Because of the lack of a widely accepted international jurisprudence governing preventive and preemptive wars, the United States tends to look to experience—our own and that of other democratic nations—with regard to such military actions. Before the invasion of Iraq in 2003, the United States had not initiated a full-scale preventive war.1 Our military had carefully studied the actions of other nations, especially Israel, because it was one of the few modern democracies that had employed preemption successfully, particularly at the beginning of the Six-Day War in 1967, and had then employed prevention in its successful attack against the Iraqi nuclear reactor in 1981. Moreover, Israel had adopted preemption and prevention as important elements in its overall defense strategy.
Perhaps because of its small size, the nature of the threat posed by its surrounding neighbors, and its inability to secure the assistance of the United Nations, Israel, more than any other modern, democratic nation, has served as a laboratory for contemporary preventive and preemptive actions. Some have been widely approved by objective scholars, while others have been condemned.2 Many fall into intermediate categories. A review of some of Israel’s anticipatory actions—the good along with the bad—will provide insights into the challenge of constructing a jurisprudence of preemption.3
From the very beginning of its history, Israel has recognized both its inability to absorb a first blow and the absence of a credible deterrent capacity to persuade some of its enemies that they will lose more than they will gain from striking such a blow.4 It has also understood that it cannot expect the United Nations to protect it from enemy attack,5 and that with regard to international law and international organizations, it lives in a “state of nature.”6 It has operated under Dean Acheson’s dictum “[T]he survival of states is not a matter of law,”7 especially when the law has not been fairly applied to a nation that has been virtually excluded from the law-making and law-applying processes.8 Accordingly, preemption has been an important part of Israel’s strategic policy since at least 1956, with little regard for UN condemnations, which have been plentiful. Israel has sought, however, to operate within the rule of law, as defined by its own legal system, and in the process has contributed to an emerging jurisprudence of anticipatory self-defense. What the late Justice William Brennan said about Israel’s contribution to the jurisprudence of civil liberties may well be applicable to the jurisprudence of preemption:
It may well be Israel, not the United States, that provides the best hope for building a jurisprudence that can protect civil liberties against the demands of national security. For it is Israel that has been facing real and serious threats to its security for the last forty years and seems destined to continue facing such threats in the foreseeable future. . . . The nations of the world, faced with sudden threats to their own security, will look to Israel’s experience in handling its continuing security crisis, and may well find in that experience that expertise to reject security claims that Israel has exposed as baseless and the courage to preserve the civil liberties that Israel has preserved without detriment to its security. . . .
I [should not] be surprised if in the future the protections generally afforded civil liberties during times of world danger owed much to the lessons Israel learns in its struggle to preserve simultaneously the liberties of its citizens and the security of its nation. For in this crucible of danger lies the opportunity to forge a worldwide jurisprudence of civil liberties that can withstand the turbulences of war and crisis.9
In the coming pages I shall consider Israel as a case study in the uses and misuses of preemptive and preventive types of action. As with nearly all such actions, they generally included both proactive and reactive elements, with the former dominant.10
The first war fought by Israel was largely reactive. On the day it declared statehood, May 14, 1948, the new nation was invaded by the armed forces of Syria, Iraq, Lebanon, Egypt, Transjordan, and Saudi Arabia. Attacks from Palestinians, which had been ongoing, persisted. Israel responded and prevailed, but with an enormous toll of “5,682 dead, 20 percent of them civilians.”11 By the time the armistice was signed, the new nation had lost “about 1 percent of the total Jewish population.”12
Attacks on Israel continued after the armistice, and between 1948 and 1956 hundreds of Israelis were killed in cross-border terrorist attacks. Israel adopted a policy of retaliation and deterrence, which had only limited success.13 In 1956, the Egyptian president Abdul Gamal Nasser announced that he had decided to nationalize the Suez Canal and to close to all Israeli shipping the Strait of Tiran, a recognized international waterway. Egyptian military forces had previously fired on and detained commercial shipping headed toward the Israeli port of Eilat, but Nasser’s announcement of a new policy constituted, in the words of Israeli Prime Minister David Ben-Gurion, a casus belli, a cause of war, entitling Israel to open the strait to its shipping by use of military force.14
Despite Egypt’s provocation, it is unlikely that Israel would have acted alone. But Britain and France also regarded Nasser’s decision to nationalize the Suez Canal as an act of aggression warranting a military response. On October 29, 1956, in a plan coordinated with Great Britain and France, Israel invaded the Sinai and easily captured it. Although the Suez Canal attack was in reaction to Egyptian provocation, the war—or campaign, as it was called—was largely preventive in nature: It was designed to stop the Egyptians from exercising control over two international waterways and to keep the Egyptian Army from using the large shipment of weapons it was scheduled to receive from the Soviet bloc.15
The results of the joint invasion were decidedly mixed: a clear military victory for Israel, which enhanced its deterrent capacity; condemnation of Israel’s actions by much of the international community, including the United States; and a diplomatic defeat for Great Britain and France, which suffered both domestic and international condemnation.16 The Strait of Tiran remained open to Israeli shipping, and relative peace prevailed across the Egyptian border for more than a decade.
As a result of its decisive military success in the Sinai Campaign, coupled with the enormous civilian casualties it had suffered during the 1948 war, the Israeli armed forces adopted a “doctrine of warfare [that] was based on two principles: first, a preemptive strike by the air force, and second, the transfer of the war into the enemies’ territories.”17 This preemption doctrine was tested during Israel’s next two wars.
