Preemption Paradox

Abstract

Using military force to destroy an adversary's nuclear program is a compelling option, so how come most nations have decided against it?

Twenty-five years ago, on a late spring afternoon, eight Israeli bombers streaked across the desert sky on a top-secret mission to destroy Osirak, Iraq's emerging nuclear reactor complex. A dramatic military action to prevent nuclear weapons proliferation, the June 7, 1981 strike left a legacy that echoes today in the “all options are on the table” drumbeat emanating from Washington and Jerusalem. The seemingly straightforward message to Iran and other would-be proliferators: Abrogate nonproliferation pledges in this post-9/11 era and risk being “Osiraked.”

These aggressive declarations, however, butt against historical reality. Save for two major exceptions–the Allied effort to scuttle Nazi Germany's nuclear program, and attempts by Iran, Israel, and the United States to wipe out Saddam Hussein's nuclear ambitions–countries with incentives to use force to halt nuclear proliferation have spent more time conducting tabletop exercises than taking action. The result: nations such as the Soviet Union, China, and Pakistan produced the Bomb despite the interest and capacity of adversaries to stop them.

A desperate race

At the beginning of the nuclear age, there was no nonproliferation norm, no concept of nuclear deterrence, and no taboos against nuclear war. All that existed was an apparent race for the Bomb in the midst of history's bloodiest conflict.

Adolf Hitler's Third Reich lacked the resources, organization, scientific understanding, and even commitment to be in the pursuit. 1 But the United States did not know this early in the war. “At best, I do not see how we can catch up with the Germans unless they have overlooked some possibilities we recognize, or unless our military action should delay them,” warned physicist Arthur Compton in a 1942 communiqué to Gen. Leslie R. Groves. 2

In response, the Allies applied a two-pronged strategy: They would try to cross the nuclear finish line first, and simultaneously, they would destroy the Nazi effort. Planners targeted German-occupied Norway's Norsk Hydro hydrogen-electrolysis plant–marking the first instance of nuclear preemption. 3 With so much at stake, the hesitation that would characterize post-World War II coercive nonproliferation had no place in this decision.

The Allies targeted the Norsk Hydro plant because it was the world's only commercial processor of heavy water (deuterium oxide), which can be used in some reactors to turn natural uranium into weapons-usable plutonium. From late 1942 to February 1944, the Allies spared no effort to destroy the plant and its contents. 4

In the first attempt, two gliders ferried 34 British commandos to their sabotage mission. But one glider crashed at sea, the other onto Norway's rocky terrain. The next Allied effort parachuted a team of Norwegian commandos into their native country; the nine saboteurs, scaling a steep, icy canyon wall, conducted a dramatic winter evening assault. The attack effectively destroyed eight weeks' worth of heavy water production, but the Germans rebuilt the installation in six weeks. The undertaking offered a cautionary note that would reverberate in post-war planning: Force can fall short; adversaries can absorb a blow and then recoup.

Desperate to finish the mission, Washington ordered hundreds of bombers to take out the plant in a single air raid. Only 12 of the 1,000 dropped bombs struck the target, but the attack was a success; Germany decided not to rebuild. A final act of sabotage sunk a transport ferry that was to carry remaining stocks of heavy water to Germany.

Though intelligence demonstrated that Germany's nuclear program had sputtered, Washington still worried. In its view, only the Reich's defeat combined with occupation, dismantlement, and destruction of nuclear facilities, and the incarceration of senior Nazi scientists would put an end to Hitler's nuclear ambitions. Preemptive military strikes alone were not enough.

Sneak attack: Norwegian newspapers documented the Allied attacks on the Norsk Hydro plant (above) and a ferry carrying heavy water to Germany.

