Keeping a Tight Lid on Pandora’s Box

The Nuclear Taboo: The United States and the Non-Use of Nuclear Weapons Since 1945
By Nina Tannenwald
Cambridge University Press, January 2008, 472 pp.

Reviewed by William Burr

In the spring of 2006, when rumors spread about a possible U.S. military strike against Iran's nuclear facilities, The New Yorker's Seymour Hersh reported that the Bush administration asked the Pentagon to include nuclear strikes among its options. The Joint Chiefs of Staff, however, strongly opposed any plans to target Iran with nuclear weapons; Hersh's sources told him that "there are very strong sentiments within the military against brandishing nuclear weapons against other countries." According to one of the sources, "[I]f senior Pentagon officers express their opposition to the use of offensive nuclear weapons, then it will never happen."

What happened next, the degree to which the Joint Chiefs prevailed in this debate about contingency plans for Iran, is unclear, but the claim that the U.S. high command strongly opposed the use of nuclear weapons resonates with the arguments made in a remarkable and long-awaited book by Brown University political scientist Nina Tannenwald. Her subject is what she sees as the "the single most important phenomenon of the nuclear age": the nonuse of nuclear weapons since 1945. (Truth-in-reviewing notice: I read and commented on an early draft of this book.) Noting that many Cold War analysts took it for granted that nuclear weapons would be fired in anger during a crisis, Tannenwald seeks to explain why the United States did not use nuclear weapons during hot wars in Asia and the Middle East. The United States was not the only nuclear-weapon state, but as a superpower with global security and economic interests, it was involved in significant military confrontations where nuclear weapons use was under consideration and where use was threatened. Despite the focus on Washington, Tannenwald believes that her findings are relevant to the experience of other countries and makes some generalizations in her conclusions.

Acknowledging that some analysts have argued that self-interest and prudence, such as the danger of nuclear war, sufficiently explain the nonuse of nuclear weapons, Tannenwald finds the realist explanations incomplete. Although strategic interests are important, Tannenwald argues that the historical record reveals that the idea of nuclear weapons use became tarred with "moral opprobrium." To explain why U.S. officials did not use nuclear weapons, Tannenwald develops a complex argument about an evolving "nuclear taboo." Drawing on anthropological theory that a taboo is "something that is not done, not said, or not touched," Tannenwald sees the nuclear taboo as a "de facto prohibition against the first use of nuclear weapons." Nuclear use, except in extremis, in retaliation to a nuclear attack, has become taboo because the weapons themselves have become stigmatized as "illegitimate and abhorrent." Violating the taboo, using nuclear weapons, would open a Pandora's box involving great danger and potentially terrible consequences. According to Tannenwald, once nuclear weapons are used, once the "bright line" is crossed, the peril unleashed would put one "immediately in a new world."

The nuclear taboo is a "normative belief about the behavior" of nonuse, that is, U.S. leaders came to believe that a taboo against first use exists. Tannenwald identifies three "pathways" that contributed to the formation of the taboo. One was the force of domestic public opinion. Another was world opinion, which exists "independently of the preferences of dominant states." The third was the personal convictions of the decision-makers themselves. All three helped create a nuclear taboo that constrained the actions of U.S. leaders during crisis and war.

How a "moral norm" proscribing the use of nuclear weapons developed during the decades after the Hiroshima and Nagasaki attacks takes up seven chapters of this book. Looking closely at the evolution of "discourse, institutions, and behavior," the author traces the beginning of the stigmatization of nuclear weapons to the years after World War II. Not unexpectedly, Tannenwald finds the roots of the taboo in the ethical tensions raised by the first use of nuclear weapons. The atomic strikes may not have been as decisive for Japan's surrender as she suggests (for example, Tsuyoshi Hasegawa's work indicates a more complex picture),[1] but Tannenwald is on firm ground when she belies the familiar view that President Harry Truman had no second thoughts about the bombings. Indeed, she finds him deeply troubled, disliking "the idea of killing...‘all those kids'" to the point that he called off further atomic attacks.

Truman was inconsistent (fire-bombing of Japanese cities also produced mass civilian casualties), but he nevertheless began putting nuclear weapons in a special category and kept them under tight civilian custody. He opposed military custody of the weapons because they were "horribly destructive" and "not for military uses." Although he eventually permitted military planning for the use of nuclear weapons and made decisions to expand the stockpile, as David Rosenberg has argued, Truman continued to see atomic bombs as weapons of last resort.[2] His reactions paralleled the development of world public opinion, which increasingly found nuclear use repugnant.