To Preempt or Not to Preempt:
A Comparison between the Six-Day and Yom Kippur Wars
In both 1967 and 1973 Israel faced the decision whether to preempt an expected imminent attack or to absorb the first blow and then respond. Each decision has, with the benefit of hindsight, been criticized as wrong,18 but at the time the decisions had to be made both may well have been correct. A comparison between them affords unusual insight into the appropriate criteria for preemption in the military context.
Israel had to decide whether to attack preemptively in 1967, when Egypt, in coordination with Syria, again closed the Strait of Tiran, expelled UN peacekeepers, massed its regular army on the border, and threatened a genocidal war. According to Nasser, the war was to be not over the Strait of Tiran but over Israel’s “existence,” and “The objective will be Israel’s destruction.”
Hafiz al-Assad ordered his Syrian soldiers to “strike the enemy’s [civilian] settlements, turn them into dust, pave the Arab roads with the skulls of Jews. Strike them without mercy.” He characterized the forthcoming attack on Israel as a “battle of annihilation.” Damascus Radio incited its listeners: “Arab masses, this is your day. Rush to the battlefield. . . . Let them know that we shall hang the last imperialist soldier with the entrails of the last Zionist.”19
Israel attacked preemptively, destroying the Egyptian and Syrian air forces on the ground, and went on to win a decisive victory in six days. Although it also feared an attack from Jordan, it did not act preemptively against that enemy to the east. Indeed, on the morning of June 5, 1967, it passed to Jordan through a UN envoy a message that it would not attack Jordanian forces if Jordan did not attack first. Instead it absorbed the first blows from Jordanian artillery against civilian targets in West Jerusalem and military targets near Tel Aviv and Ramat David.20 Jordanian ground forces also took control of Government House in Jerusalem, threatening Mount Scopus. Israeli authorities believed that “if [the Jordanian Army] could take Mount Scopus and encircle Jewish Jerusalem,”21 it would hold a strong military position. Israel then responded with ground and air attacks. It defeated Jordan as well, though with considerable casualties.22
Much of the world accepted the necessity of, and justification for, Israel’s preemptive attack of 1967, because Israel was seen as an underdog surrounded by hostile Arab nations threatening its destruction. Moreover, Egypt had been the first to commit a casus belli, an act of war, by denying Israeli shipping access to an international waterway and then threatening a large-scale attack by expelling UN peacekeepers, and by massing its troops on the border. The Egyptian Air Force also flew over Israel’s nuclear facility in Dimona, raising the specter of an aerial attack on Israel’s nuclear capacity.23 Nasser had earlier threatened that “the Arabs would take preemptive action” in order to stop Israel from developing nuclear weapons.24 It was Israel, however, that fired the first shot, because as Israeli Prime Minister Levi Eshkol told his cabinet, a war in which “the first five minutes will be decisive” was inevitable, and “[t]he question is who will attack the other’s airfields first.”25 The Israeli Air Force attacked the Egyptian airfields first, and Israel’s air superiority was assured. Its swift and decisive ground victory resulted in fewer civilian casualties than in any comparable modern war.26 The number of Israeli civilians and soldiers who might have been killed in an initial attack from the combined Arab armies was estimated to be quite high.27 It is not absolutely certain, however, that the Egyptians would necessarily have attacked despite their provocative actions.
In his influential book Just and Unjust Wars, Michael Walzer has noted that Nasser intended to make Israel believe that a catastrophic attack was imminent, so as to require it to call up its reserves and destroy its economy as well as its self-confidence. He cited Egyptian documents captured by Israel during the war that suggested a plan by Nasser to maintain “his army on Israel’s border without [actual] war.” This would have achieved “a great victory” because it would have kept the Strait of Tiran permanently closed to Israeli shipping and because “of the strain it would have placed on the Israeli defense system.”
Walzer pointed to the “basic asymmetry in the structure of forces: [T]he Egyptians could deploy . . . their large army of long-term regulars on the Israeli border and keep it there indefinitely; the Israelis could only counter their deployment by mobilizing reserve formations, and reservists could not be kept in uniform for very long. . . .”28 He also pointed to the failure of international diplomacy that made clear “the unwillingness of the Western powers to pressure or coerce the Egyptians. . . . Day by day, diplomatic efforts seemed only to intensify Israel’s isolation.” Finally, he cited the psychological factors: “Egypt was in the grip of a war fever, familiar enough from European history, a celebration in advance of expected victories. The Israeli mood was very different, suggesting what it means to live under the threat: rumors of coming disasters were endlessly repeated; frightened men and women raided food shops, buying up their entire stock, despite government announcements that there were ample reserves; thousands of graves were dug in the military cemeteries; Israel’s political and military leaders lived on the edge of nervous exhaustion.”29
Despite the after-the-fact uncertainty whether Egypt would actually have attacked, Walzer believed that Israeli leaders experienced “just fear” and that Egyptian leaders intended “to put it in danger.”30 Accordingly, Walzer concluded: “The Israeli first strike is, I think, a clear case of legitimate anticipation.”31
The Six-Day War ended with the capture and subsequent occupation of the West Bank, Gaza, and the Golan Heights, a situation that changed international opinion and made future preemptive actions more problematic.
Walzer is plainly correct as a matter of law. In the context of both domestic individual self-defense and international military preemption, the action must be judged by what was known and reasonably believed at the time the action was taken, not by what was later learned or even what was “objectively” true.32 By that criterion Israel’s preemptive attack was a lawful instance of anticipatory self-defense in response to a deliberate provocation and threatened attack by Egypt and Syria.
