DOE Shield DOE Openness: Human Radiation Experiments: Roadmap to the Project
ACHRE Report
Roadmap to the Project
HomeRoadmapWhat's NewSearch HREXMultimediaRelated SitesFeedback
ACHRE Report

Part II

Chapter 10


Human Research at the Bomb Tests

The Bomb Tests: Questions of Risk, Records, and Trust


Chapter 10: Human Research at the Bomb Tests

The Defense Department's Medical Experts: Advocates of Troop Maneuvers and Human Experimentation

As we saw in the introduction, in 1949, when AEC and DOD experts met to consider the psychological problems connected to construction of the proposed nuclear-powered airplane, the NEPA project, there was a consensus that America's atomic war-fighting capability would be crippled unless servicemen were cured of the "mystical" fear of radiation.[3] When routine testing of nuclear weapons began at the test site in Nevada in 1951, the opportunity to take action to deal with this problem presented itself. DOD officials urged that troop maneuvers and training exercises be conducted in connection with the tests. Whole military units would be employed in these exercises, and participation, as part of the duty of the soldier, would not be voluntary. DOD's medical experts simultaneously urged that the tests be used for training and "indoctrination" about atomic warfare and as an opportunity for research. The psychological and physiological testing of troops to address the fear of radiation was the first of the research to take place; this testing was largely conducted as an occupational rather than an experimental activity.

In a June 27, 1951, memorandum to high DOD officials, Dr. Richard Meiling, the chair of the secretary of defense's top medical advisory group, the Armed Forces Medical Policy Council, addressed the question of "Military Medical Problems" associated with bomb tests.[4] The memorandum made clear that troops should be placed at bomb tests not so much to examine risk as to demonstrate relative safety.

"Fear of radiation," Dr. Meiling's memorandum began, "is almost universal among the uninitiated and unless it is overcome in the military forces it could present a most serious problem if atomic weapons are used." In fact, "[i]t has been proven repeatedly that persistent ionizing radiation following air bursts does not occur, hence the fear that it presents a dangerous hazard to personnel is groundless." Dr. Meiling urged that "positive action be taken at the earliest opportunity to demonstrate this fact in a practical manner."[5]

He continued, a "Regimental Combat Team should be deployed approximately twelve miles from the designated ground zero of an air blast and immediately following the explosion . . . they should move into the burst area in fulfillment of a tactical problem." The exercise "would clearly demonstrate that persistent ionizing radiation following an air burst atomic explosion presents no hazards to personnel and would effectively dispel a fear that is dangerous and demoralizing but entirely groundless."[6]

Dr. Meiling's proposal to put troops at the bomb tests in order to allay their fears may well have been an echo of what the military already had in mind. The Army's 1950 "Atomic Energy Indoctrination" pamphlet, a primer for soldiers, showed that the military was concerned that misperception of the effect of an air burst could be damaging in combat. "[L]ingering radioactivity will be virtually nonexistent in the case of the normal air burst,"[7] it reassured the soldiers. The greater danger, it told them, was the probability that "an unreasoning fear of lingering radioactivity" would take "an unnecessary toll in American lives."[8]

While the tests provided an opportunity to allay fears, they simultaneously provided the opportunity to gather data. In this regard, Dr. Meiling appeared to be ahead of his military colleagues in expressing concern that the military was not taking adequate advantage of the bomb tests as an opportunity for "biomedical participation." In February 1951, in fact, following tests in Nevada, he had urged the DOD to incorporate "biomedical tests" into plans for future bomb tests.[9]

Meiling's suggestion that planning for biomedical tests be undertaken wound its way through the secretary of defense's research and development bureaucracy and fell into the lap of the civilian-chaired Joint Panel on the Medical Aspects of Atomic Warfare.[10] Under the chairmanship of Harvard's Dr. Joseph Aub, the Joint Panel was the gathering place for the small world of government radiation researchers and their private consultants. Its periodic "Program Guidance Reports" laid out the atomic warfare medical research agenda, summarizing work that was ongoing and that which remained. At its meetings, participants heard from the CIA on foreign medical intelligence, debated the need for human experimentation, and learned of the latest developments in radiation injury research, of the blast and heat effects of the bomb, and of instruments needed to measure radiation effects.

