Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
The Medical Implications of Nuclear War, Institute of Medicine. @) 1986 by the National Academy of Sciences. National Academy Press, Washington, D.C. Sources of Human Instability in the Handling of Nuclear Weapons HERBERT L. ABRAMS, M.D. Stanford University, Stanford, California All men are liable to error The last error shall be worse than the first. John Locke Matt. 27:64 The resolution of superpower conflict to the point where nuclear war is no longer a threat is a 30- to 50-year project. In the interim, our survival depends on the degree to which we are able to manage short-term risk. No rational leader of the United States or the Soviet Union would embark on a nuclear adventure by design because they understand too well the impact of weapons of annihilation. Nuclear war, if it comes, is far more likely to be unintentional or accidental, based on miscalculation, misun- derstanding, or misperception. To focus on the short term is not to accept the nuclear stockpiles as reasonable or desirable avenues to national goals. Instead, it recognizes contemporary reality and attempts to come to grips with the means whereby nuclear peace may be extended for the foreseeable future. Human errors account for most failures in major weapons and space vehicles, more so than mechanical or technical failure (Dumas, 1980; p. 151. In the first year of a missile test program, 43 percent of all human *From the Center for International Security and Arms Control, Stanford University, and the Departments of Radiology, Stanford University School of Medicine and Harvard Med- ical School. 490
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 491 factor errors were due to assembly and installation mistakes (Britten, 1983; p. 171. Errors, accidents, or unintentional launches in the nuclear weapons systems of the great powers represent important potential triggers of hos- tilities during periods of international tension. This paper examines the issue of personnel reliability in the handling of nuclear weapons. It explores and summarizes the character of nuclear weapons duty and its impact on behavior; drug use, alcohol abuse, and psychiatric problems in the military and the degree to which they comprise a security risk; efforts to ensure reliability in nuclear weapons personnel; and the weaknesses of the Personnel Reliability Program (PRP). It con- cludes with suggestions for strengthening the PRP and improving the conditions of PRP personnel. THE CHARACTER OF THE WORK AND ITS IMPACT Nuclear weapons duty is known to be conducive to serious behavioral problems. The isolation that is often experienced at sensitive military bases and the boredom that accompanies many tasks tend to induce stress and may degrade job performance. The nature of service life in general has also been thought to encourage drug use. Over a nodal 4-year duty tour, intercontinental ballistic missile (ICBM) launch crew members spend up to a year's time in launch control centers. These underground bunkers, with their attendant inactivity, may come to resemble solitary confinement (Dumas, 1980; p. 28) and induce high levels of anxiety and alienation (Schmidt, 19811. Crews on long submarine patrols face the stress of leaving family behind for 2 or 3 months at a time (Serxner, 1968; p. 26~. They must also cope with the demanding routine of the 1-month refit period, which is when delicate, complicated repairs must sometimes be carried out under deadline pressure (Serxner, 19681. On patrol, isolation, monotony, and confine- ment affect all but the most stable (Weybrew and Molish, 19794. Depres- sion in the wives of submarine personnel is a problem that places added strains on them when they return home from patrol (Beckman et al., 1979~. During emergencies, sleep deprivation and heavy responsibilities may cause inaccuracy in judgment, hostility, or paranoia (Black, 19831. These effects are apparent even in those who appear to be at risk the least. It has been observed that "the most potentially dangerous situations in the navy have involved personnel who demonstrated no sign of psychiatric disturbance at the time of their initial assignment to militarily sensitive duties" (Christy and Rasmussen, 19631. Among the many efforts to enhance the survivability of nuclear weap- ons, the air force has apparently considered constructing a deep under
492 IMAGES AND RISKS OF NUCLEAR WAR ground base, 2,500-3,000 feet (about 762-914 meters) below the desert surface, with a 400-mile (about 644-kilometer) network for basing mis- siles. The base is designed to survive a nuclear attack and to be self- sufficient for at least a year. It would house enough personnel to tunnel their way to the surface and drive out the launchers to fire the missiles if ordered to do so (Halloran, 19841. Such personnel, if confined for long periods, would be particularly prone to behavior changes, especially if all communications were cut off. Prolonged isolation has been found to induce negative social and psychological symptoms among members of a group. One study noted an increase in boredom, irritability, depression, and hostility among military men during 105 days of social isolation (Rockwell et al., 19761. Others have found that group isolation brought out increased territorialism, along with incompatibility and withdrawal from group in- teraction (Altman and Haythorn, 1967; Altman et al., 19681. Lack of sleep, coupled with prolonged work, leads to fatigue, thirst, and physical complaints in 24 hours. After 100 hours, visual hallucina- tions, balance disturbances, slowed movements, and lack of vigilance have been observed (Opstead et al., 19781. In a prolonged nuclear alert, with missile crews working double shifts, bomber crews working on sus- tained airborne alerts, and submarine crews working beyond their regulated limit, the likelihood of error would surely increase (Morrison, 19831. Disruptions in internal timing systems and biological clocks are now recognized as a serious public health concern. In industries with round- the-clock work shifts, accidents and errors increase between the hours of 3 and 5 a.m., a time when normal circadian rhythms are at an ebb. Pilots flying aircraft simulators make more errors at this time, and eight times the number of single-vehicle truck accidents take place near 5 a.m. than at other times of the day. The Three Mile Island accident occurred at 4 a.m. with a crew that had just gone on the night shift (Moore-Ede, 19841. Work schedules themselves may cause degraded job performance. Cir- cadian cycles in humans rotate on a 24- to 25-hour basis. Within a day, people can adjust to time shifts of 1 to 2 hours; beyond this limit, ad- justment may take days. Weekly rotations, which force people to adjust to 8-hour time shifts each week, barely give adequate time to settle into a work pattern. The feeling is one of perpetual jet lag (Moore-Ede, 19821. The duty routines of the navy's nuclear submarine crews are organized around an 18-hour cycle, composed of three 6-hour shifts. Crew members work one shift and then take 12 hours, or two shifts, off before working another shift. Although this would appear to give them ample time for sleep and general rest, the cycle itself is not in keeping with the 24-25- hour circadian system; adjusting daily to a 6-hour change in work shift is humanly impossible. Among naval crews, this constant desynchrony man
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 493 ifests itself in a high incidence of insomnia, emotional disturbance, and impaired coordination. Furthermore, the turnover rate among the U.S. submarine crews is extremely high-up to 33-50 percent per voyage do not return for a second mission (Moore-Ede et al., 1983~. Repetitive tasks may impair job performance because of the physical effects of sensory deprivation or simply because of boredom (Dumas, 19801. We block out sounds after we have grown accustomed to their drone, or we stop seeing the familiar in our daily lives. Military personnel, who sit for hours watching lights bleep across an electronic screen, or guards, who stand watch day after day at a sensitive military post, react similarly. Dissatisfaction in such work may be enhanced by the belief of many who work in missile operations that the military would take no action to have them transferred even if they were placed in the wrong job (Parrott, 1973~. Stress in general, no matter what the source, has been associated with accidents of all types (Alkov and Borowsky, 1980; Connoly, 1981; Padilla et al., 1976~. In the air force much attention has been devoted to stress and pilot error (Alkov et al., 1982; Green, 1977; Yanowitch, 1977~. Aviators responsible for aircraft mishaps have often been found to cope poorly with stress. The stress itself is commonly associated with difficult life situations and with life-change events (Alkov and Borowsky, 19801. Crisis situations are among those in which stress is considerably height- ened and the chances of operator error are enhanced. Although moderate anxiety may improve performance, the rate of error for a given task is increased when high levels of apprehension or fear are experienced (La- gadec, 1982; p. 343~. Events perceived as threatening and uncertain aug- ment stress levels physiologically (Ursin et al., 1978; p. 6), which in turn can impair performance. One study performed on personnel responsible for maintaining Titan II missiles on a 24-hour alert demonstrated that continuous intense stress not only promotes fatigue but may ultimately be associated with collapse of performance (LeBlanc, 1977~. The nuclear reactor accident at Three Mile Island (TMI) is an example of multiple errors associated with a crisis situation. At the onset, the operators in the TMI control room had trouble in correctly interpreting the nature of their problem. They failed to obsene that an important valve in the reactor's pressurizer was open when it should have been closed, creating a hidden leak in the primary system. On the basis of this mis- understanding, the operators took several actions that increased the severity of the accident (Lewis, 19801. Part of the lack of insight has been attributed to an information overload in reading the barrage of emergency signals in the power plant control room. The problem has been described as one of "how to sort out the
494 IMAGES AND RISKS OF NUCLEAR WAR 100 odd alerts that rang out in a cacophony of hoots and bells" (Lagadec 1982; p. 111). The ambiguous nature of the warning signals aggravated the quandary. While clues that signaled that the crucial valve was open were available, they were not clear (Perrow, 1980-1981; p. 251. The operators' training added to their dilemma, rather than helping them cope. Lacking a theoretical background or perspective on their work, they functioned as button pushers instead; they were drilled for an accident in which only one thing went wrong at a time (Kemeny, 1980-1981; p. 51. A multiple failure (common-mode) accident was not anticipated by safety planners or stressed in operator training programs; the operators were not prepared for it. Information overload exacerbated the Three Mile Island accident, but information underload may have played a role as well. In heavily auto- mated control rooms, the lack of anything but routine work may lead to complacency and inattentiveness. "The burning question of the near fu- ture," one expert on aviation safety has said, "will not be how much work a man can do safely, but how little" (Weiner, 19771. Human error is also responsible for faulty components and systems failure. On June 3 and 6, 1980, false alerts on the air force missile attack warning system were triggered by a bad computer chip. Random failures within the early warning computer systems (EWS) can stem from unde- tected coding errors in the program software. The possibility of residual software errors is high because the complexity of such programs may prohibit them from ever being fully tested before use (Bereaunu, 19821. A further limitation in early warning programming lies in the inability of programmers to foresee all future circumstances and combinations of program inputs. It is virtually impossible to write a complete program in such a complex system. Computer errors inevitably result from this in- herent shortcoming of the system (Bereaunu, 19829. The missile alerts of June 3 and 6 did not trigger a nuclear incident because the control room officers correctly evaluated the warnings as false (U.S. Congress, Senate, 1980~. The question remains whether human judgment will always be this reliable. If an actual alert occurs in a manner or setting different from a planned alert scenario, will operators be able to act positively or will uncertainty override appropriate action? Perhaps the most disquieting weakness in command, control, and com- munication (C3) lies in its actual transition to nuclear alert. C3 personnel would face many of the same stresses that arose in the crisis at Three Mile Island. In peacetime, C3 operators' daily routines include monitoring the worldwide military situation, providing warning information, main- taining continuous communications, and participating in training exer- cises. Ball has emphasized that "the routine of peacetime operations leads
SOURCESOFHUMANINSTABILI7'YlN HANDLING NUCLEAR WEAPONS 495 to boredom and carelessness: messages are often filed without having been read, communications are often misrouted, and connections are carelessly interrupted. Most operators are simply not trained to make the rapid tran- sition from indifference to critical awareness that would be required of them in an emergency" (Ball, 1981, pp. 3-4; Bracken, 19831. Whether in C3 or in silos or submarines, there are factors other than boredom, inactivity, anxiety, and altered sleep patterns that profoundly affect behavior. Drug and alcohol abuse, a problem that is serious enough in civilian life, takes on a new dimension when it occurs among those who have access to nuclear weapons or military plans. ALCOHOL, DRUG USE, AND PSYCHIATRIC DISORDERS While the information base for this paper is for the U.S. armed forces, additional comments on the Soviet and British armed forces are important in recognizing the breadth of the problem. The U.S. Military Drug use was not considered an important problem for nuclear weapons personnel until 1970 (Laws, 1970; p. 50), when the military apparently realized that it had become a widespread phenomenon. In 1980, the De- partment of Defense (DOD) conducted a thorough survey of drug and alcohol use in the U.S. military, questioning over 15,000 randomly se- lected personnel (Burt, 1981-19821. The study revealed that 27 percent of the respondents had used drugs within the last 30 days (Table 11. Manjuana was used most often; 19 percent smoked it at least once a week (Table 2) (U.S. Congress, House, 19821. The 18-25 age group were most involved, with 50 percent reporting drug use in the last year and nearly 40 percent reporting it in the last 30 days (Table 2 and Figure 1) (U.S. Congress, Senate, 1982; p. 6~. The navy had the most severe problem among the four services. On questions of work impairment and drug use, 10 percent of the 18- 25 age group experienced lowered performance, while 19 percent had TABLE 1 Percentage of Population Using Drugs in the U.S. Military . . Drug Use Total Marine Air Period Beyond: DOD Army Navy Corps Force 30 days 27 29 33 37 14 12 months 36 38 43 47 23 SOURCE: Derived from Table 3 in Burt (1981-1982, p. 425).
