ACHRE Report

Part II

Chapter 13


National Security and Governmental Prestige: The Legal Tradition Inherited by Cold War Agencies

The Practice of Secrecy

The Records of Our Past


Chapter 13: The Practice of Secrecy

The AEC Addresses Secret Manhattan Project Experiments

When it began operation in 1947, the AEC was heir to two traditions: one in which official secrets could extend beyond national security to matters of prestige and another in which the interest in promoting openness and limiting secrecy to matters of national security was recognized. In public, AEC biomedical officials and advisers advocated the latter policy. In secret they embraced the former and even expanded it to encompass "embarrassment." Through as late as 1949, the declassification of reports on human experiments involved their review for public relations and legal liability implications. Documents revealing the dual tracks of public policy making and the secret review process did not become public until 1994. Important pieces of the story remain unclear, including the way in which AEC officials and advisers reconciled seemingly contrary principles.

As described in chapter 5, when Manhattan Project medical official Hymer Friedell recommended in late 1946 that one of the reports on the plutonium injection experiments be declassified, officials inside the new AEC reacted strongly. On March 19, 1947, AEC Medical Division chief Major B. M. Brundage countermanded the declassification decision, on grounds of "public relations." The plutonium report produced the strongest reaction, but it was not the only report on human data at issue. Brundage's March 19 memo also stated that further reports ("Studies of Human Exposure to Uranium Compounds" and "Uranium Excretion Studies") should remain classified. On March 21, an AEC declassification officer confirmed the reclassification on the ground that "these documents may involve matters prejudicial to the best interests of the Atomic Energy Commission in that experiments with humans are involved." The memo expressed hope that "a definite policy in this matter will be announced or explained in the near future."[21]

In April 1947 that hope was partly fulfilled when Colonel O. G. Haywood of the Corps of Engineers wrote to H. A. Fidler, an AEC information officer, that "it is desired that no document be released which refers to experiments with humans and might have adverse effects on public opinion or result in legal suits. Documents covering such work should be classified as secret.[22]

Shortly thereafter the AEC seemingly embraced both of the contradictory traditions to which it was heir. In June 1947, the AEC approved the basic policy of the 1945 Tolman report as an “interim policy.[23] In August 1947 General Manager Carroll Wilson publicized that approval in a letter appearing in the Bulletin of the Atomic Scientists. The letter indicated that the AEC endorsed the Tolman report, quoting sections that advocated declassification of nuclear weapons information that posed no "danger to our military security."[24]

Also in June 1947, Chairman David Lilienthal's blue-ribbon Medical Board of Review issued its recommendations on the biomedical program. "Secrecy in scientific research," the board declared, "is distasteful and in the long run is contrary to the best interests of scientific progress." The board recommended that "in so far as it is compatible with national security, secrecy in the field of biological and medical research be avoided."[25] The endorsement of the Tolman report and the broad statement of the Medical Board would seem to indicate that high-level AEC officials and biomedical advisers were opposed to secrecy not required by national security.

But these broad statements left unaddressed the specific response to continued requests to declassify Manhattan Project human experiments. In a June 5 response to researcher Robert Stone, General Manager Wilson suggested that any experiments involving "unwitting subjects" should remain classified as they "might have an adverse effect on the position of the Commission" in "the eyes of the American people and the medical profession in general."[26] In an August 12 letter to Stone, Wilson indicated that the Medical Board of Review had considered the question of secrecy and human experiments in mid-June, but the matter had been deferred.[27]

On August 9, John Derry, serving as acting general manager, evidently in Wilson's absence, proposed a set of guidelines that restated the proposition that secrecy could be based on reasons other than national security. The definition of Confidential that he proposed went beyond the Army and Manhattan Project rules:

CONFIDENTIAL: Documents, information or material, the unauthorized disclosure of which, while not endangering the National security, would be prejudicial to the interests or prestige of the Nation or any Governmental activity, or individual, or would cause administrative embarrassment, or be of advantage to a foreign nation shall be classified CONFIDENTIAL [emphasis added].[28]

The Derry memo called for review by a classification board assembled from the AEC's regional sites. In September, this board assembled in Oak Ridge. The available documentation does not show that Derry's proposed rules went into effect, but does show that the Classification Board blessed the illustrations of matter that "should be graded" Secret or Confidential. The former category included "certain selected human administration experiments performed under MED [Manhattan Engineer District]."[29] The latter category contained a broad catch-all:

All documents and correspondence relating to matters of policy planning and procedures, the given knowledge of which might compromise or cause embarrassment to the Atomic Energy Commission and/or its contractors [emphasis added].[30]

Following the Classification Board's meeting, Oak Ridge officials wrote to Washington headquarters in search of policy guidance on human subject research. Oak Ridge explained that researchers were eager to have their work declassified. "However, there are a large number of papers which do not violate security, but do cause considerable concern to the Atomic Energy Commission Insurance Branch and may well compromise the public prestige and best interests of the Commission." A problem arose, for example, "in the declassification of medical papers on human administration experiments done to date. Again many of these radioactive agents have been of no immediate value to the patient but rather a much needed opportunity for tracer research."[31]

The problem, Oak Ridge pointed out, was not limited to data from human experiments, but also included health risks that radiation posed for workers and for the public:

Papers referring to levels of soil and water contamination surrounding Atomic Energy Commission installations, idle speculation on future genetic effects of radiation and papers dealing with potential process hazards to employees are definitely prejudicial to the best interests of the government. Every such release is reflected in an increase in insurance claims, increased difficulty in labor relations and adverse public sentiment.[32]

Indeed, the Insurance Branch had already reviewed some papers that were slated for declassification. It had advised against publishing papers that suggested health hazards to the public. In the case of one paper, for example, the Insurance Branch wrote in June 1947:

