“The Arms Control Association and all of the staff I've worked with over the years … have this ability to speak truth to power in a wide variety of venues.”
Intelligence: The Achilles Heel of the Bush Doctrine
There is not yet a clearly articulated “Bush doctrine” of national security. Yet the pointers so far, especially the victory in Iraq, suggest the shape of one that is stunning in its ambition. Focused on terrorism and weapons of mass destruction (WMD), the emerging Bush doctrine is anticipatory, pre-emptive, and, if need be, unilateral. Yet the emerging doctrine is bedeviled at its core by legitimacy and capacity, including, critically, the capability of U.S. intelligence. Although the United States has the military power to take out whatever miscreant state it chooses, it still lacks the ability to precisely locate and pre-emptively target WMD, despite all the technical wizardry of its intelligence. Indeed, even determining whether a potential adversary, such as Iraq, is developing and deploying nuclear, chemical, and biological weapons will continue to prove difficult. Taking out a foe’s real or suspected WMD is likely to continue to require taking out the foe.
Parsing the Bush Doctrine
In his 2002 national security strategy, President Bush was explicit about acting first:
We must be prepared to stop rogue states and their terrorist clients before they can threaten or use weapons of mass destruction against the United States and our allies and friends. …To forestall or prevent such hostile acts by our adversaries, the United States will, if necessary, act pre-emptively.
Or, as he put it more colorfully in his speech to the nation on March 19:
We will meet that threat now, with our Army, Air Force, Navy, Coast Guard, and Marines, so that we do not have to meet it later with armies of fire fighters and police and doctors on the streets of our cities.
He had foreshadowed the new strategy in his speech at West Point in June 2002:
By confronting evil and lawless regimes, we do not create a problem; we reveal a problem. And we will lead the world in opposing it.
In making its case for war, the administration did not point to a specific set of deployments or threats that would have constituted the grounds for “anticipatory self-defense” under international law. Instead, the administration argued that, given its nature, Iraq would pose a threat to international peace if it came to possess WMD—an argument that hinged on the link between the nature of the Iraqi regime and its internal and external behavior. As Bush said in his 2003 State of the Union address, “The gravest danger facing America and the world is outlaw regimes that seek and possess nuclear, chemical, and biological weapons.” (Emphasis added.)
In other words, democratic France might be trusted with nuclear weapons, but Saddam Hussein surely could not. He could not be deterred with any certainty. Nor could Saddam be trusted not to transfer weapons to other rogue states or terrorist groups, even though the evidence connecting Saddam to terrorism was weak at best. Thus, he had to be denied access to them. In Bush’s words: “We must work together with other like-minded nations to deny weapons of terror from those seeking to acquire them.”
The Limits to Muscular Pre-emption
Although it is logical to meet the WMD threat now with military force abroad so that first responders at home do not have to, the emerging Bush doctrine of pre-emption or preventive war places stresses on intelligence that it cannot bear. America’s capacity for “ISR”—intelligence, surveillance, and reconnaissance—is unparalleled, truly in a class by itself. It is also improving rapidly. However, its shortcomings actually mirror the techniques used in enemy WMD programs. Existing ISR is not good at detecting objects that are hidden under foliage, buried underground, or concealed in other ways. Nor is it good at precisely locating objects by intercepting their signals. Would-be proliferators can exploit these weaknesses, taking pains to conceal their facilities or change the pattern of activities at weapons sites, as India did before its 1998 explosion of a nuclear weapon.
None of the limitations on U.S. intelligence-gathering capacity will ease dramatically, at least not soon. Progress is most apparent in locating moving objects using satellites and, especially, unmanned aerial vehicles (UAVs)—and, soon, expendable optical sensors launched from airplanes—though sorting out such objects from other “traffic” or ground clutter will continue to remain demanding. Predator and Global Hawk UAVs came of age in Iraq.1 They flushed out Iraqi air defenses, targeted missiles, and provided real-time video surveillance of every mission. The armed version of the smaller, lower-flying Predator fired more than a dozen Hellfire missiles, and it was a Predator operated by the CIA that blasted a car in Yemen last fall, killing a suspected al Qaeda operative and five others.
