Also In This Issue Intelligence Gap [an error occurred while processing this directive]

By Sydney J. Freedberg Jr. National Journal
© National Journal Group Inc.
Friday, March 18, 2005

In March 2003, U.S. marines stormed the bridges over the Euphrates at Nasiriyah. The blitzkrieg on Baghdad was going well, but not quite in the way that U.S. intelligence had predicted. The threats most studied before the war -- barrages of poison gas and organized resistance by the Iraqi regular army -- had not materialized. Instead, an obscure paramilitary force, the Saddam Fedayeen, was savaging U.S. supply lines, especially past the choke point at Nasiriyah. The marines were ordered to mop the Fedayeen up. "Don't expect anything but small-arms fire" from AK-47 rifles and the like, a Marine officer remembers being told in his intelligence brief. But as his unit moved into town past Arab men in civilian clothes, the officer recalls, "all of a sudden, they'd reach down into a ditch or behind a wall and pull out an RPG" -- a rocket-propelled grenade, packed with enough explosive to blow a Humvee apart. Eighteen dead marines later, the surviving Americans had a new skepticism for their own intelligence. They also gained a new respect for the unconventional threat that would be the cause of death for the majority of the now more than 1,500 U.S. troops killed in Iraq. The threat was not the indiscriminate use of "weapons of mass destruction," but the deliberate firepower of determined guerrillas.

Learning this kind of lesson the hard, and bloody, way is precisely what intelligence is supposed to prevent. The U.S. spends an estimated $40 billion annually on intelligence, slightly more than the gross domestic product of Iraq. The Defense Department controls more than 80 percent ($34 billion) of the intelligence budget; more than 50 percent ($20 billion) is devoted not to national strategy -- determining, for example, if a foreign government is building WMD -- but to the immediate operational and tactical needs of U.S. military forces around the world (see chart, p. 833). The bulk of American intelligence work is done not by CIA spies going undercover to unearth lethal secrets, but by military attaches, staff officers, and scouts, usually in uniform, grinding away at the mundane details that make planning any attack possible. That means understanding the local terrain and weather, assessing key ports and bridges, and determining the number of enemy rifles and radars, as well as the location and organization of units in the enemy "order of battle."

Starting in World War II, when Allied signals intelligence (sigint) cracked Axis codes, and continuing through the Cold War, when imagery intelligence (imint) from spy planes and satellites pinpointed Soviet missile silos from Siberia to Cuba, America has been the master of high-tech intelligence collection. But it is human intelligence -- "humint," the age-old art of one human being getting another to talk -- that is most needed against low-tech adversaries, whether in Vietnam, Afghanistan, or Iraq. And human intelligence has been America's great weakness.

"Our ability to conduct humint has atrophied because of the overwhelming fixation on technology," said one Army officer who served in Iraq. As the insurgency swelled, guerrillas harassed his unit by setting up mortars in residential areas, launching a few rounds, and then vanishing. "We'd send up a [drone] to spot the mortarmen running away, then send in counterbattery fire with M-109 howitzers or sometimes an F-15 dropping a 1,000-pound bomb," said the officer. "By the time you got the response, these guys were long gone. We were dropping these things in people's backyards," sometimes hurting innocent civilians. The officer urged his commander to try a low-tech solution: Put intelligence and civil-military affairs specialists in the neighborhoods, get tips from locals about the guerrillas' patterns, and then lay ambushes at likely mortar set-up sites.

"You're not talking about espionage; you're really talking about rumor-mongering on an organized scale," said retired Marine Col. Gary Anderson, a leading advocate of what the military calls "cultural intelligence." In Somalia, Haiti, the Balkans, and now Iraq, the best warning of attack may not be a radio intercept or the detection of a tank, but a new splash of anti-American graffiti, or a row of shops suddenly shuttered, or a word whispered in the streets -- but you need the language skills and cultural context to understand these signs. For Americans in a hostile foreign culture, Anderson noted, the critical intelligence may be "what everyone knows but you."