More or less simultaneously, the UN Commission for Conventional Armaments stigmatized the bomb by developing the concept of "weapons of mass destruction" (WMD), which linked nuclear weapons to biological, chemical, and other unconventional weapons that had "characteristics comparable in destructive effect to those of the atomic bomb." Tannenwald sees the emergence of the WMD category as providing fertile soil for a nuclear taboo to "take root."

The Korean War provided a test case for the developing and still tentative taboo. As the crisis unfolded, the Truman administration secretly deployed nuclear weapons components in the western Pacific, and after Beijing intervened militarily, Truman unleashed a furor by stating publicly that nuclear weapons were "under active consideration." Although military planners identified suitable targets, ethical concerns made Truman again loath to consider nuclear weapons use, and top Department of State advisers worried about the political costs of using nuclear weapons for a second time against Asian people. World opinion would put the "mark of Cain" on the United States for years to come.

Tannenwald sees normative and political considerations against nuclear weapons use in Korea as reinforcing a policy of restraint (limited war) to prevent escalation to global conflict with North Korea's Chinese and Soviet partners. This understanding of White House policy is persuasive, but Tannenwald may go too far in downplaying the concern that nuclear weapons use could escalate the war. She argues that Truman's advisers were divided about the risks of a wider war, but it is not clear that those divisions were all that important because the person whose opinion really mattered to the president, Secretary of State Dean Acheson, was worried about the escalation danger of nuclear weapons use. (Interestingly, she presents no evidence on Secretaries of Defense George C. Marshall or Robert Lovett, who must have weighed in on nuclear issues.) Thus, cost-benefit explanations, which Tannenwald deems "materialist," are likely to have been more relevant to Truman's decisions on Korea than she concedes.

World opinion was a political constraint, Tannenwald argues, that inhibited President Dwight Eisenhower from considering nuclear weapons use in Korea. Eisenhower and Secretary of State John Foster Dulles believed that nuclear weapons were far more usable than Truman did and presided over nuclear contingency planning for use if the armistice talks broke down. Nevertheless, they perceived, in their words, a "tabu" against nuclear use that was difficult to overcome. As Dulles acknowledged with respect to Korea, "[I]n the present state of world opinion we could not use an A-bomb." As much as Eisenhower and Dulles wanted to reverse the nuclear taboo and make nuclear weapons as usable as other munitions, Tannenwald sees these two as constrained by outside forces, against their own wishes.

Beginning in the mid-1950s, Tannenwald sees important developments, especially the global movements against nuclear weapons and weapons testing, strengthening the taboo: "by castigating nuclear weapons as abhorrent weapons and calling for a halt to the nuclear arms race, the peace groups helped to stigmatize nuclear weapons and to delegitimize them as acceptable weapons of war." This is an important part of her argument; by holding that social pressure and public opinion were crucial to the creation of the nuclear taboo, Tannnenwald challenges the view that institutional norms are "created mainly by and for the powerful." Thus, by the end of the 1950s, despite White House efforts to influence public opinion, top policymakers such as Secretary of Defense Neil McElroy acknowledged that world opinion was so strongly against nuclear weapons that it could make them "unusable in war, as proved to be the case with chemical weapons."

During the 1960s and the following years, Tannenwald sees the "institutionalization" of the taboo. Not only did the Kennedy and Johnson administrations initiate "flexible response" strategies designed to avoid or, at least, defer nuclear weapons use in a European confrontation, Kennedy presided over the creation of the Arms Control and Disarmament Agency, which provided a voice for nuclear arms restraint in the federal government. Moreover, Secretary of Defense Robert McNamara privately advised Kennedy and President Lyndon Johnson never to initiate the use of nuclear weapons, advice that both apparently accepted. On a global scale, the stigmatization of nuclear weapons continued: Latin Americans in 1967 created the world's first nuclear-weapon-free zone, and states began signing the nuclear Nonproliferation Treaty (NPT) in 1968.

In this context, the hot war in Vietnam became what Tannenwald sees as another important test for the taboo. In an arresting chapter, Tannenwald looks at the role (or nonrole) that nuclear arms played in policymaking during the Johnson and Nixon administrations. Top officials in each administration were willing to make occasional veiled nuclear threats against Hanoi, but apparently Johnson also found nuclear use morally unacceptable and was determined to have no Hiroshimas on his watch. Thus, when some Pentagon and White House officials began to look at nuclear options during the 1968 siege of Khe Sanh, Johnson put a halt to all contingency planning. As Tannenwald observes, Johnson's view was that "not only should nuclear weapons not be used, nuclear options should not even be studied." Although RAND Corporation analyst Sam Cohen strongly argued that nuclear weapons had military utility in Vietnam, no one would pay any attention to him.