We question the advisability of publishing this document unless the contractor involved is able to establish that the amounts of fissionable material leaving the area is in no way a health hazard to the people living down stream.[33]

In an October memo to Washington, Oak Ridge suggested that the Insurance Branch should routinely review declassification decisions for liability concerns:

Following consultation with the Atomic Energy Commission Insurance Branch, the following declassification criteria appears desirable. If specific locations or activities of the Atomic Energy Commission and/or its contractors are closely associated with statements and information which would invite or tend to encourage claims against the Atomic Energy Commission or its contractors such portions of articles to be published should be reworded or deleted. The effective establishment of this policy necessitates review by the Insurance Branch, as well as the Medical Division, prior to declassification.[34]

Oak Ridge explained that its acting medical adviser, Dr. Albert Holland, Jr. (whose contribution had been praised in the June 1947 report of the Medical Board of Review), would be in Washington on October 11 to discuss the matter further.[35] On that date the Advisory Committee for Biology and Medicine met and concluded that the "important" policy questions raised by Oak Ridge would require "more study."[36]

While the discussion of Oak Ridge's inquiry did not resolve the question of classification, the matter was otherwise addressed at the October 11 meeting. The draft of the secret minutes of the meeting record the discussion of yet another letter from Dr. Robert Stone, regarding the release of "classified papers containing information on human experiments with radioisotopes conducted within the AEC program."[37] The ACBM concluded that the "problem" was addressed by "the recommendations of the Medical Board of Review and that papers on this subject should remain classified unless the stipulated conditions laid down by the Board of Review are complied with."[38]

What were the recommendations of the Medical Board of Review that the ACBM referred to? Recall that its public report did not address human experiments but briefly declared the importance of limiting secrecy. The matter is cleared up by two letters written by General Manager Wilson on November 5--the first to Stone (this is the "second Wilson letter" discussed in chapter 1) and the second to ACBM Chair Alan Gregg.[39] Consistent with the October 11 ACBM minutes, the letter to Stone explained that all classified research not in compliance with certain conditions laid down by the Medical Board would remain classified. These conditions, as we discussed in chapter 1 included written "informed consent" from the patient and the next of kin. This requirement, Wilson further explained, was contained in an "unpublished and restricted" draft report of the Medical Board of Review, which had been read to the Commission in June. The letter to Gregg, who had served on the Medical Board of Review, indicated that the ACBM need not consider the matter further because the Medical Board of Review's statement was sufficient.

Thereafter, documents show that the AEC continued to review reports for possible public relations and liability consequences and, as Oak Ridge had recommended, called on the AEC Insurance Branch to vet reports for public relations and liability implications.

In 1948 former Manhattan Project researchers pressed the AEC to declassify data from human experiments for inclusion in a history of Manhattan Project medical research as part of a group of publications called the National Nuclear Energy Series, or "NNES." In February 1948, the University of Rochester's Harold Hodge complained about classification officers gutting his chapter on uranium toxicology. "I would like," Hodge wrote, "to advance the argument that Chapter XVI does not report experiments with humans, and should never have been classified on this basis in the first place."[40]

The researchers sought a "final policy" decision on reports regarding plutonium and uranium from the Division of Biology and Medicine and its advisory committee. In a March 15 letter to a participant in the NNES project, Oak Ridge's Holland reported that it was "the feeling" of these groups that the reports should not be declassified. "While I am sure we both fully appreciate the desirability of declassification, I feel certain that the various individuals concerned will also understand and appreciate the reasons for this decision."[41] (The minutes of the March 10, 1948, ACBM meeting, themselves declassified in 1994, do not refer to the policy decision.)

The policy of classifying reports for reasons of public relations and liability was not limited to human experiments conducted under the Manhattan Project; it extended to at least one human experiment conducted under the AEC. In late 1948, Division of Biology and Medicine chief Shields Warren stated his "complete agreement" with Oak Ridge's Holland that a report on a 1948 University of California experiment with zirconium (the research has since become known as the "CAL-Z" experiment; see chapter 5) had to be kept under wraps.[42] The report had to remain secret because "it specifically involves experimental human therapeutics" and could not be rewritten in a way that "would not jeopardize our public relations."[43]

In addition, data on workers, as well as sick patients, was vetted for labor relations and legal concerns. In chapter 11 we discussed the exposure of Los Alamos workers involved in the "RaLa" intentional releases. In late 1948 the AEC Declassification Branch reviewed a study entitled "The Changes in Blood of Humans Chronically Exposed to Low Level Gamma Radiation." The document, a memo from the Declassification Branch recorded, "has been issued as an unclassified report by Los Alamos, since it clearly falls within the open fields of research." While agreeing with Los Alamos, the Declassification Branch sent the document to the Insurance Branch, at the suggestion of the medical adviser.[44]

In a December 20, 1948, memo to the Declassification Branch, the Insurance Branch recorded its alarm over the study's finding that accepted gamma radiation safety levels "may be too high." In calling for "very careful study" before making the report public the Insurance Branch declared:

We can see the possibility of a shattering effect on the morale of the employees if they become aware that there was substantial reason to question the standards of safety under which they are working. In the hands of labor unions the results of this study would add substance to demands for extra-hazardous pay . . . knowledge of the results of this study might increase the number of claims of occupational injury due to radiation and place a powerful weapon in the hands of a plaintiff's attorney.[45]

While the Insurance Branch reviewed declassification decisions it did not automatically veto the release of all human experimental data. In an October 1947 memo, Holland approved a report ("The Effect of Folic Acid on Radiation Induced Anemia and Leucopenia") for publication "since purportedly the human work was done in the Department of Medicine of the University of Chicago," and not, presumably, an AEC or Manhattan Project facility.[46] Even when publication might result in bad public relations or might encourage litigation, information was sometimes released.[47]

Thus, while the evidence of formal policy-making that can be recovered is fragmentary, it appears that even though the AEC biomedical officials and advisers publicly advocated limiting secrecy to matters of national security, they secretly endorsed a different policy and followed the secret one. The AEC employed the concepts of "prejudicial to the best interests of the government" and "administrative embarrassment" in determining what information to withhold on human experiments. This course was crafted and administered in secret and remained a secret for decades. Its full reach remains unknown.