The reluctant superpower

At war's end, the United States stood atop the nuclear pinnacle. But it could sense the Soviet Union striving to reach the same heights. In 1946, the Kremlin rejected the U.S.-proposed Baruch Plan that would have placed the most worrisome elements of the nuclear fuel cycle under the auspices of an International Atomic Development Authority. Coupled with tensions over occupation rights in Germany and the Soviet Union's consolidation of its hold on Eastern Europe, some public figures and members of the nuclear scientific establishment–including Winston Churchill, Bertrand Russell, Leo Szilard, and John von Neumann–called for military action to stop the Kremlin before it got the Bomb. 5

General Groves, the head of the Manhattan Project, reflected: “If we were ruthlessly realistic, we would not permit any foreign power with which we are not firmly allied … to make or possess atomic bombs. If such a country started to make atomic weapons, we would destroy its capacity to make them before it had progressed far enough to threaten us.” 6 But President Harry S. Truman seemed resigned to the inevitability of proliferation. In a September 18, 1945 conversation with White House confidant Joseph E. Davis, Truman said, “Are we going to give up these [atomic] locks and bolts which are necessary to protect our house? … Clearly we are not. Nor are the Soviets. Nor is any country if it can help itself.” 7

The idea of preemption also found resistance in the military. “It might be desirable to strike the first blow, [but] it is not politically feasible under our system to do so or to state that we will do so,” Pentagon planner George Lincoln acknowledged in 1945. 8 As Cold War historian John Lewis Gaddis has observed, U.S. reluctance was partly rooted in self-image: “America did not start wars.” 9

The practical question of whether the United States could actually “win” a preemptive war against the Soviet Union also loomed large. In a statement that presaged the dilemma currently confronting U.S. advocates of preemption, Truman's defense secretary, James Forrestal, observed, “Conquering the Russians is one thing, and finding what to do with them afterward is an entirely different problem.” 10

Truman's successor also contemplated the merits of preemption. President Dwight D. Eisenhower endorsed a December 1954 National Security Policy paper that concluded, “The United States and its allies must reject the concept of preventive war or acts intended to provoke it.” 11 But the principle would never be as immutable as the policy document suggested. The Eisenhower administration generated a plethora of conflicting statements about the use of preemptive force against a coiled Soviet Union, suggesting a policy of ambiguity that neither embraced nor rejected the option. 12

By 1962, the United States had learned to live with a nuclear-armed Kremlin for 13 years. But “geographic proliferation” on Washington's doorstep–the placement of Soviet nuclear missiles in Cuba–would be another matter. 13 While there would be no repetition of the Norsk Hydro attacks, during the Cuban Missile Crisis President John F. Kennedy forced Soviet leader Nikita Khrushchev to back down by crystallizing Washington's preemptive intentions. Given the balance of interests and military power favoring Washington–coupled to a side agreement to remove U.S. missiles from Turkey–the Soviets decided that they would not risk Moscow for Havana.

The idea of preemption found resistance in the military. U.S. reluctance was partly rooted in self-image: “America did not start wars.”

No good options

China began its nuclear weapons research in the 1950s. By December 1960, U.S. satellite imagery and U-2 intelligence pieced together the fragmentary basis for a National Intelligence Estimate that forecast a Chinese nuclear weapon test by as early as 1963. 14

That China might get the Bomb generated tremors among U.S. policy makers. 15 In June 1961, the Joint Chiefs of Staff warned that Beijing's “attainment of a nuclear capability will have a marked impact on the security posture of the United States and the free world.” 16 In their subsequent review of diplomatic, economic, and military options–which included sabotaging nuclear facilities, an invasion by Chinese nationalists, a maritime blockade, renewing the Korean War, and U.S. conventional or tactical nuclear weapon strikes on the nuclear plants–the Joint Chiefs made no specific recommendation. Yet their aversion to preemption was evident. “Many of the [military] actions … above are obviously acts of war, [and] should be initiated only after all other means have been exhausted, and then only after full and careful consideration of the implications at the time,” they argued in a 1963 memorandum. 17

The State Department arrived at a similar cautionary conclusion. Robert Johnson, the department's leading analyst on the issue, sought to dampen the sense of mounting panic within the U.S. bureaucracy, arguing that the “great asymmetry in Chinese Communist and U.S. nuclear capabilities and vulnerabilities” made it unlikely that Beijing would ever contemplate nuclear first-use, except in the event of an attack that “threatened the existence of the regime.” 18 And even if U.S. military strikes were successful–an uncertain outcome given incomplete intelligence on China's nuclear facilities–preemption would, at best, “buy some time.” 19 If anything, such an attack could strengthen Beijing's resolve to obtain a nuclear deterrent, and a hardened successor program could be constructed in as few as four or five years.

