For twenty years, scholars have argued that moral concerns play a decisive role in inhibiting states from using certain weapons of mass destruction. Taboos on chemical, biological, and nuclear weapons have been identified, and the literature helped provide an intellectual framework for the movement to ban landmines.Footnote 1 These scholars have argued that taboos are more than just provisions in international law. They are a special breed of socially constructed norms that, once internalized, make first use of these weapons odious and unthinkable for fully integrated members of the state system.
Although Moon's work on what are now identified as “taboos” antedated the constructivist project, taboos have become a central element in the constructivist security studies literature. In purporting to show that norms can keep states from reaching for the most powerful weapons in their arsenal, the literature on taboos made an unexpected and compelling claim for the power of ideas and identity. It was unexpected because it revealed a powerful phenomenon that restrains states and their leaders from using potent weapons despite the pressures of war. It is compelling in the sense that it seems to solve a puzzle that more conventional literatures had never fully answered.
Much of the causal force of the weapons taboos comes from the psychological and emotional costs of violating these special norms. While taboo talk often invokes the language of psychology, constructivist taboo theorists have not taken account of psychological research that indicates that when deciders believe that taboo violation will save other values or ideas sacred to them, the choices before them will seem “tragic” rather than taboo. This tragic emotional and psychological context eliminates the psychological deterrents to taboo violation.
An analysis of U.S. General George Marshall's 1945 proposal to use chemical weapons against Japan in World War II provides clear evidence that this kind of normative conflict can cause opponents of forbidden weapons to advocate their usage. Although the proposal was ultimately blocked by Fleet Admiral William Leahy, the case demonstrates that even well-respected military and state leaders, when confronted with a conflict between two taboos they embrace, may reluctantly conclude that a tragic violation of one of them is necessary. Marshall's disagreement with Leahy also provides insight into how states and state leaders address competing norms. Even powerful officials are constrained by the social character of these norms, because they require support from other stakeholders before their proposals become policy. Those stakeholders will be more easily persuaded when they share the proposers' normative commitments and interpretations of the trade-offs at hand. When their commitments and perceptions diverge, however, stakeholders will have more difficulty in persuading that a tragic violation is necessary, thus making a state decision to violate a norm less likely.
More Than Just Hard Choices
Constructivists describe taboos as a special species of normative rule. Taboos, Tannenwald states, are “particularly forceful kind[s] of prohibition[s],” “concerned with the protection of individuals and societies from behavior that is defined as or perceived to be dangerous” and that “typically [refer] to something that is not said, not done, or not touched.”Footnote 2 Quester identifies the psychological roots of their particular forcefulness: taboos are “something we do not even think about doing, something about which we do not weigh costs and benefits but that we simply reject.”Footnote 3 Paul notes the “revulsion” against violating the nuclear taboo.Footnote 4 While Price's claims about the power of taboos are somewhat more attenuated—that they are not “all powerful” norms alone determining use or nonuse—his argument is still predicated upon the existence of a “stigma” against their use, capable of narrowing the range of circumstances in which they might be used.Footnote 5 Legro, Price, and Tannenwald's discussions of taboos are all replete with the discourse of civilized identity and emotion-laden reactions to suggestions that prohibited weapons be used.Footnote 6
Most normative rules in international relations lack a moral foundation and psychological force;Footnote 7 states and other actors are deterred from violating rules by the possibility that others will perceive them as threats or “rogues.” Because of their emotional character, however, taboos also influence action by shaping the way that decision makers think and feel about the options they confront. That taboos influence decision makers in this additional way seemed to resolve theoretical doubts about the feasibility of using norms to explain behavior. In the seminal constructivist security studies text, The Culture of National Security, Kowert and Legro identified the “ubiquity of norms” problem, caused by the fact that “multiple norms can influence actors—with competing or even contradictory prescriptions.” These contradictory prescriptions can give rise to “indeterminate predictions” about how actors will behave.Footnote 8 The psychological consequences of even thinking about breaking taboos seemed to eliminate this problem by creating a hierarchy of norms topped by taboos.
Even as they advance psychological effects as the basis of a hierarchy of norms, however, constructivist taboo theorists fail to identify the psychological underpinnings of their arguments. They are not alone in this: constructivist international relations scholars have tended to theorize from a holist position and have not developed a consensus model of individual decision making.Footnote 9 Some seem to embrace a “logic of the appropriate” model that, taken to the extreme, can cast individuals as something akin to social automatons. Others simply avoid making micro-foundational, individual-level commitments despite using arguments that implicate psychology.Footnote 10 Most do not engage with modern research psychology, which suggests that individuals construct their decisional calculus in complex ways and respond to it with nuance.Footnote 11 Without clear micro-foundations, however, explanations of how discourses, norms, or social beliefs translate into human decisions and actions have an ad hoc character.Footnote 12 This is evident in the literature on weapons taboos: no clear explanations of what a stigma is or how it works, or how weapons use produces revulsion, are offered.
Psychological studies of values and value trade-offs provides a micro-foundational analysis of how taboos influence decision making. As the constructivists expect, strongly negative emotional responses, generally associated with prospective changes in self-concept, can cause people to refuse to even consider choices violating certain normative prohibitions. But, contrary to constructivist expectations, this work also finds that when two sacred values or prohibitions are pitted against each other, that emotional resistance vanishes.Footnote 13 While these findings validate the general argument that taboos exist and are given force by their emotional consequences, they also make clear the force of the taboo is situation-dependent. This situational dependence is unsurprising, given that emotions are elicited by the perception of situations with defined characteristics and differently structured situations evoke different emotions.