496 IMAGES AND RISKS OF NUCLEAR WAR TABLE 2 Percent Prevalence of Drug Use among U.S. Military Personnel and Comparable Civilians (ages 18 to 25) Comparable Military Civilians Drug Use Beyond: (N = 8,224) (N = 2,022) Marijuana/Hashish 30 days 40 42 12 months 52 54 Amphetamines or Other Uppers 30 days 10 4 12 months 21 12 Cocaine 30 days 7 10 12 months 18 23 SOURCE: Summarized from Table 11 in Burt (1981-1982, p. 432). been high while working (almost half of these on 40 or more days during the past year). In the navy, 26 percent were high during work (U.S. Congress, House, 1982; p. 259) (Table 3~. In another survey of U.S. military stations in Italy and Germany (June and July 1981), on-duty drug use was described in 43 percent of the army, 17 percent of the air force, 35 percent of the marines, and 49 percent of the navy. On-duty drug use was slightly over 60 percent on the USS Forrestal. A total of 27 percent used uppers on duty in the navy, while 24 percent smoked marijuana on duty in the army (U.S. Congress, House, 1982~. The consumption of alcoholic beverages was widespread throughout all services (Table 41. Drinking on duty was also common: 28 percent in the army; 21 percent in the navy; 19 percent in the marines; and 15 percent in the air force (U.S. Congress, House, 1982; pp. 277-278~. The highest TABLE 3 Percentage of Personnel with Work Impairment due to Drug Use Total Marine Air Impairment DOD Anny Navy Corps Force Total 21 22 28 28 9 High while working 19 21 26 25 8 Lowered performance 10 12 15 13 3 SOURCE: Reprinted from Burt (1982, p. 431), by courtesy of Marcel Dekker, Inc.
SOURCES OF HUMAN NEST IN HIDING NUCLEI WEAPONS 497 90 80 70 60 50 40 30 20 10 o Alcohol DRUG USE AMONG MILITARY PERSONNEL AG ES 18-25 PAST 30 DAYS Marijuana/Hashish Amphetamines 1 0% Cocaine 6 Hallucinogens Barbiturates 4% Heroin ~ 1% 1";;'';:;:;2':-':::: 1 ~= ~,l 1980 WORLDWIDE DoD STUDY FIGURE 1 Drug use during the previous 30 days among military personnel ages 18-25. Source: U.S. Congress, Senate (1982~. prevalence of drinking was reported by the senior officers (Burt, 1981- 1982~. Alcohol dependence was found in 7 percent of all military per- sonnel. Another 27 percent suffered some degree of work impairment due to alcohol (U.S. Congress, House, 1982~. An accident aboard the USS Nimitz brought at least part of the drug problem to light. On May 26, 1981, a Marine EA-6B aircraft crashed on the deck of the USS Nimitz, killing 14 people, injuring 44, and damaging 20 other aircraft. At autopsy, traces of marijuana were found in the blood of six deckhands who died. A House Subcommittee pointed out that "the relevance of this information becomes more pronounced when considered with the fact that the USS Nimitz had been at sea for 11 days prior to the accident" (U.S. Congress, House, 1982; p. 2801. The pilot of the plane that crashed was found to have 6-11 times the normal blood level of brompheniramine, an antihistamine, which had not been prescribed by a military physician. Such a level may cause sedation, dizziness, double
498 IMAGES AND RISKS OF NUCLEAR WAR TABLE 4 Percentage of Worldwide U.S. Military Population Using Alcohol During the Previous 30 Days Total Manne Air DOD Army Navy COIPS Force 83 80 86 86 82 SOURCE: U.S. Congress, House (1982). vision, and tremors. The naval report concluded that "the presence of this drug, combined with other stress factors, precipitated the pilot error which caused this accident" (U.S. Congress, House, 1982; p. 281~. The same investigation also uncovered the regular use of amphetamines by crew members to sustain themselves during 1 8-hour workdays. Twenty- eight percent of navy personnel used amphetamines and did so habitually, rather than casually or for recreation purposes. The accident prompted a large-scale test of urine samples of naval personnel for marijuana, cocaine, opiates, and other drugs; this was fol- lowed by a sharp decline in the percentage of those shown to be using marijuana (New York Times, November 8, 19841. The Naval Safety Center reported that alcohol and hangover effects were involved in 15-20 percent of the major aircraft accidents in 1979 (U.S. Congress, House, 1981~. In six army accidents during a 5-year period, drugs were found in the blood samples of military aviators. These drugs included marijuana, cocaine, methaqualone, and, in one case, in- sulin prescribed by a civilian physician without knowledge of the military. The marines also reported eight instances of drug abuse that involved aviators (U.S. Congress, House, 1982; p. 4391. The official policy of the DOD is to prohibit the induction of drug- or alcohol-dependent individuals. Each service implements this policy in- dependently, but all "appear flexible enough to enlist anyone except those convicted of trafficking or declared drug dependent by a medical author- ity" (U.S. Congress, House, 1982; p. 432~. In recent years, the armed forces have experienced a measurable de- crease in drug use, according to the Research Triangle Institute of North Carolina (New York Times, January 21, 19861. Nevertheless, the problem remains an urgent one for nuclear forces. In 1982, the House Appropri- ations Investigative Subcommittee found a large number of drug incidents and arrests involving personnel working with nuclear weapons. In one case, four marijuana pushers in an army nuclear missile battalion were identified. An army husband and wife team managed the enterprise and indicated that 125 others in the unit were users. Twenty-three were in the
SOURCES OF HUMANINST~INH^DLING NUCLEI WEAPONS 499 nuclear weapon PRP and remained there because of the lack of hard evidence to deny them security clearances. At another army nuclear missile site, a company commander disregarded for 3 months a subordinate's request for a drug detection program. He knew that the likelihood of PRP personnel replacements was slim and therefore ignored the army's policies on drug enforcement until some of his soldiers were arrested for drug possession. A nuclear weapons guard on a base in Germany was arrested for smoking hashish while walking his post (U.S. Congress, House, 1982~. As recently as last year, 1,400 individuals were decertified from nuclear weapons handling because of drug and alcohol abuse. The submarine force maintains a zero drug tolerance policy, with of- fenders being transferred to another post. During 1980, 589 enlisted men 75 percent of all navy transfers were removed from the Atlantic sub- marine force, and for many this was because of drug abuse. "Cost and time to train replacements obviously has a readiness impact when the total submarine force constitutes 32 percent of all navy war ships" (U.S. Con- gress, House, 1982~. The evidence is compelling that all of the factors associated with drug use in the civilian population and in the military at large are operative- at times in accentuated form among the forces responsible for nuclear weapons. PSYCHIATRIC PROBLEMS The annual incidence of psychosis in the armed forces around the world has been estimated at 3 per 1,000 in peace and war (Group for the Ad- vancement of Psychiatry, 19641. The figures for neurosis and personality disorders are higher. Psychiatric hospitalization rates for navy men from 1966 to 1968 were 689 new cases per 100,000 per year. Thus, nearly 0.7 percent were so obviously unstable that they required in-patient treatment. Of these, 78 per 100,000 were psychotic, 184 were neurotic, 334 suffered from character and behavior disorders, and 93 exhibited "situational mal- adjustment" (Schuckit and Gunderson, 19731. Psychiatric disorders can manifest themselves in a number of different ways, including alcoholism or discipline problems. Unstable individuals may be separated from the service through administrative actions, such as general court-martial, without ever being seen by a psychiatrist (Arthur, 1966~. Mentally ill crew members may be referred to naval installations far from the home base, creating a care drain (Satloff, 19671. Commanding officers may be unwilling to cite psychiatric disorder as a cause for transfer because of negative attitudes about such a label. All of these factors may render the data base uncertain.
soo IMAGES AND RISKS OF NUCLEAR WAR Nevertheless, the magnitude of the problem becomes apparent when the distribution of psychiatric diagnoses in the armed forces is analyzed. Dunng World War II, a total of 160,285 men were discharged from the navy on psychiatric grounds. The psychiatric patients typically came from unstable families, had below average score work performances, and com- plained of frequent physical illnesses (Arthur, 19661. In a longitudinal study of 11,000 naval enlisters, Plag et al. (1970) found that approximately 1 enlister in 12 (~.7 percent of the total) was discharged during his first enlistment because of psychiatric illness. From 1980 to 1984, over 45,000 individuals in the army had psychiatric disorders (Table 5~. Approximately 4,500 were schizophrenic, while an additional 3,000 had other psychotic disorders. Dependence on, abuse of, or psychosis from drugs and alcohol accounted for 30,000 cases. Another 4,400 had neurotic disorders and 10,000 had personality disorders (De- partment of the Army, 19851. Psychiatric instability, together with the tensions of military life, has been reflected in the number of assaults on officers in the armed forces. From 1968 to 1972, 550 such assaults are said to have occurred, resulting in 86 deaths (Gillooly and Bond, 19761. The suicide rate in the U.S. Army (10.1 per 100,000 soldiers [Redman and Walter, 19851) is probably not related only to the stress of military TABLE 5 Distribution of Psychiatric Diagnoses in U.S. Army (1980 1984) Diagnosis19801981198219831984Total Alcoholic psychoses10613315812598620 Alcohol dependence2,6242,7722,5732,2392,27412,482 Drug psychoses7662585122269 Drug dependence329199144145101918 Drug abuse1,1321,0519417671,0154,906 Schizophrenia1,4651,0648736574704,529 Neurotic disorders8476214194083852,680 Personality disorders1,2231,1096986777564,463 Other psychotic states5254785615116132,688 Mental disorders following organic brain damage12511683105121550 Other diagnoses2,5082,1852,1292,1402,22311,185 Total10,9609,7908,6377,8258,07845,290 SOURCE: U.S. Army Patient Administration Systems and Biostatistics Activity: Disposi- tion and Incidence Rates Active Duty Army Personnel, Psychiatric Cases, Worldwide, CY 80-84.