Military action entailed other risks: It could prompt Chinese retaliation against Taiwan or U.S. allies in East Asia, entangle the Soviets (who were unwilling to join in such an attack), degrade Washington's international prestige and alliances, and reduce the prospects for arms control initiatives to constrain China. Johnson concluded, “Action with no justification other than a general argument that the U.S. was seeking to preserve the peace of the world through depriving a potential aggressor of nuclear weapons” could not be defended. 20

The arguments resonated. On September 15, 1964, the secretaries of State and Defense and the national security adviser met with President Lyndon B. Johnson in the Cabinet Room to present their recommendation: “We are not in favor of unprovoked unilateral U.S. military action against Chinese nuclear installations at this time. We would prefer to have a Chinese test take place than to initiate such action now. If for other reasons we should find ourselves in military hostilities at any level with the Chinese Communists, we would expect to give very close attention to the possibility of an appropriate military action against Chinese nuclear facilities.” 21 Johnson, running on a peace platform in the upcoming election, concurred. On October 16, 1964, China conducted its first nuclear test and became the world's fourth nuclear weapon state.

Israel's gamble

In the 1960s, the United States was willing to adopt a “wait and see” approach with China. Decades later, Israel would not give Iraq the same benefit. On September 8, 1975, Iraqi Vice President Saddam Hussein declared publicly that Baghdad's purchase from France of a 70-megawatt test reactor called Osiris and a small Isis reactor–collectively called Osirak–would be “the first actual step in the production of an Arab nuclear weapon.” 22

Jerusalem's 1981 attack against Osirak marked the culmination of a failed multiyear campaign to halt the construction of Iraq's reactor by means other than military action. The government of Prime Minister Menachem Begin first lobbied Paris and “Washington. Rebuffed, Israel tried assassination, but the deaths of three scientists affiliated with the nuclear program barely hindered Iraq's efforts. Israel attempted sabotage next, but the demolition of Osirak's core as it awaited shipment and the fire-bombing of an Italian firm contracted to provide reprocessing technology delayed Osirak only marginally. 23

The decision to use military force was not made easily; Israel debated the option intensely. In cabinet discussions, Israel's deputy prime minister, defense minister, health minister, chief of military intelligence, and head of Mossad all preferred the “wait and see” approach. 24 They feared that an unprovoked attack could spark war, stimulate region-wide nuclear proliferation, and make Israel a global pariah. Furthermore, some doubted that Iraq even had the scientific ability to build the Bomb.

The preemption option itself posed serious logistical problems. 25 A commando raid in which agents were inserted into and extracted from hostile territory was too daunting. But an air strike was also risky, as demonstrated by a botched Iranian aircraft strike on the plant in September 1980, shortly after the start of the Iran-Iraq “War. Some Israeli officials questioned whether Israel's Phantom or F-15 aircraft could travel 600 miles over hostile territory to do the job.

As it turned out, these aircraft did not have to take on the task. In the aftermath of the 1979 Iranian Revolution, Shah Reza Pahlevi's order of U.S. F-16s did not go through. Superior to Israel's Phantoms and F-15s, Jerusalem scooped up the F-16s after “Washington put them on the block. Still, it was not the acquisition of the aircraft that tipped Israel's scales in favor of a preemptive strike; it was the determination of Prime Minister Begin.