While our discussion thus far has focused on the AEC, it was not alone in its concerns that data on human radiation exposure could cause public relations or legal liability problems. As we saw in chapter 10, in 1947, former Manhattan Project head General Groves, and the chair of the new AEC's Interim Medical Advisory Committee, Stafford Warren, were evidently among those who counseled the Veterans Administration to keep secret records in anticipation of potential claims from servicemen. In both cases, the impulse to keep such information secret was accompanied by the decision to create a highly publicized program of radioisotope research, which resulted in numerous human radiation experiments that were not secret.

The practice (and any policy) of keeping information secret on grounds of embarrassment or potential legal liability should have ended no later than 1951, and perhaps as early as 1949.[48] In its 1949 "Policy on the Control of Information," the AEC recognized that secrecy must be balanced against not only the value of the progress of science but also the value of a well-run democracy. Limiting secrecy, the AEC said, ensures "that people may be able to judge the action of their representatives and officials and to participate in public policy decisions. Information about a public enterprise of such consequence as the atomic energy program should be concealed only for reasons soundly based upon the common defense and security."[49] In 1951 President Harry Truman issued a new executive order on classification.[50] While the order expanded government secrecy by giving every department and agency the authority to classify information, it limited the reasons for classification to national security. Today, the governing executive order expressly prohibits classification of information "in order to: (1) conceal violations of the law, inefficiency, or administrative error; (2) prevent embarrassment to a person, organization, or agency; (3) restrain competition; or (4) prevent or delay the release of information that does not require protection in the interest of national security." The order also prohibits classification of "basic scientific research information not clearly related to national security."[51] As we shall see later in this chapter, while the law has long since begun to draw a line against the keeping of classified secrets for reasons other than national security, the boundary between national security and public relations rationales remains murky.

Human Radiation Experiments In the 1950s: Experiments Are Not Classified, but Some Secrets Remain

The 1947-1948 AEC declassification controversy may have taught Shields Warren and other AEC biomedical officials that secrecy and human radiation experimentation were a troubling mix, to be avoided if possible. The search efforts of the Human Radiation Interagency Working Group and the Committee located very few human radiation experiments in the post-Manhattan Project period that were classified secrets. Nonetheless, important information relating to many experiments was still intentionally concealed from the public.

When the AEC and DOD debated the need for human experiments for the proposed nuclear-powered airplane (NEPA) in 1950, Warren and the Advisory Committee for Biology and Medicine counseled the Defense Department that there would be "serious repercussions from a public relations standpoint" if human experiments were conducted by an agency that did some of its work in secret.[52] As we saw in chapter 1, in March 1951, Los Alamos asked Warren to state the policy on human experimentation. In transmitting to Los Alamos excerpts from General Manager Wilson's November 1947 letter to Stone, which cited the requirement for "informed consent," Warren added further counsel against secrecy. Warren cited the Medical Board of Review's public declaration that secrecy should only be countenanced when required by national security. He then quoted ACBM chairman Alan Gregg: "The secrecy with which some of the work of the Atomic Energy Commission has to be conducted creates special conditions for the clinical aspects of its work in that the public is aware of this necessity for secrecy and of the subsequent difficulty of probing into it."[53] When in 1952 the DOD's Joint Panel on the Medical Aspects of Atomic Warfare called for renewed discussion of human experiments, Warren reportedly advised "that studies of this type under the Joint Panel's purview should be conducted by the Public Health Service or some agency where security restrictions would not lead to misunderstanding.”[54]

Thus, Warren and Gregg's statements convey a profound concern for the public's perception of human experiments, particularly where human experiments are conducted by agencies that also conduct activities in secret.

Under Paul Aebersold, the AEC isotope distribution program--the provider of the source material for many hundreds of human experiments--became a showcase for public research (see chapter 6). At the Defense Department as well, biomedical human radiation experiments--even when there was clear military purpose--were typically not classified. For example, post-Manhattan Project total-body irradiation research sponsored in part by the military, in the wake of the controversy that raged when similar human experiments were proposed for the NEPA project, was not conducted in secret (see chapter 8).

But if the experiments themselves were not secret, important decision-making context for them was sometimes secret, and hidden rules or practices may have also limited what the public was told about particular experiments. The ability of the public and the press to probe experiments connected to secret programs was limited, making it difficult for the public to critically assess the practices of its government.

For example, the 1950-1952 meetings in which DOD biomedical officials discussed the need for an ethical code to govern human experiments were classified.[55] So were the meetings of the Joint Panel on the Medical Aspects of Atomic Warfare. Similarly, meetings of the ACBM were often conducted in part or whole in secret. These meetings, as we have seen from the review of the 1947-1948 secret keeping, included seminal discussions of the ethics of human experimentation and the rules governing declassification of experimental data.[56]

To some degree experiments sponsored by civilian agencies such as the National Institutes of Health were also rooted in this secret context. The 1952 letter that reported Warren's belief that human experiments should be separated from secret programs communicated the willingness of NIH and PHS to cooperate in conducting research needed for military purposes. These civilian agencies were themselves participants in DOD biomedical planning for atomic warfare, and their research was also listed in the secret digests (which included classified and nonclassified research) of atomic warfare-related research that the DOD's Committee on Medical Sciences provided to the Joint Panel on the Medical Aspects of Atomic Warfare.[57] Also in 1952, an internal report on “Defense Activity of the National Institutes of Health (1950-52) noted that “a second major activity of the NIH relating to radiation research has been participation in the medical and biological aspects of atomic bomb tests. A large share of this activity has been borne by the Armed Forces Special Weapons Project. The substance of this work is classified.[58]