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 501 life, although 40 percent of those who committed suicide were said to have difficulties with their work in the army. A large percentage also have marital or family problems (Rothberg et al., 1984~. The rate is slightly lower in the military than in the civilian population (Redman and Walter, 19851. A recent drop in suicide rate has been associated with the shift in the white-black ratio from 3.7 to 2.2 and is probably explained by the lower suicide rate among blacks and their relatively larger number in the armed forces (Date] and Jones, 1982~. The risk of mental illness is especially acute in nuclear armed sub- marines, in which crews remain for months at a time. In a comparison of submarine crews with a control group of surface fleet personnel subject to the same health screening, the neuropsychiatric illness rate during 1968- 1973 was twice as high among those in submarines (Tansey et al., 19791. Up to 5 percent of the crew that were on Polaris submarine patrols for 2 months required treatment for psychological problems, ranging from minor anxiety to acute psychosis (Serxner, 19681. In another study, 3.8 percent of nuclear submarine crews were referred for psychiatric consultation. Of these, 1.9 percent had disorders that were serious enough that they could not return to the submarines. Eight percent of those referred were psychotic (SatIoff, 1967~. The Soviet Military Accurate information on sources of human instability in the Soviet armed forces is not readily available. Emigre accounts often have a strong anti-Soviet bias, and accessible Soviet government sources are generally lacking in objective, quantitative data. Nevertheless, it is clear that Soviet military service may be associated with several kinds of stressful expe- riences. The very first encounter of the recruit is frequently a shock: Draftees are packed off in railroad cars to unknown destinations without the slightest ideas of what they will be doing .... The abbreviated Gaining course, which puts soldiers under tremendous physical and psychological pressure, is made even more overwhelming by their geographical isolation: Hey are stationed in faraway places, often in unfamiliar climates (Tarasulo, 19851. Living conditions in the barracks, or kazarma, are a source of constant stress. The chief problem is overcrowding: 50 to 100 men are packed into a room, with an estimated 2 square meters allotted per man. There is no opportunity for privacy. The barracks are often freezing in winter and excessively hot in summer; they tend to be rancid and poorly ventilated. There is a shortage of hot water, and the few toilets are usually outdoors, even in freezing climates.
502 IMAGES AND RISKS OF NUCLEAR WAR Food is often inadequate and tasteless. The prevalence of illnesses such as running sores, acne, dental problems, night blindness, oversensitive eyes, or eye infections may at least partially attest to a vitamin-deficient diet. Troops attempt to supplement their military diet by growing produce and raising livestock, especially pigs. There are differences between services in the quality and quantity of food served, with navy, air force, and border guards generally receiving better rations. In any case, poor nutrition must to some extent undermine soldiers' physical and mental potential (Gabriel, 1980; Cockburn, 1983; Tarasulo, 1985~. Ethnic strains pose another problem. Barely half of the Soviet population is Russian, while the 90 or more ethnic groups speak 130 languages. Not only do non-Russian soldiers experience learning and social difficulties because Russian is the standard language in the military, but they face discrimination in placement and in their dealings with their fellow soldiers. Such tensions are probably less of an issue in the strategic missile forces. These elite positions are more ethnically homogeneous, composed mostly of Russians (R. Anderson, University of California, Berkeley, personal communication, 1985~. Systematic hazing of younger recruits is a feature of Soviet military life. Early in their 2-year term of duty they are forced to give up food rations, exchange new uniforms for older ones, perform unpleasant tasks intended for older recruits, and hand over part of their monthly allowance. This caste system is enforced by older soldiers through physical violence and has even led to death by beating (Myagkov, 1976; Suvorov, 19824. The Soviet military has been described as "an organization riven with hatred and strife to an extent that occasionally borders on anarchy" (Cock- burn, 1983; p. 781. Stress manifests itself in different ways. In a survey of Soviet emigres, 49 percent felt that soldiers in their unit were absent without leave (AWOL) often or very often. High AWOL rates may be of particular concern as a reflection of low morale and discipline problems. The frequency with which military leaders are physically assaulted by their own men has also been noted. Of those surveyed, 36 percent had witnessed assaults on officers and 63 percent had direct knowledge of assaults on noncommissioned officers. This kind of insubordination and violence embodies an undue measure both of stress and of disaffection. Suicide is another index and there appears to be a high rate in the Soviet armed forces (Gabriel, 1980; p. 1781. Reliability may also be affected by the way in which Soviet troops are trained. For the most part, this takes place in military academies and
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 503 training corps. It consists of drills, marching in formation, instruction about their weapons, and field experience with older or cheaper weapons and equipment than would actually be used in combat. Because of the limited amounts of weapons and supplies available, soldiers may fire an average of only three bullets a year (Tarasulo, 19851. One incident reported in the Soviet military press involved a crew intent on increasing its launching speed for the sake of socialist competition. The men ignored the requirement that ladders be used to mount the mobile launcher and missile and, consequently, damaged the missile. Another account described an antiaircraft unit's eagerness to impress the senior commander with their speed at readying the missile for firing. The launch crew focused so much on speed that they forgot accuracy and coordination (Goldhamer, 1975; pp. 121-1221. While much of the information presented above is based on emigre sources, the problems of alcoholism are treated directly in the Soviet literature. Alcoholism is a health problem of epidemic proportions in the Soviet Union. The high death rate from acute alcohol poisoning indicates the extent of abuse. There are approximately 45,000 deaths annually from acute ethanol poisoning, a figure that is 100 times as high as in the United States (400 per year) (Treml, 1982b). Over the past 20 years the per capita consumption of alcohol has more than doubled (Davis, 1985~. Alcohol abuse throughout the ranks of the Soviet military has also been documented (Gabriel, 1984, p. 119; Goldhamer, 1975, p. 152) and is higher than in the civilian population (Davis, 1985~. A typical soldier cannot afford and is not legally allowed to obtain state-produced vodka. Many drink instead surrogate alcohol, such as shaving lotion, varnishes, cleaning fluid, antifreeze, brake fluid, shoe polish, and glue. The toxicity of these home brews, consumed throughout the country, explains in part the high acute alcohol-poisoning rate. According to one estimate, alcohol dependence affects one-third of the Soviet military (Wimbush, 1981~. The effect of this epidemic, combined with extremely poor nourishment and the stress of overcrowding in the barracks, is to undermine Soviet defense capabilities. Furthermore, it is thought that the Soviet officer corps and noncommissioned officers may "account for more than three-quarters of alcoholics and heavy drinkers" (Davis, 1985~. Such factors have a definite, though undetermined, impact on the issue of human reliability and nuclear weapons in the Soviet Union. Little has been written about the abuse of drugs other than alcohol in the Russian military. Hard drugs are probably not a major problem; the annual death rates from these are low in the USSR, far below those in the United States (Treml, 1982a). On the other hand, large numbers of
504 IMAGES AND RISKS OF NUCLEAR WAR soldiers returning from Afghanistan are said to have used hashish and to have transported it into the Soviet Union in significant quantity (R. Pipes, Harvard University, personal communication, 15~84~. Although there is virtually no discussion of nuclear weapons handling in the available Soviet literature, Soviet military doctrine clearly acknowl- edges the possibility of human error in causing nuclear war. In a July 1982 Pravda article, Defense Minister Dimitri Ustinov (1982) stated that one pressing military issue involved setting up a still more strict framework in the training of troops and staff, the determination of the composition of arms, Me organization of still more strict control, for the assured exclusion of the unsanctioned launch of nuclear weapons from the tactical to strategic. The chain of command in the Soviet military flows from the Politburo through the Defense Council to the main Military Council, which governs the general staff, the direct controlling agent for strategic nuclear forces (Ball, 1981~. If the general staff granted release authority, theater nuclear weapons would fall under the control of the theater command or directly under the control of the front command, which is in turn responsible for dispersing the nuclear weapons to lower commands and authorizing the launch of these weapons. When the lower commands receive authorization, they would have a considerable degree of flexibility in deciding when and whether to use the weapons under their control. One top Soviet militant official has observed that Transfer of nuclear weapons to the disposal of strategic, operational, and tactical command echelons gives each great independence and enables them to choose for themselves the means and methods of military operations within the zones of their responsibility and within the bounds of their authority (Zav'yalov, 1971~. Although strategic and tactical weapons have been fitted with electronic locks to guard against unauthorized launching (Meyer, 1985; p. 188), with the devolution of authority in the field, the issue of human instability requires serious consideration. The British Military Alcohol has had a long history in the British military, especially in the navy. Since the seventeenth century, sailors have been able to draw on a daily ration of rum, or grog. Tot-time was not abolished until 1970, although attempts to control alcohol consumption, such as diluting the grog and reducing rations, were initiated periodically.
SOURCES OF HUMAN NEST IN H^DLING NUC~E~ WEAPONS 505 Alcohol abuse affects all ages and ranks. One physician has noted that many alcohol users in the military "are highly (and expensively) trained, yet deteriorations in functions are often allowed to continue until a major medical or disciplinary crisis makes avoidance of their recognition im- possible" (Blunder, 1981~. While the abolition of the rum ration marked the first major step toward controlling alcoholism in the navy, alcohol treatment units throughout the military have now been established, and preventive education programs are stressed (Hires, 1980, 1981~. In consonance with the U.S. military, the British have identified boredom, underemployment, lack of job sat- isfaction, lack of group cohesion, and stress among the factors that lead to alcohol abuse. The military's traditional acceptance of alcohol, its ready availability, and its widespread use among the recruiting population add to the problem. For soldiers stationed in northern Europe, alcohol is even more available; there are few licensing restrictions, and it may be pur- chased cheaply, at duty-free prices (Wood, 19801. Information on drug abuse is scarce. In a parliamentary response, 73 cases of drug abuse in the British military in 1983 were acknowledged (Thompson, 1985; p. 59~. Within a 2-month period at one military hos- pital, soldiers were observed "abusing heroin intravenously, cannabis, amphetamines, lysergic acid diethylamide (LSD), barbiturates, analgesics, antiparkinsonian agents, benzodiazepine, and glue" (Blunder, 1981~. Marijuana is readily available and widely used in the British armed ser- vices. (Opiates are rare because they are difficult to conceal.) Barbiturates and amphetamines are not as readily available to soldiers. The abuse of prescribed drugs, which can be purchased legally, is also common. If the American experience is relevant, the patterns in the British mil- itary, in general, are applicable to nuclear weapons personnel as well. Not long ago, the British discovered and court-martialed a heroin ad- dict who also used hallucinogenic drugs. He had served on two nuclear submarines and apparently was often stoned while on duty (Boston Globe, October23,19811. No description of the screening procedures for personnel reliability in the British nuclear forces has been made available. SUBSTANCE ABUSE AND PSYCHIATRIC DISORDERS AS A SECURITY RISK What kind of security threat is posed by psychiatric and drug-related disorders in the military? The nature of the risk depends on the kind of psychiatric disorder or drug use, the individual's responsibilities, and the extent to which a state of crisis is pending or exists.