In September 1951 the Joint Panel considered a draft report on "biomedical participation" in bomb tests.[11] "It is, of course obvious," the report noted, "that a test of a new and untried atomic weapon is not a place to have an unlimited number of people milling about and operating independently." Planning was therefore in order. There were, the document explained, basic criteria for "experimentation" at bomb tests. For example, "Does the experiment have to be done at a bomb detonation; is it impossible or impractical in a laboratory?"[12]

The document turned to "specific problems for future tests." The list of twenty-nine problems was not intended to be all-inclusive, but was "designed to show the types of problems which should be considered as a legitimate basis for biomedical participation in future weapons tests." The term human experimentation was not used, and most of the items could be performed without humans.[13] However, the list included several examples of research involving human subjects:

11. Effects of exposure of the eye to the atomic flash . . .

24. Measurements of radioactive isotopes in the body fluids of atomic weapons test personnel . . .

27. The efficiency and suitability of various protective devices and equipment for atomic weapons war . . .

28. Psychophysiological changes after exposure to nuclear explosions.

29. Orientation flights in the vicinity of nuclear explosions for certain combat air crews.[14]

By the end of the decade, human research would be conducted in all these areas.[15]

At the same September meeting, the Joint Panel also considered a "Program Guidance Report" on the kinds of atomic warfare-related research that needed to be conducted, in the laboratory as well as in the field. The areas singled out for immediate and critical attention included the initiation of "troop indoctrination at atomic detonations" and "psychological observations on troops at atom bomb tests."[16]

A section on "Biomedical Participation in Future Atomic Weapons Tests" concluded that the next step should be

4.1 To complete present program and plan for participation in future tests in light of results from Operation GREENHOUSE [a prior atomic test series]. These plans should include studies on the effect of atomic weapons detonations on a troop unit in normal tactical support [emphasis added].[17]

Thus, while it was well known at the time that troops participated at the bomb tests and were subjected to psychological testing, it now is evident that the DOD's medical advisers advocated the presence of the troops at the tests for both training and research purposes. The doctors were not alone in attaching high priority to such research. The Joint Panel's September guidance punctuated, perhaps echoed, the Armed Forces Special Weapons Projects's midsummer 1951 call for a "systematic research study . . . [to] provide a sound basis for estimating troop reaction to the bomb experience and . . . the indoctrination value of the maneuver."[18]

The HumRRO Experiments

Just two months later, in November 1951, at a bomb test in the Nevada desert, the Army conducted the first in a series of "atomic exercises."[19] This exercise was designed primarily to train and indoctrinate troops in the fighting of atomic wars. The exercise also provided an opportunity for psychological and physiological testing of the effects of the experience on the troops.

Desert Rock was an Army encampment in Nevada adjacent to the nuclear test site. At the exercise named Desert Rock I, more than 600 of the 5,000 men present would be studied by psychologists from a newly created Army contractor, the Human Resources Research Organization (HumRRO). HumRRO's research was directed by Dr. Meredith Crawford, who was recruited by the Army from a deanship at Vanderbilt University.[20] The identity of all the participants involved in the "HumRRO experiments," and the further DOD research discussed later in this chapter, is not known. The numbers of those who participated must be reconstructed from available reports.[21]

The highly publicized bomb test was well attended by military and civilian officials. "Las Vegas, Nevada," Time magazine reported, "had not seen so many soldiers since World War II. . . . The hotels were jammed with high brass. . . . [o]ut on the desert, 65 miles away 5,000 hand-picked troops were getting their final briefing before Exercise Desert Rock I--the G.I.'s introduction to atomic warfare."[22] The detonation, Representative Albert Gore (father of the current Vice President), told the New York Times, was "the most spectacular event I have ever witnessed. . . . As I witnessed the accuracy and cataclysmic effect of the explosion, I felt the conviction that it might be used in Korea if the cease-fire negotiations broke down."[23]

To render the experience more realistic, the observers and participants were told to imagine that aggressor armies had invaded the United States and were now at the California-Nevada border. An atomic bomb would be dropped, with the troops occupying a position seven miles from ground zero. After the detonations they would "attack into the bombed area."[24]

At their home base, two groups of troops--a control group that would stay at home base and an experimental group that would go to Nevada--had listened to lectures and seen films intended to "indoctrinate" them about the effects of the bomb and radiation safety. Both groups were administered a questionnaire to determine how well they had understood the information provided. Dr. Crawford explained in a 1994 interview that "indoctrination," which today has a negative connotation, was not intended to suggest misrepresentation of fact, but "had more to do with attitude, feeling and motivation."[25] At Desert Rock, the experimental group was given a further "non-technical briefing." They were "reminded that no danger of immediate radiation remains 90 seconds after an air burst; that they would be sufficiently far from ground zero to be perfectly safe without shelter; and that with simple protection they could even be placed quite close to the center of the detonation, with no harm to them."[26]

After the blast, a questionnaire was again administered to most of the experimental subjects, and physiological measurements including blood pressure and heart rates were taken. The questionnaire was designed to test the success of the "indoctrination."[27] For example, questions included (answers in parentheses were those the HumRRO report stated were correct):

l. Suppose the A-bomb were used against enemy troops by exploding it 2000 feet from the ground and suppose all enemy troops were killed. How dangerous do you think it would be for our troops to enter the area directly below the explosion within a day? (Not dangerous at all). . . .

6. If an A-bomb were exploded at 2000 feet, under what conditions would it be safe to move into the spot directly below, right after the explosion? (Safe if you wore regular field clothing.)[28]

These answers were not correct. Answers to questions like the above depend on weather conditions, the yield of the weapon, and the assumptions about the degree of risk from low levels of exposure. For example, while an airburst (where the fireball does not touch the ground) may result in little fallout in the immediate area of the blast, it does not result in none; if rain is present, a substantial amount of fallout may be localized.