The country's research resources should have been available to serve national security needs. But, as Warren and Gregg suggested, when human research and national security are intertwined, care must be taken to ensure that the public has means to separate out secret and nonsecret purposes with confidence. At this time it is not clear what, if any, classified human radiation experiments were conducted by DHHS's predecessors and what was said in secret about otherwise public human radiation experiments.[59]

Similarly, while most AEC biomedical radiation research was not classified, some was. From available records, it appears unlikely that much of the secret research involved humans. But, given the secrecy and the absence of clear records, certainty is impossible.[60]

Moreover, even if little human subject radiation research itself was classified, information about the research could be concealed by less formal means. As we discussed in the Introduction and chapter 10, in July 1949, the NEPA advisory group met with a group of psychologists and psychiatrists to discuss the psychology of radiation risk. The participants were told:

This is not a closed meeting. Some of our advisers . . . have not been cleared. Ordinarily, medical and biological discussions are not, of course classified. We shall ask you, however, to refrain from discussing these matters on the outside, since of course we do not want newspapers to know of these discussions at this time.[61]

Moreover, the determination to render information formally secret could be applied in a manner that was invisible and arbitrary, as illustrated by the following case. At midcentury, the Medical College of Virginia (MCV) performed research on the effect of thermal burns for the Defense Department. MCV's research, conducted with animals, prisoners, and medical students, initially appears to have been a matter of public record. In January 1951, following inquiry by a reporter from the Richmond Times-Dispatch, MCV investigator Dr. Everett I. Evans grew alarmed that press reports decrying the use of dogs would "greatly harm the work we are doing on the experimental burn in relation to atomic bomb injuries."[62] Evans called on the chairman of the Army's Medical Research and Development Board to classify the work so that "I would have legal means of preventing a public newspaper discussion of these experiments. . . ."[63]

The Army immediately provided a declaration that all work under the MCV contract "will be classified RESTRICTED."[64] The Army decreed that a bureaucratic obstacle course would have to be overcome before information was released, including "coordination" with the experimenters, and evaluation by "the other branches of the Armed Forces, the Federal Civil Defense Administration, the National Security Resources Board, the Atomic Energy Commission, and the National Research Council."[65] This rigor was essential because "individual releases may be mistaken for official advice to civil defense groups and result in confusion of training and procedure, the stockpiling of unnecessary or inappropriate materials, etc."[66] Finally, perhaps on the possibility that the local reporter might be uniquely dogged, the Army added that it "is also the policy of the Department of Defense that public releases to the press are made simultaneously to all national news services, and that the releases are not made to individual reporters or newspapers."[67] While the secrecy was prompted by revelations on animal experiments, in late 1951 Dr. Evans invoked it to close the curtain on the use of prisoner volunteers at the state penitentiary.[68].") The prison assured Evans that inmates and staff were informed that "no publicity should be given to the experiment being carried on at the Medical College."[69]

In the case of research related to chemical and biological warfare, the military issued a secret edict that published articles be cleansed of any reference to military purpose.[70] In many cases the opportunity to obscure the full purpose of research by careful wording was obvious. As a DOD document put it, "the term 'radiobiology' is so flexible semantically that, depending upon the investigator's point of view, any project could be classified as 'clinical' or 'basic' or 'nuclear weapons effects.'"[71] In 1961, the U.S. Department of Agriculture issued an extensive bibliography of research on strontium and calcium. The preface made clear the publication was relevant to those researching fallout (radioactive strontium being a major fallout concern).[72] However, Advisory Committee staff review of many of the articles on human experiments included in the bibliography revealed few indications of fallout as a purpose for the research.[73]

The difficulty of determining what was secret is compounded because the government sometimes actively deceived or lied. Most remarkably, the AEC continually told inquiring members of the public that it did not perform human experiments--even when its isotope division very publicly supported them. In 1948, for example, the AEC wrote to a member of the public that "there is no possibility, at present or projected, of human experimentation with atomic energy."[74] In 1951, when the press pursued a rumor that the AEC was sponsoring an experiment with prisoners, the AEC's chief public information official assured the Associated Press that the AEC "has never sponsored a medical research project where human beings were being used for experimental purposes."[75] In 1953 the AEC wrote to members of the public that it "does not deliberately expose any human being to nuclear radiation for research purposes unless there is a reasonable chance that the person will be benefited by such exposure."[76] At the same time an internal AEC memo from the public information office noted that "any experimentation on humans has obvious and delicate public relations aspects. Any project involving such experimentation must have careful prior consideration by both the field and Washington, particularly as to content of any public statements."[77]

As we saw in chapter 12, uranium miners were not adequately informed about the purpose of research regarding their exposure to radon in the mines. Above and beyond lack of disclosure, there is evidence that deception was not unusual in data gathering on AEC workers, as illustrated by a 1955 exchange between the University of Rochester's Dr. Louis Hempelmann and the AEC Division of Biology and Medicine regarding a proposed study evidently designed to measure the occurrence of lung cancer among a group of former workers. "You will have to find a good excuse so as not to worry the person you are contacting," Hempelmann wrote to DBM chief Charles Dunham. "This isn't very clever but, perhaps, you could say in some convincing way that you, or rather the person conducting the study, represents a life insurance company studying the health of people employed by the Harshaw Company during a certain period."[78] Dr. Hempelmann apologized for his lack of imagination:

I don't know whether these ideas are helpful at all. It is more difficult to find an excuse for these individual workers than it is in the case of patients who were treated for something or other at a hospital. I think that someone with imagination might come up with a better idea than I have had to date.[79]

This last comment implies that it was not only workers, but also patients, who were deceived about their participation in research, and more easily at that. The statement is particulary striking when it is recalled that Dr. Hempelmann was, as an adviser to Robert Oppenheimer, a proponent of the plutonium injection experiments, and, following the war, became professor of experimental radiology at the University of Rochester, a major AEC biomedical contractor. Thus, if the statement is a reflection of the readiness to deceive patients, it is one mady by a doctor at the center of the AEC biomedical community and, indeed, was made directly to the head of the AEC's Division of Biology and Medicine.