506 IMAGES AND RISKS OF NUCLEAR WAR "The psychotic individual (manic, depressed, or schizophrenic) rep- resents a severe actual or potential danger to himself, to those around him, and of course to the military mission of his unit" (Kleinmann and Krise, 1957; pp. 1008-10091. Such individuals must be removed from sensitive duties. Because the symptoms of psychosis may appear gradually, it is rare to find an actual breach of security by a psychotic. Neuroses present another category of psychological disorder that may be damaging to military operations. Although the neurotic may sometimes be unreliable, he is not necessarily a security risk; "frequently, he is a very conscientious worker transferring much of his concern and compul- siveness to his job" (Kleinmann and Krise, 1957; p. 1008~. Instances of neurosis are best evaluated individually, on a case-by-case basis. Never- theless, many feel that the neurotic should be removed from a position that involves highly sensitive duties and transferred to a less critical job. Character and behavior disorders represent the most common form of psychiatric disability (Arthur, 1966; p. 3601. The danger from such in- dividuals may lie in their difficulty in practicing sound judgment, a prob- lem which is especially threatening in times of crisis. Disorders of this nature are difficult to detect and may not be apparent until after some breach of security. The individual with suicidal tendencies presents a unique threat to se- curity, especially in work with nuclear weapons. Attempted suicide, the air force states, is "particularly significant when attempted through the deliberate detonation of a weapon or crashing of an airplane" (Department of the Air Force, 1983; p. 121. The impact of drug and alcohol abuse also varies with the substance used and the pattern of use. While the more social drugs such as marijuana or alcohol generally slow reaction time and impair judgment, harder drugs such as LSD, heroin, or PCP may cause more dangerous behavior. The more profound alteration of perceptions presented by these hard drugs probably presents a greater security risk than do the effects of softer drugs. Air force regulations reflect this disparity: If the investigation [during the Recertification process] reveals the use of a hal- lucinogen either posing the threat of flashback (for example, LSD) or severe behavioral effects (for example, PCP), the certifying official must impose a per- manent Recertification (Department of the Air Force, 1983; p. 39~. But excessive use of any substance may dangerously diminish relia- bility. The air force distinguishes between problem drinkers, who are temporarily decertified and entered into a rehabilitation program, and chronic alcoholics with little chance for rehabilitation, who are perma
SOURCES OF HUMAN INSTABlL17~Y IN HANDLING NUCLEAR WEAPONS 507 nently decertified. The type and pattern of use materially affects the nature of the security threat. The risk also depends on the individual's responsibilities. Personnel in the nuclear release system are of critical importance in weapons security. Unreliable personnel with immediate access to strategic nuclear weapons, such as launch capsule officers for silo-based ICBMs, are a potential hazard. Checks and safeguards on them are rather tight: once the author- ization code is sent, a total of four officers in two separate launch control capsules in the same squadron still need to decide individually to launch the ICBMs before they can be fired. When a launch has been ordered, any other capsule can send out an inhibit command (Senate Armed Services Committee, 1977; p. 68451. While the consequences of an accidental or unauthorized ICBM launch would be disastrous, the likelihood of such a launch is correspondingly low. Unreliable crew members on ballistic missile submarines also present a security risk. Because the navy maintains no permissive action links (PAL) on its submarine-launched ballistic missiles (SLBMs), the captain and several officers on board can decide to launch a nuclear weapon without higher authorization. A drugged or mentally disturbed individual could not decide to launch a weapon alone; the procedure requires the agreement of several crew members. The navy claims that a minimum of 30 submarine personnel must cast their votes before a weapon can actually be fired, though others contend that the true figure is lower (Meyer, 1984b). PALs are also lacking on aircraft carriers and other vessels that carry nuclear weapons (Center for Defense Information, 19851. Perhaps the greatest threat to security involves unstable individuals who have access to theater nuclear weapons,- including bombs and depth charges on nonstrategic aircraft; air, sea, and land-based cruise missiles; nuclear land mines; artillery projectiles; and short-range ballistic missiles. While short- and medium-range weapons can do less damage than strategic weap- ons, the controls over these weapons are also looser. During a crisis, such controls are apt to loosen even more as commanders release authorization codes to personnel in order to avoid delay if the weapons are to be used. Such Revolution of control would put many in a position to launch theater nuclear weapons. While the two-man rule is applied to theater weapons, the possibility for their unauthorized launch is considerably higher. In some military services, the small pool of applicants or a high attrition rate may increase the likelihood that suboptimal personnel will be se- lected e.g., those with psychiatric disorders or drug habits. For instance, the submarine service in the late 1960s had a selection ratio of 80 to 95 percent, due to an attrition rate of about 20 to 25 percent (Kinsey, 19681.
508 IMAGES AND RISKS OF NUCLEAR WAR The service also suffered from a lack of volunteers for long nuclear sub- marine patrols (D. Morrison, Center for Defense Information, personal communication, 19851. Not only is the training very rigorous, but in times of personnel shortage, individuals may be selected for submarine duty even when it is their last choice (Serxner, 19681. Line commanders facing a personnel shortage may utilize an individual for short-term duties, even with the knowledge that the person is mentally unstable. "While theoretically this should not occur, in reality it has happened and undoubtedly will in the future" (Christy and Rasmussen, 1963; p. 543~. The ultimate decision on whether someone presents a human reliability risk lies not with the physician or psychiatrist but with the line commander. The possibility of one serviceman "running amuck at the controls" is far less significant than the problem of psychological disorders leading to reduced efficiency and reliability. Similarly, frequent or habitual use of recreational drugs or alcohol takes a toll, with reduced performance while under the influence or hung over and long-term degradation of health and mental alertness. Although the impact of drugs and alcohol on military readiness has not been formally studied, the Deputy Secretary of Defense stated in June 1981 that a military unit has never been declared noncombat ready because of drug or alcohol abuse. This may be because the DOD reporting system uses the quantity not the quality of such items as equipment and per- sonnel to define readiness (U.S. Congress, House, 19821. In the belief that officers closest to the troops have the most compre- hensive view of the drug problem, 167 military commanders in Europe, the United States, and the Pacific were surveyed. Forty-seven percent of the commanders recognized some lowering of combat readiness due to drugs. Nevertheless, only 10 percent believed that drug abuse had a spe- cific or identifiable impact on readiness. Most often it was seen as a casual matter and primarily a discipline problem. "Several commanders felt that the marijuana user posed no problem until he was caught, after which the administrative process began" (U.S. Congress, House, 1982; p. 2841. By strictly employing enforcement measures, a commander may lose necessary personnel. Through frequent drug searches, he risks alienating his troops and lowering morale. There are good reasons to tolerate drugs rather than enforce discipline. Many commanders felt that alcohol abuse posed a larger threat to readiness because senior officials were more likely to be involved. What seems unassailable is that the navy would have no way of pre- venting a determined submarine crew from launching its missiles and that
- . SOURCES OF HUMAN INSTABILITY lN HAND KING NUCLEAR WEAPONS 509 it "must rely on the sanity of the skipper and his crew to save us from Armageddon" (Volz, 1983). THE PERSONNEL RELIABILITY PROGRAM Approximately 120,000 individuals have access to U.S. nuclear weap- ons. The military takes special precautions to ensure that these people are suitable for their positions. The heart of the effort is the Personnel Reli- ability Program (PRP), which is designed to screen out unstable individuals and "to ensure that such weapons are not subject to loss, theft, sabotage, unauthorized use, unauthorized destruction, accidental damage or jettison . . . only those personnel who have demonstrated unswerving loyalty, integrity, trustworthiness, and discretion of the highest order shall be employed in nuclear weapon PRP positions" (Department of Defense, 1981; p. 11. Among the critical positions that fall under PRP jurisdiction are com- manders of nuclear weapons delivery units, pilots and crews of delivery aircraft, and delivery unit personnel with access to and technical knowl- edge of nuclear weapons. In addition, so-called controlled PRP positions include security guards, storage and supply personnel, and launch per- sonnel in nuclear missile silos. Qualifying standards for the PRP include the following: · Physical competence, mental alertness, and technical proficiency · Evidence of dependability in accepting responsibilities and accom- plishing duties, and flexibility in adjusting to changes in the working environment · Evidence of good social adjustment, emotional stability, and ability to exercise sound judgment in meeting adverse or emergency situations · Positive attitude toward nuclear weapon duty The initial screening procedure includes a background investigation and a security clearance. A medical evaluation is also required, to be completed by "competent medical authority or other medical personnel who are specifically trained and designated to perform this function" (Department of Defense, 1981, Enclosure 5, p. 3~. It may begin and end with a review of the candidate's medical documents. Only if these are considered in- adequate will an actual examination be performed. Although consideration of psychological problems is a part of the basic evaluation, again, no interview is required. The process also embodies a review of the candidate's personnel files and a personal interview advising the individual as to the nature of the
510 IMAGES AND RISKS OF NUCLEAR WAR PRP. The interview is intended to determine the candidate's attitude toward the job. On completion of a course of study relevant to the position, a proficiency certificate is issued. After certification, medical and personnel records are marked, alerting physicians and others to the sensitive nature of the work. Supervisors are advised that the individual is in the PRP. Factors that might impair per- formance-medication, for example, or family problems (such as a wife's miscarriage or the death of a close relative) can then be acted on when recognized, generally by temporary removal of the person from nuclear weapons duties (Department of Defense, 1981, Enclosure 5, p. 3; L. Meyer, 1984a). There is no formal monitoring system or regular review of PRP mem- bers. Those in the program are responsible for reporting any factors that might impair their own performance or that of their peers, while super- visors, managers, and other authorities are expected to observe them daily and continue to assess their reliability (Department of Defense, 1981; Department of the Air Force, 1983, p. 41. Failure to meet the PRP qualifications theoretically results in temporary or permanent decertif~cation, depending on the situation. Personnel are disqualified for the following reasons: · Alcohol abuse · Drug abuse (isolated or experimental use of marijuana or hashish need not be automatically disqualifying; in each case, the certifying official must judge the impact of the drug use on the individual's reliability) · Negligence or delinquency in perfo'~ance of duty · Court-martial, nonjudicial punishment, serious civil convictions or behavior indicative of contemptuous attitude toward law or authority · Aberrant behavior: mental, physical, or character traits that would lead to unreliable performance · Poor attitude or lack of motivation Relatively little specific information is available on the Soviet coun- terpart of the PRP i.e., individuals involved with nuclear weapons. Presumably, their selection and training procedure produces soldiers of greater reliability and skill than the average draftee. Military schools that train cadets to become lieutenants in the Strategic Rocket Forces have a special entrance requirement: the students' political reliability must be officially certified by a military commissariat officer (Scott and Scott, 19841. Nevertheless, the PVO (Nation Air Defense) suffers from the same kinds of problems as the other services. The Soviet Navy which carries a special tradition and prestige has difficulty attracting sufficient num- bers to serve on missile submarines. Many are unwilling to endure the
SOURCES OF HUMAN INSTABILITY IN HANDLING NUCLEI WEAPONS 511 strains of long patrols. Consequently, while the U. S. Navy has two rotating crews for each submarine, the Soviet Navy maintains only one crew per boat (Cockburn, l983~. The Soviet Union is clearly aware of the psychological-physiological stress to which nuclear weapons personnel are subject. Classes on psy- chological will are given at the training center for officers responsible for launch controls. An article in Pravda quotes an officer on the shift: "When an input comes in, in your soul there's an anxiety: is this combat, or is it training?" (Gorokhov, 1985~. A recent article in the Soviet military medical journal emphasizes the importance of psychological testing in an era of complex military tech- nology (Bodrov, 19841. The psychological selection process involves ex- amination of the individual's analytic abilities, motivation, capacity to communicate, leadership qualities, and many other factors. Those who score high are much more successful in completing flight training than those with lower scores. Such tests have been used for many years in the USSR, and they also assess emotional stability, perceptiveness, and at- tentiveness (Solov'ev et al., 19711. Personnel in whom these psychological measures are less satisfactory often experience tension, confusion, fatigue, and greater tendencies to accidents. In all likelihood the Soviet screening program for nuclear weapons personnel shares many features with that of the U.S. armed forces. But because of the high degree of secrecy that surrounds the Soviet military, it is impossible to be certain. WEAKNESSES OF THE PERSONNEL RELIABILITY PROGRAM A system designed and implemented by humans cannot be 100 percent effective. In 1957-1958, the air force engaged a group of psychologists, psychiatrists, and others specializing in human resources and accident prevention to address the question of human fallibility and nuclear safe- guards. The committee concluded that effective screening might reduce the likelihood of an unauthorized launch, but that because of human fallibility it was impossible to safeguard all acts involving nuclear weap- ons (Laws, 1970; p. 251. Because secrecy surrounds many PRP assignments, medical officers may have little knowledge about the positions for which they interview personnel. Lack of experience with the type of work and the role expected of the applicant may make it difficult for the physician to assess his suitability for the position. Furthermore, the initial screening need not include a medical or psychiatric examination. A record review may suffice.