Similarly, whereas the 1946 Bikini bomb tests at Operation Crossroads in the Pacific had caused contamination so severe that many of the surviving ships were scrapped, the question and answer provided said:

Some of the ships in the Bikini tests had to be sunk because they were too radioactive to be used again.


In a 1995 review of the 1951 questionnaire, the Defense Department found that "changes/corrections/clarifications" would be in order for nine of the thirty questions.[30]

In January 1952, the Army surgeon general expressed "continuing interest in the conduct of psychiatric observations," offering funds for "Psychiatric Research in Connection with Atomic Weapons Tests Involving Troop Participation."[31] In March 1952, however, the Army and the Armed Forces Special Weapons Project (AFSWP), which coordinated nuclear weapons activities for the DOD, provided critical reflections on Desert Rock I. "[O]ne is inevitably drawn to the conclusion," the Army reported, "that the results, though measurable, were highly indeterminate and unconvincing. The limitations of evaluation were inherent in the problem. Handicapped by a preconceived notion that there would be no reaction, it took on the form of a gigantic experiment whose results were already known. No well controlled studies could be undertaken which could presume even superficial validity. . . ."[32] In a letter to the AEC, the AFSWP went further. Owing to the "tactically unrealistic distance of seven miles to which all participating troops were required to withdraw for the detonation,"[33] troops might get the wrong impression about nuclear warfare.

In 1994, Dr. Crawford reflected on the logic of testing for panic in an environment that was thought to be too safe. "No troops," Dr. Crawford recalled, "were exposed anywhere where anybody thought there was any danger, so you might ask the question, so what? I've asked that question myself and I've thought about it. It was the first HumRRO project. It was really pretty well agreed upon before I got up here from Tennessee . . . so we did what we could."[34]

Despite the reservations about the 1951 study, on May 25, 1952, the Army conducted its second HumRRO experiment at the exercise called Desert Rock IV. It was similar in methodology to the first experiment and involved about 700 soldiers who witnessed the shot and 900 who served in the control group as nonparticipants.[35] This time, to add more realism, the troops witnessed the blast, an 11-kiloton weapon that was set off from the top of a tower, from four miles from ground zero. By the end of the second research effort, there was even more reason to question the utility of the experiments. HumRRO's report on Desert Rock IV stated that while knowledge increased as a result of the indoctrination, the actual maneuver experience did not produce significant improvement in test scores and decreases were actually reported on some questions.[36]

In both Desert Rock I and Desert Rock IV, the Army hoped that the troops who witnessed the blasts would disseminate information to the troops who stayed at home base. However, the troops who participated in the exercises were warned that discussion of their experiences could bring severe punishment, and the researchers found that communication was at a minimum.[37] Moreover, those who stayed home, HumRRO found, "showed no evidence of great interest, of extensive discussion, or of any important benefit from dissemination as a result of the atomic maneuver."[38] Meanwhile, the experience that the participants had been warned not to discuss and that was of little interest to their comrades was front-page news throughout the country. "When they returned to camp," Time reported of the first Desert Rock exercise, "the men were quickly herded into showers. Some were given test forms to fill out. Did you sweat? Did your heart beat fast at any time? Did you lose bladder control? Most of the answers were no."[39]

Without any direct comment on the results of the Desert Rock I and IV experiments, in September 1952, the Joint Panel urged that the psychological research continue:

It is possible that inclination to panic in the face of AW [atomic warfare] and RW [radiological warfare] may prove high. It seems advisable, therefore, to increase research efforts in the scientific study of panic and its results, and to seek means for prophylaxis. . . . The panel supports the point of view that troop participation in tests of atomic weapons is valuable. As many men as possible ought to be exposed to this experience under safe conditions. Psychological evaluation is difficult and results can be expected to appear superficially trivial, but the matter is of such extreme importance that the research should be persisted in, utilizing every opportunity.[40]

Indeed, a third set of experiments was carried out in April 1953, at Desert Rock V; this time, the number of participants is unknown.[41]

The final HumRRO bomb test study was conducted in 1957 at Operation Plumbbob.[42] No formal report was prepared, but the experience was recorded in a personalized recollection by a HumRRO staffer.[43] Weather-related delays, the departure of HumRRO staff, the continued redesign of the exercises, and the failure of a fifth of the troops to return from a weekend pass in time for the events took their toll. The researchers were not given the script used in the indoctrination lectures to the troops. Thus, it was impossible for the researchers to know whether incorrect responses were due to "lack of inclusion of the topic in the orientation or to ineffective instruction."[44] The research was to include exercises such as crawling over contaminated ground.[45] But, yet again, the researchers found that the safety rules in force precluded important study: "shock . . . and panic . . . would not be observed."[46]