Connecting Norms with Individuals
Although the nuclear taboo concept is sometimes credited to Schelling, it has matured in the context of constructivist theory.Footnote 14 Constructivist international relations theory, like social theory more broadly, focuses on interrelationships between discourse, practice, and identity. This tendency is evident in the literature on forbidden weapons. States are members of a social system of states; taboos are part of the social fabric composed of international law, diplomacy, and bureaucratic practice, and the constitutive discourse of civilization and modernity that influences those members.Footnote 15 Taboos become part of this social fabric as states and state bureaucracies change their practices and ways of talking about the weapons.Footnote 16
Social theory provides a solid foundation for analyzing these norms as an international phenomenon. However, social theory alone is not enough to explain how these ideas will influence the behavior of states led by particular leaders. While both cause (taboos) and effect (nonuse) may exist at the level of the state, the mechanisms that make taboos work exist at the individual level. There are two reasons for this. First, both the state and its subsidiary organizations are abstract constructs, and as such neither is capable of action. Like nations,Footnote 17 states and their institutional elaboration are fundamentally imagined things. They live only as individual human beings see themselves as part of a state organization. They act only when individual human beings, perceiving a particular status within that state organization (president, soldier, police officer, judge, citizen), act accordingly. While everyone attached to a state is influenced by the discourses and practices of that state, and are collectively influenced by its political, economic, and security situation, agents within the state are not interchangeable. These individuals are analytically separable from the sets of ideas associated with the state because no person is perfectly socialized, and therefore even members of a single organization can differ in causal beliefs, interpretations of events, and fundamental values. Indeed, despite the influence of organizational selection mechanisms, there is often great diversity in the beliefs and values of the individuals who come to lead states. Recent work on norm decline has made this evident by demonstrating that when individuals who disagree with systemic norms have come to power in important states, the norms themselves can be altered.Footnote 18
Second, only individuals can have the emotional responses central to the taboos phenomena. When individuals entangle a collective identity into their personal identity, that identification can create the basis for individual emotional responses to group experiences. However, they do not lose their personal identity in the process, and their emotional responses are only influenced rather than determined by group experiences.Footnote 19 Because the emotions principally associated with taboo violation—shame, guilt, and self-directed anger—all involve personal responsibility judgments, nondecision makers will not experience these emotions as intensely, if at all.Footnote 20 The centrality of the individual decision maker has long been evident in the empirical element of the taboo literature: most of the key evidence for the existence of and changes in weapons taboos comes from state leaders' (and their advisors) perceptions of forbidden weapons and emotional responses to their prospective use.Footnote 21
Given the importance of the individual decision maker, taboos can be said to be fully internalized by a state only when they have become part of state leaders' self-concepts.Footnote 22 This integration can be the product of multiple processes. As individuals are socialized from an early age, they learn norms—guidelines that help determine correct behavior. When these norms determine not only correct behavior, but whether a person is good and bad, they are usually explained and understood in terms of underlying moral values. Given the imperfect nature of socialization and individual differences in needs for individuation and other personality traits, individuals' construction of their self-concept largely but incompletely replicates the dominant constructs of their society, which can include norms generated by interactions between states.Footnote 23 Beyond early formation, moral argument provides an important, possibly central, role in the continued development of an individual's normative self-concept. Experience, particularly vivid or emotionally searing experiences, can also shape the individual's self-concept and their understanding of what is good, bad, and indifferent in idiosyncratic ways. Through the messy processes of socialization, moral argument, and experience, individuals may come to identify some behaviors that, if engaged in trivially, render a positive self-concept all but impossible. These are “tabooed” behaviors. Crucially, the origin of these rules and values is irrelevant for their effects. Once they become part of an individual's self-concept, international norms, the ethics of a nation, religious edicts, personal morality, and a possibly wide range of other normative systems all confront the individual as fundamentally similar things.
Once part of an individual's self-concept, taboos can create self-perceptions that generate negative emotions. More specifically, a recognized disjuncture between behavior and an individual's beliefs about their character can elicit self-conscious emotions like shame, guilt, and self-directed anger.Footnote 24 They are principally focused on regulating individual behavior and ensuring compliance with the norms comprising their self-concept.Footnote 25 Like other emotions, they are elicited by the individual's perception of their situation—the product of either conscious or automatic processingFootnote 26—but their elicitation schemes are more complex than those of the basic emotions like frustration, anxiety, or contentedness.Footnote 27 The most important elements in the elicitation scheme for these emotions are an individual's self-concept, their perceptions of the context of action, and beliefs about the moral status of particular actions.Footnote 28
The emotional responses generated by these perceptions provide the emotional force associated with taboo violation. These responses influence individual thought and action through a number of physical systems. Depending on the specific emotion elicited, emotional responses can activate a range of muscular and cardiovascular systems, alter the subjects and processes of cognition, and produce sentimental feelings.Footnote 29 When an individual considers violating a taboo, unbidden effects on cognition can cause them to reinterpret the situation they confront, introducing concerns about the effects of breaking the taboo on their character.Footnote 30 Strongly unpleasant feelings, caused at least in part by changes to their muscular and other physical systems, become evident. By altering how they think about their choices and about themselves, and by creating unpleasant sensations,Footnote 31 the emotional response to prospective taboo violation actively discourages violation. Actual violation of the taboo can produce both recurrent negative emotional responses and unpleasant changes in self-perception. Major taboo violations are unlikely to be easily forgotten, and because emotional responses are reelicited every time an infraction is recalled, individuals are likely to experience these negative emotional responses repeatedly.