512 IMAGES AND RISKS OF NUCLEAR WAR But in some cases, records give no indication of serious flaws that render an individual unfit for nuclear weapons duty. Psychological prob- lems are a case in point. For reasons of confidentiality, mental health clinic records are kept separate from the rest; thus, unless the main record includes a referral for psychiatric treatment, there may be no indication that an individual ever required or received it. In bases that maintain mental health clinics, the psychiatric records are available for review, but not all bases operate such clinics. Certain indications of psychological instability may be evident from medical documents. These include frequent appearances on sick call with vague and questionable health complaints; overdependence on medications and tranquilizers; repeated entries reflecting accident proneness or alcohol abuse; and stress-type headaches, chronic diarrhea, or other psychological symptoms of stress (Kentsmith, 19811. Besides the initial screening, systematic, meaningful follow-up by Pained personnel might seem a logical next step. Instead, managers or superiors are relied on to report actions that indicate a lack of dependability (De- partment of Defense, 1981), and coworkers are expected to evaluate each other (L. Meyer, 1984a). Needless to say, inertia, camaraderie, and peer pressure may hamper honest evaluations of reliability. The occasional violations of PRP security rules that come to light are not reassuring. In 1969, an air force major allowed three men under his command to continue their nuclear weapons security work even after they were identified as having dangerous psychiatric problems. One of the guards allegedly lost control with a loaded carbine at the base. The major testified that he had known of the men's psychiatric problems, but he had kept them on because he was understaffed (New York Times, August 18, 19691. The Walker spy case represents a striking example of the problem of ensuring reliability in the military in general and in nuclear weapons personnel in particular. John Walker, Jr., was in the Naval Nuclear Weap- ons PRP as a petty officer with communications training and a high-level security clearance. He served as a communication specialist on a Polaris submarine. As leader of the espionage ring that sold classified documents to the USSR from 1968 on, he represented one of the most serious security breeches in recent naval history. Clearly, the PRP was unable both to detect and to predict this radical aberration from accepted norms of reliable behavior (Magnuson, 1985; McGrath, 19851. The data on PRP Recertification are disturbing. From 1975 to 1984, an annual average of 112,000 individuals, all screened by the PRP, were employed to handle nuclear weapons (Table 6) (Department of Defense, 1975-1977, 1978-19841. During that period, 51,199 individuals were decertified, which is about 5,100 per year. If the assumed length of service
SOURCES OF HUMAN INSTABILITY IN HANDLING NUCLEAR WEAPONS TABLE 6 Total Number of PRPs and Number Decertified ( 1975-1 984)a Year Total PRPs Decertifications 1975 119,625 5,128 1976 115,855 4,966 1977 118,988 4,973 1978 116,253 5,797 1979 119,198 5,712 1980 114,028 5,327 1981 109,025 5,235 1982 105,288 5,210 1983 104,772 5,085 1984 103,832 3,766 aAverage, 112,686; total, 51,199. SOURCE: Department of Defense (1975-1977, 1978-1984). 513 is 1 year, that amounts to 5 percent of the total; if the length of service is 2, 3, or 4 years, it adds up to 9, 14, or 18 percent, respectively (Table 71. Drugs and alcohol accounted for a large fraction, but psychological problems were also important (Table 81. Over the 10-year period, 33 percent were decertified for drug abuse, and 9 percent were decertified for alcohol abuse (Table 9~. This is a larger percentage than in the more loosely screened U.S. Army in Europe, where 27.5 percent were dis- charged because of excessive drug or alcohol use (Manning and Ingraham, 1981~. If the air force-which is responsible for the major portion of the strategic nuclear arsenal is looked at separately, the data are similar (Table 10) (Shealy, 19851. To summarize a 7-year period, 26 percent TABLE 7 Average Annual Rates of PRP Decertiblcations (1975-1984)a Length of Service (years) Percent Decertified s 2 3 4 9 14 18 aAverage annual Recertification total, 5,120; average annual PRP total, 112,686. SOURCE: Department of Defense (1975-1977, 1978- 1984).
514 Do 1 _' o Cal Cal ._ C=) Cal o oo m 00 _ Cal 00 _ 00 _ 00 to 00 Or Cal - 00 - AN - - - o ca O O~ 00 O ~00 O _ ~ 00 _ ~ ~ 3 o _ ~ o ~ ~Ed o o ooo O ~ ~O ~ca C ~o o o o oo oo_ _ oo ~X ~o _ ~ o ~, ., , ~ 1 C, , m _j C~ ~0 _ C) ._ `: u~ ~ c ~ as - 0 O ~t ~r ~~ ~_ ·:> .o s~ oo ~ - i ~ c ~O r~ ~ 0 ~ ~ct 00 ~._ _ - n ~ ~ ~ ~ ~oo oo ~o c~ ~ oo ~oo ~v - - ~- ~ ~ ~ oo ~n oo oo ~ ~ x ~ c ~ t- c-) - - ~ ox o ~ ~ ~oo ~ ~ o ~c~ - ~ t~ - - ~ o 3 ~ ~ o ~ - ~ ._ .° ~ ~ . ~3 _ ~ . ~_ o D ~ o ° e=,~ ~ ^° ~ ca 4) c o 00 ~> - e~ - 00 - 00 - - _ ~ C) C~ :: Ct ~ 0 ~_ 0-= ~ C~ C) .. ~ ~ V C~ ~ ~ ~ 5 o^ o 3 c~ Ct
SOURCES OF HUMAN INSTABILITY IN HANDLING NUCLEAR WEAPONS 515 TABLE 9 Summary: PRP Decertiblcations (1975-1984) ReasonTotal No. Decertified Percent Alcohol abuse4,519 9 Drug abuse17,136 33 Negligence or delinquency4,107 Court-martial or civil conviction; behavior contemptuous toward thelaw8,185 16 Physical, mental, or character Gait or aberration10,541 21 Poor attitude6,711a 13 (l9a) Total no. decertified51,199 aValues are for 1978-1984. SOURCE: Department of Defense (1975-1977, 1978-1984). were decertified because of drug abuse and 11 percent because of alcohol abuse (Table 114. For the PRP as a whole and for the air force specifically, 22 percent were separated for psychological, behavioral, or physical traits or aberrations. The decertified individuals were 5 percent of the total air force PRP, but if a 2-year enlistment on average is assumed, then 11 percent of all PRP personnel were at risk (Table 121. It is difficult to judge precisely the effectiveness and sensitivity of the PRP. Independent evaluations are not available; it is not even clear whether the military has conducted such studies. A relatively large number of personnel have been permanently decertified from PRP work. This implies not only that the screening process is imperfect but also that at any one time many unstable individuals are charged with the day-to-day respon- sibility for handling nuclear weapons. Furthermore, among those who are approved for nuclear weapons work, it may be that only the most blatantly unstable or unlucky are later recognized as security risks and are decer- tified. STRENGTHENING THE PRP Analysis of the present structure and function of the PRP suggests a number of discrete measures that could improve the screening process and diminish the number of potentially unstable people handling nuclear weapons.
516 IMAGES AND RISKS OF NUCLEAR WAR TABLE 10 Total Number of PRPs in Air Force and Number Decertified (1977-1983) Reason 1977 1978 1979 1980 1981 1982 1983 Alcohol abuse Drug abuse Negligence or delinquency Court-martial or . ., ClV11 conviction; behavior contemptuous toward the law Physical, mental, or character trait or 52 340 656 303 415 1,020 964 335 81 91 103 588 352 417 371 303 536 524 452 956 848 988 290 317 aberration594714809752793699 496 Poor attitude 508822908884865 764 Total number decertified2,2302,6803,4523,5013,5753,337 Total number of PRPs55,75153,65053,52453,35352,16452,826 SOURCE: Shealy (1985). TABLE 11 Summary: Air Force PRP DecertiDlcations (1977-1983) Total No. Reason Decertified Percent Alcohol abuse 2,411 11 Drug abuse 5,758 26 Negligence or delinquency 1,453 6 Court-martial or civil conviction; behavior contemptuous toward the law 2,638 12 Physical, mental, or character trait or aberration 4,857 22 Poor attitude 4,751a 24 Total no. decertified 21,868 aValue is for 1978-1983. SOURCE: Shealy (1985).