The Pentagon is taking steps to redress its decades-long neglect of human intelligence. "Our ability to use human beings to collect, process, and produce information is clearly a shortfall relative to our ability to use technical measures," said Stephen Cambone, who is Defense Secretary Donald Rumsfeld's right hand and the first-ever undersecretary of Defense for intelligence. The very creation of Cambone's job, in March 2003, showed the Defense Department's new seriousness about coordinating and reforming its eight intelligence agencies (see table, p. 833). Some of the reform effort has focused on strengthening humint-gathering skills and emphasizing awareness of local cultural factors with a Defense-wide plan to bolster language proficiency, a Marine-led "Cultural Awareness Working Group," and, most controversial, a "Strategic Support Branch" to run clandestine humint-gathering teams abroad. Other parts of the reform effort are organizational and include proposals to enhance intelligence planning and to elevate in-theater intelligence-gathering from a lower-level staff function to a higher-up operational capacity run by somebody who is co-equal to the military's air, land, and sea commanders. And another part of the reform effort -- the most expensive -- is specifically technological, with tens of billions of dollars slated for laser communications satellites and a "Space-Based Radar" that could track moving targets on the ground in real time instead of just taking still photographs of them. "We have to sustain and improve technical collection, even as we repair the deficit on the human side," said Cambone. "We can't allow the same imbalance [in reverse]."

But critics argue that Rumsfeld, Cambone, and company remain enthralled with technology of high cost and dubious benefit. "Iraq forced them to focus more on human intelligence, because the deficiency was so obvious; but their initial vision was largely driven by technology," said Loren Thompson, an analyst with the Lexington Institute who consults for defense contractors and who is hardly a Luddite. "Space-Based Radar [alone] will cost more than every other military satellite program in the budget today, combined." To develop, launch, and maintain a partial system of nine radar satellites, which would still leave major gaps in coverage around the globe, would cost an estimated $34 billion, a figure so high that Congress ordered big cuts to the program. "You could populate northern Iraq with agents for that kind of money," said Thompson.

But cost is not the only factor. Human intelligence is best in synergy with technical means, not in place of them -- a point made by several Iraq veterans National Journal interviewed. Asked which was more useful in Nasiriyah, his aerial drones or human-intelligence teams, the Marine officer said they were equally useful, but "in different phases" of the battle. The high-tech bird's-eye view of clashing forces was critical in the opening assault, but the human expertise on the ground was critical afterward to secure the city. The challenge is not just to rebalance resources toward the human side, but to weave intelligence from all sources into a single picture of the war.

Strengthened Special Operations
Since its creation, the Defense Department has dominated intelligence. Its 80 percent-plus share of the intelligence budget dwarfs the CIA's 12 to 13 percent. And "in big wars, which is what you have now, the armed forces get impatient with the CIA and start to ignore them," said retired Col. Patrick Lang, a veteran of Army intelligence in Vietnam and a founder of the Defense Intelligence Agency's Defense Human Intelligence Service, which was created in the early 1990s. Certainly, recent intelligence lapses in Iraq and Afghanistan, and the aggressive personality of ex-wrestler Rumsfeld, all play a role in the Defense Department's current expansion of its intelligence operations into turf long claimed by the CIA, but this change is also driven by forces rooted deep in institutional cultures. Indeed, the very first conflict in which the Defense Department and the CIA were up and running, Korea, was an intelligence disaster. Chinese troops unexpectedly came to the rescue of the Communist North Koreans and forced advancing U.S. troops back with heavy losses. "The CIA was not providing [Gen. Douglas] MacArthur with satisfactory intelligence on Chinese intentions," said George Friedman, chairman of the analysis firm Stratfor. "It wasn't MacArthur's job to know what the Chinese were going to do. But MacArthur got hammered." The lesson that commanders still draw today, said Friedman, is, "I need to have an intelligence organization that I can give an order to. I don't need an intelligence organization that I have to make deals with."

The decades-old tension between Defense and the CIA reflects their different institutional priorities. The CIA was "serving masters in Washington; I was serving the commander [of our forces] in Europe," recalled retired Maj. Gen. Edward Atkeson, who led Army intelligence in Europe in the late 1970s and early 1980s. Atkeson wanted to send more agents into East Germany to track Soviet troop movements. "The CIA said, 'That's not important enough for you to take those risks.' I said, 'Stick it in your ear.' "

Retired Air Force Lt. Col. Rick Francona, who ran military intelligence agents in Iraq in the mid-1990s, still fumes over his dealings with the CIA. "Every time DOD wanted to conduct an intelligence operation, or you met someone who might be useful, we had to ask CIA -- some station chief or somebody sitting at Langley who doesn't know anything about the military," Francona said. Often, the CIA would say, "We have prior interest," or "We don't think it's worthwhile," or even say nothing and sit tight -- any of which, he added, was effectively a veto. But today, with the creation of a separate national intelligence director post, the head of the Central Intelligence Agency no longer has authority over the entire intelligence community. And under Rumsfeld, the military has reinterpreted the requirement to "coordinate" with the CIA on any operation as simply calling for advance notice. If the CIA does not respond, the military carries out the mission alone.