The hard-nosed President Richard Nixon and Secretary of State Henry Kissinger may not have worried about the morality of using the bomb in Vietnam and possibly looked at nuclear options during 1969. Nevertheless, the nuclear taboo functioned as an "instrumental not internalized constraint" on their actions; "anticipated domestic and world public condemnation" continued to keep nuclear weapons off the table.

Whatever Nixon's and Kissinger's personal views about using nuclear weapons were, they strengthened nonuse norms through arms control negotiations with Moscow. The 1972 Anti-Ballistic Missile (ABM) Treaty was especially important because it amounted to a "strategic no-first use agreement" by codifying mutual deterrence. With the treaty leaving each country undefended from strategic attack, mutual survival depended on the prospect that neither side purposefully intended to launch their bomber or missile force. Tannenwald argues that "superpower self-interest" was critically important to strengthening the taboo.

In Tannenwald's account, the revived anti-nuclear controversies and movements of the late 1970s and early 1980s produced what she sees as "de facto denuclearization." The furor unleashed by the U.S. government's unsuccessful effort to produce and deploy the neutron bomb (Enhanced Radiation Warhead [ERW]), a weapon designed to destroy people but not structures, showed the continuing deep public antipathy to nuclear weapons. That Pentagon officials saw the ERW as a more credible threat was precisely what made it objectionable to others, the risk that it would make it easier to use nuclear weapons.

Also reinvigorating anti-nuclear sentiments and pressures for arms control was the Reagan administration's military buildup and public anxiety over the dangers of a reinvigorated Cold War. Pressures for a nuclear freeze and condemnations of the morality of nuclear first use and nuclear threats (deterrence) by the National Conference of Catholic Bishops and secular organizations were signs of the time. So was a critique that linked morality and interests by arguing that a strategy based on monstrous capabilities to destroy human existence was inconsistent with a realistic conception of security. Tannenwald finds that this thinking was in tune with public opinion, which, although not supporting nuclear disarmament, opposed first use of nuclear weapons.

The account of the 1980s as a decade that strengthened the taboo would be even stronger if it took into account the research of Paul Lettow and others, which portrays an anti-nuclear President Ronald Reagan.[3] Although Tannenwald acknowledges that Reagan contributed to anti-nuclear sentiment with his statements about making nuclear weapons "obsolete," why he felt that way deserves fuller explanation. After all, this president discussed with Mikhail Gorbachev proposals to abolish nuclear weapons. Like many of his predecessors, Reagan had a strong aversion toward nuclear weapons, evidence for which former associates have provided when they recalled him saying that nukes were "horrible" and "inherently evil." Such stigmatizing language strongly suggests that Reagan had internalized the nuclear taboo.

The discussion of the nonuse of nuclear weapons during the 1991 Persian Gulf War makes a strongly suggestive, if not definitive, case for the taboo persisting beyond the Cold War. Tannenwald provides significant evidence that, across the board, top military commanders found nuclear weapons irrelevant and unusable during the war. Some of them raised moral objections and argued that civilized states did not use nuclear weapons, but using the information that the author provides, it is possible to construct other explanations. Some of the statements by the commanders suggest more pragmatic, cost-benefit calculations, for example, that the United States would lose the "moral high ground" and that nuclear weapons were fundamentally impractical for the situation in the Gulf. Most prominently, Colin Powell, then chairman of the Joint Chiefs of Staff, ruled out using nuclear weapons because "we're not going to let that genie loose."[4] Powell's implication may have been that he wanted to avoid setting a bad precedent, for example, that nuclear use might encourage even more proliferation and produce a more dangerous world. Although Tannenwald reasonably argues that norms shape thinking about precedent, Powell's remarks did not necessarily imply any particular ethical discomfort.

The lack of strong evidence on the thinking of civilian policymakers during the 1990 Gulf conflict makes it difficult to make authoritative statements about the impact of the nuclear taboo. Tannenwald shows that President George H. W. Bush, Secretary of State James Baker, and Secretary of Defense Dick Cheney were certainly willing to make ambiguous nuclear threats against Saddam Hussein, but to what extent had they internalized the taboo? Perhaps they did, but drawing firm conclusions requires more interviews, memoirs, and documents.