Locating and targeting moving objects better will surely be important at the opening of any war, especially one involving the possible use of WMD. That capability, though, will not greatly help the United States to pre-emptively destroy nascent WMD facilities. Other technical innovations in intelligence will help identify suspicious facilities in the future. Hyperspectral imagery, for instance, can contribute to what is called MASINT (measures and signatures intelligence) by permitting analysts to identify the composition of facilities and their emissions. But such capabilities remain limited today.
Reading the Intelligence Record
Iraq and North Korea point to the limits of the administration’s emerging national security strategy. Months of scouring have yet to produce more than possible husks of proscribed WMD in Iraq, demonstrating the limits of strategic intelligence. The United States’ tactical wartime intelligence was impressive, however. As in Afghanistan, with absolute air supremacy, U.S. intelligence had layers of sensors, from satellites to UAVs to the tactical intelligence aboard warplanes, supporting both advance special operations forces and advancing main force units. John P. Abizaid, whom President Bush has nominated to head U.S. Central Command, told the Senate Armed Services Committee on June 25, “Intelligence was the most accurate that I’ve ever seen on the tactical level, probably the best I’ve ever seen on the operational level, and perplexingly incomplete on the strategic level with regard to weapons of mass destruction.”
Secretary of State Colin Powell’s presentation to the United Nations before the war contained indications of the range of U.S. sources, especially imagery and intercepted communications between Iraqi leaders.2 In intercepted communications, Iraqi officials spoke of concealing “forbidden ammo” and made references to “nerve agents.” Powell showed satellite photographs of buildings, said to be chemical and biological weapons bunkers, with “decontamination trucks” parked outside. Another set of aerial photographs, said to have been taken two days before inspections began in November, showed a convoy of trucks and a crane, which Powell said indicated pre-inspection “housecleaning.”
The latest advance in what used to be called “all source analysis”—that is, putting together indicators from the various intelligence sources, or INTs—and what later was called “fusion” is now “multi-INT.” It involves teams of computer-savvy analysts, using today’s robust communications capabilities, to very quickly put together satellite and aircraft imagery (or IMINT) with intercepted signals (or SIGINT) and any human-source intelligence (or HUMINT), such as defector reports or interviews with recently captured Iraqis.
One intelligence tip on the eve of the war resulted in the attack on Baghdad, which was targeted at Saddam—though that appears to have been a “single-source” tip from an individual. Throughout the war, the communications problems that had hampered U.S. operations in earlier conflicts, including Afghanistan, were much less in evidence. There was much better intelligence coordination between ground and air forces, enabling air strikes against enemy ground forces with fewer casualties to friendly forces.3 In the fog of war, American forces were occasionally surprised and sometimes made mistakes, but U.S. intelligence told them where enemies were and allowed them to target foes with precision weapons to a degree unprecedented in the annals of warfare.
Still, however the debate over prewar intelligence turns out, it was plain that U.S. intelligence was far from good enough to identify, let alone target, specific Iraqi biological, chemical, or nuclear weapons with any precision. Whether Iraq successfully hid evidence of its WMD, moved the weapons on the eve of the invasion, or didn’t have many to begin with, the United States could not locate weapons of mass destruction—before or after the war.
And, in many respects, Iraq was a convenient case if not an easy one. Not only had the United States and its intelligence been working on the country solidly for more than a decade, it also had been Iraq’s ally during Baghdad’s war with Iran. Iraq’s prominence among U.S. national security concerns ensured regular collection of all kinds against Iraqi targets, and U.S. analysis had a constancy and depth during the 1990s that distinguished Iraq from many others. Moreover, while weapons inspectors with the United Nations Special Commission, or UNSCOM, left Iraq in 1998, their years of work provided at a baseline for later efforts by the UN Monitoring, Verification and Inspection Commission, or UNMOVIC.