Norms and Decisions
The literature on value trade-offs helps define more precisely when political decisions can trigger these negative self-conscious emotions. Baron's work with “protected values” is most clearly allied to the taboos literature. A value is “protected” if individuals treat it as if it has an infinite value and is protected from being traded for regular, economic values. In his experiments involving protected values, respondents evince quantity insensitivity when considering them and associate them with moral obligation. These protected values give rise to deontological rules, which when violated cause individuals to respond with anger. Baron's experiments suggest that taboos are deontological rules prohibiting certain behaviors generated by relevant protected values.Footnote 32
Tetlock's account is somewhat more nuanced. While the existence of nonprotected values is implicit in Baron's work, Tetlock argues that most people have two categories of values—secular and sacred. Secular values are those that are easily traded in an economic context; sacred values possess “infinite or transcendental significance that precludes comparisons, tradeoffs, or indeed any mingling with bounded or secular tradeoffs.”Footnote 33 Tetlock then addresses how these different kinds of values can influence decision making by creating a typology of trade-offs. Ordinary trade-offs involve exchanges of secular values, and are easily made. Taboo trade-offs involve the sacrifice of a sacred value for a secular value. Most individuals resist making taboo trade-offs. They have strong negative reactions against people who do so, describing them as morally deformed, even evil. They view themselves, and anyone who proposes that they engage this kind of a trade-off, with anger and hostility. Should they later recognize that they accidentally committed a taboo trade-off, they experience shame, guilt, and feelings of moral contamination or disgust.Footnote 34 Shame, for instance, is elicited when an individual perceives that they have a major character flaw—as would clearly occur if they have made a taboo trade-off. Guilt is the product of perceiving that one has committed a wrong action.Footnote 35
Both Baron's work on protected values and Tetlock's work on taboo trade-offs are consistent with the existence of weapons taboos. For instance, proposals that these values be sacrificed are generally rejected with prejudice, as are proposals to use nuclear weapons in peacetime. Likewise, that most people believe that they will never even consider violating them is consistent with the “unthinkability” noted by Quester.Footnote 36 However, this literature differs in one important regard: it finds that when conflicts between protected values or sacred values do occur, individuals are able and willing to sacrifice one for the other without negative emotional experiences or degrading their self-esteem.
Normative Conflict and the Limits of Normative Prohibitions
Most existing explanations of norm violation in international relations actually explain how and why individuals and states violate norms that they do not fully accept. Personality characteristics, rationalization, and national and organizational culture can all prevent norms from being internalized or create “rules of exception.”Footnote 37 While Kowert and Legro note that the “ubiquity” of norms can create situations in which they contradict each other, they merely suggest that the consequences of such a situation are indeterminate.Footnote 38 No author has investigated the consequences of a conflict of taboos.
The problem of normative conflict is particularly relevant to the use of forbidden weapons because the cauldron of war often evolves in strange and surprising ways, putting states and the things their people value at risk unexpectedly and, in so doing, transforming the trade-offs leaders face. In addition to the lives of their soldiers and citizens, countries—particularly those on the verge of defeat—sometimes find their way of life and their most fundamental political values in danger. Sometimes these risks are evident at the outset of a conflict, but in others the war develops with a ferocity and speed that exceeds state leaders' expectations.Footnote 39
The findings in the psychological literature suggest that these normative conflicts elicit emotional responses that are very different from those associated with simple taboos or taboo trade-offs. Baron finds that the resistance to violating protected values goes away when individuals seriously confront a conflict between two protected values.Footnote 40 Tetlock's typology explicitly incorporates this condition by identifying tragic trade-offs, which arise when individuals must choose between two sacred values.Footnote 41
The perception that compliance with one taboo puts another taboo or sacred value at risk elicits an emotional context different from that associated with the violation of an unconflicted taboo. While the broader circumstances that force the problem onto the decision maker may engender a range of negative emotions, including anger, the decision problem itself has been associated with sadness.Footnote 42 As sadness is associated with lethargy, the perception of threat and uncertainty about response—the recipe for anxiety, a highly motivating emotion—may be required for individuals to confront the difficulty associated with the decision problem and act.Footnote 43 This is consistent with findings that tragic trade-off decisions are often experienced as difficult and stressful.Footnote 44 In the aftermath of these decisions, decision makers are likely to experience a mix of sadness, but they may also feel authentic pride.Footnote 45 Others who share the decision makers' perception of the trade-off may perceive them as wise.Footnote 46 In short, in this tragic context, taboos can be violated without negative emotional or attributional consequences.
Using functional Magnetic Resonance Imaging (fMRI) scans to study moral decision making, Greene and his collaborators have found evidence consistent with this account. In their studies, personal moral dilemmas exist when the direct effects of an individual's decision can cause a particular individual or group of people to suffer serious bodily harm; impersonal dilemmas exist when harm may come through only an indirect effect of their actions. Dilemmas are “easy” if there is a moral case for only one option, and “hard” if a moral case can be made for both options. Greene found that when participants responded to impersonal dilemmas, areas of the brain most associated with conscious cognition were active.Footnote 47 When easy moral dilemmas were confronted, areas of the brain associated with automatic processing and emotion were active, and decisions were made extremely quickly. However, when participants were confronted with hard moral dilemmas and they decided that a moral violation was necessary, areas of the brain associated with both emotion and conscious cognition were active, and individuals demonstrated long reaction times.Footnote 48 The evidence that emotion and automatic processing were the primary active processor in unconflicted moral dilemmas is strongly suggestive that taboos are processed in ways strikingly different from most other decisions. The evidence that moral conflicts cause a different processing pattern, one that is both emotional and consciously cognitive, is likewise striking evidence that people respond to normative conflicts in yet different ways.