Dunham's assistant evidently agreed that workers should be deceived, but "we have racked our brains for any useful subterfuge in carrying out the study but none came to mind which could possibly hold water for any length of time."[80] The AEC opted for subtle deception:

The attack with which we are going to start the study will be to inform the old Harshaw employees that our interest in them is only part of an over-all program to make sure that the safety controls in the atomic energy business are absolutely perfect. To be sure, such an approach might cause some alarm but this should not be too great, I hope, because it is essentially a negative one; namely, the Commission is sure that there will be no injury to its workers but it needs to document this fact for the record.[81]

The AEC official agreed that "routine physical examination would be relatively fruitless since the ultimate objective is to determine the incidence of lung cancer, which can be obtained best with a post-mortem examination. On the other hand," the official noted, "the attitude of the Western Reserve group [with whom the AEC was proposing to contract for the study] is that physicial examinations are a useful means for maintaining close contact with people and will improve the chances of getting post-mortem information."[82]

In sum, after the Manhattan Project the governing presumption, to which the Advisory Committee found little exception, has been that biomedical human radiation experiments should not be classified. But the presumption included important qualifications, some of which were hidden at the time, and others of which may be beyond our ability to retrieve and reconstruct. These qualifications are shortcomings and legitimate cause for public concern, especially when held up to the ideals publically espoused by the AEC's initial leaders.

Human Data Gathering Connected with Bomb Tests and Intentional Releases: National Security, Secrecy, and Public Opinion

The view that a line needed to be drawn to ensure that human radiation experiments were not too closely associated with secret keeping was not easily translated to settings where entire groups of people were placed at risk by environmental releases of radiation. In March 1951, as we have just noted, Shields Warren advised Los Alamos to avoid secrecy in human experimentation. Warren and other AEC officials also told the military of their concern for public repercussions if human experiments were conducted in close proximity to government secret keeping. At the same time, however, Warren and other AEC biomedical experts were called on to advise on nuclear weapons activities that might place entire populations at risk. Here, the question of public disclosure was more difficult to resolve. In May 1951 for example, as discussed in the Introduction, Warren chaired a secret meeting in Los Alamos to consider the safety concerns of the first underground test of a nuclear weapon. The record of the meeting shows that Warren and other experts worried that fallout from the tests could endanger citizens around the Nevada Test Site. The public was not given access to the discussion of testing that the participants were concerned might endanger surrounding communities.[83] Press information stressed the absence of public danger.[84]

As we saw in the discussion of intentional releases (chapter 11), little or no information was contemporaneously made public about the radiological warfare tests at Dugway, the RaLa tests at Los Alamos, or the Green Run at Hanford. National security required some degree of secrecy; but whether more could or should have been disclosed is unclear in retrospect. In the case of at least the Dugway tests, secrecy was fueled by concern that the public might not understand the tests and might question the program.

Atmospheric nuclear weapons tests were, in contrast to the intentional releases and underground nuclear weapons tests, much more difficult to keep secret. In chapter 10 we saw that activities could simultaneously have elements of deep secrecy and appear as front-page news. A then-secret report on the Desert Rock exercises observed, "It was a constant source of amusement at the camp that the newspapers carried accounts of the atomic tests which included information, usually accurate, which the men had been expressly forbidden to reveal."[85] At the same time that the bomb tests were highly publicized, basic information on the risks to participants was not public. "Secrecy," summarized Barton Hacker, author of a DOE-sponsored history of the bomb test program, "so shrouded the test program . . . that such matters as worker safety could not then emerge as subjects of public debate."[86]

Once bomb tests became routine, fallout presented a further opportunity and obligation for the government to sponsor data gathering, including human subject data gathering. It did so on a global scale. As discussed in chapter 12, the research on the Marshall Islanders to measure fallout effects began in secret. "Due to possible adverse public reaction, "the director of the research project was counseled, those involved should limit discussions of the research to those with a "need to know."[87] The Marshall Islands research was only one component of a worldwide data-gathering program that was constructed and operated in substantial secrecy until the latter part of the 1950s. The Advisory Committee was not created to study atomic bomb testing or the related debate about the effects of fallout. However, the human subject research related to bomb-test fallout also presents questions about openness and secrecy in human research and the ethics of human data gathering.

The Fallout Data Network: Projects Gabriel and Sunshine

The study of fallout began with the effects of the first atomic bomb test in New Mexico in 1945.[88] In 1949 the AEC commissioned Project Gabriel, a study to determine how many atomic weapons could be detonated before radioactive contamination of air, water, and soil would have a long-range effect upon crops, animals, and humans.[89] The AEC soon created a worldwide network for the collection and measurement of fallout (typically by permitting it to fall on a horizontal gummed paper or plastic sheet).[90] By 1954 Gabriel included about seventy investigations supported by the Division of Biology and Medicine, involving 325 person years of labor per year and costing $3.325 million annually.[91]

In the early 1950s the Defense Department created its own fallout research program, under the auspices of the Armed Forces Special Weapons Project. The Public Health Service joined with the AEC and the DOD in monitoring fallout around the Nevada Test Site.[92]

In 1953, under contract to the AEC and the Air Force, the Rand Corporation convened a review of Gabriel.[93] The study was directed by Dr. Willard Libby, a University of Chicago radiochemist who would receive the Nobel Prize in 1960 for the development of the radioactive carbon dating method. The resulting report concluded that strontium 90 (Sr-90) was the most dangerous long-term, global radioactive product of bomb testing and that a global study of strontium 90 fallout was needed.[94]