Pre-empting Against North Korea
The North Korea case is a harder one still for the would-be pre-emptor. As one illustration, U.S. intelligence has judged since the mid-1990s that North Korea had enough plutonium to build one or two hidden nuclear weapons.4 But it has had little idea where those weapons, if they exist, might be located in North Korea’s mare’s nest of underground tunnels.
The most recent North Korean crisis also serves as a reminder of how hard it is for intelligence to know of, let alone locate and still less target, incipient WMD programs. Over the summer of 2002, U.S. intelligence concluded that, in addition to its known plutonium facilities, North Korea was operating a covert uranium-enrichment program. The program apparently began in the late 1990s, but U.S. intelligence only confirmed its existence during 2001 by monitoring activities, such as North Korea’s extensive purchases of materials for construction of a gas-centrifuge enrichment facility. The CIA contended in November 2002 that the facility was at least three years from becoming operational, but analysts believed that a completed facility could ultimately produce sufficient fissile material for “two or more nuclear weapons per year.”5
Sheer numbers and warning time compound the problem of taking out North Korea’s WMD. For delivery vehicles, it has an estimated 12,000 artillery tubes and 2,300 multiple rocket launchers that, from their current emplacements, are capable of raining 500,000 shells per hour on U.S. and South Korean troops. Five hundred long-range artillery pieces are able to target Seoul, which is only about 20 miles from the demilitarized zone that separates North and South Korea.6
By one estimate, much of North Korea’s forward-based force is protected by over 4,000 underground facilities in the forward area alone, including tunnels under the demilitarized zone that would enable the North Koreans to rapidly insert forces behind the defenders. Warning times for U.S. and South Korean forces would be short—24 hours or less—if North Korea invaded using this forward-leaning posture.
Not surprisingly, recent history is also cautionary about pre-emption. The last major nuclear crisis on the Korean Peninsula erupted in 1993, when North Korea was caught extracting bomb-making plutonium from spent reactor fuel produced by its 5-megawatt research reactor at Yongbyon. The United States came close to war, and there was much talk in Washington and Seoul about “surgical strikes” against these nuclear facilities. In the end, the Clinton administration took the path of negotiation. Given the proximity of the North and its weaponry, the death toll from war could have run into the hundreds of thousands, with large-scale casualties among the 37,000 U.S. soldiers stationed in South Korea. The eventual result was the Agreed Framework of 1994, under which the United States agreed to provide fuel oil and two light-water reactors in return for North Korea suspending its nuclear program.7 The Bush administration, however reluctantly, is likely to be forced down a similar negotiating path when dealing with Pyongyang.
International Inspections
The cases of North Korea and Iraq suggest both the value and the limits of on-site inspections, such as those conducted by UNMOVIC and the International Atomic Energy Agency (IAEA), in buttressing national intelligence. On the down side, no system of international inspection can be foolproof, not least because nations can dismiss the inspectors, as North Korea did with the IAEA late last year. And inspectors will almost always be too few in number and too limited in their ability to conduct surprise inspections anywhere in a country. UNSCOM’s years of inspections in Iraq in the 1990s were a cat-and-mouse game, a constant struggle between Iraq’s restrictions and UNSCOM’s struggle against those restrictions.
Indeed, according to one analyst, it would not be possible to verify a North Korean commitment to freeze or dismantle its uranium program.8 Instead of running 3,000 centrifuges at one site to produce several bombs’ worth of uranium per year, groups of centrifuges could be hidden in some of the country’s thousands of caves. Unlike North Korea’s declared plutonium production facilities, whose locations are known and whose operation can be detected by satellite, much of North Korea’s uranium enrichment program appears to be out of sight at indeterminate underground locations. With centrifuge enrichment technology, there is much less need to centralize production at a single site than is the case for plutonium production, so it is more difficult to determine whether a country has acquired the requisite equipment.