In 1973 Israel again faced a coordinated attack from Egypt and Syria. The intelligence concerning its enemies’ intentions was not as good as in 1967, and the threat was not perceived as so grave because of the buffer provided by the territories conquered in 1967 and because the threat was not as overt. Perhaps the knowledge, obtained in the aftermath of the Six-Day War, that Nasser may not have intended to attack in June 1967, led some to believe that Sadat did not intend to strike. Moreover, in this case Sadat intended to make Israel believe that it was not planning to attack. Nonetheless, there was a brief opportunity for a preemptive attack against the Egyptian and Syrian air forces, despite the reality that they were better protected than they had been in 1967. At 4:00 A.M. on October 6, 1973, Israeli intelligence received a report from a high-level Egyptian official who was spying for Israel that an Egyptian-Syrian attack was imminent. Several military leaders proposed preemptive action. Indeed, Israel’s Chief of Staff David Elazar “ordered the commander of the air force . . . to prepare a preemptive strike and issued a standby order to attack at 11:00 in the morning.”33 But Defense Minister Moshe Dayan “adamantly opposed a preemptive strike,” observing that “we’re in a political situation in which we can’t do what we did in 1967.”34
In the hours leading up to the Egyptian-Syrian attack, the proposal for preemption was debated in front of Israel’s prime minister, Golda Meir. The case for preemption was presented by General Elazar, who argued that it would save many lives: “Elazar entered Mrs. Meir’s office on that Saturday morning with some persuasive arguments in favor of a pre-emptive air-strike, namely, that a first strike could disrupt and retard the enemy offensive, allow Israel’s army more time to mobilize, destroy at least part of the enemy air-defense systems, and limit Israeli casualties.”35
The negative case was argued by Dayan, who said that a preemptive attack by Israel would make it more difficult to secure American support if needed. He believed that the “political damage Israel would incur by striking first would be far greater, and accordingly he opposed pre-emption.”36 When the presentations were completed, “the prime minister hemmed uncertainly for a few moments but then came to a clear decision. There would be no preemptive strike. Israel might be needing American assistance soon and it was imperative that it not be blamed for starting the war. ‘If we strike first we won’t get help from anybody,’ she said.”37
At the close of the war, U.S. Secretary of State Henry Kissinger told Dayan that Israel “had been wise not to stage a preemptive strike on Yom Kippur. If it had . . . it would not have been received so much as a nail from the United States.”38 Because Israel was willing to absorb the first blow, and because it suffered enormous losses—in both life and in equipment39—the United States agreed to supply it with considerable replacement armaments. Most of the American weapons arrived too late for actual use in the war, but the knowledge that they were on the way allowed the Israeli Army to use all available weapons and ammunition without fear of running out.40
The immediate military cost to Israel of having not preempted is impossible to calculate, but some experts have estimated that the number of casualties suffered by Israel—2,656 dead and 7,250 wounded—would have been considerably lower.41 During the 1967 war, fewer than 800 Israeli soldiers were killed and approximately 2,500 wounded.42 More significant, the ratios of dead and wounded between Israel and its enemies was very different in the two wars. The ratio in the Yom Kippur War was between four to one and seven to one in Israel’s favor, depending on whose figures are credited,43 whereas the ratio during the Six-Day War was “approximately 25 to 1 in Israel’s favor.”44
Military historians and analysts have studied the Israeli decision not to preempt and have concluded that the cost of not preempting was high. U.S. Air Force specialists estimated that “had the air force been permitted to pre-empt, the destruction of 90 per cent of the SAM [surface-to-air missile] sites could have been accomplished in a period of three to six hours for the loss of under ten aircraft.” This would have reduced “Israeli losses on the ground” considerably: “Pre-emptive air strikes on front-line missiles and troop concentrations could certainly have been expected to disrupt the enemy offensive and its communications. Some authoritative writers have argued that the IAF [Israeli Air Force] could have delivered three thousand tons of bombs on enemy targets before the Arab attack reached full strength.”45
One reason why it is impossible to come up with precise estimates of how many lives might have been saved by a preemptive attack is that such an attack could have changed the entire course of the war, perhaps even prevented the attack by Egypt and Syria.46 This must remain highly speculative, especially with regard to the Yom Kippur War because Egypt’s goal was not to win a full-blown military victory in traditional terms but rather to restore its honor by inflicting heavy casualties on the Israeli Army and regaining some of its lost territories before a cease-fire could be imposed. That is why, despite its eventual military defeat, “for Egypt, the war was a towering accomplishment,”47 and its president, Anwar el-Sadat, the man who started the war, emerged as “the clearest victor.”48 Egypt’s “victory” in the Ramadan War (as the Yom Kippur War is called in the Arab world) is still celebrated, despite massive military losses.
It is precisely with regard to wars that are not subject to the usual deterrent calculi of military costs and benefits that preemption becomes an attractive option. Deterrence, at least in theory, is always a preferable way of preventing a war since it relies on the sword of Damocles’s hanging rather than dropping. However, for enemies who do not fear the sword dropping, preemption may be the only realistic option. It will probably always be a matter of degree since almost no enemy is completely oblivious to the threat of retaliation.
The calculations that went into the decisions regarding the Yom Kippur War went well beyond short-term military advantages and disadvantages. Israel would almost certainly have gained considerable short-term military benefits from a preemptive strike against Egypt and Syria.49 But the longer-term risks would have been considerable, especially in regard to its relations with its major ally, the United States.50 One important reason why the risks of preemption would have been so high for Israel was that it had successfully employed that tactic six years earlier and had won an overwhelming victory. Because it now was perceived as so strong, and because it had been criticized in some quarters for starting the shooting in the Six-Day War, Israel was in a difficult position with regard to preempting once again. It feared—understandably, as it turned out—that the international community would not believe that it was in fact acting preemptively once again to ward off an inevitable and imminent attack, but rather that it was using the excuse of preventive self-defense to wage an aggressive war.