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 517 TABLE 12 Average Annual Rates of PRP Decertification, Air Force (1977 1983)a Length of Service (years) Percent Decertified 4 11 17 23 aTotal number decertified from 1977 to 1983, 21,868; average annual decertif~cation total, 3,124; average an- nual PRP total, 53,821. SOURCE: Shealy (1985). 1. Beyond a review of their medical records, all candidates should be examined by a trained physician, without exception. 2. The examining physician should be aware of the individual's re- sponsibility for nuclear weapons and the exact nature of his work. For example, he should be advised if the position will require long periods of isolation and confinement, in order to assess the candidate's suitability for the job. 3. A personal interview should be included that is designed to determine and assess the psychiatric or emotional stability of a PRP candidate. Traits to watch for in such an interview include suspicion, hostility, impulsive- ness, anxiety, and level of motivation. Such an interview could readily be combined with the one that advises the individual of the nature of the PRP and could determine his or her attitude toward nuclear weapons duties. 4. Standardized psychological testing should be incorporated as part of the PRP screening procedure. A test such as the Minnesota Multiphasic Personality Inventory (MMPI) takes approximately 2.5 hours, can be easily administered and scored, and is appropriate for group comparisons. It requires no highly trained professional personnel. It has proved effective in identifying rule breakers and those with character disorders in a general army population (Leventhal, 19601. The Recruit Temperament Survey, a questionnaire devised for navy enlisters, has been shown to predict not only which individuals are likely to be hospitalized for psychiatric reasons but also a specific diagnosis of psychosis, neurosis, or personality disorder (Bucky and Edwards, 19741. This survey might effectively be used in conjunction with or instead of the MMPI. 5. Records of the individual's past emotional, social, and vocational history should be available to examiners. Such information "probably is
518 IMAGES AND RISKS OF NUCLEAR WAR the best single indicator of his future performance" (Christy and Ras- mussen, 1963; Gunderson and Nelson, 1965; Roff, 1960~. To help de- te~mine preservice elements that might be indicative of psychological problems, Roff has devised a priority list of sources on a candidate's background. These sources include personnel of schools, camps, and scout organizations; case workers or psychiatrists; family members; and state- ments by the individual regarding social adjustment (Roff, 19601. 6. The candidate's family records should be reviewed for pertinent information such as psychiatric problems, alcoholism, a tendency toward diabetes or high blood pressure, and a history of drug abuse. Such family data may be important in predicting responses to stress or isolation. 7. Candidates from academically oriented high schools and those with some postsecondary education should be preferentially recruited. An air force study has demonstrated that high-performing nuclear weapons tech- nicians are more likely to come from such an educational background (Sauer et al., 1977~. 8. Age limits should be established for admission to particular groups. The 20- to 38-year-olds have been found to have the lowest incidence of "the kind of serious psychiatric problem which might tend toward acci- dental or deliberate nuclear detonation" (Eggertson, 1964; p. 2161. This age group was- found to have a relatively low frequency of psychosis, with schizophrenia appearing earlier and depressive and paranoid condi- tions appearing later. Severe behavioral disorders, as well as alcohol problems, occurred both earlier and later. While psychoneurosis increased steadily after the age of 20, this condition appeared to have less bearing on personnel reliability. 9. Periodic evaluations of reliability and work quality currently are not scheduled in the PRP. Systematic monitoring of all personnel should be established. Because of the deficiencies of the screening process itself, such monitoring deserves great emphasis. Surveillance should be per- formed by outside consultants, especially trained psychiatrists. Com- manders may be subject to the same stresses as the individuals they are trying to assess, and therefore, they may make less sound judgments about a person's stability than would outside professionals (Christy and Ras- mussen, 19631.
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 519 IMPROVING THE WORKING CONDITIONS OF PRP PERSONNEL 1. In assigning PRP personnel, greater emphasis should be placed on group composition. Team cohesiveness is probably the single most im- portant factor affecting task performance (Sauer et al., 1977; p. 47~. 2. The fatigue and boredom involved in nuclear weapons duties should be limited as much as possible. At very low stress levels, when a task is dull and unchallenging, performance tends to be poor; it reaches its peak at a low to medium stress level (U.S. Nuclear Regulatory Commission, 1975; p. III-601. Nuclear weapons tasks should therefore include a certain amount of variety and responsibility, in addition, measures such as task rotation or changes in the work schedule can reduce fatigue and boredom. 3. Squadron changes aimed at improving group morale should be em- ployed consistently. This will augment the team cohesiveness that is so essential to performance quality (Sauer et al., 1977; pp. 47 and 54~. 4. Further incentives, such as status, financial reward, or employment benefits, should be developed to make certain sensitive jobs more ap- pealing, in order to increase the pool of applicants from which to choose. Work aboard nuclear submarines and missile launch capsules is both sensitive and demanding, and both jobs suffer from a lack of enthusiastic volunteers. 5. Tighter control over drug use and alcohol abuse is required for those already in the PRP program. 6. More thorough data on nuclear weapons task errors should be col- lected, and additional research on human reliability factors should be undertaken (Sauer et al., 1977; p. 544. In focusing on the problems of human instability in the microstructure- those involved in day-to-day handling of the nuclear weapons-there is a danger that we may exaggerate their impact. How important are tension, drug and alcohol abuse, psychosis, irrationality, boredom, performance lag, and the other behavioral aberrations observed in nuclear weapons forces? Can such sources of human instability provoke a major nuclear mishap? Human error is far less critical in times of tranquility than in periods of intense crisis. Given time, even the accidental launching of a nuclear weapon at the city of an adversary can be analyzed and explained. When tensions are high and time for decision making is short, accidents, false alerts, paranoias, misunderstandings, and miscalculations assume different significance and may result in irrevocable actions of grave dimensions.
520 IMAGES AND RISKS OF NUCLEAR WAR APPENDIX NUCLEAR MISSIONS OF EACH MILITARY SERVICE A brief summary of the nuclear missions and numerical distribution of each military service is useful in understanding the breadth of the problem and in appraising the potential impact of the many factors underlying human instability among the forces. Much of the information that follows is derived from Cochran et al. (1984) and from the Military Balance, 1984-85 (International Institute for Strategic Studies, 1984~. When other sources have been used, they are specifically cited. As of 1980, there were a total of 722 military units with nuclear ca- pability among the various military services. Nuclear units must pass a certification inspection: personnel should know both the capabilities and the safety and control features of the nuclear weapons in their charge. U.S. Military Services Air Force Nuclear Role The air force plays the most pivotal role in U.S. nuclear weapons, with its strategic offensive forces comprising an estimated 90 percent of the United States' total megatonnage strategic capability. Weapons include nuclear-equipped bombers, land-based intercontinental ballistic missiles (ICBMs), and tactical nuclear fighters. As of July 1, 1984, the Strategic Air Command's (SAC) arsenal included 1,037 land-based ICBMs. SAC also maintains some 356 combat aircraft, including long- and medium- range bombers. Many of the air force's 3,700 combat aircraft are equipped as tactical nuclear fighter bombers. Certified nuclear-capable units usually consist of squadrons which are subordinate to a wing or a group. A squadron includes 15-24 aircraft, 18 Titan missiles, or 50 Minuteman missiles. Nuclear weapons are kept in the custody of the munitions maintenance unit for each wing or group. The SAC's bomber squadrons each control about 150 nuclear weapons. Strategic bombers are located at 20 air force bases around the United States, with one in Guam. Land-based Titan and Minuteman missiles are deployed in nine underground silo locations, mostly in the Midwest. In addition to these nuclear forces, the air force controls a number of tactical units of the U.S.-based tactical air command, Pacific Air Force, and United States Air Force Europe (USAFE). These are equipped with nuclear weap- ons and are located primarily in Europe. It is probable that there are nuclear-certified units in the Pacific area and in the United States as well.