Central to the military's ability to move ahead without the CIA is the Defense Intelligence Agency's new "Strategic Support Branch." But critics in Congress see this human-intelligence-gathering outfit as a military end run around laws requiring the CIA to report covert actions abroad. The branch has also had start-up problems, including the resignation of its founding director. But as the Pentagon depicts it, Strategic Support is designed simply to accompany Special Operations units abroad to help them gather intelligence for specific missions. The Defense Intelligence Agency has always loaned personnel to Special Operations for some kinds of missions, but often in ad hoc pickup teams. The Strategic Support Branch will create a standing force of human-intelligence teams -- mostly civilian case officers, interrogators, debriefers, and analysts -- ready to deploy to the field as needed.

Although it is called the Strategic Support Branch, this new unit is mainly tactical -- focusing on specific short-term military missions. It is, however, "strategic" in one crucial sense: It allows the military's Special Operations Command (SOCOM) to operate in a new, worldwide role, in theory striking against elusive terrorist targets anywhere around the globe. (The ways these strikes can go wrong keep congressional overseers up at night -- video of U.S. soldiers held as hostages comes to mind.) In the past, SOCOM was primarily what the military calls a "supporting command" and a "force provider." It trained, equipped, and organized Special Operations units, then handed them over to regional military commanders who were in charge of the actual missions. But SOCOM has always had the capacity, rarely discussed, to send highly secret units with code names such as Delta Force, Grey Fox, and SEAL Team Six on covert missions at the orders of the president. Rumsfeld has apparently decided to make such "direct action" a far larger part of the war on terror, and he has elevated SOCOM to a "supported command," which means the Pentagon has to supply it with more authority, money, planning staff, and -- most essential in a shadow war -- intelligence.

But not only will the newly strengthened Special Operations Command require intelligence feeds to do its job -- its highly trained troops will also gather intelligence. Special Operations troops are increasingly sent around the world to train local allies, scout potential targets, and capture prisoners, so they already come across a lot of local intelligence. "Our focus in the past was on being a consumer of intelligence -- as soon as somebody arrived with a box full of intelligence, we would go take action," said an officer familiar with SOCOM activities. "Special Operations forces have a role in gathering intelligence as well: getting out and mingling with and learning from the population and the environment."

One World, One Battlefield
With his new Strategic Support Branch and strengthened Special Operations Command, Rumsfeld is trying to create a human-intelligence and covert-action force that can land anywhere around the world. With his reforms of the U.S. Strategic Command, he is trying to create a global ability to collect and process technical intelligence and to use it to support long-range, precision-strike capabilities. In other words, he wants all of that intelligence gathered from space and spy planes to help pick the targets on the ground that he can hit from long range with missiles fired from Navy ships or submarines or Air Force planes.

STRATCOM was created in 1992 to coordinate all long-range nuclear weapons -- those on Air Force bombers and those mounted on intercontinental ballistic missiles based on land or in missile submarines. (It replaced the old Strategic Air Command, which during the Cold War controlled long-range Air Force planes carrying nuclear bombs.) But Rumsfeld has de-emphasized STRATCOM's nuclear aspect and focused on its ability to launch precise conventional attacks -- using long-range cruise missiles and the like, armed with normal explosives, to hit targets that are hard for U.S. troops to reach. And he has merged this organization with the previously separate Space Command, which oversees spy satellites, and given it jurisdiction over high-tech surveillance aircraft, cyberspace, missile defense, and the "global-strike" mission. Although STRATCOM's transformation has just begun, Rumsfeld's implicit vision is of a global high-tech network, able to see and destroy targets anywhere on Earth.

The new STRATCOM's global technical responsibilities, like Special Operations Command's global human ones, are superimposed over the existing regional military commanders -- the generals and admirals who head the U.S. European Command or U.S. Pacific Command, for example. This principle of organization is not new. Almost every military commander in history has had his own scout forces operating throughout the territory under his command. The innovation is the scale. For the first time in history, the entire planet is being treated as a single area of operations -- one world, one battlefield.

It is from this lofty perspective that Space-Based Radar satellites become intriguing. They offer the prospect, albeit at tremendous expense and technical difficulty, of seeing the planet as a whole. Similarly, the Pentagon's second high-tech, high-cost satellite program, the $23 billion "Transformational Satellite Communications System," could link every U.S. military unit on Earth -- even the most isolated -- into a single high-speed network that would share intelligence of every kind, from human observations to radar data. "Their ultimate vision," said Loren Thompson, "is 'ubiquitous intelligence': the capacity to monitor the entire world on an ongoing basis."