Taken together, this evidence suggests that when pitted against another taboo or fundamental value, taboos may lose their stigma. This fact alone does not resolve the ubiquity of norms problem, inasmuch as it does not provide an a priori way of predicting which taboo will be maintained and which taboo will be violated. The desperation to save something may overwhelm constraints against using forbidden weapons,Footnote 49 but this may not always be the case. Psychological analysis does, however, provide further insight into how decision makers may respond to this decision problem. First, Baron, Greene, and their colleagues all argue that these dilemmas produce decision making with a utilitarian character.Footnote 50 Second, maintaining the tragic character of the trade-off requires maintaining an interest in the values underlying each taboo. Third, the emotions elicited when an individual confronts a tragic trade-off—particularly if it is unexpected—may encourage further analysis of the problem. Together, these factors suggest that decision makers confronting competing taboos may craft options that try to jointly maximize the values encompassed by the competing taboos, even as a taboo is violated. This may be done by trying to minimize the violation in some way or by trying to mitigate its effects. For instance, if a leader believes that they confront a choice between losing sacred land and using a forbidden weapon, in these tragic circumstances a means of dividing it that nevertheless preserves its core “sacredness” may be found to be acceptable. Alternatively, if the forbidden weapon is to be used, decision makers may try to limit the extent of its use. Tragic violations are likely to be the minimal violation decision makers believe will allow them to retain both their political goals and their character. Because it may often seem easier to mitigate the effects of taboo weapons, tragic trade-offs may tend to favor weapons taboos.Footnote 51 When dealing with weapons of mass destruction, however, even a minimal violation would not have minor effects.
The Development of Moral Dilemmas During Conflict
This psychological work suggests that state leaders may choose to tragically violate taboos even as they embrace them as morally desirable. Their usage of these forbidden weapons or methods of war is not an indication of a failure to internalize the norm—at either the individual or the public level—nor is it an effort to revise a norm or challenge its usage. Rather, from the perspective of state leaders, tragic taboo violation is a response to a difficult moral dilemma. These dilemmas are commonly realized in wars when leaders fighting for a transcendent value confront the prospect of loss or defeat. As decision makers try to identify alternative courses of action, the risk to their transcendent value may allow them to consider alternatives that had been dismissed earlier. Forbidden weapons or methods of fighting may be introduced as alternatives to losing the protected value.
This process of strategy revision is rife with pressures on decision makers. Some pressures derive from the situation itself, which is likely to be characterized by anxiety. Some pressures are social and political: decision makers put their jobs at risk if they are seen to fail or engage in unnecessarily costly strategies.Footnote 52 Still other pressures are a mixture of the social and the individual, as decision makers may perceive a responsibility or duty to others to secure their war aims or minimize costs. These pressures collectively drive the strategy revision process and motivate leaders to consider previously unthinkable options. At the same time, as they recognize that the alternatives they are considering involve normative violations, relevant individual and social taboos become salient and leaders may feel shame for considering tabooed actions, or fear being reviled, humiliated, and perhaps tried for war crimes. This is particularly true if the normative conflict is unexpected—if leaders never expected to have to consider taboo and tragic trade-offs. While the gravity of failure may provide some psychological protection against these effects, it is unlikely to eliminate them. Ultimately, the process yields a hard moral dilemma if decision makers conclude that the only alternative to an unthinkable loss is a taboo-violating strategy.
The “Tragic” Use of Prohibited Weapons
In sum, when conflicting taboos present leaders with an apparently tragic trade-off, taboos against the first use of nuclear weapons or chemical weapons are unlikely to retain their constraining effects, creating a permissive condition for use. The core prediction, then, is that when state leaders believe they face a choice between conflicting taboos or between violating a taboo and sacrificing a value of transcendent value, these prohibitions will lose their psychological force. It is important to note that the decision makers considering violating these taboos are not people with “nothing left to lose,” gambling with one last throw of the dice; on the contrary, they are motivated by the recognition of a competing taboo or transcendent value. While they may still choose not to violate the taboo—perhaps being restrained by deterrent threats or attempting to create a new option, for instance—the decision calculus will become more intensively subject to utilitarian considerations. This means that they will pursue what they believe to be the greatest or most important value while seeking to mitigate the effects of violating the conflicting norm or value.
George Marshall, Chemical Weapons, and the War with Japan
During World War I, all major belligerents on the Western Front used chemical weapons. Cook has documented its effects: no mere nuisance during the war, gas was responsible for 27.4 percent of American Expeditionary Force (AEF) casualties.Footnote 53 By Armistice Day, chemical warfare was seen as a powerful and effective tactical element, one feared and hated by infantrymen.Footnote 54 During World War II, however, it was largely unused against combatants. To some extent, this can be explained through deterrence: American and Allied leaders feared that using chemical weapons, even against the Japanese, would give the Germans permission to use them. However, as Moon has shown, deterrence is not enough to explain nonusage after the Germans surrendered, given that the American leaders did not anticipate that the Japanese would or could use chemical weapons against them.Footnote 55 Indeed, so little did the Americans expect a Japanese chemical attack that by June 1945, Pacific Theater forces lacked the capability to effectively retaliate in kind.Footnote 56 Instead, Moon and others have argued that chemical weapons' status as immoral or forbidden weapons is responsible for their nonuse.