SOURCES OF HUMAN INSTABILITY IN HANDLING NUCLEAR WEAPONS 521 In terms of strategic defense, several fighter interceptor squadrons are nuclear certified and equipped. In 1981 there were 6 active squadrons and 10 Air National Guard squadrons with nuclear missions based around the country. As of December 31, 1984, SAC contained 105,623 personnel of a force of 594,500 in the air force (C. Skill, SAC Public Affairs Headquarters, personal communication, 19851. Among these, there were 53,255 mili- tary, federal, and civilian air force personnel in the nuclear weapons Personnel Reliability Program (P. Delaney, SAC Public Affairs Head- quarters, personal communication, 1985~. Army Nuclear Role The nuclear weapons under the control of the army are generally short range and are more integrated into overall military capabilities than they are in the air force. Special ammunition ordnance units have charge of all nuclear weapons until they are needed by a nuclear-certified delivery unit. The nuclear-capable U.S. forces are responsible for receiving nuclear weapons, maintaining them in a safe and secure environment, and deliv- ering them against specified targets when directed. Offensive nuclear weapons under army control include the West Ger- man-based Pershing 1A missile, targeted at specified Eastern European sites; the Lance missile, intended as general support to U.S. conventional ground forces in Central Europe; and nuclear-capable artillery. Defensive weapons include the Nike-Hercules surface-to-air missile system capable of firing nuclear warheads and nuclear land mines known as atomic dem- olition munitions (ADMs). These last are extensively deployed, especially in West Germany. By the end of 1982, there were a total of 16,733 army personnel in the PRP, with 780,000 personnel in the army as a whole. Navy Nuclear Role Navy nuclear capabilities consist of strategic warfare through missile- f~ring submarines, tactical and theater land-attack warfare, defensive anti- air warfare, and antisubmarine warfare. Surface ships, submarines, and ship- and land-based aircraft encompass the navy's nuclear forces. The navy's strategic capabilities rely on its 35 ballistic missile nuclear submarines (SSBN), which carry Trident or Poseidon missiles. In the tactical realm, most surface ships are nuclear capable and can launch
522 IMAGES AND RISKS OF NUCLEAR WAR nuclear Terrier surface-to-air weapons or ASROC antisubmarine rockets. Attack submarines are equipped with the short-range SUBROC antisub- marine rocket, while land- and ship-based aircraft are certified to carry nuclear bombs for land attack and antisubmarine warfare. As of December 31, 1982, there were 34,871 navy (and marine force) PRP personnel of a total of 564,800 (Moore, 1980; p. 6131. By December 31, 1984, the navy had a total of approximately 22,000 personnel on 130 SSBN and attack submarines (P. Johnstone, Chief of Information Office of the Navy, personal communication, 1985~. Marine Corps Nuclear Role The marine corps maintains no nuclear weapons of its own in peacetime. During hostilities, the navy would transfer air-delivered weapons to the Marine Wing Weapons Unit in charge of nuclear weapons. Nuclear ar- tillery and atomic demolition munitions would be delivered to the nuclear ordnance platoon of the Marine Division for use by marine ground forces. The marine corps has two aircraft and several types of amphibious ships which are certified to carry nuclear warheads. As in the army, the marine corps views nuclear weapons as an extension of conventional forces rather than as an independent capability. Allied Nuclear Roles Besides the nuclear missions assigned to each service, the United States has nuclear weapons agreements with nine North Atlantic Treaty Orga- nization (NATO) countries, including Belgium, Canada, Greece, Italy, The Netherlands, South Korea, Turkey, the United Kingdom, and West Germany. Bilateral agreements known as programs of cooperation (POCs) address the deployment of nuclear weapons in NATO countries and the transfer and certification of nuclear-capable delivery vehicles. There are over 600 allied dual-capable tactical fighters and medium bombers avail- able for nuclear duties. The U.S. Air Force, Army, and Navy each operates a custodial unit, composed of U.S. personnel, to maintain nuclear weapons for allied use. No foreign military personnel are in the PRP (U.S. Congress, House, 1978~. Soviet Military Services The Soviet armed forces consist of five services: the Strategic Rocket Forces (SRF), Ground Forces, Troops of National Air Defense, Air Forces, and Navy. Soviet strategic nuclear forces fall under the command of the
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 523 Strategic Rocket Forces, which are responsible for long-range land-based ballistic missiles; the navy, which is in charge of the Soviet ballistic missile submarine fleet; and the long-range aviation (LRA) element of the air forces, which operates long- and medium-range bombers. The National Air Defense Troops (Voyska-PVO) are responsible for strategic nuclear defense (Ball, 1981; p. 43~. SRF comprises the most important element of Soviet strategic nuclear capability. With an estimated 415,000 personnel, SRF maintains roughly 1,400 ICBMs and some 600 intermediate- (IRBMs) and medium-range ballistic missiles (MRBMs). However, ICBMs are kept at a relatively low level of readiness compared with those in the United States, parallelling low alert rates of other Soviet strategic forces (Ball, 1981; p. 451. Like the other five services, its headquarters are located in Moscow. Soviet missile silos contain four men. Two SRF officers are responsible for launching a rocket, while two KGB officers would actually arm the nuclear warhead. Less than 30 percent of Soviet nuclear forces are sea based, compared with over 50 percent in the United States (Thompson, 1985; p. 68~. The Soviet navy currently maintains 981 submarine-launched ballistic missiles (SLBM) in 79 submarines. KGB officers on board strategic submarines may perform a function parallel to their role in missile silos; namely, they may act as a safeguard against inadvertent launch by the military and assume a key position in the chain of nuclear command. The Soviet navy does not send more than about 15 percent of its nuclear-capable submarines to sea at any one time (Meyer, 1985; pp. 184-1851. This has been said to reflect the higher command's concern over the increased difficulty in maintaining control over the weapons once the submarines are away from base (Ball, 1981; p. 45~. There are 490,000 personnel in the navy (Cock- burn, 1983; p. 295), but no figures are available on the number responsible for handling nuclear weapons. The strategic elements of the air force maintain 752 combat aircraft, 143 long-range bombers, 475 medium-range bombers, as well as recon- naissance aircraft, tankers, air-to-surface missiles, and electronic coun- termeasures (ECMs). An estimated 100,000 personnel are involved, with a total of 400,000 in the air force. The National Air Defense Troops, known as Voyska-PVO, were es- tablished as a separate military force in 1948. They number some 370,000 soldiers (Cockburn, 1983; p. 358) and are responsible for a vast antibom- ber network of radars, interceptors, and missiles. These brief descriptions of the matching forces convey some sense of the large number of men that are involved daily in the handling of nuclear weapons. They are spread in various forms throughout the forces, and the
524 IMAGES AND RISKS OF NUCLEAR WAR tactical weapons, at least, may ultimately be under the control of those in the field if war should come and weapons are threatened with capture or destruction. In the heat of armed conflict, human instability will surely become more pronounced, and whether Trough fear or anger, the pos- sibility of the use of nuclear weapons will be heightened. ACKNOWLEDGMENTS Supported by grants from the Carnegie Corporation, the John D. and Catherine T. MacArthur Foundation, the W. Alton Jones Foundation, and the New Prospect Foundation. Elizabeth Polin played a major role in organizing much of the data on which this article was based, and did so with perceptiveness, persistence, and wonderful good sense. Annette Makino also made a large contribution to the data-gathering process, working wig a keen intelligence and a fine capacity to sift the important signals from the noisy background. Margaret Marshal and Ann McGrath typed the manuscript at different stages wig skill and grace. I am indebted to all of ~em. REFERENCES Alkov, R. A., and M. S. Borowsky. 1980. A questionnaire study of psychological back- ground factors in U.S. Navy aircraft accidents. Aviat. Space Environ. Med. 51(9):860- 863. Alkov, R. A., M. S. Borowsky, and J. A. Gaynor. 1982. Stress coping and the U.S. Navy aircrew factor mishap. Aviat. Space Environ. Med. 53(11):1112-l l lS. Altman, I., D. Taylor, and L. Wheeler. 1968. Stress relations in socially isolated groups. J. Personality Soc. Psych. 9(4):369-373. Altman, I., and W. Haythorn. 1967. The ecology of isolated groups. Behavioral Sci. 12(3): 169-182. Arthur, R. J. 1966. Psychiatric disorders in naval personnel. Military Medicine 131(4):354- 360. Ball, D. 1981. Can nuclear war be controlled? Adelphi Paper 169. London: International Institute for Strategic Studies. Beckman, K., A. J. Marsella, and R. Finney. 1979. Depression in the wives of nuclear submarine personnel. Am. J. Psych. 136(4B):524-526. Bereaunu, B. 1982. Self-activation of the world nuclear weapons stockpile. Revue Rou- maine de Mathematiques Pures et Appliques 27:652. Black, A. W. 1983. Psychiatric illness in military aircrew. Aviat. Space Environ. Med. 54(7):594-598. Blunder, S. 1981. Drugs and the soldier. J. R. Army Med. Corps 127:72-79. Bodrov, V. A. 1984. Basic principles of the development of a system for the occupational psychological selection of servicemen and its performance. Voenno-Meditsinskii Zhurnal. 9:41-43. Boston Globe. October 23, 1981. British drug addict served on N-subs. Bracken, P. 1983. The Command and Control of Nuclear Forces. New Haven, Conn.: Yale University Press. BriKen, S. 1983. The Invisible Event. P. 17. London: The Menard Press.
SOURCES OF HUMAN INSTABILITY IN HANDING NUCLEAR WEAPONS 525 Bucky, S. F., and D. Edwards. 1974. The recruit temperament survey (RTS) as it dis- criminates between psychoses, neuroses, and personality disorders. J. Clin. Psychol. 30(2): 195-199. Burt, M. R. 1982. Prevalence and consequences of drug abuse among U.S. military per- sonnel: 1980. Am. J. Drug and Alcohol Abuse 8(4):419-439, Marcel Dekker, Inc., New York. Center for Defense Information. 1985. Who Could Start a Nuclear War? The Defense Monitor. Washington, D.C. 14(3). Christy, R. L., and J. E. Rasmussen. 1963. Human reliability implications of the U.S. Navy's experience in screening and selection procedures. Am. J. Psych. 120:540-547. Cochran, T. B., W. M. Arkin, and M. M. Hoenig. 1984. Nuclear Weapons Databook, Vol. 1. U.S. Nuclear Forces and Capabilities. Cambridge, Mass.: Ballinger Publishing. Cockburn, A. 1983. The Threat: Inside The Soviet Military Machine. New York: Vintage Books. Connoly, J. 1981. Accident proneness. Br. J. Hosp. 26(5):470-481. Datel, W. E., and F. D. Jones. 1982. Suicide in United States Army personnel. Military Med. 147:843-847. Davis, R. B. 1985. Alcohol abuse and the Soviet military. Armed Forces and Society 11(37):399-411. Department of the Air Force. 1983. Military Personnel: Nuclear Weapon Personnel Reli- ability Program. AF Regulation 35-99, May 6? 1983. Washington, D.C. Department of the Army. 1985. Disposition and Incidence Rates, Active Duty Army Personnel, Psychiatric Cases, Worldwide, CY 80-84. U.S. Army Patient Administration Systems and Biostatistics Activity. Washington, D.C. Department of Defense. 1981. Nuclear Weapon Personnel Reliability Program. Directive Number 5210.42. April 23, 1981. Washington, D.C. Department of Defense, Office of the Secretary of Defense. Calendar Year Ending De- cember 31, 1975, 1976, 1977. Annual Disqualification Report, Nuclear Weapon Per- sonnel Reliability Program. RCS DD-COMP(A) 1403. Washington, D.C. Department of Defense, Office of the Secretary of Defense. Year Ending December 31, 1978, 1979, 1980, 1981, 1982, 1983, 1984. Annual Status Report, Nuclear Weapons Personnel Reliability Program. RCS DD-POL(A) 1403. Washington, D.C. Dumas, L. 1980. Human fallibility and weapons. Bull. Atomic Sci. November 1980. Eggertson, P. F. 1964. The dilemma of power: Nuclear weapons and human reliability. Psychiatry 27(3):211-218. Gabriel, R. A. 1980. The New Red Legions: An Attitudinal Portrait of the Soviet Soldier. Westport, Conn.: Greenwood Press. Gabriel, R. A. 1984. The Antagonists: A Comparative Combat Assessment of the Soviet and American Soldier. Westport, Conn.: Greenwood Press. Gillooly, D. H., and T. C. Bond. 1976. Assaults with explosive devices on superiors. Military Med. 141:700-702. Goldhamer, H. 1975. The Soviet Soldier: Soviet Military Management at the Troop Level. New York: Crane, Russak & Company. Gorokhov, A. 1985. Behind the controls of the strategic missiles. Pravda, May 29, 1985, p. 6. Green, R. G. 1977. The psychologist and flying accidents. Aviat. Space Environ. Med. 48(10):922-923. Group for the Advancement of Psychiatry. 1964. Gunderson, E. K. E., and P. D. Nelson. 1965. Biographical predictors of performance in an extreme environment. J. Psychol. 61:59-67.