There is no question that HumRRO activities were research involving human subjects; the projects involved an experimental design in which soldier-subjects were assigned either to an experimental or a control condition. Available evidence suggests, however, that the Army did not treat HumRRO as a discretionary research activity but as an element of the training exercise in which soldiers were participating in the course of normal duty. The HumRRO subjects were apparently not volunteers. Dr. Crawford in 1994 said of the HumRRO subjects, "Whether they were requested to formally give their consent is pretty unknowable because in the Army or any other military service people generally do what they're asked to do, told to do."[47] Indeed, as HumRRO's initial report stated, the primary purpose of the atomic exercise was training; "research was necessarily of secondary importance."[48] However, Dr. Crawford felt confident that the risks were disclosed. Because of the "number and intensity of briefings . . . [n]o soldier, to our knowledge, went into the test situation with no idea about what to expect. They were adequately informed."[49]

We now know that in 1952 the Defense Department's medical experts were simultaneously locked in discussion of the need for psychological studies and other human research at bomb tests and, as we saw in chapter 1, the need for a policy to govern human experimentation related to atomic, biological, and chemical warfare. In October of that year, the Armed Forces Medical Policy Council recommended that the Nuremberg Code be adopted, as it was by Secretary Charles Wilson in 1953. What is still missing is information that might show how, as seems to be the case, the same experts could have been having these discussions without communicating the essence of them to those responsible for conducting the human research at the tests. There is no evidence that the investigators responsible for HumRRO were informed about the Wilson memo. Dr. Crawford, for example, when queried in 1994, reported that he did not know of the 1953 Wilson memorandum. It is possible that HumRRO was not viewed as being subject to the requirements stated in the Wilson memo despite the fact that it was human research relating to atomic warfare. Although the experimental variable was participation at a bomb test, arguably, the troops would have been present at the test in any event, along with many thousands of other soldiers who were not subjects in the HumRRO research.

Atomic Effects Experiments

At the same time that the third set of HumRRO experiments was being conducted, in April 1953 at Desert Rock V, the Army called on several dozen "volunteers for Atomic Effects Experiments."[50] According to the Army, all were officers familiar with the "experimental explosion involved" and were able to personally judge "the probability of significant variations in [weapon] yield." They were instructed to choose the distance from ground zero they would like to occupy in a foxhole at the time of detonation, as long as it was no closer than 1,500 yards. If the surviving documentation is the measure, these officers, and perhaps officer volunteers in the subsequent Desert Rock series, were the only subjects of bomb-test research who signed forms saying that they were voluntarily undertaking risk.[51] The exposures were meant to set a standard for developing "troop exposure programs and for confirming safety doctrine for tactical use of atomic weapons."[52]

An Army report on the volunteers at Desert Rock V concluded that there would be "little more to be gained by placing volunteer groups in forward positions on future shots."[53] An April 24, 1953, Army memorandum recommended termination of the program "as little will be gained in repeatedly placing volunteers in trenches 2000 yards from ground zero."[54] However, officer volunteers were called on again at the next Desert Rock exercises at the 1955 nuclear test series called Operation Teapot. Following Teapot, the Army recommended that further experiments be conducted in which the volunteers would be moved closer to ground zero, "until thresholds of intolerability are ascertained."[55] This "use of human volunteers under conditions of calculated risk," the Army told the AFSWP, "is essential in the final phase of both the physiological and psychological aspects of the overall program."[56]

In response, the AFSWP pointed out that the injury threshold could not be determined "without eventually exceeding it."[57] The Army was essentially proposing human beings be exposed to the detonation's blast effects to the point of injury. The proposal, an AFSWP memo explained, would not pass muster under the rules of the Nevada Test Site and was otherwise unacceptable:

In particular, it is significant that the long range effect on the human system of sub-lethal doses of nuclear radiation is an unknown field. Exposure of volunteers to doses higher than those now thought safe may not produce immediate deleterious effects; but may result in numerous complaints from relatives, claims against the Government, and unfavorable public opinion, in the event that deaths and incapacitation occur with the passage of time.[58]

If the Army wanted more data on blast effects, AFSWP declared, it should proceed with laboratory experiments, for which money would be made available. The AFSWP was not opposed to the kinds of activities that had previously taken place at Desert Rock. But those activities, AFSWP's memo observed in passing, "cannot be expected to produce data of scientific value."[59]

The Desert Rock experience was apparently repeated, again with officer volunteers, in the next Nevada test series, the 1957 Operation Plumbbob. Although the total number of officers involved in all of the "officer volunteer" experiments is not known, it is probably fewer than one hundred.

The Flashblindness Experiments

Beginning with the 1946 Bikini tests, experiments with living things became a staple of bomb tests. At Operation Crossroads, animals were penned on the decks of target ships to study the effects of radiation. In the 1948 Sandstone series at the Marshall Islands Enewetak Atoll, seeds, grains, and fungi were added. In 1949, the AEC and the DOD began to coordinate the planning of the biomedical experiments at tests and set up a Biomedical Test Planning and Screening Committee to review proposals.[60] Presumably, the human experiments at bomb tests should have been filtered through this or some other review process designated to consider experiments. Yet, in only one case--flashblindness experiments--did this happen.