The report noted how atmospheric testing had, as an unintended side effect, introduced tracers into the world's ecosystem: "Until comparatively recently it would have been extremely difficult, if not impossible, to obtain a measure of a number of the parameters. Today we are afforded the opportunity of doing a radioactive-tracer chemistry experiment on a world-wide scale."[95] The group recommended that "studies then current be supplemented by a world-wide assay of the distribution of strontium 90 from the nuclear detonations which have occurred. This assay has been designated Project Sunshine." The name for the project would be variously attributed to the project's gestation in Santa Monica, California, (where Rand was headquartered) and to the determination to measure the presence of strontium in "sunshine units." Three laboratories were engaged to analyze samples of strontium 90: one at Libby's research center at the University of Chicago, another at the Lamont Geological Observatory of Columbia University, and a third at the New York office of the AEC.

The long-term goals of the full-scale study would be to (1) determine if a hazard had already been created by fallout; (2) determine the number of bombs that could be exploded without creating a hazard, and (3) determine the mechanisms by which radioactive materials might become concentrated.[96]

Secrecy and Deception in Fallout Studies: Project Sunshine's Collection of Human Bones

Project Sunshine was born in secrecy.[97] The decision to keep the existence of the worldwide assay Secret "limited the freedom with which suitable combinations of samples might be obtained from foreign countries."[98] For the pilot program, the report suggested that twelve human samples (bone and teeth) be drawn from each of six regions around the world. In addition, samples would be drawn from livestock, foodstuffs, water, and soil.[99] The discussion of collecting individual samples was limited to means of ensuring uniformity in practice, without mention of the ethical relationship between investigators and human subjects. An early effort concerned the collection of baby bones.

In an October 1953 letter to Dr. Libby, Robert A. Dudley of the DBM explained that the collection process would proceed "through personal contacts with foreign doctors" and groups like the Rockefeller Foundation, which had many overseas contacts. Because the chief of the DBM, Dr. John Bugher, advised that "security specifications" needed to be maintained, a cover story would be employed.[100]

The stated purpose of the collection is to be for a survey of the natural Ra [Radium] burden of human bones . . . there are still enough uncertainties regarding threshold dose for injury . . . to provide a plausible explanation for further surveys. . . . As for the emphasis on infants, we can say that such samples are easy to obtain here, and that we would like to keep our foreign collections comparable.[101]

Dudley explained that the AEC wanted to be kept "out of the picture where possible," but to be helpful "I would still be prepared to do all the work except for providing the signature."[102]

One week later Dudley wrote to Shields Warren in Boston. Dudley, noting that the effort was proceeding "pretty much on the lines you suggested," sought Warren's assistance in contacting another Boston doctor who might not be in on the full story. Dudley offered that "while the real purpose will of course remain secret . . . we do expect to make radium analyses on at least some of the samples, so our story is merely incomplete, not false."[103]

On the same day, Dudley wrote to his father, the director of a missionary organization, also in Boston. The letter explained the public purpose of the data gathering and solicited assistance.[104] On November 10, evidently from a referral from his father, the AEC official wrote to the Christian Medical Association in Nadya Pradesh, India, also soliciting assistance. Finally, the DBM sought assistance from civilian organizations that already had well-developed contacts at the local level in foreign countries.[105]

What was the "real purpose" that had to be kept so carefully concealed, even from those who were actually assisting the project? On December 9, Dudley sent a letter to a doctor at the AEC's project at the University of Rochester that explained "for you alone" the AEC's real interest:

This letter will explain in a little more detail than I was able to do over the phone our interest in obtaining infant skeletons from Japan.

The Division of Biology and Medicine is engaged in a project to evaluate the long range radiological hazard which might result from the large scale use of atomic weapons. . . . In order to help in the evaluation of the hazard, we are providing for the direct measurement of the world-wide Sr-90 distribution which has resulted from the 40 or 50 nuclear detonations in the last few years. One type of sample on which we are concentrating is the bones of infants, either stillborn or up to a year or two of age. We have found that stillborn bones are easy to obtain in the United States, and are trying to extend our collection to foreign countries. It appears that the ABCC [Atomic Bomb Casualty Commission] would be a logical contact in Japan. We could use perhaps 6 or 8 skeletons from that area.

It has been decided, for various reasons including public and international relations, to classify this project SECRET for the present. Hence, the unclassified description of our purpose in obtaining these bones is for Ra analyses.[106]

The July 1954 Gabriel report summarized the "human, animal and animal product samples" that had been analyzed.[107] The list included stillborns from Chicago (fifty-five), Utah (one), Vellore, southern India (three), and human legs from Massachusetts (three).[108]

Soon, the DOD was also engaged in fallout data gathering. In the fall of 1954, the Armed Forces Special Weapons Project established a "Fall-out Study Group" following a request for information from the Joint Chiefs of Staff.[109] In 1954 DOD planned a secret project to collect human urine and animal milk and tissue samples following the 1955 Operation Teapot tests in Nevada. The work was coordinated by the Walter Reed Army Institute for Research, with review from researchers at the Harvard Medical School and the National Institutes of Health. The purpose of the effort was to establish a baseline for forthcoming Pacific tests.[110] The military data gathering also involved a cover story. A December 16, 1954, memorandum from the chief of the Armed Forces Special Weapons Project stated, at least in regard to the animal sampling:

The actual data obtained are SECRET and the sample collection should be discreetly handled. It is suggested that a statement be included in the instructions to the effect that these samples are being collected for nutritional studies.[111]

In January 1955 the Gabriel-Sunshine program was the subject of a classified "Biophysics Conference" convened by the Division of Biology and Medicine. The spring 1954 Marshall Islands disaster had, the attendees were told, added new urgency to their task. "I keep reading," noted one participant, "the articles by the Alsops and others [journalists] of the high level groups which are frantically trying to find the answer to how many bombs we can detonate without producing a race of monsters."[112]