The full story is more complicated, however. After the invasions of Okinawa and Iwo Jima resulted in extremely high casualties, and with a difficult invasion looming, chemical weapons gained an important advocate—General Marshall, the army chief of staff. His advocacy may seem out of character: one of only three generals to win the Nobel Peace Prize, Marshall was and is respected for his character and integrity.Footnote 57 Yet, despite being legally and morally aware of the taboo, Marshall was willing to break it in hopes of minimizing U.S. casualties. Although Fleet Admiral LeahyFootnote 58 and President Harry S. Truman vetoed his proposal, Marshall's willingness to break the preeminent taboo of the prenuclear era sheds light on when taboos can fail to prevent the use of normatively prohibited weapons.
Two Taboos: Chemical Weapons and Excessive Casualties
The Chemical Weapons Taboo in America
Despite cultural and legal roots predating the war, the chemical weapons taboo fully matured only after the widespread use of chemical weapons during World War I.Footnote 59 The experience of the war itself, which made vivid the horror of gas, had much to do with its development into a taboo.Footnote 60 Many leading soldiers opposed using chemical weapons, consistent with Grossman's argument that the more horrific the means of killing, the more difficult it is for those responsible to retain a positive self-image.Footnote 61 The development of the airplane as a prospective delivery system intensified concerns that it might be used against civilians and added to the popular horror of these chemical weapons.Footnote 62
In the United States, most American political and military leaders supported outlawing chemical weapons. They grounded their opposition in moral terms. Elihu Root, McKinley's Secretary of War and Theodore Roosevelt's Secretary of State, presented the 1922 Washington Conference with a treaty banning chemical weapons in hopes that it would “stigmatize” and “denounce the use of poisoning gases and chemicals in war, as they were used to the horror of all civilization in the war of 1914–1918.”Footnote 63 An advisory committee led by Army Chief of Staff (and former American Expeditionary Force Commander) General of the Armies John J. Pershing endorsed the move: “Chemical warfare should be abolished among nations, as abhorrent to civilization. It is a cruel, unfair, and improper use of science. It is fraught with the gravest danger to non-combatants and demoralizes the better instincts of humanity.”Footnote 64
General Pershing remained an opponent of chemical warfare, and later campaigned for the Geneva Protocol.Footnote 65 The General Board of the Navy agreed, stating that “gas warfare threatens to become so efficient as to endanger the very existence of civilization” and “the General Board believes it to be sound policy to prohibit gas warfare in every form and against every objective.”Footnote 66 During the period, the only Americans in public life to embrace chemical weapons were associated with the chemical industry or the Army Chemical Weapons Service (ACWS). Their intense lobbying was enough to prevent Senate ratification of the Geneva Protocol at that time.Footnote 67 Still, despite not being party to an active treaty forbidding chemical weapons, U.S. leaders considered themselves bound to a no-first-use agreement.Footnote 68
President Franklin D. Roosevelt opposed their use throughout World War II. In an 8 June 1943 warning against Axis uses of chemical weapons, he declared that the
use of such weapons has been outlawed by the general opinion of civilized mankind. This country has not used them, and I hope that we never will be compelled to use them. I state categorically that we shall under no circumstances resort to the use of such weapons unless they are first used by our enemies.Footnote 69
Allied policy was consistent with this position. In 1942, the U.S./UK Combined Chiefs of Staff agreed to CS 106, which authorized the procurement of chemical weapons for deterrent purposes and required mutual agreement prior to any use of chemical weapons.Footnote 70 In early 1945, FDR rejected a proposal by the director of the Office of Strategic Services to use gas weapons against the defenders of Okinawa and Iwo Jima, even though the threat of retaliation was passing.Footnote 71 It is clear that prior to and during World War II, there existed a chemical weapons taboo in America.
