence.
That's a serious omission, says Arther Hulnick, associate professor of international relations at Boston University. According to Hulnick, the current intelligence gathering system is rooted in the Cold War paradigm, which is "based on the notion that we could target certain individuals who had secret information that we needed, and we could recruit them to cooperate with us."
The problem is that, "Now we're facing a new kind of enemy," he says. "They're not recruitable in the same way that state actors and officials used to be. Terrorist cells are very hard to penetrate."
Woolsey concurs. "It's much harder to penetrate terrorist groups than a large bureaucratic institution like the Soviet military and Soviet intelligence," he says, noting, "Terrorists don't come to embassy receptions."
Developing agents under nonofficial cover (undercover agents) requires extensive training, including language skills and developing of a cover story, which will allow them to blend in with their host country, says Woolsey. Thus, these type of agents can take many years to develop.
While not structurally changing the collection paradigm, the National Intelligence Reform Act did make some changes to the law that will make it easier to spy on potential terrorists, says Turner. He cites an amendment to the Foreign Intelligence Surveillance Act (FISA) as an example.
Originally passed in 1978, FISA provides a framework for the use of electronic surveillance for foreign intelligence gathering. Subsequent legislation expanded federal laws dealing with this type of foreign intelligence gathering to address physical searches, pen registers and trap-and-trace devices, and access to certain business records. The statute has been expanded several times since 9-11.
The latest change grants the government broader authority to surveil anyone who is a noncitizen and suspected of being a threat even if they are not tied to a terrorist group. This gets at the so-called lone-wolf terrorist who is perhaps inspired by, but not associated with, a group such as al Qaeda. (The power does not extend to surveillance of U.S. citizens, however.)
Civil liberty advocates have raised concerns about the expanded FISA authority. Turner notes that the FISA provisions have a sunset clause. Once the time limit is up, the legislation can either be renewed or left to expire. "If it shows that nothing was particularly gained, then we ought to let those clauses lapse. But if they produce some helpful intelligence on terrorism, then we probably will want to tolerate them."
Open source and outsource. The act calls on the intelligence agencies to recognize the value of open-source information. In this connection, Rothkopf suggests that the government take advantage of existing expertise in the private sector through outsourcing.
"If you're looking for information, for instance, on oil fields, the best analyst works at an oil company because he's getting paid twice as much as he could get paid in the government. Wouldn't it be better if the government could go and draw on that skill set?" he asks.
Rothkopf notes that the information the analyst would be providing in that case would not be proprietary or secret. It would be from open sources but useful because it was collected by an expert who knew what to look for. "The U.S. government should move as quickly as possible to take advantage of the fact that 95 percent of the intelligence that they gather is available in the open source and this could be gathered by outsourcing," he advises.
The intelligence bill has several provisions aimed at improving analysis. The most significant is the already mentioned NCTC, which will build on the work of the Terrorist Threat Integration Center (TTIC), created by the Homeland Security Act and headed by the CIA. The NCTC will be the primary organization charged with analyzing and integrating intelligence. It will also serve as the knowledge bank of known and suspected terrorists and groups (including information on goals, strategies, capabilities, and networks of contacts and support).
Another component of better analysis is better sharing of information. The act calls for improved information-sharing technology and protocols. It calls for "common information technology standards," but gives no time frame within which a truly integrated information-sharing system would have to be in place.
It also calls for the intelligence community to "ensure that sufficient relationships are established between intelligence collectors and analysts," and to "ensure that analysis is based upon all sources available." How these objectives will be met is left to the DNI.
In a section on the CIA, the bill does set a timeline of not later than 180 days after enactment for the director of the CIA to submit to the DNI and Congress a report on how the CIA will improve analysis and human intelligence- collection capabilities, including "a plan for recruitment, training, equipping, and deployment of such personnel."
Part of the challenge as the new NCTC gears up will be finding enough qualified people to fill the new positions. But it does not necessarily take years before an analyst is good at the job, says Turner. "You don't always have to be a total expert to see that there's something shallow about an analysis; you need a good analytic mind."