The Secret transcript of the conference, declassified from Restricted Data status only in 1995, shows that the AEC and its researchers assigned a high priority to what was referred to as "body snatching." No AEC program, explained Dr. Libby, who had become an AEC commissioner, was more important than Sunshine. There were great gaps in knowledge and human samples were essential to fill them. "[H]uman samples are of prime importance and if anybody knows how to do a good job of body snatching, they will really be serving their country."[113] In the 1953 Rand Sunshine study, Libby recalled, an "expensive law firm" was hired to study the "law of body snatching." The lawyers' analysis showed "how very difficult it is going to be to do it legally."[114]

Nonetheless, "excellent sources" were available from several places, including New York, Vancouver, and Houston. In Houston, said Columbia University's Laurence Kulp, "they intend to get virtually every death in the age range we are interested in that occurs in the City of Houston. They have a lot of poverty cases and so on."[115]

Where good personal relationships with medical sources existed, Dr. Kulp offered, "the men did not require you to tell them anything except that they realized it was something confidential. They could guess, and they probably didn't guess very wrong, but they were willing to cooperate just on the basis that this was an important thing."[116] With a connection "through one of the top medical people who is internationally known, it will not be hard at all to be able to establish the sites that we should establish." The DBM's Dr. Bugher explained that the AEC was exploring the possibility of a special clearance ("L") so that medical professionals who did not want to "fill out any forms" could be briefed on a limited basis. "You are," he stated, "dealing with directors of hospitals and pathologists and persons in general who have an understanding of the seriousness of the project in which we are engaged."[117]

Libby hoped to declassify at least the existence of the Sunshine program. "Whether this is going to help in the body snatching problem, I don't know, I think it will. It is," he said, "a delicate problem of public relations, obviously."[118]

The efforts bore fruit. A report on Sunshine's 1955-1956 operations recorded that during that period hundreds of human bone samples were collected by dozens of stations abroad and by researchers in Boston, Denver, Houston, and New York.[119]

In addition to the Sunshine-related research, the AEC sponsored further efforts to gather human tissue in order to study the effects of radiation on weapons complex workers, as well as fallout on citizens. In a June 1995 report, the General Accounting Office summarized fifty-nine studies, most of which were conducted and terminated in the 1950s and 1960s. While many, probably the great majority, were not secret programs, the GAO found that typically no information can now be located about the consent practices that were followed. Today, the Department of Energy sponsors a program under which those with documented exposures to certain radioactive elements may donate their tissues for research. The operations of the transuranium and uranium registries are subject to review by an institutional review board, and donors must sign a consent form.[120]

In sum, during the 1950s the AEC promoted human tissue sampling for studies on fallout and other research, and its efforts involved secrecy and deception. The AEC evidently considered the legal aspects of "body snatching," but there is no evidence that it sought to consider any independent ethical requirements for disclosure to the families of the subjects (or the subjects themselves, where alive) whose tissue was sampled. While further rationale for keeping the data gathering secret may have existed, in surviving documents concern for public relations emerges as the dominant motivation. At the same time, the AEC recognized that secrecy hampered the conduct of research that it believed central to the public interest.

Secrecy, Public Opinion, and Credibility

On reviewing the transcript of the 1955 Biophysics Conference in 1995, Dr. Merril Eisenbud, a former AEC official who participated in the session, expressed surprise that the document had been classified in the first place.[121] There was, he observed, nothing that merited national security classification; if anything, perhaps it merited the category of Official Use Only, which instructs officials not to publicize the document but is not a category in the formal classification system.[122] As in the case of the AEC's 1947-1948 decision to keep experimental data secret, however, information on fallout data gathering appears to have been classified out of concern that public opinion (in the United States, but also elsewhere) might imperil U.S. weapons development programs.

In November 1954 AEC officials met for lunch with Secretary of Defense Charles Wilson, the signator of the 1953 memorandum discussed in chapter 1, to discuss civilian evacuation in case of atomic warfare and the related question of what the public should be told about fallout. "Secretary Wilson," an AEC record of the meeting summarizes,

stressed the importance of not arousing public anxiety in this country or abroad by public official discussions of the dangers of atomic warfare, particularly with reference to fall-out. He expressed the view that much too much had already been said publicly about fallout, and he urged that before the Government reveals the full extent of the dangers to be expected the Government work out the answers to a lot of questions as to what our citizens could do in the event of an atomic blitz.[123]

"Obviously," records a history of the AEC by AEC and DOE historians, "estimates of the biological effects of fallout on large human populations were more likely to arouse fear and controversy than were small-scale experiments on laboratory animals. Thus, it was not surprising that initial studies of large-scale effects were highly classified and unknown to the public.[124]

Within a very short period, however, much of the secret research was disclosed, but under circumstances where, as the AEC itself came to recognize, its credibility as an information source was seriously impaired.

The Marshall Islands disaster, and the attendant controversy related to the irradiation of the crew of a Japanese fishing boat in the area, marked the beginning of a worldwide debate on fallout that would end with a ban on atmospheric testing.[125] Following this event, ban-the-bomb protests began in Britain.[126] Two years later, in 1956, presidential candidate Adlai Stevenson called for an end to nuclear testing. Soon thereafter, the closely held fallout research began to become public. In October, Libby, addressing the American Association for the Advancement of Science at the dedication of its new headquarters in Washington, reported that the amounts of strontium 90 entering the bodies of children were well below the maximum permissible concentration.[127] In February 1957 Dr. Kulp and his associates presented the results of their study of 1,500 human bones from around the world. The report made the front page of the New York Times.[128] In June, the National Academy of Sciences issued a report noting that strontium 90 and genetic effects were two potentially long-term hazards from nuclear testing.

