August 5, 2009
Human Intel Or Technical Intel?
By Greg Grant
Some of the leading doyens of the Washington national security set recently returned from Afghanistan where they were part of new Afghan commander Gen. Stanley McChrystal’s strategy review. CSIS’s Anthony Cordesman reported back last week with a generally pessimistic take on the state of affairs on that front.
One point Cordesman made in his briefing to Washington reporters really jumped out: the surprisingly poor intelligence we have on the enemy. How is it that eight years into this war we don’t have better intelligence on exactly who we’re fighting?
Part of the problem is the complexity of the tribal milieu in Afghanistan, characterized by constantly shifting loyalties and patronage networks. RAND’s Seth Jones, speaking to a Washington audience last month, said the intricacies and complexities of the tribal layers and relationships in Afghanistan means its nearly impossible for the development community to work in the troubled southern and eastern parts of the country; the challenge of deciphering the “human terrain” is only worse for the military.
Another reason may be that the U.S. military is becoming overly reliant on technical means of intelligence gathering. A RAND research team, examining lessons from the Vietnam-era Phoenix program for contemporary counterinsurgencies, said that since the wars began this decade, the U.S. military has developed two different models to spying on and attacking insurgent networks: what they call the “U.S.-centric approach,” uses signals collection and overhead surveillance with aerial drones and satellites; the second approach is the down in the trenches human intelligence, which essentially means running a network of informants.
Using massed and persistent intelligence, surveillance and reconnaissance (ISR) over Iraq produced a number of successes: including taking down bomb making cells, largely by catching insurgents in the act of burying IEDs, and killing high value targets, such as Abu Musab Al-Zarqawi in 2006. The problem with the overhead surveillance approach is that it takes a massive amount of resources to establish that “unblinking eye” over the battlefield that commanders so desire. It also requires catching insurgents doing insurgent stuff; ISR is not all that useful in uncovering the insurgent “shadow government” in some remote mountain village.
The second approach, according to RAND, is exemplified by the Sunni “Awakening” movement where U.S. troops leveraged contacts in Iraq’s Sunni community to successfully target Al Qaeda. The strengths of the second approach are “essentially the opposite” of the ISR approach: it doesn’t cost very much. Paying off informants is a lot cheaper than building and operating a fleet of aerial drones and electronic snooping aircraft. It’s also an intelligence gathering method that is not force structure dependent. The downsides are that people’s loyalties, particularly people who come with a price, are notoriously prone to shifting.
The RAND researchers favor the human intelligence approach. As an example of its superiority they cite the fact that the ISR heavy approach was used in Iraq’s al Anbar province from 2003 to 2006, during which time the insurgency grew dramatically. However, once the Awakening movement began to spread after 2006, tip offs and intelligence on insurgent networks increased and Al Qaeda was decimated.
Running a network of informants does require some consistency in terms of people on the ground, which is why long dwell CIA personnel are better at the job than military personnel who are constantly rotating through an area; of course that’s also what CIA agents are trained to do.
As the RAND team points out, it takes a long time to establish an effective informant network. ISR can spot Taliban “flying columns” of fighters, particularly if they move in any real numbers. The more challenging task is unmasking that insurgent shadow government lurking in local communities. That will require the Afghan population trust that U.S. forces are there to protect them and that they won’t be leaving any time soon. Convincing the Afghan people that we are in it for the long haul will demand a lot more than payoffs.
+++++++Phi Beta Iota Editorial Comment+++++++
This is the point that was made in ON INTELLIGENCE: Spies and Secrecy in an Open World, and every other book we've authored, edited, or published since then. We make the point again in our most recent monograph, HUMAN INTELLIGENCE (HUMINT): All Humans, All Minds, All the Time. It is not a point that the Department of Defense (DoD) is willing to hear or act upon at any level up to and including the Secretary of Defense and the Undersecretary of Defense for Intelligence. The military-industrial complex does not make money from honest HUMINT, and the military is not willing to shift money from toys for boys to training for men–it takes a long time to develop a skilled HUMINT person, and even longer to develop localized skill sets. Everyone now leading DoD is a technocrat; there is no bench in DoD for HUMINT, OSINT, or anything remotely resembling cultural awareness, inter-agency operations, or multinational multifunctional collaborative planning, programming, budgeting, and campaigning (PPBC). Such a pity–so easy to do, so inexpensive….all it takes is leadership.