Yet the contrast between the two countries also suggests the value of on-site inspection. There is little baseline data on Pyongyang’s nuclear activities. In contrast, although the UNSCOM inspectors were harassed, they did fan out across Iraq for seven years, from l991 through l998, visiting both declared and undeclared sites. In contrast, IAEA inspectors conducted only one routine inspection of North Korea’s declared nuclear facilities, and that was 10 years ago.
Other circumstances no doubt will circumscribe how closely U.S. intelligence can cooperate with international inspectors, but the experience in Iraq drives home the desirability of doing so when possible.9 As the prospect of war loomed, the earlier sensitivities about information sharing between U.S. intelligence and a UN body, UNMOVIC, diminished. U.S. U-2s, along with other allied aircraft, began flying reconnaissance for UNMOVIC, giving the inspectors much more capacity to see developments at suspected facilities over time.
If the United States contemplates preventive or pre-emptive action, in principle it will want the widest possible international support and authorization for doing so. Yet, as the Iraq example demonstrated, that is precisely what it cannot get. The problem arises not from the fecklessness of the UN but rather from asking nations to take hard, potentially dangerous decisions about dealing with threats that have not yet materialized, and whose imminence is a matter of judgment.
In those circumstances, the United States will want to make the best case it can. Ideally, it will want an “Adlai Stevenson moment,” a moment like that in 1962 when the U.S. ambassador to the UN brandished incontrovertible images of Soviet missile bases in Cuba taken from a U-2 spy plane. Otherwise, even if intelligence is good enough to undertake the military pre-emption, the United States will run the risk of looking like a bully who wants rules to apply to others but not itself.
1. Eric Schmitt, “In the Skies Over Iraq, Silent Observers Become Futuristic Weapons,” The New York Times, April 17, 2003. The various UAV programs are comprehensively surveyed in a new CRS report. See Elizabeth Bone and Christopher Bolkcom, “Unmanned Aerial Vehicles: Background and Issues for Congress,” (Washington: Congressional Research Service, April 25, 2003).
2. Karen DeYoung and Walter Pincus, “Satellite Images, Communications Intercepts and Defectors’ Briefings,” The Washington Post, February 5, 2003.
3. Ronald O’Rourke, Iraq War: Defense Program Implications for Congress, (Washington: Congressional Research Service, June 4, 2003), pp. 59-60.
4. See U.S. National Intelligence Council, “Foreign Missile Developments and the Ballistic Missile Threat Through 2015,” December 2001, at https://www.dni.gov/files/documents/missilethreat_2001.pdf.
5. “CIA Report to the U.S. Congress on North Korea’s Nuclear Weapons Potential,” November 19, 2002, as published at http://www.fas.org/nuke/guide/dprk/nuke/cia111902.html.
6. Willis Stanley, “From Vietnam to the New Triad: U.S. Nuclear Weapons and Korean Security,” March 11, 2003 at http://www.nautilus.org/VietnamFOIA/analyses/StillValid.html#Stanley.
7. For background on the framework, see Jonathan D. Pollack, “The United States, North Korea, and the End of the Agreed Framework,” Naval War College Review, Summer 2003.
8. Henry Sokolski, “Contending With a Nuclear North Korea,” December 23, 2002, available at http://nautilus.org/fora/security/0228A_Sokolski.html.
9. Damian Carrington, “Spy Planes ‘Significant’ Boost to Weapons Inspections,” The New Scientist, February 17, 2003, available at http://www.newscientist.com/news/news.jsp?id=ns99993399.
Gregory F. Treverton is a senior analyst at RAND and associate dean of the RAND Graduate School. He was vice chair of the National Intelligence Council in the first Clinton administration, and his Reshaping National Intelligence for an Age of Information was published in 2001 by Cambridge University Press.