This suggests another important potential cost of preemption: that once done—even successfully—it becomes difficult to repeat. This is so for a variety of practical reasons. First, an enemy who is aware of the prior preemptive act can better prepare for a repetition. Sadat went to great efforts to mask his intentions on the eve of the Yom Kippur War so as to deny Israel the opportunity to take effective preemptive measures. (We shall see how Israel’s preemptive attack against Iraq’s nuclear reactor in 1981 caused Iran to take precautions against a similar attack later.) Second, a successful preemptive attack makes it difficult to justify—on moral, legal, political, and diplomatic grounds—a subsequent attack. We saw this in Dayan’s argument against a repeat in 1973 of what Israel had done successfully in 1967. Finally, and related to the second reason, if a nation repeatedly preempts following an initial successful preemption, the propriety of its initial preemption will be retrospectively challenged. This too was part of Dayan’s concerns. We shall see that the United States’ preventive attack on Iraq in 2003 (and its subsequent occupation) have made it more difficult to justify a preventive attack against Iran’s nuclear facilities, even though Iran appears to be much closer to developing nuclear weapons than Iraq was.
Another related factor that may have contributed to Israel’s decisions to preempt in 1967 but not in 1973 is that in 1967 Nasser was overtly threatening to attack and took action designed to make Israel, and the world, believe that an attack was imminent (even if it was not). The world would thus be more likely to accept Israel’s preemptive justification than it would in 1973, when Sadat intended that Israel, and the world, believe that he was not planning an attack (even though he was). Accordingly, a preemptive attack may be more acceptable in the face of an overt threat, even if intended as a bluff, than in the face of a well-concealed sneak attack.
In sum, therefore, an analysis of the Six-Day and Yom Kippur wars suggests that a major factor in deciding whether to undertake a preemptive attack will generally be the short-term military advantage secured by striking the first blow. As Shimon Peres, the former prime minister and defense minister of Israel, has written, it is the duty of a leader to “meet [war] under the least dangerous conditions,” which necessarily must include the option of a preemptive attack.51 But there will be other considerations as well, including political, diplomatic, legal, moral, humanitarian, and prudential. Moreover, there is always the possibility that firing the first shot—even on the basis of the best intelligence and with the highest of motives—may turn out to provoke a war that would not otherwise have begun.
The burden of justification should always be on the nation that acts preemptively or preventively. In satisfying that burden, the anticipatory actor can point to the certainty of an attack, its imminence, and the extent of the damage it would suffer from absorbing the first blow, both militarily and to its civilian population. Other critical factors include the nature of the preemptive action (a single decisive first strike against a military target as distinguished from a full-scale war with massive casualties and a long-term occupation); the limited damage to noncombatants it would inflict with its preemption (especially as compared with absorbing the first blow and then retaliating); the likelihood of a shorter, less damaging war, if it preempts; and other ineffable factors. In the end the decision will always be dependent on the quality of intelligence and an assessment of probabilities.52 It will also turn on the comparative values placed on the lives of a nation’s own civilians and soldiers and those of its enemies. It will always be a matter of degree and judgment. But no law or rule of morality will ever succeed in prohibiting all preemptive military actions. Nor should it.53
The Rescue of Hostages at Entebbe
In June 1976 Palestinian terrorists hijacked an Air France commercial jet flying from Athens to Paris. This flight had originated in Tel Aviv. The terrorists then flew the airplane to Entebbe, Uganda, where they threatened to kill the Israeli passengers unless their demand for the release of prisoners was met. Although the Israeli decision to attempt to rescue the hostages was not strictly a preemptive action since it was taken in response to an ongoing act of aggression, it was designed to prevent the killing of the passengers. It thus included preemptive elements. Moreover, the decision to authorize a military attack on foreign soil was based on the same kinds of complex factors that frequently inform purely preventive actions. And some critics of Israel within the international community accused Israel of waging an unlawful preventive aggression against the territorial integrity and sovereignty of Uganda.54 It is thus useful to include this military action among the preemptive decisions to be evaluated. Moreover, the type of military action Israel took—precise, targeted, and singular—will probably have to be reprised by other countries as hostage taking increases, as it is in Iraq, Chechnya, and other hot spots around the world. The United States now has well-trained teams, modeled on the team sent into Entebbe, ready to rescue hostages in the right circumstances. Its most visible effort—the attempt to rescue its embassy personnel held captive by Iranian radicals in Tehran—failed as the result of technical problems.
The prime minister of Israel at the time was Yitzhak Rabin. His daughter, Dalia, told me that prior to the commencement of the rescue operation Rabin wrote a letter in which he offered to resign if the mission failed. He defined failure as twenty or more deaths. Obviously these did not include the deaths of the terrorists or the Ugandan soldiers who were protecting the terrorists. I do not know whether the figure of twenty included Israeli soldiers, Ugandan civilians, or only the hostages, but it probably was limited to Israelis and other innocents.55
Whether or not Rabin and the others who made the risky decision engaged in an explicitly quantitative and qualitative cost-benefit analysis, it seems likely that they weighed several factors, including the likelihood that some or all of the hostages would be killed if nothing were done; the likelihood that Israel would be forced to make concessions if some of the hostages were killed and the others threatened; the probability that some or all of the hostages, plus some or all of the rescuers, would be killed or injured if the rescue was attempted and failed; the likelihood that some of the hostages and rescuers would be killed or injured even if the rescue “succeeded”; the benefits to the prestige and deterrent credibility of Israel in the event of “success”; and the loss of deterrent credibility and prestige in the event of failure.