Evidence suggests that General Marshall embraced the taboo. He had served in France and saw first-hand the misery and horror of the attacks. Indeed, in one battle he planned, American forces suffered 200 gas casualties.Footnote 72 General Pershing, a great opponent of chemical weapons, was Marshall's mentor and close confidant until his 1948 death.Footnote 73 During the 1930s, Marshall called on the public to remember the infantryman who, amid “cold and mud, under shell and machine gun fire, was bombed and gassed” but ultimately was responsible for victory.Footnote 74 Marshall also oversaw the adoption of CCS 106, the Allied agreement to abjure the first use of chemical weapons. This document stated that any consideration of chemical weapons involved “the utmost gravity.”Footnote 75 In 1943, his trusted assistant chief of staff, Major General Thomas Handy, soundly rejected a proposal by the ACWS to use chemical weapons.Footnote 76
The “Casualty Issue”
The chemical weapons taboo was not the only normative code constraining Marshall's decision making, however. Marshall was highly sensitive to his soldiers' deaths. This sensitivity—an effective taboo against risking any more than the absolute minimum required to accomplish a mission—was evident in most generals of his cohort. As Heufler has made clear, the “casualty issue” permeated American military culture both during and after World War I.Footnote 77
Marshall's primary military socializing experience was World War I, where he witnessed casualty rates unprecedented for the U.S. army. He personally saw battle as a junior officer and was involved in planning and reviewing actions as a staff officer. In these capacities he saw officers pursue high-casualty operations that he sometimes judged were not worth the cost in lives.Footnote 78 He came to see these “stressingly heavy” casualties as negligent or immoral, which is not surprising given the relationships of loyalty and trust that link soldiers and effective officers.Footnote 79 Public and congressional responses to these losses further confirmed these impressions: the AEF's leadership was harshly criticized in newspaper editorials and congressional investigations for “wasting” the lives of American soldiers.Footnote 80
Marshall's approach to casualties was also shaped by his mentor, General Pershing. While commander of the AEF, Pershing's difficulty with giving the orders that lead to the death and maiming of his soldiers was apparent to his immediate staff, which from early 1918 included Marshall. At one point, a visit to a surgical ward left Pershing weeping in his staff car.Footnote 81 More concretely, Pershing acted repeatedly, regardless of old friendships or their political connections, to remove commanders whose orders produced excessive casualties. When challenged for ruining military careers, he responded bluntly: “I don't care. Men's lives are at stake.”Footnote 82 He attempted repeatedly but often in vain to improve the tactical practice of a poorly trained American military in an effort to mitigate its excessive casualties.Footnote 83 Many of these traits transferred to Marshall. During the interwar period, Marshall worked across a variety of fields to eliminate the incompetence and unpreparedness that drove up casualties among the AEF.Footnote 84 On the eve of World War II, he moved to push aside those generals he found unsuitable for commanding men, often at the cost of decades-long friendships.Footnote 85 To remind his commander in chief of the human cost of the war, he sent President Roosevelt weekly reports with fatalities typed in red.Footnote 86 After the war, like Pershing before him, he served as chairman of the American Battle Monuments Commission, ensuring that his soldiers were properly buried and their graves well tended.
Marshall also had personal experience with casualties. Lieutenant Allen Brown, a stepson he had helped raise from the age of twelve, was killed by a sniper in Italy on 29 May 1944.Footnote 87 In a striking letter to Allen's widow, written weeks after his stepson's death, General Marshall described his efforts to meet members of Allen's unit and—though it required multiple attempts over the course of two days—to fly over the spot where he was killed.Footnote 88 In correspondence with an officer's widow in July 1945, he not only referenced his personal loss, but wrote that “the suffering experienced by American families as a result of casualties caused by this terrible struggle is most distressing to me personally.”Footnote 89
Though Marshall did not select American war aims in World War II, he was fully committed to them. His task, therefore, was to win the war. But his personal and professional experiences also generated in him a strong norm that, while winning the war, American military casualties must be minimized. Failing to do so would mean wasting something sacred to him—the lives of his soldiers—and would constitute a betrayal of the trust reposed in him by front-line soldiers, an act of disloyalty that would be deeply painful to a man who so prized his integrity.Footnote 90 This meant that anticipated casualties were the measuring stick against which militarily viable strategic options were judged.Footnote 91
The Norms Conflict
As the war in the Pacific approached the Japanese homeland, Japanese forces increasingly engaged in delaying tactics, particularly using fortified caves. Dealing with these positions—even after the main battles were decided—was often costly in American lives.Footnote 92 The battles on Okinawa and Iwo Jima caused severe casualties; American casualties (though not fatalities) on Iwo Jima exceeded Japanese casualties there. Everyone expected the invasion of Japan to be as bad, or worse; Truman described it as a prospective “Okinawa from one end of Japan to the other.”Footnote 93 This invasion seemed likely to involve far more casualties than had been experienced to that point, often in situations—such as dealing with bypassed caves and fortifications—which would not materially affect the military outcome of the war. Chemical weapons would be effective in these “mopping up” actions, almost certain to reduce American casualties. Because Marshall and, more broadly, the American government were alert to the taboo against chemical weapons, they were the only existing class of weapon that had not been seriously considered for use against Japan.Footnote 94 Marshall thus found himself in a situation of normative moral conflict. He knew of the generally immoral status of chemical weapons, but he also believed that failing to use chemical weapons when they were tactically appropriate would increase American casualties, possibly significantly.
Given the existence of this conflict, the psychological work suggests that he should perceive the situation as tragic. While he never actually says so, his repeated references to flamethrowers while arguing for the use of chemical weapons provide important evidence that he felt the tragic character of the decision. For Marshall, flamethrowers and chemical weapons had important commonalities. First, they killed in similar ways. When used against deep cave defenses on Okinawa and elsewhere, flamethrowers principally killed defenders by consuming the oxygen in the cave and thereby suffocating everyone inside. Second, for Marshall and others, these weapons were, like chemical weapons, inhumane and appalling. Nevertheless, he and other war decision-makers had made a tragic decision to use them so as to limit U.S. casualties. These sentiments are clear in a 1944 radio address:
I speak with an emphasis that I believe is pardonable in one who has a terrible responsibility for the lives of many men, because I feel that here at home we are not yet facing the realities of war, the savage, desperate conditions of the battlefronts. Vehement protests I am receiving against our use of flame fighters do not indicate an understanding of the meaning of our dead on the beaches at Tarawa. Objections to this or that restriction are inconsistent with the devoted sacrifices of our troops … The situation, however, demands a determination which will divorce the individual from his own selfish weaknesses and ulterior motives. Our soldiers must be keenly conscious that the full strength of the nation is behind them, they must not go into battle puzzled or embittered over disputes at home which adversely affect the war effort. Our small sacrifices should be personal even more than financial.Footnote 95
In these comments, Marshall identifies a conflict between two competing values—one against the use of cruel weapons and one against the wasteful sacrifice of American soldiers' lives. He acknowledges the emotional costs of bearing his “terrible responsibility for the lives of many men.” In urging them to stop their “vehement protests” and make a small “personal” sacrifice, he is encouraging them to recognize the necessity for trade-offs in this “savage” war and be willing to accept the tragic necessity of using flamethrower. These statements make clear that he did see these trade-offs as real and challenging, and he felt keenly the bite of violation even as he believed it was a tragic necessity.