With Dr. Meiling's 1951 call for renewed DOD effort at biomedical experimentation came a revival of the DOD-AEC joint biomedical planning. From the start, the AEC doubted DOD's willingness to cooperate.[61] In a January 1952 letter to Shields Warren, Los Alamos's Thomas Shipman complained that the committee was limited to reviewing proposals from civilian groups, and not the military: "[I]f," he wrote, the "AEC can not exercise a measure of control in this matter, they might better withdraw from the picture completely and permit the military to continue on its own sweet way without the somewhat ludicrous spectacle of an impotent committee's snapping its heels like a puppy dog."[62] In retrospect, Shipman wrote to Warren's successor in June 1956, the military's refusal to participate "reduced that committee to impotence."[63]

Whatever its effectiveness, in 1952 the biomedical research screening group did consider at least one of the military's flashblindness experiments.[64] Flashblindness--the temporary loss of vision from exposure to the flash--was a serious problem for all the armed services, but particularly for the Air Force. Pilots flying hundreds of miles an hour in combat could not afford to lose concentration and vision even temporarily.[65]

The flashblindness experiments began at the 1951 Operation Buster-Jangle, the series that included Desert Rock I, with the testing of subjects who "orbit[ed] at an altitude of 15,000 feet in an Air Force C-54 approximately 9 miles from the atomic detonation. . . ."[66] The test subjects were exposed to three detonations during the operation, after which changes in their visual acuity were measured.[67] Although these experiments were conducted at bomb tests that potentially exposed the subjects to ionizing radiation, the purpose of the experiment was to measure the thermal effects of the visible light flash, not the effects of ionizing radiation.

When another experiment was proposed for Operation Tumbler-Snapper, the 1952 Nevada tests, the AEC sought a "release of AEC responsibility" on grounds that "there is a possibility that permanent eye damage may result."[68] It is not clear how the military responded, but the experiment proceeded. Twelve subjects witnessed the detonation from a darkened trailer about sixteen kilometers from the point of detonation.[69] Each of the human "observers" placed his face in a hood; half wore protective goggles, while the other half had both eyes exposed.[70] A fraction of a second before the explosion, a shutter opened, exposing the left eye to the flash.[71] Two subjects incurred retinal burns, at which point the project for that test series was terminated.[72]The final report recorded that both subjects had "completely recovered."[73]

At the 1953 tests, the Department of Defense engaged in further flashblindness study.[74] During this experiment, "twelve subjects [dark-adapted] in a light-tight trailer were exposed to five nuclear detonation flashes at distances of from 7 to 14 miles."[75]

The flashblindness experiments were the only human experiments conducted under the biomedical part of the bomb-test program and the only human experiments where immediate injury was recorded. They were also the only experiments where there is evidence of any connection to the 1953 Wilson memorandum applying the Nuremberg Code to human experimentation.

Recently recovered documents show that upon a 1954 review of a report showing the injuries at the 1952 experiment, AFSWP medical staff immediately declared that "a definite need exists for guidance in the use of human volunteers as experimental subjects."[76] Further inquiry revealed that a Top Secret policy on the subject existed. That policy detailed "very definite and specific steps" that had to be taken before volunteers could be used in human experimentation. But, the AFSWP wrote, "No serious attempt has been made to disseminate the information to those experimenters who had a definite need-to-know."[77]

Nonetheless, some form of consent was obtained from at least some of the flashblindness subjects. In a 1994 interview, Colonel John Pickering, who in the 1950s was an Air Force researcher with the School of Aviation Medicine, recalled participating as a subject in one of the first tests where the bomb was observed from a trailer, and his written consent was required. "When the time came for ophthalmologists to describe what they thought could or could not happen, and we were asked to sign a consent form, just as you do now in the hospital for surgery, I signed one."[78] There is no documentation showing whether subsequent flashblindness experiments, which followed upon the issuance of the secretary of defense's 1953 memorandum, required informed and written consent. However, given the recollection of Colonel Pickering and the military tradition of providing for voluntary participation in biomedical experimentation, this may well have been the case. (A report on a flashblindness experiment at the 1957 Plumbbob test uses the term volunteers;[79] a report on 1962 "studies" at Dominic I provides no further information.)[80]

In early 1954 the Air Force's School of Aviation Medicine reported that animal studies and injuries at bomb tests (to nonexperimental participants) had shown that potential for eye damage was substantially worse than had been understood.[81] Studies of flashblindness with humans continued in both field and laboratory tests through the 1960s and into the 1970s. These experiments tested prototype versions of eye protection equipment, and the results were used to recommend requirements for eye protection for those exposed to atomic explosions.[82]

Research on Protective Clothing

In late 1951, following the first Desert Rock exercise, the government conducted Operation Jangle, a nuclear test series that detonated two nuclear weapons, one on the surface and one buried seventeen feet underground. The two Jangle shots were tests where the weapon's fireball touched the ground. When a nuclear weapon's fireball touches the ground it creates much more local fallout than an explosion that bursts in the air. Consequently, these tests posed some potential hazard to civilians who lived near the test site and to test observers and participants.