The daring military action, a secret flight by a large transport loaded with jeeps full of well-trained Israeli commandos who caught the Ugandans and the terrorists by surprise, succeeded and almost certainly prevented the deaths of more hostages, but at some cost in terms of the deaths of three hostages and one Israeli soldier.56 Among the benefits, in addition to the lives probably saved, was Israel’s enhanced deterrence against terrorism outside its borders.
There was also some loss in terms of diplomatic criticism of Israel for violating “Ugandan sovereignty,” despite hard evidence that Uganda’s murderous leader, Idi Amin, was actively assisting the hijackers. The secretary-general of the UN, Kurt Waldheim, condemned Israel. France, whose plane had been hijacked, did not even thank Israel. And many UN member nations joined a chorus of one-sided criticism directed against the successful efforts of the Jewish state to rescue 103 of its citizens who had been threatened with execution. An “Afro-Arab resolution condemning Israel’s alleged violation of Ugandan sovereignty and territorial integrity” did not, however, garner sufficient votes and was ultimately withdrawn, but not before Israel was absurdly accused of invading Uganda as part of a plot to implement an old plan, dating to 1903, to establish a Jewish state in Uganda.57
Several years after the successful rescue mission, an interesting analysis of the decision-making process was published in the Journal of Conflict Resolution.58 It disclosed that Israel had considered several options, including negotiation with the terrorists, who were demanding release of Palestinian prisoners in Israel, Germany, France, and Kenya. At the beginning Israel also considered trying to persuade Idi Amin to secure release of the hostages, but it soon became clear that he was fully cooperating with the terrorists and would not help Israel. It was also clear that the specific group in charge of the hostages would not hesitate to kill them all; they had previously killed hostages, including an American diplomat, and they had separated Israelis from the others in preparation for beginning to kill the Israelis.59
Rabin was initially opposed to a rescue operation because the chief of staff of the army told him that none of the military options “had a good chance to succeed, due to lack of information regarding some critical details.” He decided instead to begin negotiations with the hijackers on the release of Palestinian prisoners. Defense Minister Peres strongly disagreed, arguing that “Israel could not afford to capitulate in the face of blunt blackmail. Such an exchange would severely damage Israel’s capacity to fight terrorism in the future and would lead to an increased number of terrorist attacks on Israeli citizens.”60
A decision to begin negotiations was made, while the Israelis were preparing for a military option. Israel intelligence soon began to fill the information gaps, and the new information led Israeli military planners to raise the likelihood of a successful outcome, especially after they learned that the terminal in which the hostages were being held had not been wired with explosives, as originally feared. Rabin then changed his opinion and told his advisers: “We now have a military option with a high probability of success. . . . I ought to stress, however, that success cannot be assured with certainty. Even if the military force can overcome the terrorists, we should expect between 15 to 20 fatalities among the hostages and the raiding forces.” He acknowledged the “grave consequences” of failure, including a very high death toll and diplomatic condemnation, but he recommended that “we take the military option, because I believe that it has a good chance of success.”61
Although the author of the article in the Journal of Conflict Resolution acknowledged that the Israeli decision makers did not perform a precise cost-benefit analysis, he found that an “analysis of individual and intraorganizational revision suggests that both individuals and planning teams closely approximated, in an intuitive fashion, Bayesian approach* to revision of opinion in light of new information.”62
The author then provided his own explicit Bayesian analysis, replete with numbers and complicated formulas, and determined that the intuitive decisions actually made were remarkably close to the Bayesian decisions that would have been made had an explicitly mathematical model been employed. He concluded, therefore: “The Entebbe case demonstrates that individuals can be vigilant explorers of multiple options, competent information processors, and comprehensive evaluators of multiple values under the most stressful circumstances. The evidence suggests awareness of and willingness to accept highly sensitive value tradeoffs and efficient revision of opinion in light of new information.”63 He then offered a caveat: “These findings run contrary to most of the evidence regarding choice processes under crisis conditions, and may suggest that decision-making in the Entebbe crisis is the exception rather than the rule.”
The Destruction of the Iraqi Nuclear Reactor in 1981
The Israeli military action most frequently cited in the current debate over preemption is the one-strike destruction of Iraq’s nuclear reactor in 1981. As Professor Michael Reisman put it in the American Journal of International Law, “[international law has been grappling with the claim of preemptive self-defense for decades. . . . The Israeli destruction of the Osirak reactor near Baghdad in 1981 was a quintessential preemptive action.”64 Although the UN Security Council unanimously condemned Israel’s attack as “in clear violation of the Charter of the United Nations and the norms of international conduct,” as we shall see, Israel’s preemptive action has—over time—become accepted as a proper and proportional example of anticipatory self-defense in the nuclear age. It is extremely relevant therefore to current options under consideration, especially with regard to Iran’s efforts to become a nuclear power discussed in detail in Chapter 6.
Five years after Israel’s successful rescue at Entebbe, a new government faced an even more daunting preventive decision: whether to bomb and destroy Iraq’s nuclear reactor. That decision is a paradigmatic example of a pure preventive (or preemptive, as we shall see) attack short of full-scale war. It also illustrates the limitations of the “imminence” and “certainty” requirements. The reactor was not yet completed or operational. Iraq possessed no nuclear devices, yet it was threatening to produce one and to deploy it against Israel.