In this strategic and moral context, Marshall began to advocate for the use of chemical weapons against the Japanese.Footnote 96 Assistant Secretary of War John McCloy recorded one instance of his advocacy, which occurred in a meeting with Marshall and Secretary of War Henry Stimson:
The General then spoke of his stimulation of the new weapons and operations people to the development of new weapons and tactics to cope with the care and last ditch defense tactics of the suicidal Japanese. He sought to avoid the attrition we were suffering from such fanatical but hopeless defense methods—it requires new tactics. He also spoke of gas and the possibility of using it in a limited degree, say on the outlying islands where operations were now going on or were about to take place. He spoke of the type of gas that might be employed. It did not need to be our newest and most potent—just drench them and sicken them so that the fight would be taken out of them—saturate an area, possibly with mustard, and just stand off. He said he had asked the operations people to find out what we could do quickly—where the dumps were and how much time and effort would be required to bring the gas to bear. There would be the matter of public opinion which we had to consider, but that was something which might also be dealt with. The character of the weapon was no less humane than phosphorous and flame throwers and need not be used against dense populations or civilians—merely against these last pockets of resistance which had to be wiped out but had no other military significance.Footnote 97
This statement makes clear that Marshall was not only concerned with public opposition to using chemical weapons but was himself alert to the moral status of chemical weapons. This is evident in his reference to flamethrowers and phosphorus and in his suggestions that weaker chemical weapons be used, and even then only used in mopping-up operations. It also makes clear that he believed that violating the norm was acceptable, even necessary, because it would save American lives that would otherwise be uselessly wasted in mopping-up operations against “fanatical” resistance. A June 1945 War Department study he shared with Admiral Ernest King put it baldly: “people must balance their present objection to this weapon which does not compare with the flame thrower, white phosphorus or the petrol bomb, as a terrible instrument of war, against the additional cost in American youth, American resources, and the length of the war.”Footnote 98 In advocating the use of gas, Marshall was effectively arguing that the same tragic trade-off that he and other war decision makers had already made with respect to inhumane weapons such as flamethrowers be applied to chemical weapons.
The Response to the Proposal
Marshall was not alone in endorsing this proposal. That it was sent to the White House implies the support, or at least acquiescence, of Secretary of War Stimson, who was present at the 29 May meeting mentioned earlier. The White House response, however, was not as favorable. Admiral Leahy, Chief of Staff to the Commander in Chief, emphatically rejected the proposal, citing FDR's 1943 statement rejecting the use of chemical warfare. President Truman, without comment, followed Leahy's advice.Footnote 99 The precise nature of Leahy's response is not clear, however, as he left a far less informative documentary footprint than Marshall. Given the divergent experiences of the army and navy in World War I, he—like many other leaders of the navy during World War II—may not have internalized the casualty norm. This would mean that he would not perceive a conflict of norms or the need for a tragic trade-off. If this is true, his rejection of chemical warfare suggests a normative consistency with his opposition to the use of nuclear weapons and attacks on civilians.Footnote 100 If Leahy did perceive a tragic trade-off, it is clear that he preferred to sacrifice the soldiers' lives; this would not be inconsistent with the theory, which does not offer general a priori predictions about which norm will be violated.Footnote 101 However, it is not clear that Leahy perceived a tragic trade-off. Like many in the navy, he was skeptical about the planned invasion of Japan, and preferred a blockade strategy that would allow for victory without invasion.Footnote 102 Given that the actual decision to invade had not yet been finalized, and perhaps doubting that an invasion would ever happen, Leahy may simply not have perceived a normative conflict. After the proposal was rejected by the White House, Marshall did not attempt to resurrect it. However, the army continued to study using chemical weapons. Had the atomic bombs been unsuccessful, it seems likely Marshall would have advocated for chemical weapons again—particularly as the difficulty of invading Kyushu, the first of the main Japanese islands slated for invasion, became more apparent.Footnote 103
Discussion
The evidence is clear that Marshall perceived the choices that confronted him as involving a trade-off between competing values. Facing this trade-off, Marshall argued for the violation of one of those norms by supporting the use of chemical warfare against the Japanese. In doing so, he did not mask the fact that this involved a normative violation. He also argued for using chemical warfare in ways that were designed to mitigate its effects. This outcome is consistent with the literature on value conflict: when a trade-off is seen as involving the sacrifice of a sacred value to save another, decision makers are able to do so without shame or other negative emotional responses, and without threatening their self-esteem. That Marshall argued for limiting the scope and lethality of their use, moreover, is consistent with a tragic sensibility that tries to limit the scope of the violation in utilitarian ways.