Eight F-16s each carrying two 2,000-pound gravity bombs flew toward Osirak; a phalanx of F-15s hovered nearby to jam Iraqi radar, intercept Iraq's air force, and provide a communication link to Israel. Search and rescue helicopters were at the ready. The attack took place before Osirak went critical to avoid the release of the plant's radioactive contents. All but two bombs hit the plant, leaving the reactor in ruins.

Israel's assault sent shock waves through the world, garnering condemnation and grudging admiration. But despite the destruction the preemptive strike wrought, it did not extinguish Iraq's ambitions. Baghdad, which had committed 400 scientists and $400 million to the nuclear program before the attack, enlarged its nuclear staff to 7,000 and upped its budget to $10 billion. 26

South Asian standoff

One country that took particular note of Israel's raid on Osirak was India. Throughout the 1970s and 1980s, India had watched apprehensively as Pakistan, its perennial adversary, followed a nuclear weapons path similar to its own. In addition to the acquisition of a “peaceful” heavy water reactor from Canada, the other telltale signs included an expanding number of reprocessing and enrichment installations. Given the diminutive size of Pakistan's nuclear power program, the fuel-cycle plants likely served only one purpose: nuclear weapons.

Adding to these concerns would be the program's principal political promoter. In 1971, in the aftermath of Pakistan's loss of Bangladesh to India's armed forces, Zulfikar Ali Bhutto became prime minister. Bhutto had been a patron of the atom as far back as 1960 when he became minister of minerals and natural resources. As foreign minister, in 1965, he laid the foundation for later nuclear weapons assistance from the Chinese. In the same year, he made his ambitions clear: “If India builds the bomb, we will eat grass or leaves, even go hungry, but we will get one of our own.” 27 True to his word, as Pakistan's leader, he was determined to move forward. On January 24, 1972, he gathered the country's top scientists for a meeting in Multan, Punjab, to set the program in motion.

Getting “Osiraked”: The remains of the bombed-out Osirak nuclear reactor sit 40 kilometers (25 miles) south of Baghdad. The site was destroyed twice, first by the Israelis in 1981 and then by the allies during the 1991 Gulf War.

Although Bhutto would never see his ambition come to fruition–in 1977 the military overthrew and later executed him–his successors continued his efforts. India watched Pakistan's development at first with disdain–it couldn't possibly build the Bomb–but, in time, with mounting anxiety. 28 By the early 1980s, Indian analysts concluded that Pakistan had enriched sufficient uranium for one or two bombs.

Israel's bombing of Osirak provided inspiration for a military solution. The Indian Air Force–energized by the procurement of new British Jaguar strike aircraft–studied the application of Israel's example to Pakistan's Kahuta enrichment plant. Military planners concluded that the attack could succeed, but at a cost of half the bomber force. Such losses would not be the gravest risk. “What will happen next?” the chief of operations asked. “The international community would condemn us for doing something in peacetime, which the Israelis could get away with but India would not be able to get away with. In the end, it will result in a war.” 29

The specter of war prompted another concern. Pakistan's retaliatory response could include striking Indian nuclear reactors and reprocessing sites situated near urban settings–effectively mounting a devastating radiological attack upon India. 30 A high-ranking official commented, “We knew we would have to live with Pakistan's nuclear capability, and there was no way around it.” 31

But living together required some insurance; overcoming the temptation for nuclear preemption was in the interest of both parties. The two countries negotiated a 1985 accord (which was fully implemented in 1993) not to attack one another's nuclear facilities. However, as years passed, events would call into question whether the agreement would stand. The countries' tit-for-tat 1998 nuclear weapons detonations, coupled with the unresolved Kashmir dispute, periodically raised the preemption specter anew.

Targeting North Korea

As South Asia wrestled with its nuclear conundrum, another would-be nuclear power emerged in the Far East. North Korea's ambitions to obtain nuclear weapons lay rooted in U.S. threats to use the Bomb during the Korean War and the North's unease over the Soviet Union's retreat during the Cuban Missile Crisis. With assistance from Moscow, Pyongyang steadily expanded its indigenous nuclear program over the next couple of decades. However, in 1985, Moscow made it clear that further support was contingent upon North Korea joining the Nuclear Non-Proliferation Treaty (NPT).