Following a failed Iranian attack on the Iraqi nuclear reactor on September 30, 1980, the official Baghdad newspaper al-Jumhuriya assured the Iranian people that the nuclear reactor posed no threat to Iran: “The Zionist entity is the one that fears the Iraqi nuclear reactor [which] constitutes a grave danger for ‘Israel.’ ” Another Iraqi newspaper was even clearer: “The Iraqi nuclear reactor is not intended to be used against Iran, but against the Zionist enemy.”65 Iraqi Deputy Prime Minister Tariq Aziz repeated the threat in an interview with a Jordanian journalist.66
The threat to use nuclear weapons against Israel, though clear, was neither imminent nor certain. Israel, with its own well-developed nuclear weapons, certainly had a significant deterrent capacity despite its official unwillingness to confirm its status as a nuclear power.67 Moreover, Saddam Hussein’s Iraq was a secular tyranny that was amenable to material threats (as distinguished from a fundamentalist religious regime whose leaders claim to welcome death as a prelude to a better world). Because Iraq was still technically “at war” with Israel—it had adamantly refused to sign on to the various cease-fires between Israel and its immediate neighbors—it proclaimed the right to continue the ongoing war at a time and place of its choice, as it subsequently did during the First Gulf War when it rained Scud missiles on Tel Aviv. Israel believed therefore that it too was legally free to attack Iraqi military targets at will.68 Politically and diplomatically, however, there were significant risks associated with any attack, since Iraq had not engaged in any casus belli against Israel for eight years (since 1973). There were also military and other risks involved in any decision to bomb a target seven hundred miles away and separated by hostile enemy territory.
In spite of the lack of imminence and certainty to the threat posed by the Iraqi reactor, the Israeli government after much consideration concluded that the potential benefits outweighed the risks because the magnitude of the threat (even discounted by the lack of imminence and certainty) was so cataclysmic.69 Israel’s prime minister, Menachem Begin, who had lost much of his family during the Holocaust, feared that “another Holocaust would have happened in the history of the Jewish people.” He pointed to intelligence estimates that Saddam Hussein could develop “between three and four Hiroshima-type nuclear bombs” and warned that “we shall not allow any enemy to develop weapons of mass destruction turned against us.”70
In other words, even if there was only a 5 percent likelihood that Iraq would deploy a nuclear weapon against Israel in the succeeding ten years, the magnitude of the potential harm—hundreds of thousands, perhaps even millions of deaths and injuries to its civilian population—was more than enough to justify a preemptive strike. Moreover, a dirty bomb might have made significant portions of Israel’s tiny landmass unlivable for generations.
The timing of the strike also involved careful calculations of the costs and benefits of waiting until the Iraqi reactor was closer to completion and the threat therefore more imminent and certain. As the reactor moved toward activation, the risks to Iraqi civilians became greater in the event of an Israeli bombing. The window of opportunity was quite narrow for destroying the reactor before the risks to civilians became too great. Prime Minister Begin stated unequivocally that Israel would not have bombed the reactor once it became “hot,” which, according to intelligence sources, was expected to occur in July or September 1981: “No Israeli government could contemplate bombing [when] such an attack would have brought about a massive radioactive fallout over . . . Baghdad [in which] tens of thousands of innocent residents would have been hurt.”71 The commander in chief of Israel’s armed forces apparently told Begin that he would not have obeyed an order to bomb the reactor under such circumstances.72 Begin went so far as to say he did not think that if Tel Aviv had been destroyed by an Iraqi nuclear bomb, he could agree to a retaliatory attack against Baghdad: “That is our morality.” As he put it in a related context, “The children of Baghdad are not our enemy.”73
Michael Walzer has observed that: “The Israeli attack on the Iraqi nuclear reactor . . . is sometimes invoked as an example of justified preventive attack that was also, in a sense, preemptive: the Iraqi threat was not imminent, but an immediate attack was the only reasonable action against it.”74
Imminence, according to this view, involves more than merely the temporal immediacy of the threat itself; it is a function of the temporal opportunity for carrying out the preventive action, as well as the seriousness of the feared harm. For many pragmatic international law scholars, the Israeli strike against the Iraqi nuclear reactor represented both a paradox and a challenge: It produced what many believed was a positive result, but it was difficult to square with existing international law. Indeed, Kenneth Adelman, who was part of the American delegation to the UN when the condemnation occurred, described the U.S. vote as a “big mistake” about which he was “mortified.” He now regards the Israeli attack as “the most wonderful example of preemption in the modern era.”75 “Thank God Israel did that, looking back at it [sic]. The idea of Saddam Hussein being the main Arab leader with nuclear weapons since 1985 is frightful.”76
The debates over whether to condemn Israel for its bombing of the Osirak reactor, both within the U.S. administration and at the UN, evidence the changes in attitudes about anticipatory self-defense that have occurred within less than a quarter century. Immediately after the bombing, hawkish Republicans, some of the same people who now support the invasion of Iraq, took the lead in denouncing Israel. They were led by Secretary of Defense Caspar Weinberger and included Attorney General Edwin Meese.77 *
Leading the chorus of European criticism of Israel was former British prime minister Margaret Thatcher who declared: “Armed attack in such circumstances cannot be justified. It represents a grave breach of international law.” She apparently forgot about that reading of international law when she urged President Bush not to “go wobbly” on Saddam Hussein. Prior to the joint U.S.-British attack on Iraq, Thatcher advocated “a major deployment of ground forces as well as sustained air strikes.”78 Her article was subheadlined: “Saddam must go. It’s bad enough that India and Pakistan have nukes.”79