Alternate explanations for this series of events are less persuasive. It is clear that the American failure to use chemical weapons against Japan was not the product of successful Japanese deterrence.Footnote 104 It also seems unlikely that Marshall was secretly waiting for an opportunity to advocate for the use of chemical weapons. He did not protest against either CCS 106, the intra-Allied agreement to refrain from first use, or Roosevelt's 1942 proclamation. Likewise, that he and his staff had no problem rejecting earlier proposals to use chemical weapons indicates that he was not pressured by subordinates into making the proposal. Rather, the evidence supports the argument that he accepted both norms, and only when he felt clearly conflicted did he support the use of gas. This finding supports the more general argument that when decision makers confront conflicting norms, they are able to violate one or another more easily than the existing literature on taboos suggests.
In addition, it is theoretically significant that the competing taboos Marshall confronted in 1945 had different origins. Despite the fact that the chemical weapons taboo originated in international law and the casualties taboo originated in American political and military culture, they nevertheless confronted him with nearly equal force. This makes sense psychologically. Regardless of their origins, both taboos—through experience, socialization, and moral discourse—had entered into his self-concept. Therefore, when he reached a situation where they seemed to conflict, it was experienced as a conflict of taboos, rather than as a conflict between an international and a national taboo. Neither the evidence in this case nor the underlying theory suggest that there is a consistent hierarchy of taboos. Instead, the outcome of each prospective tragic trade-off seems to be the product of the particular situation and the human deciders involved.
While the case demonstrates the effects of competing taboos on decision makers, Marshall's failure to alter U.S. policy also shows the limited influence of a single individual who is not an ultimate decision maker. He failed to do so because he was unable to persuade Leahy that breaking the gas taboo was a tragic necessity. This failure of persuasion may have been the product of divergent normative commitments (Leahy not sharing, or choosing not to privilege, the casualty norm) or different strategic beliefs (Leahy not believing that the United States would need to invade). Regardless, this failure is suggestive of the interactions between individuals, the social distribution of norms and values, and institutions in state decision making. The case suggests that decision makers' ability to convince others that their tragic conclusion is correct—and in some cases, that taboo violation is therefore necessary—depends on those others sharing a similar constellation of norms, taboos, and values. Accordingly, the extent of diversity within states matters in normative decision making. When there is greater diversity in decisions makers' normative constellations, persuading other stakeholders that a tragic violation is necessary may be more difficult. When there is broad consensus about norms, taboos, and values, however, persuasion—and thus violation—will be easier.
This lesson from the case is important. The need to persuade others that a hard moral dilemma exists, and that a tragic violation is the correct way to resolve the dilemma, is the crucial disjuncture between an individual decision maker's belief that a tragic normative violation is necessary and a state actually committing the violation. Decision makers must then convince others that their tragic conclusion is correct. This is particularly true if the decision maker lacks the ability to effectuate the decision alone—be it because they are not the ultimate decision maker or because other decision makers hold effective vetoes. Even if no one can veto their orders, they may still need to devise justifications for the public and other stakeholders.
Marshall's persuasive failure is also instructive about the importance of socialization and selection processes. The distribution of norms and values among decision makers and the public is to a great extent the product of socialization and selection processes. A counterfactual outcome is easy to imagine, since the replacement of a single man, the then-seventy-year-old Leahy, with a senior army officer would likely have meant the approval of Marshall's plan. This is because making Marshall's plan persuasive does not seem to have been a matter of de-internalizing (publicly or privately) the gas taboo, but of internalizing the casualties taboo and anticipating a bloody invasion of Japan. The casualties taboo seems to have been more advanced in the army relative to the navy during World War II, and the generals were more likely than the admirals to believe that an invasion would be necessary.
Implications
This study yields important findings for the study of forbidden weapons norms. It demonstrates that the taboos described by Tannenwald, Price, and others are psychologically valid concepts. It also shows, however, that under conditions of normative conflict, violating them may cease to be unthinkable. In circumstances experienced as tragic, decision makers may then violate them without negative psychological consequences. Together, these findings suggest the wisdom of policies that support weapons taboos, counterproliferation, and deterrence. While states possessing forbidden weapons may protest that they will abide by the norms forbidding use, these commitments may not be reliable should they find themselves confronting conflicting normative imperatives. This makes counterproliferation particularly important. At the same time, the possibility of a tragic violation means that threats to use forbidden weapons facing serious loss should be taken seriously. In these circumstances, deterrence may retain an important role in preventing the use of these weapons. Arguably, the existence of nuclear taboos supports both policies. Taboos make proliferation more controversial, because a state may be contemplating the unthinkable.Footnote 105 Taboos may also make deterrence more credible since the apparent evilness of a violator may make harsh responses seem justifiable to some decision makers and other stakeholders.Footnote 106
The study also makes more general contributions to the study of normative decision making in international relations. First, these findings clearly demonstrate that normative decision making is different from other kinds of decision making, and clearly different from the expectations of traditional rational decision theory. Second, the finding that the psychological effects of taboos on decision makers are more situationally dependent than earlier work had suggested suggests that attention to psychological research can enrich and add greater nuance to constructivist analyses. Third, the study helps clarify some of the dynamic interactions between individuals with distinct normative constellations and the social context implicit in group decision making. This may help further analyses of the theoretically tricky relationships between individual agents and socially constructed norms and structures. Each of these contributions may help further the emerging literature that seeks to improve psychological and social constructivist scholarship in international relations by building bridges between them.Footnote 107