From the beginning, signs were not good for international control. Seven years would pass before Pyongyang signed the International Atomic Energy Agency's (IAEA) Safeguards Agreement. And afterward, the North persistently denied IAEA inspectors access to sites of concern. In spring 1994, matters came to a head when North Korea began removing spent fuel from its five-megawatt reactor in Yongbyon.

Not an option: Clinton administration officials considered attacking North Koreas nuclear facilities, which include the Yongbyon-1 nuclear reactor (top) and spent fuel pool, but they decided it was too risky.

Diplomacy was at the heart of the Clinton administration's effort to curb North Korea, but the president's advisers could not ignore the military option. The administration's senior national security staff evaluated the prospects in June 1994, against the backdrop of Pyongyang's announced intention to withdraw from the NPT and expel inspectors. 32 But armed force posed too many risks. 33 A commando or cruise missile attack might stop the future extraction of weapon-grade material, but it would not guarantee the destruction of material the North may have removed in advance. Attacking the reactor would substantially set back Pyongyang's program (albeit risking a radiological release), but doing so could prompt international opprobrium. Also, a strike against the North could spark a full-scale war. In the view of Gen. Gary Luck, commander of U.S. forces in South Korea, “If we pull an Osirak, they will be coming south.” This could result in as many as 750,000 U.S. and South Korean military casualties alone. 34

U.S. officials decided, at least for the time being, that sanctions offered a far more attractive option than preemption. Air Force Chief of Staff Gen. Merrill McPeak commented, “We can't find nuclear weapons now except by going on a house-to-house search,” suggesting that, when it came to military options, he felt that only the occupation of North Korea, rather than limited preemptive strikes, would succeed. 35

The matter became moot when former President Jimmy Carter visited Pyongyang in June 1994. Laying the foundation for what would become the Agreed Framework, the United States consented to lead an international consortium to provide the communist state with a light water reactor and heavy oil to meet its immediate energy needs.

In January 2003, confronted with U.S. intelligence that it had violated its nonproliferation vows by pursuing a secret enrichment program, President Kim Jong II bolted from the NPT. On February 10, 2005, Pyongyang officially claimed that it had manufactured nuclear weapons.

Cautionary tales

This recounting has summarized the “big” events in the history of nuclear preemption, save one: Washington's 2003 attack on Iraq. 36 Built on the false premise that Saddam had an active nuclear program, America's invasion was an exercise in preemptive overkill. Baghdad's nuclear program was already dead–not because of Saddam's aversion to the Bomb, but due to an earlier war that marked nuclear preemption by serendipity. 37 But for Iraq's defeat in 1991–or had Saddam delayed his invasion of Kuwait by a year or more–Baghdad might have joined the nuclear club.

While this and other cases are different from one another, viewed comparatively they suggest some tentative findings to answer a nonproliferation conundrum: Why has history not witnessed more consistent application of the Osirak template?

First, nuclear preemption in peacetime poses profound political risks. In confronting the prospects of a nuclear-armed Soviet Union and China, the United States was reluctant to be seen as the aggressor. India had similar reservations. Even Israel's security establishment–well aware of Iraq's declared ambition to obtain nuclear weapons–was mindful of the potential political costs of instigating an unprovoked attack.

A second lesson is that nuclear states appear reluctant to preempt emerging nuclear powers if the latter have a significant capacity to strike back. The geographic proximity of adversaries has long accentuated this dilemma. India feared ruinous reprisals from neighboring Pakistan. U.S. military planners worried that China would lash out across the Taiwan Straits and that North Korea would head south against Seoul. In today's world, strike-back could also include spectacular surrogate terrorist acts or military counterattacks against vital economic lifelines, such as petroleum production or distribution.