The defense of the Israeli action was led by such Democratic senators as Edward Kennedy and Alan Cranston but included several Republican senators as well. One Republican senator, Larry Pressler of South Dakota, changed his mind after hearing the evidence and concluded his statement with a prescient observation: “I began in these hearings to be somewhat of a critic of Israel. But as time has passed, I have come to believe, based on the information Israel had, it probably did the only thing that a country could do, and probably something our country down the road will do at some point.”80
Some experts on international law offered analogies to other preemptive actions, such as the U.S. blockade of Cuba and the Israeli rescue of the hostages at Entebbe. They pointed to Israel’s limited purpose of destroying a nuclear reactor that would have placed the very “existence of a small state such as Israel . . . in jeopardy” and distinguished this attack from one on “a plant that manufactured tanks or conventional artillery, because of the enormous destructive potential of nuclear weapons.”81
Former Supreme Court Justice Arthur Goldberg weighed in with an influential legal analysis that justified the Israeli attack not only as preemptive, but also as part of an ongoing war: “In light of the fact that by its own decision Iraq deems itself to be at war with Israel, the State of Israel, under established rules of international law, has the right to take military action, including bombing, against installations in Iraq which potentially may assist Iraq in its proclaimed war-like designs. It is not necessary, in my understanding of applicable rules of international law, for Israel to prove that the nuclear installation in question is producing nuclear bombs. It is sufficient that this nuclear installation potentially may be of assistance to Iraq in its announced program designed to undermine the security of the State of Israel.”82
Notwithstanding these and other arguments, both the U.S. administration and the UN condemned the attack on the reactor in what Israel characterized as a “festival of hypocrisy.” Perhaps the most hypocritical of the arguments was that Israel could not be absolutely certain that delaying the attack until after the reactor became hot would actually have killed thousands of Iraqi civilians.83 It was answered by Congressman Jonathan Bingham: “. . . the critics of Israel are now arguing that the danger to Baghdad, if the reactor was bombed while operational, was not so great. If Israel had waited until that reactor was hot and had then bombed it, those same critics would have increased the decibels of their criticism by about tenfold, for the heartlessness of Israel in bombing a live, hot reactor.”84
The decision to bomb the Osirak reactor demonstrates that preemption, if implemented cautiously, often carries the prospect of killing fewer innocent civilians than does deterrence. By its nature, preemption is generally directed against military targets (with the ever-present possibility of inadvertent civilian casualties), whereas deterrence is by its nature often directed at civilian targets. The Cold War concept of mutually assured destruction promised tit-for-tat retaliation: If you bomb our population centers, we’ll bomb your population centers. The deterrent threat would be far less effective if it merely promised retaliation against military targets for an attack on civilian targets. Moreover, mutually assured destruction gives an advantage to amoral totalitarian states over moral democracies. The former would not hesitate to retaliate against a first strike by dropping a nuclear bomb on an enemy city. A moral democracy, on the other hand, might hesitate on ordering such tit-for-tat retaliation, as evidenced by Menachem Begin’s statement about the children of Baghdad’s not being his enemy. The threat of mutually assured destruction worked for the United States during the Cold War because it had proved its willingness to drop nuclear bombs on enemy cities at the end of World War II. It might work less well for Israel, because the Israeli Air Force has never deliberately targeted a large civilian population center,85 and its leaders have said its morality would not permit it to do so.86
Deterrence, however, has the virtue of greater certainty and fewer false positives. The retaliation does not occur until after the enemy attacks, thereby reducing the risk of error. Preemptive action, on the other hand, must always be based on predictions, and predictions always carry with them the risk of false positives. Thus preventive attacks may well occur in situations in which what is sought to be prevented might never have occurred.87 Moreover, the line between prevention and deterrence in the context of military action will not always be clear, since the threat of acting preventively may itself serve as a deterrent. It may also serve as a stimulus to belligerent action. The issues are extraordinarily complex, as we shall soon see.
The Invasion of Lebanon in 1982
Following Israel’s successful use of a decisive, single-strike preemptive attack against a uniquely dangerous potential weapon, Israel engaged in a very different sort of preventive action, the large-scale, long-term occupation of southern Lebanon in an effort to prevent terrorist attacks against its northern towns and to end the Palestinian Liberation Organization (PLO) presence in Lebanon. The war in Lebanon, which began in June 1982, is among Israel’s most controversial military actions, both domestically and internationally. It remains controversial within Israel because many Israelis believe that Ariel Sharon, who was the defense minister, misled his own prime minister, Menachem Begin, as well as the Israeli people, about the intended scope and duration of the invasion. The original mission, as approved by the cabinet, was extremely limited:
“(a) The IDF is entrusted with the mission of freeing all the Galilee settlements from the range of fire of terrorists, their Headquarters and bases concentrated in Lebanon. (b) The operation is called ‘Peace for Galilee.’ (c) During the implementation of the decision the Syrian army should not be attacked unless it attacks our forces. . . .” The goal, according to Sharon, was “to remove the ‘terrorists’ from firing range of Israel’s northern border, ‘approximately 45 kilometres.’ ” Beirut was “out of the picture.” The entire operation was supposed to last forty-eight hours.88
It turned out very differently, with Israeli troops surrounding Beirut. The siege of Lebanon’s capital lasted seventy days. The Israeli Air Force also engaged the Syrian Air Force, first destroying its ground-to-air missile system and then shooting down ninety-six Syrian F-15s and F-16s, without the loss of a single Israeli plane. The PLO eventually abandoned Beirut, with nearly fifteen thousand Palestinians, along with Yasser Arafat, departing by the end of August 1982. At the same time Bashir Gemayel, a Lebanese Christian, was elected president of Lebanon and signaled his intention of signing a peace treaty with Israel following his scheduled inauguration on September 23, 1982.