Two weeks before Jangle the DOD requested an additional 500 observers at each of the Jangle shots, to acclimate the troops to nuclear weapons. The AEC advised against the additional participants, declaring that "[t]his [the first detonation] was an experiment which had never been performed before and the radiological hazards were unpredictable." In the AEC's view, no one should approach ground zero for three or four days after the surface shot.[83]

The AEC seems to have been successful in persuading the Department of Defense not to include the extra observers, but the DOD did not agree to the AEC's suggestion on approaching ground zero. Four hours after the first shot, the DOD conducted research involving troops who were accompanied by radiation safety monitors.[84] Eight teams of men walked over contaminated ground for one hour to determine the effectiveness of protective clothing against nuclear contamination.[85] Similar tests were conducted after the second shot at Jangle, but this time after a longer period. Five days after the shallow underground shot, men crawled over contaminated ground, again to determine the effectiveness of protective clothing.[86] Other men rode armored vehicles through contaminated areas to check the shielding effects of tanks and to check the effectiveness of air-filtering devices.[87] According to the final report, the protective clothing was "adequate to prevent contact between radioactive dust and the skin of the wearer."[88]

The information on this research is limited. The only mention of the subjects in the report reads, "The volunteer enlisted men, too numerous to mention by name, who participated in the evaluation of protective clothing were of great assistance which is gratefully acknowledged."[89] It is likely that at the time these men were not viewed as subjects of scientific research but rather as men who had volunteered for a hazardous or risky assignment. We know nothing about what these men were told about the risks or whether they felt they could have refused the assignment if they had an interest in doing so.

The Jangle activities are a good illustration of difficulties in drawing boundaries in the military between activities that are research involving human subjects and activities that are not. Although the Jangle evaluation was likely not considered an instance of human research at the time, it has many similarities to ground-crawling activity conducted several years later, not in conjunction with a nuclear test, that was treated as research involving human subjects. In 1958 ninety soldiers at Camp Stoneman, in Pittsburg, California, were asked to perform "typical army tactical maneuvers" on soil that had been contaminated with radioactive lanthanum.[91] The soldiers were then monitored for their exposure to study beta contamination from this nonpenetrating form of radiation. In 1963 soldiers were again asked to maneuver on ground contaminated with artificial fallout, this time at Camp McCoy in Wisconsin.[91]

The plans for the 1958 maneuvers, which were administered by the Navy's Radiological Defense Laboratory, had been submitted for secretarial approval, as was required for biomedical experiments involving Navy personnel.[92] In accordance with the Navy rules, the soldiers signed "written statements of voluntary participation."[93] During the 1963 experiments the Army processed the activity under its 1962 regulation on human experimentation (AR 70-25).[94] This rule, a public codification of the secretary of defense's 1953 Nuremberg Code rule, also required secretarial review and written consent.[95]

Cloud-Penetration Experiments

What are the dangers to be encountered by the personnel who fly through the cloud?--How much radiation can they stand?--How much heat can the aircraft take?--Can the ground crews immediately service the aircraft for another flight?--If so, what precautions are necessary to insure their safety?[96]

The Air Force felt that it was essential to answer these questions. To do so, it carried out experiments, including some with animals and a few with humans.

At the first atomic tests the military used remote-controlled aircraft, called "drones," to enter and gather samples from atomic clouds in order to estimate the yield and learn the characteristics of the weapon being tested. Military pilots did, however, "track" mushroom clouds, gathering information and plotting the cloud's path in order to warn civilian aircraft. During a 1948 test, a cloud tracker piloted by Colonel Paul Fackler inadvertently got too close to a cloud. But after the accident, Colonel Fackler commented, "'No one keeled over dead and no one got sick.'"[97] Colonel Fackler's experience, an Air Force history later recorded, showed that manned flight through an atomic cloud "would not necessarily result in a lingering and horrible death."[98]

Some of the trackers had "sniffers" on their aircraft to collect small samples. The Air Force conducted experimental sampling missions at 1951 tests and later permanently replaced the drones with manned aircraft because drones were difficult to use, and they often did not get the quality samples of the atomic cloud that Atomic Energy Commission scientists desired. By Operation Teapot (1955), the AEC considered the testing of a nuclear device "largely useless" unless sampler aircraft were used to obtain fission debris that would be used to estimate the nuclear weapon's performance.[99]

As the sampling mission became routine, a new mission in the clouds began. At Teapot the Air Force performed the first manned "early cloud penetration." The phrase was used by the Air Force to refer to missions conducted as soon as minutes after detonation of the test weapon. The main purpose was to discover the radiation and turbulence levels within the cloud at early times after detonation.