Radiological risks add novel concerns. India was not alone in its apprehension about retaliation against its nuclear sites. Japan, for example, feared that U.S. preemption of North Korea could result in attacks on its nuclear power plants. 38 Even without this prospect, successful destruction of Pyongyang's plants risked radioactive releases. 39 (By contrast, anxiety over nuclear contamination actually accelerated Israel's decision to bomb Osirak. Prime Minister Begin pressed for the strike before the reactor became “hot.”)

Facilities that have been destroyed can be rebuilt. As former U.N. weapons inspector David Kay once pointedly asked, “How do you roll back knowledge?”

The use of an atomic weapon by a preemptor would have even greater consequences. Even the lowest-yield nuclear “bunker buster” promises significant local damage and contamination. 40 The National Academy of Sciences estimates that high-yield nuclear attacks on hardened sites could result in hundreds of thousands of casualties depending on the depth of the burst, weather patterns, and the proximity of populations. The resulting fallout–which would be even greater if the target were a nuclear facility–could reach beyond the borders of the country that was attacked to include a neighboring initiator or its allies. 41

In Israel's raid on Osirak, geography played a different role. The distance separating the adversaries made Jerusalem's decision easier, since Baghdad could not easily retaliate. However, in the 2003 Iraq War, geography was less important for another reason: the proximity of Washington's regional allies to Baghdad did not deter preemption because American planners believed that U.S. military superiority and destructiveness would more than compensate.

Another lesson from history is that preemptive military strikes are a logistical nightmare. Absent complete intelligence, a nation cannot be certain that a military attack will sufficiently devastate a rival country's nuclear program. In that respect, Osirak was an anomaly, in that the key elements of Iraq's nuclear infrastructure were clustered within one site. Other would-be nuclear proliferators have learned that lesson well, and today an adversary's nuclear program is likely to be scattered across several hardened facilities. As General McPeak acknowledged regarding North Korea in the early 1990s, nothing less than a “house-to-house search” would suffice. The effectiveness of international inspectors in Iraq after the 1991 war demonstrated that point.

But, even then, another dilemma remains: What has been destroyed can be rebuilt, a fact first evidenced in the initial “successful” sabotage of Norsk Hydro. As former U.N. weapons inspector David Kay once pointedly asked, “How do you roll back knowledge?” 42 During the Cold War, U.S. analysts recognized that China could reconstitute a nuclear program within a few years. Israeli bombers were able to destroy Osirak, but they could not destroy Iraq's technical and scientific infrastructure. Consequently, Iraq was able to reconstitute its nuclear program by the beginning of the 1990s. 43

If nations are unable to deny adversaries the means and know-how to develop nuclear weapons, then the preemptive option with the best chance of long-term success is the World War II template of regime change. Yet, as the current U.S. occupation of Iraq reveals, this is not an option to be considered lightly. The costs of foreign rule and nation building–in terms of lives and economic resources–can be daunting.

The final lesson from the history of preemption is that surgical military strikes can only buy time. That said, buying time should not be as readily dismissed as some argue. 44 Israel's 1981 attack clearly set back Iraq's program. Indeed, time bought by military or other means allows international incentives, disincentives, and/or domestic political changes to curtail nuclear ambitions. Libya's 2003 nuclear surrender proves that, given sufficient time, a country can decide to relinquish its nuclear ambitions. 45 But as North Korea and Iraq in 1981 demonstrated, time purchased, either through diplomacy or coercion, is no sure path to stem nuclear ambitions.

Ultimately, the legacy of Osirak lies in the fact that it effectively legitimized military means as a way to halt proliferation if other measures fail. But as history reveals, preemption is no easy solution. It is just one of many competing tactics in dealing with a predicament that has proven time and again to have few good options. 46

Supplementary Material

Memorandum for the Secretary of Defense: Study of Chinese Communist Vulnerability

Supplementary Material

A Chinese Communist Nuclear Detonation and Nuclear Capability: Major Conclusions and Key Issues

Supplementary Material

Treaty on the Non-Proliferation of Nuclear Weapons