Another concern is finding "officers with integrity who will fight against partisan agendas distorting intelligence," says Michael Smith, professor of intelligence studies at American Military University.
Another key in intelligence analysis is the way management approaches the information. "We don't want to have only one interpretation of intelligence data that comes in. We always want to be sure the person with the minority view at least gets to voice it," says Turner.
Whether the new structure, with a strong DNI at the top of the intelligence community, will make it more or less likely that minority views are heard depends on whom you ask.
The position was created in part to ensure that the CIA was not the dominant voice. For example, says Turner, George Tenet, who was both head of the CIA and head of all intelligence agencies before 9-11, relied overly on the CIA interpretation of weapons of mass destruction and "didn't listen to the State Department," for instance. He adds, "In short, when the bossman runs one agency, he tends to rely heavily on it."
But I.M. "Mac" Destler, director of the University of Maryland's Program on International Security and Economic Policy, sees a potential problem with intelligence analysis that could emerge from the centralized role of the DNI. "There's a tension between a central control and getting good analysis," he says, adding that there is the risk that the DNI "will want the analysis to lead to certain conclusions that may or may not be justified."
The bill does attempt to prevent that from happening. To that end, it says that the DNI should "ensure that differences in analytic judgment are fully considered and brought to the attention of policymakers."
The act further establishes a National Intelligence Council made up of "senior analysts within the intelligence community and substantive experts from the public and private sector." The council is directed to produce national intelligence estimates "including alternative views held by elements of the intelligence community."
But the problem is not just at the top. It extends far down into the bureaucracy, where, according to Hulnick, "analysts are still oversupervised. Analysts in most of the agencies don't have a chance to put their individual evaluations on a product without having to go through several layers of review."
Although that process is designed to ensure that "the CIA product speaks for the CIA and not for just some individual analyst," it "tends to stifle the type of imaginative thinking that people are saying we need in the new environment," he says.
Another key resource failure troubling Rothkopf and others is the lack of trained translators available in the intelligence community. According to Smith, "we know there are simply not enough linguists around to translate what we're getting today, much less what we'll probably be collecting in five years time." Smith blames an insufficient budget because "the people in the community did not foresee...a need for increase across the board, including those with the Middle East."
Woolsey says the translation issue is a problem that he and others have been trying to overcome for some time. "I made a request over a decade ago to substantially enhance Arabic and Farsi language training and it was turned down by the Senate Intelligence Committee," he says. According to Woolsey many members of the committee didn't believe the reforms were necessary.
Rothkopf again favors outsourcing the problem. "If the U.S. government wants to solve the problem quicker, outsourcing is a vital tool," he says.
Smith agrees and says, "I don't think the federal government has to personally own the linguists."
Whether the work is in-house or outsourced, another issue is whether the linguist is also an analyst, says Turner. The problem he says, is that "at some point, you've got to have the language and the regional expertise in the same person. You can't really be a good expert in Pakistani politics, for example, unless you can understand Urdu."
After information is collected and analyzed, the next issue is whether it is appropriately disseminated. To that end, the Intelligence Reform Act creates an Information Sharing Council and directs the President to establish what it calls an information-sharing environment (ISE) that "allows users to share information among agencies, between levels of government, and, as appropriate, with the private sector." The Department of Homeland Security retains the responsibility of disseminating information about threats to state and local government officials and private sector entities.
Whatever improvements to collection and analysis that result from this intelligence reform, policymakers and the executive branch need to understand that intelligence is not perfect and that mistakes will always be made, says Rothkopf.
"The closest thing to an ideal intelligence community," Smith says, "is one that is flexible enough to address issues that its not ideally structured to deal with."
In the end, however, "It's not a given that you'll somehow know the time and date" of any attack, says Turner. That is one fact that no amount of reform is likely to change.
Marta Roberts is staff editor at Security Management.