Rumsfeld and Cambone have been leading advocates of this high-tech vision at least since they worked -- as chairman and staff director, respectively -- on the 2000-2001 Commission to Assess United States National Security Space Management and Organization. But they hardly invented the idea. It was President Clinton's vice chairman of the Joint Chiefs of Staff, Navy Adm. William Owens, who argued that a high-speed network of sophisticated sensors could "lift the fog of war," as the title of his 2000 book suggests. And the Air Force has long trumpeted this vision, as well. In the official 2004 "U.S. Air Force Transformation Flight Plan," the air service goes so far as to speak of "predictive battlespace awareness," a level of intelligence collection and analysis so sophisticated and lightning-fast that U.S. forces can "predict and pre-empt adversary actions when and where we choose ... regardless of the adversary, location, weather, or time of day." Some Air Force officers boil this down to "catching the bank robber before he robs the bank."

It is significant that these apostles of orbital omniscience include senior civilians, Air Force generals, and Navy admirals, but precious few Army or Marine leaders. Air and water are transparent and uninhabited; the land is neither. Before Operation Anaconda in Afghanistan, the raid in which American troops sought Qaeda diehards, U.S. satellites and scout planes extensively mapped the mujahedeen defenses in the mountainous terrain. "Then the first U.S. group landed right on top of an al Qaeda fighting position they couldn't see from the satellite," said Michael Scheuer, a former CIA analyst and the author of Imperial Hubris. By the battle's end, more than half of the enemy positions the troops encountered were ones that the high-tech intelligence had missed. "The enemy is smart, and they know how to use the terrain," said Scheuer. "You have to have the human intelligence in there."

The Ground Truth
For thousands of years, the only intelligence was human intelligence. Then in World War II, the Allies introduced two wonder technologies: radar to see the enemy coming and radio eavesdropping to know what he was planning. U.S. intelligence has spent six decades seeking to replicate that revolutionary success. But today's brutal, low-tech conflicts are throwbacks to more primitive times. Radar cannot distinguish between a car full of children and a car bomb, and radio cannot listen to two men talking in a darkened room. Against guerrillas, the best intelligence collectors are not sophisticated machines in the air, but sophisticated people on the ground.

As the marines were fighting to secure the supply lines through Nasiriyah in March 2003, an Army support unit took a wrong turn into an ambush. Nine soldiers died, and several, including Pfc. Jessica Lynch, were captured. An Iraqi man, Mohammed Odeh al-Rehaief, saw her under guard in the local hospital, and he walked to the marines' front lines to tell them about it. "Initially, we didn't believe him," the Marine officer recalled. "We said, 'Yeah, right,' said, 'Go get us the room number and the layout,' and sent him back." When al-Rehaief returned with the details, the surprised marines quickly passed the information, and its bearer, along to higher headquarters. Lynch was rescued, and al-Rehaief's family was given asylum in the U.S.

The Americans were doubly lucky. First, they had not shot al-Rehaief at the checkpoint by mistake. Only after the war began did U.S. forces realize, to their horror, that the American gesture for "stop" -- one hand held up, palm forward -- is the Iraqi gesture for "welcome." Second, their informant spoke English. The marines in Nasiriyah had so few translators -- about two dozen for more than 4,000 troops -- that they were constantly shuffling them among units to meet peak needs. Even today, said Anderson, "the biggest single complaint you hear from the troops is the lack of interpreters."

But translators were not the only human resources that U.S. forces were (and are) short of. Besides interpreters, said the marine who fought in Nasiriyah, "I could've used some more counterintelligence guys -- the ones you put with the guys conducting patrols -- because they recognize things the average 19-year-old would never even consider."

The military has raced to ramp up its numbers of trained experts at the front lines, adding more interrogators for its prison camps, more tactical human-intelligence teams for its patrols, more civil-affairs specialists to work with local communities, and more translators everywhere. But preparing them takes more than a few months of boot camp. The typical American soldier needs three years at language school to learn workable Arabic. Civil-affairs troops are usually reservists who bring the experience of an entire civilian career in city management or public services. And "it takes five to 10 years to develop a good humint [collector]," said the Marine officer. "We're trying to do this in two or three years."

Nor is the military bureaucracy set up to nurture such expertise. No less a figure than Rumsfeld has declared war against the rigidities of the personnel system. "Our system is not structured to provide the kind of people we need," said Lt. Col. Ralph Peters, a retired Army intelligence officer-turned-author. "You can't use an Industrial Age assembly-line approach to intelligence personnel." Current military career regulations are set up to train enlisted technical specialists and broadly educated officers; troops typically rotate from job to job and place to place every two or three years. Such a system has little room for people with the talent and desire to learn about a particular culture in depth. Even those specifically trained as foreign area officers specializing in a given region generally find that the required years of language training and embassy work come at the expense of regular promotions in rank, while colleagues who stick to mainstream combat postings move ahead faster.