The public fallout debate was on, pitting scientists against one another. "Test ban advocates," a historian of the fallout controversy recounted, "always stressed the great potential hazard from fallout over a long period of time; their opponents minimized the danger by pointing to similar or greater risks that people routinely accepted, such as luminous wristwatches and medical X-rays."[129]

In May and June 1957, Congress's Joint Committee on Atomic Energy held its first public hearings on the dangers of fallout. The initial 1953 Sunshine report, "Worldwide Effects of Atomic Weapons--Project Sunshine," was apparently declassifed on May 25, two days before the hearings began.[130] Most of the debate focused on the dangers of strontium 90. In June, Commissioner Libby responded to a proposal from an NIH official for the use of children's milk teeth to measure strontium 90. The idea was good, but he advised (in the immediate aftermath of the first highly publicized hearings on fallout), "I would not encourage publicity in connection with the program. We have found that in collecting human samples publicity is not particularly helpful."[131]

In October 1958, a moratorium on nuclear testing began, and in May 1959 the Joint Committee on Atomic Energy held a second series of hearings on fallout. The hearings concluded that the risk was worth the returns to national security; but the public debate continued.[132]

As AEC documents on the fallout debate have become available in the intervening years, it has become clear that the government's effort to manage public opinion was rooted in a sensitivity to its importance. For example, in 1953, following the spring Nevada test series, ranchers in Utah began to report the deaths of their sheep from what, it appeared, might be radiation burns from the tests.[133] The AEC convened a panel to consider the continuation of testing at the Nevada Test Site. The panel concluded that continued testing was justified by the national interest, although risks were inevitable.[134] The tests to date were relatively safe, but there were serious problems with "public reaction."[135] The panel found that "a sufficient degree of . . . public acceptance has not been achieved."[136] Radiation remained a "mysterious threat."[137] But the government had surrounded the program with an aura of secrecy, its own statements were not clear, and statements by former AEC experts or officials had caused "near-panic concern."[138] The public, "which is expected to accept a certain degree of hazard, has not been adequately informed of the extent and nature of the hazard."[139] An extensive program of public education was called for.[140]

The AEC study found the problem was not only with the public; there was a "lack of agreement and acceptance, first, within AEC and test management, and second, among health, medical, and other scientific individuals and groups." The problem was exacerbated by "lack of knowledge of this new subject, by lack of definition, by the extreme sensitivity of the subject, and by the resulting nervousness of the various levels of management."[141] As shown by the secrecy surrounding the ongoing Project Sunshine, however, the public was not let in on the uncertainty of knowledge or on the steps being taken to answer questions of admitted import to all citizens.

AEC insiders recognized that credibility was a problem. In a December 1954 letter to DBM's director, Charles Dunham, Los Alamos Health Division Leader Thomas Shipman touted the importance of Sunshine and suggested a possible role for Los Alamos. At the same time he lamented the lack of credibility possessed by those too closely associated with the AEC:

There is also the fact that Los Alamos may be regarded as a rather biased institution. Some people may feel that we are interested parties. I certainly am only too well aware of a resistance, particularly in the Press, to accept pronouncements and conclusions coming out of the AEC itself. Strangely enough, they were quite willing to accept the conclusions of the National Academy of Sciences, completely forgetting that the subcommittees were in very large measure composed of AEC or AEC contractor representatives. They were the same guys wearing different hats.[142]

In the late 1950s the AEC itself came to question whether its data-gathering efforts were serving the purposes of scientific knowledge and public understanding, as had been hoped. Sunshine, internal memos recorded, lacked coordination and clear goals, and the confusion of roles cost credibility. "[T]he primary reason," wrote Hal Hollister, an AEC fallout expert, "the AEC is now in the soup with respect to Congress, the public, and the fallout problem is that all three of these relationships with the public (reporting data scientifically, getting it across to the public, and telling official interpretations of it) have been inextricably mixed up. This has continued to be true after the hearings, and the future promises more of the same."[143]

In 1959 President Dwight Eisenhower acted to take responsibility for radiation safety away from the AEC, placing it in the hands of a new Federal Radiation Council, chaired by the secretary of the Department of Health, Education, and Welfare.

By the mid-1960s the possibility that data gathering could only get the AEC into more trouble became an incentive to "not study at all." In 1965 Dwight Ink, general manager of the AEC, advised against conducting proposed studies on the detrimental effects of nuclear testing partly because of liability concerns: "[P]erformance of the above U.S. Public Health Service studies will pose potential problems to the Commission. The problems are: (a) adverse public reaction; (b) law suits; and (c) jeopardizing the programs at the Nevada Test Site."[144]

In his DOE-sponsored history of the AEC and nuclear testing safety, Barton Hacker, laboratory historian at DOE's Lawrence Livermore Laboratory, concludes that while AEC officials did not doubt that testing could be done safely if precautions were taken, there was divergence about what to tell the public, and reassurance won out over information:

[T]he people in the field, those involved in the test program directly, tended to favor telling the public just what the risk was and stressing that whatever risk testing might impose was far outweighed by the national importance of the test program. Openness, they argued, would retain public trust and ensure continued testing.[145]


AEC officials in general, headquarters staff members in particular, mostly preferred to reassure rather than inform. Convinced that trying to explain risks so small would simply confuse people and might cause panic, they feared jeapordizing the testing vital to American security. Their policy prevailed. A formal public relations plan became as much a part of every test as the technical operations plan. Carefully crafted press releases never to my knowledge lied, though they sometimes erred. Yet, by the same token, they rarely if ever revealed all. Choices about which truths to tell, which to omit, could routinely veil the larger implications of a situation. . . .[146]

"Reluctance to acknowledge any risk, the policy that mainly prevailed in the 1950s," Hacker concluded, "undercut the AEC's credibility when the public learned from other sources that fallout might be hazardous."[147]

back table of contents forward