Intelligence: From Secrets to Policy (18 page)

Read Intelligence: From Secrets to Policy Online

Authors: Mark M. Lowenthal

BOOK: Intelligence: From Secrets to Policy
8.86Mb size Format: txt, pdf, ePub
The P&E imbalance has become a political issue when Congress makes budget decisions. As noted, the intelligence committees find it difficult to put money into new collection systems when they are told that only as many images or signals will be processed and exploited as was the case for the previous generations of collectors. Although there may be valid explanations for this outcome. Congress—as might be imagined—would rather see increasingly expensive systems result in more collected intelligence that can be used by analysts.
 
COMPETING COLLECTION PRIORITIES. Given that the number of collection platforms, or
spies,
is limited, policy makers must make choices among competing collection requirements. They use various systems to set priorities, but some issues inevitably get shorter shrift, or may be ignored altogether, in favor of those that are seen as more pressing.
Both policy makers and the intelligence officers acting on their behalf request increased collection on certain issues. However, their requests are made within a system that is inelastic in terms of both technical and human collectors. Every collection request that is fulfilled means another collection issue or request goes wanting; it is a zero-sum game. That is why a priority system is necessary in the first place. Moreover, the system has little or no surge capacity: few collection systems (airplanes, drones, and ship-based systems) or spies are waiting in reserve for an emergency. Even if additional satellites have already been built, launching them requires a ready rocket of the appropriate size, an available launch pad, and other resources. (The Soviet Union used a different collection model. Soviet satellites lacked the life spans of their U.S. counterparts. During crises, the Soviets supplemented current collection assets with additional, usually short-lived, satellites, which were kept on hand with launch vehicles ready.) Similarly, one does not simply tap a spy and send him or her off to a new assignment. Cover stories need to be created, along with the inevitable paraphernalia; training may be necessary; and a host of other preparations must be made. Inelasticity of resources makes the priorities system difficult at best.
The shifting—or nonshifting—of collection resources in the face of novel situations or emergencies is always subject to 20/20 hindsight. For example, in May 1998 the newly elected government of India resumed testing nuclear weapons, as it had promised in its election campaign. The U.S. intelligence community had not detected the test preparations. As a result, Director of Central Intelligence (DCI) George J. Tenet (1997-2004) asked retired admiral David Jeremiah to review the intelligence community’s performance on this hard-target issue—preventing the proliferation of nuclear weapons.
Jeremiah reported several findings, including the fact that—given the Indian government’s avowed intention to test, which required no clandestine collection to learn—intel—ligence performance could have been better. But he noted that collection assets that might have picked up indications of the impending test were focused on the Korean demilitarized zone (DMZ), at the request of the commander of U.S. forces in Korea. As an NSA director put it, the Korean DMZ was the only place in the world in the late 1990s where someone else could decide if the United States would go to war. Although the Korean DMZ remains a constant concern, for a brief period in 1998, Indian test activities perhaps should have been accorded a higher priority.
 
COLLECTION SWARM BALL. A major problem that has occurred in managing collection is the phenomenon known as collection
swarm ball.
This refers to the tendency of all collectors or collection agencies to collect on an issue that is deemed to be important, whether or not they bring anything useful to the table or can offer an appropriate type of collection. It is called “swarm ball” because it resembles the tactics of small children playing soccer, in which both teams converge on the ball en masse regardless of their assigned positions. Swarm ball has usually involved high-priority issues. For example, if a high-priority issue was the cyber attack capabilities of a hostile state, little value would be gained by imagery, although imagery collection managers might be tempted to contribute to the issue based solely on its priority. The impetus for swarm ball is clear: It allows collectors to show that they are working on high-value issues, regardless of their contribution, which will be important for their continued support in the next round of budget allocations.
The solution to swarm ball is twofold. First, agreement must be reached on which INTs are responsible for collecting on specific issues or priorities. This is not a difficult agreement to reach, although it is time consuming, as the attributes of most issues can be delineated (locations, facilities. people involved, likelihood of communications, types of intelligence that is needed, and so on) and then matched against current or impending collection capabilities. Second, the agreement must be rigorously enforced, and agencies must not be penalized for not collecting against issues not suited to them regardless of the issues’ importance and must be recognized for concentrating on the issues about which they can collect needed intelligence.
 