Like the first sampling missions, the first early cloud-penetration missions were conducted by unmanned drone aircraft. In 1951 Colonel (now General) E. A. Pinson, an Air Force scientist who had earlier conducted tracer experiments on himself and other scientists, placed mice aboard a drone aircraft; in 1953 he flew mice, monkeys, and instrumentation in drone aircraft through atomic clouds. Pinson concluded that the radiation risk from flying manned aircraft through atomic clouds could be controlled by monitoring the external gamma dose.[100] But the Air Force was not convinced and asked Pinson to follow up the animal experiments with studies with humans during Operation Teapot (1955) and Operation Redwing (1956) to confirm the results. This research appears to have involved a small number of subjects, perhaps in the range of a dozen or so.

Pinson designed the human experiments to "learn exactly how much radiation penetrates into the human system"[101] when humans flew through a mushroom cloud. The Air Force had pilots swallow film contained in small watertight capsules. The film was attached to a string held in their mouths, so that it could be retrieved at the end of the mission.[102] When the film was retrieved, the researchers compared the exposures measured inside the human body with those measured on the outside. They found that the doses measured outside the body were essentially identical to the doses inside the body; this was a critical finding, because it meant that surface measurements would be "representative of the whole-body dose."[103]

For the experiment, the AEC test manager for Teapot waived the AEC's test-exposure limit of 3.9 roentgens and permitted four Air Force officers to receive up to 15 roentgens whole-body radiation.[104] The exemption was "based on the importance of [the project] to the Military Effects Test program and the fact that radiation up to 15 R may be necessary for its successful accomplishment."[105] When the air crews entered the atomic clouds, they measured dose rates of radiation as high as 1,800 rad per hour. Since the crews were in the cloud for such a short period of time, however, the actual doses were much lower than 1,800 R.[106] The maximum reported dose received on a single mission was 17 R,[107] higher than the 15 R authorized for the project. Since the air crews flew on several missions, two of the crew members received more than 17 R.[108]

A year later, at Operation Redwing, where the atomic and hydrogen bombs were tested, the Air Force conducted another series of experimental cloud penetrations. Part of the Redwing experiment was to measure the hazard from inhaling or ingesting radioactive particles while flying through a mushroom cloud. When mice and monkeys were flown through clouds during earlier tests they were placed in ventilated cages to determine the hazard from inhaling radioactive particles. The studies found that the hazard from inhalation was less than 1 percent of the external radiation hazard. As General Pinson put it, "In other words, if the internal hazard were to become significant, the external hazard would be overwhelming."[109] To confirm this finding, Pinson undertook a similar experiment with humans, and again, as with the Teapot experiment, Pinson was a subject as well as a researcher. To perform the experiment, no filters were installed in the penetration aircraft.[110] Again, it is estimated that about a dozen subjects were involved.

The military this time set the authorized dosage (the maximum dosage to which Pinson could plan to have people exposed) at 25 R and a limiting dosage (in which case a report had to be filed) at 50 R.[111] During the experiment "maximum radiation dose rates as high as 800 r/hr were encountered, and several flights yielded total radiation doses to the crew of 15 r."[112] (To measure the internal dose of radiation the scientists analyzed urine samples and used whole-body counters.)

The project, as Pinson's final report noted, marked the transition from animal experimentation to human measurement:

Although a considerable amount of experimentation had been done with small animals which were flown through nuclear clouds, the early cloud-penetration project of Operation Redwing was the first instance in which humans were studied in a similar situation.[113]

The results confirmed those of the animal experiments. The internal hazard of radiation was insignificant relative to the external hazard. Consequently, the researchers recommended "that no action be taken to develop filters for aircraft pressurization systems nor to develop devices to protect flight crews from the inhalation of fission products."[114]

Experimental Purpose: Military Tactics, Money, and Morale

Why was the Air Force interested in showing that atomic clouds could be penetrated soon after a detonation?

Most important, the military wanted to assure itself that it was safe for combat pilots to fly through atomic clouds, if need arose during atomic war. But the research did not make much of a scientific contribution. Researchers had already established the levels of radiation in atomic clouds by flying drone aircraft through them, and there was nothing pathbreaking about humans being exposed to levels of radiation under 25 R. General Pinson later noted, "there are no research people that I know of that gave a damn [about manned early cloud-penetration experiments], because this is . . . a negligible contribution to research and scien[ce]--scientifically, you know, this contributes less than I suspect anything I've ever done . . . its only virtue is the practical use of it."[115]

From the scientific perspective the data would not likely be of great use; from an immediate practical perspective human data were felt to be essential for reassurance. Should the Air Force have been satisfied with the wealth of data it had from the drone experiments? In retrospect Pinson found the question difficult. "There's reason to say, 'Well, you should have been satisfied with the data that had been gathered with the drones.' But, you know, these are hard-nosed, practical people that--that put their life on the line and in military combat . . . where the hazards are far greater than in this modest exposure to radiation."[116]

The budget also played a key role in cloud-penetration research, as well as the related decontamination experiments, which will be discussed shortly. The Defense Department declared that the knowledge gained through its cloud-penetration experiments would save "the taxpayers thousands upon thousands of dollars" because there would be no need to develop special protective clothing or equipment, which had been thought to be necessary.[117]

As in the case of the HumRRO experiments and the troop maneuvers, indoctrination and morale were important forces behind the experimentation. "Perhaps the most important problem of all," a popular men's magazine of the day wrote about the Teapot experience, "might be a psychological resistance of combat pilots and crews flying into the unknown dangers of hot, radio-active areas."[118] The press, therefore, depicted the Teapot experiment as a message to the world--pilots can fly through atomic clouds safely.