With nuclear proliferation becoming a focal issue in the post-Cold War world, Tannenwald sees a more complex environment for the nuclear taboo during the 1990s. Although the non-nuclear-weapon states did not challenge the taboo, they did question the inequalities of the nonproliferation regime, which limited nuclear deterrence and nonuse norms to a few. This unequal status proved galling to the non-nuclear-weapon states, which pushed hard to strengthen the NPT's disarmament requirements, obligations that the nuclear-weapon states have made weak commitments to fulfill. Nevertheless, efforts to delegitimize nuclear weapons proceeded, as non-nuclear-weapon states expanded nuclear-free zones and key NATO governments unsuccessfully introduced proposals to review first-use policy. Significantly, anti-nuclear activists began putting the abolition of nuclear weapons on the table, although polling data cited by Tannenwald suggests that U.S. public opinion was divided on the merits of abolition and nuclear weapons use. Tannenwald's history of these developments, including the successful push for a World Court advisory opinion on nuclear weapons, is invaluable and instructive, not least because it helps put recent events, such as the nuclear abolition initiative proposed by former senior officials, including former Secretary of State George Shultz, in historical context.[5]

Tannenwald emphasizes that, as an "evolving" phenomenon, the taboo is not yet "fully robust." Thus, her story is not one of assured progress: plans for nuclear war still exist, and nuclear threats continue to be made. Nevertheless, Tannenwald argues that the taboo has become so entrenched in official thinking that first use by the United States has become for all intents and purposes "unthinkable." Moreover, with respect to other countries, she argues that the taboo has become "widespread,"' not least because most other Western democracies are "more antinuclear than the United States." Although Tannenwald does not believe that the taboo in this country is facing a serious threat, she sees challenges that could weaken it, such as pressure for the use of nuclear "bunker busters" against terrorist or rogue-state underground complexes or an aggressive nationalistic policy, which she calls a "Leviathan" state-essentially current policy-that might seek to validate the use of nuclear weapons. With its unilateral emphasis, the Leviathan approach strongly supports missile defenses to maximize freedom of action. In this respect, Tannenwald is troubled by President George W. Bush's withdrawal from the ABM Treaty because it removed a prop from the system of mutual deterrence that had validated the practice of strategic nonuse for years.

Certainly, Hersh's reporting of the White House request for nuclear options against Iran suggests some weakening of the nuclear taboo in the Bush administration, as do its efforts on behalf of nuclear bunker busters. To prevent any erosion and to strengthen the taboo, Tannenwald believes it essential to keep alive the fear of nuclear war, move forward on a variety of institutional approaches to reduce the risk of nuclear use, and ensure that nuclear weapons remain in the category of "unacceptable" weapon of mass destruction. In this connection, she does not see abolition in the cards because of such difficulties as verification and the risk that some nation might secretly prepare to break out of an agreement. Instead, Tannenwald finds it more feasible to move toward "virtual abolition," to be achieved through changes in "habit, attitude, norm, law." Some of those changes-a wholly democratic world and the "obsolescence of war"-can be achieved only through sweeping historical development, but Tannenwald believes they could make nuclear weapons irrelevant. Besides these visionary prognostications, the author offers more tangible proposals to stigmatize nuclear weapons further: no-first-use agreements or a ban on nuclear weapons use altogether. The author's take on the abolition question should produce some debate, especially in light of the proposals by Shultz and his associates.

Tannenwald's commitment to political science's international relations (IR) theory, for example, her abstruse discussion of "material" versus "non-material" explanations, makes parts of The Nuclear Taboo a demanding read. Nevertheless, readers who are less interested in IR theory will benefit greatly from reading the historical chapters. The author does make some factual errors (e.g., misidentifying Eisenhower's role in 1950, misdating the first test of a deliverable hydrogen bomb, and confusing Kennedy with Johnson), but that may be inevitable when producing such a wide-ranging study. These complaints aside, Tannenwald has made a persuasive case for the existence of a nuclear taboo. Her book is a great accomplishment that will influence thinking about nuclear history and nuclear weapons policy for years to come.

William Burr is a senior analyst at the nongovernmental National Security Archive and directs its nuclear history documentation project.

Are you interested in purchasing this book? You can help support the Arms Control Association by visiting one of our partners.







1. Tsuyoshi Hasegawa, Racing the Enemy: Stalin, Truman, and the Surrender of Japan (Cambridge: Harvard University Press, 2005).

2. David Alan Rosenberg, "The Origins of Overkill: Nuclear Weapons and American Strategy, 1945-1960," International Security, Vol. 7, No. 4 (Spring 1983), p. 11.

3. Paul Lettow, Ronald Reagan and His Quest to Abolish Nuclear Weapons (New York: Random House, 2006).

4. Colin Powell with Joseph E. Persico, My American Journey (New York: Random House, 1995), p. 486.

5. See, for example, George P. Shultz, William J. Perry, Henry A. Kissinger, and Sam Nunn, "A World Free of Nuclear Weapons," The Wall Street Journal, January 4, 2007, p. A15; and "Toward a Nuclear-Free World," The Wall Street Journal, January 15, 2008, p. A13,