And this tendency for specialization to become ghettoization doesn't just cramp careers; it hobbles operations. Commanders trained as generalists may not know how to use specialists, and specialists may not know their place in the general plan. That is why commanders ended up sending aerial drones, not Arabic speakers, to chase hit-and-run guerrillas, and why the military's Iraq experts failed to teach everyone else the proper way to signal "stop."

Learning the Lessons
Nothing concentrates the mind like the prospect of being shot at. So the military is busy, in its venerably pragmatic way, trying to learn new ways to do human intelligence. "Our intelligence community is continually learning," said Marine Maj. Patrick Carroll, an Arabic-speaking foreign area officer now on his second tour in Iraq. "There are now Americans, for instance, who understand the overarching tribal structure of Iraq as well as many Iraqis do." The intelligence team that found Saddam Hussein, for example, used commercially available law enforcement software to track his relatives and known associates. Young officers go online to PlatoonLeader.Army.mil and CompanyCommand.Army.mil -- Web sites started without official sanction, but now sponsored by the Army -- to pose urgent questions to Iraq veterans and to share advice. Today's troops are learning under fire in Iraq, and with their lives at stake, they are not waiting for the Pentagon to rewrite the manuals.

"There's really a lot of innovation outside the chain of command," said Michael Noonan, a captain in the Army Reserve and deputy director of the national security program at the Foreign Policy Research Institute in Philadelphia. Web sites, on both the civilian Internet and the classified SIPRNET, allow "rapid dissemination of lessons learned, in a much more horizontal manner."

Information technology makes such ad hoc adaptation easier. But it remains ad hoc. Smart leaders learned a lot the hard way in Vietnam, too -- but the military never embraced the lessons. So new troops coming into Southeast Asia kept bloodily reinventing the wheel, and after Saigon fell, the lessons were filed and forgotten. Those who would not repeat the past must institutionalize its lessons.

In obscure offices in Northern Virginia, though, a military working group is today drafting new doctrine that will try to institutionalize appreciation for, and use of, "cultural intelligence" -- the knowledge of local ethnicity, tribes, religions, and leaders, combined with knowledge of the local physical terrain and infrastructure, that can give troops an understanding of the whole into which pieces of intelligence fit best. The group envisions training every soldier and marine to have at least a basic understanding of culture's influence on military operations, whether in an insurgency like Iraq, in humanitarian relief efforts, or in an all-out war against a foe with alien values, such as the Imperial Japanese with their banzai charges and kamikaze suicide attacks. Everyone deploying to a specific region will learn 50 to 100 essential words and phrases in the local language -- plus gestures. Commanders and intelligence analysts will be trained to a higher level, of course, but without having to spend their careers as linguists or foreign area specialists.

Awareness of other cultures, said Marine Maj. Ben Connable, an Iraq veteran who is now with the working group, is more than not offending people: "The last thing a commander needs is sensitivity training. [This is about] mission, mission, mission, all the way, cutting through the fog of war" by giving troops a greater appreciation of the local context. Culture cannot be an issue just for intelligence specialists, Connable added, "or you're briefing people who don't understand why it's important: in one ear, out the other." The challenge is to make cultural intelligence everybody's job.

The Army chief of staff, Gen. Peter Schoomaker -- a Special Forces veteran himself -- preaches that "every soldier is a sensor." That is particularly true in guerrilla warfare, when the only way the outgunned insurgents can even fight is by avoiding detection. Any piece of information that reveals a guerrilla's potential hiding place is useful. "It used to be, we graded our military operations on how many things we broke," said an officer familiar with SOCOM's activities. "Now we often grade our combat operations by how much intelligence they generate."

So can the U.S. really pull this off? Every superpower in history has had its Achilles' heel, said retired Army War College commandant Maj. Gen. Robert Scales. "For us, it has been the failure to know our enemy," he observed. "We're just not good at it." Then again, if it is any comfort, "everybody is bad at this," said Stratfor's Friedman. "The Germans -- who pulled no punches -- could never wipe out the partisans in Ukraine or Yugoslavia. The Soviets could not take out the Afghan mujahedeen. And the British were never able to get control of Afghanistan."

But the longevity of the British Empire also shows that there are ways to do intelligence well enough to win. The fog of war remains; 100 percent omniscience is an illusion. But the military hopes that for its 80 percent of the intelligence system, it has found an 80 percent solution.

[an error occurred while processing this directive]

Need A Reprint Of This Article?
National Journal Group offers both print and electronic reprint services, as well as permissions for academic use, photocopying and republication. Click here to order, or call us at 877-394-7350.