PROTECTING SOURCES AND METHODS. The details of collection capabilities—and even the existence of some capabilities—are among the most highly classified secrets of any state. In U.S. parlance, classification is referred to as the protection of
sources and methods.
It is one of the primary concerns of the entire intelligence community and a task specifically assigned by law to the director of national intelligence.
Several levels of classification are in use, reflecting the sensitivity of the intelligence or intelligence means.
(See box, “Why Classify?”)
The security classifications are driven by concerns that the disclosure of capabilities will allow those nations that are collection targets to take steps to prevent collection, thus effectively negating the collection systems. However, the levels of classification also impose costs, some of which are financial. The physical costs of security—guards, safes, and special means of transmitting intelligence—are high. Added to these is the expense of security checks for individuals who are to be entrusted with classified information (see chap. 7 for details).
Critics maintain that the classification system is sometimes used inappropriately and even promiscuously, classifying material too highly or, in some cases, classifying material that does not deserve to be classified. Critics are also concerned that the system can be abused to allow the intelligence community to hide mistakes, failures, or even crimes.
Beyond the costs of the classification system and its potential abuse, the need to conceal sources and methods limits the use of intelligence as a policy tool. For example, in the late 1950s Khrushchev broke a nuclear test moratorium and blustered about the Soviet Union’s growing strategic nuclear forces. President Dwight D. Eisenhower, bolstered by the first images of the Soviet strategic forces, knew that the United States enjoyed a strong strategic superiority. But, to protect sources and methods, Eisenhower did not reply to Khrushchev’s false boasts. What might have been the results if the United States had released some imagery to counter the Soviet claims? Would the release have spurred the Soviets to greater weapons-building efforts? Would it have severely undercut Soviet foreign policy? Would it have affected U.S. intelligence capabilities, even though the Soviets already knew their country was being overflown by U-2s and later by satellites? These questions are not answerable, but they provide a good overview on the problem.
More recently, the U.S. intelligence community has grown concerned about protecting intelligence sources and methods during post-cold war military operations that involve cooperation with nations that are not U.S. allies. Even among allies the United States employs gradations of intelligence sharing, having the deepest such relationship with Britain, followed closely by Australia and Canada. Intelligence relations with other North Atlantic Treaty Organization (NATO) allies are close, albeit less so than with the “Commonwealth cousins.” But some operations, such as in Bosnia, have involved military operations with nations that are viewed with lingering suspicion, such as Russia and Ukraine. In these cases the need to protect intelligence sources and methods must be balanced against the need to share intelligence—not only for the sake of the operation but also to ensure that military partners in the operation are not put in a position in which their actions or inactions prove to be dangerous to U.S. troops.
WHY CLASSIFY?
 
Numerous critics of the U.S. classification system have argued—not incorrectly-that classification is used too freely and sometimes for the sake of denying information to others who have a legitimate need for it.
However, a rationale and some sense are behind the way in which classification is intended to be used. Classification derives from the damage that would be done if the information were revealed. Thus, classifcation related to intelligence collection underscores both the importance of the information and the fragility of its source—something that would be difficult to replace if disclosed.
The most common classification is SECRET (CONFIDENTIAL is rarely used any longer), followed by TOP SECRET. Within TOP SECRET are numerous TOP SECRET/CODEWORD compartments—meaning specific bodies of intelligence based on their sources. Admission to any level of classification or compartment is driven by an individual’s certified need to know that specific type of information.
Each classification level is defined; current definitions are found in Executive Order 13292 of March 25, 2003.
• CONFIDENTIAL: information whose unauthorized disclosure “could be expected to cause damage to the national security ”
• SECRET: information whose unauthorized disclosure “could be expected to cause serious damage to the national security ”
• TOP SECRET: information whose unauthorized disclosure “could be expected to cause exceptionally grave damage to the national security.”
 
Higher levels of access are useful bureaucratic levers for those who have them in contrast to those who do not.
 