526 IMAGES AND RISKS OF NUCLEAR WAR Halloran, R. 1984. Air Force seeks missile base deep underground for 1990's. New York Times, October 3, 1984. Hiles, F. M. J. 1980. The prevention and treatment of alcoholism in the Royal Navy- 1. A policy. J. R. Naval Med. Service 3:180-185. Hiles, F. M. J. 1981. The prevention and treatment of alcoholism in the Royal Navy- 3. Preliminary report on a pilot study of naval alcoholics. J. R. Naval Med. Service 2:70-76. International Institute for Strategic Studies. 1984. The Military Balance 1984-85. London: International Institute for Strategic Studies. Kemeny, J. 1980-1981. Political fallout. Society 18:5-9. Kentsmith, D. K. 1981. The physician's determination of personnel reliability in sensitive occupations. Aviat. Space Environ. Med. 52(1):45-49. Kinsey, J. L. 1968. Selection of personnel for system effectiveness. J. Occupat. Med. 10(5):238-240. Kleinmann, M. V., Jr., and E. F. Krise. 1957. Mental illness and classified information. U.S. Armed Forces Med. J. 8(7):1007-1016. Lagadec, P. 1982. Major Technological Risk: An Assessment of Industrial Disasters. New York: Pergamon Press. Larus, J. 1970. Safing the Nukes: The Human Reliability Problem. Stockholm: Stockholm International Peace Research Institute. LeBlanc, N. D. 1977. The Measurements of Job Stress on an Operational Unit of the Military. Pp. 45, 46. Technical Report 117, Center for Human Appraisal. Wichita, Kans.: Wichita State University. Leventhal, A. M. 1960. Character disorders, disciplinary offenders, and the MMPI. U.S. Armed Forces Med. J. 11(6):660-664. Lewis, H. 1980. The safety of fusion reactors. Sci. Am. 242(3):53-65. Magnuson, E. 1985. Very serious losses. Time, June 17, 1985, pp. 18-22. Manning, F. J., and L. H. Ingraham. 1981. Personnel attrition in the U.S. Army in Europe. Armed Forces Soc. 7(2):256-270. McGrath, P. 1985. A family of spies. Newsweek, June 10, 1985, pp. 32-36. Meyer, L. 1984a. The men who would finish World War III. The Washington Port Ma~- . . ~ ~ ^ A ~ ~ % ^ ~ ~ ~ azine, June 3, 1984. Meyer, L. 1984b. Fail-safe and subs: Should we trust the Navy to trust itself? The Wash- ington Post Magazine, September 30, 1984. Meyer, S. M. 1985. Soviet perspectives on the paths to nuclear war. Pp. 167-206 in Hawks, Doves, and Owls: An Agenda for Avoiding Nuclear War. G. T. Allison, A. Carnesale, and J. S. Nye, Jr., eds. New York: W. W. Norton. Moore, J., ed. 1980. Jane's Fighting Ships 1980-81. London: Jane's Publishing Company United. Moore-Ede, M. 1982. Sleeping as the world turns. Nat. Hist. 91(10):28-36. Moore-Ede, M. 1984. The body's inner crocks. P. 173 in Health and Medical Horizons. New York: MacMillan Education Co. Moore-Ede, M., C. Czeisler, and G. Richardson. 1983. Circadian time keeping in health and disease. Part 2: Clinical implications of Circadian rhythmicity. N. Eng. J. Med. 309(9):534. Morrison, D. 1983. Weapons wired for war. Nuclear Times. Myagkov, A. 1976. Inside the KGB. New Rochelle, N.Y.: Arlington House Publishers. New York Times. August 18, 1969. Three atom guards called unstable; major suspended. New York Times. November 8, 1984. Navy defends its method in fighting drug abuse. New York Times. January 21, 1986. Survey finds sharp drop in marijuana use in the military, Pentagon says.
SOURCES OF HUMAN NEST IN H^DLING NUCLEI WEAPONS 527 Opstead, P. K., R. Ekanger, M. Nummestad, and N. Raabe. 1978. Performance, mood, and clinical symptoms in men exposed to prolonged, severe physical work and sleep deprivation. Aviat. Space Environ. Med. 49(9):1065-1073. Padilla, E., D. Rosenow, and A. Bergman. 1976. Predicting accident frequency in children. Pediatrics 58(2):223-226. Parrott, G. S. 1973. A Career Attitude Survey of Officers Serving on Titan or Minuteman Missile Crews. Technical Report 108. Center for Human Appraisal. Wichita, Kans.: Wichita State University. Perrow, C. 1980-1981. Normal accident at Three Mile Island. Society 18:25. Plag, J. A., R. J. Arthur, and J. M. Goffmann. 1970. Dimensions of psychiatric illness among first-term enlisters in the United States Navy. Military Med. 135:665-673. Psychiatric Aspects of Prevention of Nuclear War. 1964. Group for the Advancement of Psychiatry 5 (Report 57):223-313. Redman, R. A., and L. J. Walter, Jr. 1985. Suicide Among Active Duty Military Personnel. Health Studies Task Force, Office of the Assistant Secretary of Defense (Health Affairs). Washington, D.C.: U.S. Department of Defense. Rockwell, D. A., M. G. Hodgeson, J. R. Beljan, and C. M. Winget. 1976. Psychologic and psychophysiologic response to 105 days of social isolation. Aviat. Space Environ. Med. 47(16):1087-1093. Roff, M. 1960. Relations between certain preservice factors and psychoneurosis during military duty. U.S. Armed Forces Med. J. 11(1):152-160. Rothberg, J. M., N. L. Rock, and F. Del Jones. 1984. Suicide in United States Army personnel, 1981-1982. Military Med. 149(10):537-541. Satloff, A. 1967. Psychiatry and the nuclear submarine. Am. J. Psych. 124(4):547-551. Sauer, D. W., W. B. Campbell, N. R. Potter, and W. B. Askren. 1977. Human Resource Factors and Performance Relationships in Nuclear Missile Handling Tasks. Advanced Systems Division, Wright-Patterson Air Force Base, Ohio, and Air Force Weapons Laboratory, Kirtland Air Force Base, New Mexico, AFHRL-TR-76-85, AFWL-TR-76- 301. Schmidt, W. E. 1981. Some near Titan base cite strains in crewmen's life. New York Times, June 3, 1981, p. Ale. Schuckit, M. A., and E. K. E. Gunderson. 1973. Job stress and psychiatric illness in the U.S. Navy. J. Occupat. Med. 15(11):884-887. Scott, H. F., and W. F. Scott. 1984. The Armed Forces of the USSR. Boulder, Colo.: Westview Press. Senate Armed Services Committee. FY 1978 Authorization Act, Part 10, April 7, 1977. Serxner, J. L. 1968. An experience in submarine psychiatry. Am. J. Psych. 125(1):25- 30. Shealy, R. W., Lt. Colonel, U.S. Air Force. Chief, Civil Affairs Branch, Community Relations Division, Office of Public Affairs. January 7, 1985. Mimeograph. Washington, D.C. Solov'ev, A. D., M. S. Liaskovskit, and G. I. Cychev. 1971. Experimental-psychological observations of pilots admitting erroneous actions. Voenno-Meditsinskii Zhurnal 9:66- 68. Suvorov, V. 1982. Inside the Soviet Anny. London: Hamish Hamilton Ltd. Tansey, W. A., J. A. Wilson, and K. E. Schaefer. 1979. Analysis of health data from 10 years of Polaris submarine patrols. Undersea Biomed. Res. 6:S217-S246. Tarasulo, Y. 1985. A profile of the Soviet soldier. Armed Forces Soc. 11(2):221-234. Thompson, J. 1985. Psychological Aspects of Nuclear War. Winchester, England: British Psychological Society and John Wiley & Sons. Treml, V. 1982a. Fatal Poisoning in the USSR. Paper presented at the International Sym
528 IMAGES AND RISKS OF NUCLEI We posium sponsored by Centro Studi sui Sistemi Socio-Economical dell'Esti, Milan, Italy, September 1982. Treml, V. 1982b. Death from alcohol poisoning in He USSR. The Wall Street Journal, November 10, 1982, p. 26. Ursin, H., E. Baade, and L. Seymour. 1978. Psychobiology of Stress. P. 6. New York: Academic Press. U.S. Army Patient Administration Systems and Biostatistics Activity: Disposition and Incidence Rates Active Duty Army Personnel, Psychiatric Cases, Worldwide, CY 80- 84. U.S. Congress, Senate. 1977. Senate Anned Services Committee, FY 1978 Authorization Act, Part 10. April 7, 1977. U.S. Congress, Senate. 1980. Recent False Alerts from the Nation's Missile Attack Warning System. Committee on Armed Services. U.S. Senate by Senators Barry Goldwater and Gary Hart. October 9, 1980. Washington, D.C.: U.S. Government Printing Office. U.S. Congress, House. 1978. Military Construction Appropriations for 1979. Section on Security of Nuclear and Chemical Weapons Storage. Hearings before a Subcommittee of the Committee on Appropriations. 95th Cong., 2d. sees., Subcommittee on Military Construction Appropriations. Part 2. Washington, D.C.: U.S. Government Printing Of- fice. U.S. Congress, House. 1981. Drug Abuse in the Military 1981. Hearing before the Select Committee on Narcotics Abuse and Control. 97th Cong., 1st sess. September 17, 1981. Washington, D. C.: U.S. Government Printing Office. U.S. Congress, House. 1982. Drug Abuse in the Military. Department of Defense Appro- priations for 1983. Hearings before the Subcommittee on Appropriations. Part 3. 97th Cong., 2d sees., pp. 253-444. April 1, 1982. Washington, D.C.: U.S. Government Printing Office. U.S. Congress, Senate. 1982. Drug and Alcohol Abuse in the Armed Services. Joint Hearing before the Subcommittee on Manpower and Personnel and the Subcommittee on Pre- paredness on the Committee on Anned Services. 97th Cong., May 18, 1982. Washington, D.C.: U. S. Government Printing Of lice. U.S. Nuclear Regulatory Commission. 1975. Human Reliability Analysis. Section 6.1 in Appendix III Failure Data in Reactor Safety Study: An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants. Wash-1400 (NUREG-75/014). Washington, D.C.: U.S. Government Printing Office. Ustinov, D. F. 1982. Otvesti ugnozu yadernoy. Pravda, July 12, 1982, p. 4. (As quoted in S. M. Meyer, 1985; p. 185.) Volz, J. 1983. Living on the nuclear firing line. New York News, October 18, 1983, pp. 8-1OF. Weiner, E. 1977. Controlled flight into terrain accidents: System induced error. Human Factors 19(2). Weybrew, B. B., and H. B. Molish. 1979. Attitude changes during and after long sub- marine missions. Undersea Biomed. Res. 6(Submarine Suppl.):S175-S189. Wimbush, E. 1981. The Red Army. PBS TV documentary. May 6, 1981. (As cited in Cockburn, 1983; p. 62.) Wood, P. J. 1980. Alcoholism in the Army: A demographic survey of an inpatient pop- ulation. Br. J. Addict. 75:375-380. Yanowitch, R. E. 1977. Crew behavior in accident causation. Aviat. Space Environ. Med. 48(10):918-921. Zav'yalov, I. 1971. The evolution of the correlations of strategy, operational art tactics. Military Thought 11:37. (As quoted in S. M. Meyer, 1985; pp. 186-187.) , and