Up to this point Israel’s (or Sharon’s) preventive war seemed to be achieving all its goals—and more. Then, on September 14, 1982, Gemayel was assassinated, and Sharon’s plans unraveled. Israeli troops poured into Beirut to prevent a civil war, and “the Maronite [Christian] Phalangist militia entered the [Palestinian] refugee camps of Sabra and Shatilla to ‘clean out’ the 2,000 PLO guerrillas who, according to reports, were still hiding there.”89 But the Falangists went beyond cleaning out the guerrillas. They exacted revenge for the murder of their elected leader by massacring between six and seven hundred Palestinians, including civilians. Although the Israeli armed forces did not participate in the massacre, a commission of inquiry imputed responsibility to Sharon for not stopping the Falangists, who he knew, or should have known, would—in the long tradition of Lebanese blood feuds—seek revenge against those they believed responsible for the murder of their leader.
Following the events at Sabra and Shatilla, the situation in Lebanon deteriorated. Israeli attitudes toward this war were different from those toward any previous war because this was widely viewed as a war of choice, rather than as one forced on Israel by impending aggression. It was the first war that provoked conscientious objections from some soldiers and widespread demonstrations throughout the country.90 This war was different because it was almost entirely preventive rather than preemptive in nature, designed to strengthen Israel’s position in the event of future attacks. The 1967 war had been preemptive, and the 1956 campaign, though also largely preventive, had come at a time when Israel reasonably believed that an attack by a well-armed Egypt was only a matter of time. Moreover, the 1956 war had reactive elements, since Egypt had blocked access to an international waterway and was, at the very least, facilitating cross-border attacks against Israeli civilians by fedayeen. The 1982 war was also at first reactive to terrorist attacks from southern Lebanon, but its scope and duration exceeded by far any reasonable response to these threats.
Summarizing the Israeli-Arab War Experience: A Swinging Pendulum of Preemption
The 1982 war was the last of Israel’s wars against foreign enemies. The remaining “wars” have been against terrorists, primarily from within the occupied territories of the Gaza Strip and West Bank. These wars did not involve large-scale attacks by regular armed forces, with airpower and artillery. Combating terrorism requires different kinds of responses, many of which are also preventive or preemptive in nature, but on a more micro rather than macro scale. Before we turn to this somewhat different, though related, set of issues, it may be useful to draw some conclusions with regard to the Israeli’s uses and misuses of prevention and preemption in fighting its external enemies.
This history can be represented by the image of a pendulum with wide swings. Israel’s first war, the War of Independence, was largely reactive. Israel fought back against a multinational aggression and in the process suffered large numbers of civilian and military casualties. Israel’s next large-scale military operation, conducted in cooperation with Great Britain and France, was undertaken in response to Egypt’s blockading of an international waterway and its facilitation of cross-border fedayeen attacks, but it also had preventive aspects, though there was no real preemptive justification. As the result of these initial experiences, the Israeli military developed a strategy that emphasized two elements: preemption, when appropriate, and fighting battles outside its territory and away from civilian population centers. This strategy was tested in the 1967 war and succeeded beyond all expectations. But despite its success, perhaps because of it, this strategy could not be employed when Israel received several hours of warning before a planned attack by Egypt and Syria in 1973. It fought an entirely reactive war after being attacked on Yom Kippur and suffered considerable casualties. In 1976 it reacted to the hijacking of an Air France commercial airliner with Israeli passengers who were threatened with execution by proactively sending in a rescue team. In 1981 it took preemption to a new level, by attacking and destroying the Iraqi nuclear facility just before it was likely to become “hot” and thus no longer vulnerable to preemptive attack. Despite unanimous condemnation by the Security Council, virtually all Israelis regarded the Osirak attack as a great success that was both morally and legally justified. Then, a year later, Israel overplayed its hand and engaged in a purely preventive large-scale war and occupation of Lebanon, which had some immediately positive but many long-term negative results and which is more difficult to justify either morally or legally.
This decidedly mixed history suggests that the distinction between preventive and preemptive military action is important. When Israel had little choice but to attack or face potential devastation, its preemptive actions were justified and generally successful. Its one failure to take preemptive action when faced with an imminent attack in 1973 proved costly and militarily wrong, though diplomatically understandable. But when it engaged in preventive wars, in 1956 and 1982, the results were more questionable and the condemnation was more warranted.
* Bayesian logic “deals with probability inference: using the knowledge of prior events to predict future events.” TechTarget Network, “Bayesian logic,” http://whatis.techtarget.com/definition/0,,sid9_gci548993,00.html.
* The same Caspar Weinberger enthusiastically supported the preventive invasion of Iraq in 2003 and condemned those who would have “ignored” reports “about Iraq’s efforts to acquire biological and chemical as well as nuclear weapons.” Yet he would have required Israel to “ignore” much more solid intelligence not only about Iraq’s “efforts” to acquire nuclear weapons but also about its stated intention to use them against Israel. As far as I can determine, he has never tried to reconcile these apparently inconsistent positions. Nor has Meese reconciled his hawkish support for the U.S. attack on Iraq with his condemnation of Israel for its attack. Even before the U.S. invasion Meese argued that al-Qaida’s terrorist attack of September 11, 2001, justified “going after [Hussein], certainly going after his headquarters, going after his military establishment and if in the course of that we have to hit him, I think that’s very appropriate.” (Fox News, Hannity and Colmes, March 11, 2003.)