Research, Consent, and Volunteerism

Like the HumRRO experiments, the cloud flythrough experiments were treated as occupational, rather than experimental, activities. None of the participants signed consent forms, and waivers to dose limits were sought, and approved, under the process followed for the nonexperimental flythrough activities. In 1995 General Pinson said that he had not been aware of the ethical standards declared in the 1953 secretary of defense memorandum. If he had been, he "would have gotten written consent from the people that were involved in this."[119]

A 1963 Air Force history of the cloud-sampling program does not describe the process of crew and pilot selection, but does provide a perspective:

The Strategic Air Command pilots picked to fly the F-84G sampler aircraft were pleased to learn that they were doing something useful, . . . not serving as guinea pigs as they seriously believed when first called upon to do the sampling.[120]

Did the personnel understand the risks? Some of them surely did. The aircraft carried airmen and scientific observers. Because the scientific observers were the very scientists who designed the experiments, they certainly understood the radiation risks as well as anyone could be expected to. In this way, the cloud flythrough experiments exemplified the ethic of researcher self-experimentation. As Pinson recalled in 1995, "If you are going to do something like this and you think it's safe to do it, then you shouldn't ask somebody else to do it. The way you convince other people that at least you think it's all right, is do it yourself."[121]

The nonscientists were briefed and informed that the risks from their radiation exposure would be minimal.[122] A pilot in the cloud-tracking activities recalled one of the briefings: "The scientists line up at a briefing session and tell you there's no danger if you will follow their instructions carefully. In fact, they almost guarantee it."[123]

But many of the pilots seemed to have been neither worried at the prospect of risk nor excited at the prospect of glory. Pinson, for example, described the attitude of the pilot who flew his aircraft as "matter of fact."[124] And at Operation Teapot, Captain Paul M. Crumley, project officer for the early cloud penetrations, stated, "We consider these flights routine. Neither the pilots nor observers are unduly concerned over the fact that no one else has flown into an atomic cloud so soon after detonation."[125]

Decontamination Experiments

In conjunction with the Teapot cloud flythrough experiment, the military also conducted an experiment on ground crews "to determine how soon these same aircraft could be reserviced and made ready to fly again."[126] The Air Force used the contaminated aircraft from the early cloud-penetration experiment.[127] The research sparked a debate between the Air Force and the AEC over the costs and benefits of safety measures, a debate that was itself resolved by further experimentation.

In one part of the "experimental procedure," personnel (the number involved is not reported) rubbed their gloved hands over a contaminated fuselage, and in another part "the bare hand was also rubbed over a surface whose detailed contamination was known and a radioautograph of the hand surfaces [was] made."[128] None of the "survey team" exceeded the AEC's gamma exposure limit of 3.9 R.[129] Concluding that aircraft did not need to be "washed down" or decontaminated after they flew through the atomic clouds, Colonel William Kieffer, deputy commander of the Air Force Special Weapons Center, proposed that decontamination procedures be eliminated except in extreme circumstances. This change in procedures might cause overexposures, Kieffer wrote, but they would be acceptable as long as "dangerous" dosages would be avoided.[130]

The proposal was not warmly received by the AEC. Los Alamos's Thomas Shipman complained that the goal should be to reduce exposures to zero.[131] Harold Plank, a Los Alamos scientist who was in charge of the cloud-sampling project and who rode along on many of the cloud-sampling missions, said, "Kieffer simply could not understand the philosophy which regards every radiation exposure as injurious but accepts minimum exposures for critical jobs."[132]

Kieffer suggested a compromise; test the proposal with only one or two sampler aircraft.[133] Plank objected, but the AEC test manager promised to "do everything possible to obtain a waiver of AEC operating radiological safety requirements."[134] The Air Force carried out the study during the 1957 Operation Plumbbob. An additional plane was flown through the atomic clouds created by five "events" to determine the hazard from the Air Force's proposed procedures.[135] The study showed that decontamination would be necessary to prevent overexposures at test sites.[136] In the end, the Air Force was unsuccessful in its attempt to change the decontamination procedures for sampler aircraft.

We do not know how the Air Force viewed this activity. Given that it did not treat the cloud flythroughs as an experiment, it is unlikely that the Air Force considered the ground personnel activity to be an experiment. There is no record of what the ground personnel were told or whether they were volunteers.

back table of contents forward