Another intelligence sharing issue arose in 2002-2003, in the months before Operation Iraqi Freedom. The United States and Britain said they would provide intelligence on Iraqi WMD to United Nations (UN) inspectors but not necessarily all available intelligence. Some controversy arose after DCI Tenet said the United States was cooperating fully but the CIA later revealed that it had shared intelligence on 84 of 105 suspected priority weapons sites, which some members of Congress felt was not what they had understood to be the agreed level of intelligence sharing.
LIMITATIONS OF SATELLITES. All satellites are limited by the laws of physics. Most orbiting systems can spend only a limited time over any target. On each successive orbit the satellites shift to a slightly different coverage pattern. (Satellites correspond to the motion of the earth, as they are trapped within Earth’s gravitational pull. Thus, satellites’ orbits move from west to east with each pass.) Moreover, satellites travel in predictable orbits. Potential targets of a satellite can derive the orbit from basic knowledge about its launch and initial orbit. For a variety of reasons, some individuals and organizations attempt to publicize this information. This enables nations to take steps to avoid collection—in part by engaging in activities they wish to keep secret only when satellites are not overhead.
Satellites that are in
geosynchronous orbit
stay over the same spot on Earth at all times. But to do this they must be placed twenty-two thousand miles above Earth. The great distance between the collectors and their targets raises the problem of transmitting collected information back to Earth. Collection can be precise only up to a point, thus explaining the vacuum cleaner problem. Satellites can also be flown in
sun-synchronous orbits,
that is, moving in harmony with the Earth’s rotation so as to always remain where there is daylight, but this produces an easily tracked orbit. Sun-synchronous orbit is better for commercial satellites than for national imagery satellites.
Another interesting orbit is the “Molniya” orbit, named after the Soviet communications satellites that first used them. The Molniya orbit is highly elliptical, coming close to the Earth over the southern hemisphere (perhaps 300 miles) and then much further away from the Earth over the northern hemisphere (perhaps 25,000 miles). In this pattern, a satellite revolves around the Earth twice in a day. It is important to remember that the Earth’s land mass is not evenly distributed; much more of it lies north of the equator than south of it. The advantage of the Molniya orbit is that it moves very quickly across the southern hemisphere, where there are likely to be fewer targets, because it is close to the Earth’s gravitational pull, but then “lingers” as it moves across the northern hemisphere when it is further away. Approximately eight of the twelve hours of one revolution will be spent over the northern hemisphere. This allows increased collection over the larger area of land. But the satellite’s greater distance over the northern hemisphere also dictates that it does broad area collection as opposed to close-in or “spot” collection.
 
THE STOVEPIPES PROBLEM. Intelligence practitioners often talk about collection “stovepipes.” This term is applied to two characteristics of intelligence collection. First, all of the technical collection disciplines—geospatial intelligence (GEOINT, formerly imagery or IMINT), signals intelligence (SIGINT), and measurement and signatures intelligence (MASINT)—and the nontechnical human intelligence (HUMINT), or
espionage,
have end-to-end processes, from collection through dissemination. (Open-source intelligence—OSINT—should have end-to-end processes, but it does not.) Thus, a pipeline forms from beginning to end. Second, the collection disciplines are separate from one another and are often competitors. The INTs sometimes vie with one another to respond to requests for intelligence—largely as a means of ensuring continuing funding levels—regardless of which INT is best suited to provide the required intelligence. Often, several INTs respond, regardless of their applicability to the problem, thus creating the swarm ball. Within the U.S. intelligence system, a variety of positions and fora have been designed to coordinate the INTs, but no single individual exercises ultimate control over all of them. During testimony about the 2004 intelligence legislation, some of the tension between the DCI and DOD over control of the National Geospatial-Intelligence Agency (NGA) and NSA was evident. These agencies are, as the names indicate, national intelligence agencies and come under the DNI (or the DCI at the time of the hearings). But NGA and NSA are also DOD agencies and are designated as combat support agencies, thus indicating a degree of control by the secretary of defense as well. The legislation creating the DNI does not clarify this situation. The stovepipes are therefore complete but individual and separate processes.

Other books

The Passion of Dolssa by Julie Berry
Acid Sky by Mark Anson
Not Quite Perfect Boyfriend by Wilkinson, Lili
A Christmas Wish by Desconhecido(a)
Deeper by Jane Thomson
PATTON: A BIOGRAPHY by Alan Axelrod