11

The Battle for Publication



EARLY IN JANUARY of 1969, the article on fetal and infant mortality was returned by the editor of Science. This was the paper that had been discredited by Abelson in his phone conversation with the reporter Stuart Brown, before the paper had arrived in the offices of Science. Copies of three reviews were enclosed. Two were clearly written by individuals in the field of public health and statistics; these merely contained suggestions for certain changes that might make the case more complete. The third reviewer, however, was totally negative. This individual went into a detailed analysis of fallout and dose levels in the U.S., making eight references to internal AEC reports. The majority of these reports had been prepared by the staff of the AEC's Health and Safety Laboratory in New York, directed by John Harley, the man responsible for the classified fallout measurement at Troy.

The argument used to discredit the paper in this review (and also in an article published by Harley in the Quarterly Bulletin of the AEC's Health and Safety Laboratory later that year) was this: According to the detailed measurements made by the AEC's laboratory, the highest levels of strontium 90 had actually been in Utah, Montana, Wyoming, South Dakota, and Nebraska, as well as in very small portions of Massachusetts and Rhode Island. On the other hand, the lowest levels were measured in the southern U.S., from southern California to Florida. This was just the opposite of what I had said in my paper. Harley therefore argued that since infant mortality had risen most in the southern United States to the east of Nevada and New Mexico, and least in the low-rainfall mountain states of the Southwest, it clearly could not be the fallout that was responsible.

This was indeed a devastating argument, supported by a vast set of detailed measurements. Yet these measurements cited by Harley were in total disagreement with the measurements reported by the Public Health Service, which I had used in my paper. According to the Public Health Service, year after year the "wet" southern and eastern parts of the U.S. showed levels of strontium in the milk two, three, or even four times as high as in the dry western mountain states. The mystery was resolved a few months later when Dr. E. A. Martel, a former U.S. Air Force fallout specialist, told me the story of the gummed films.

The technique that had been used by Harley's lab to measure the fallout involved the use of a sheet of plastic about a foot square, coated on one side with a sticky substance very much like that used on flypaper. These plastic squares were mounted on a stand with the gummed side facing upward so as to catch the fallout particles as they descended. Every few days the films were collected, and shipped to the laboratory, where the radioactivity was measured.

In a detailed study later carried out by scientists at the Battelle Memorial Institute in Ohio, it was discovered that in the dry states of the west, the winds constantly picked up the radioactive dust again and again, so that the exposed gummed films, acting just like the flypaper in a room full of flies, ended up collecting much more radioactivity than was typical for the soil of the area. On the other hand, in the high-rainfall areas east of the Mississippi, the rains soaked the fallout deep into the soil and kept the dust levels low. Thus, by the early 1960s, it was widely realized that the so-called "gummed film" measurements of fallout had given levels much too high for the dry mountain states, and too low for the East and South. In fact, in the 1962 United Nations Report on Atomic Radiation, of which John Harley was one of the authors, there was a note from Harley on page 225 indicating that the "gummed film" procedure "may lead to an overestimate of the tropospheric fallout," the troposphere being the lower part of the atmosphere containing the clouds of rain and radioactive dust.

Yet seven years later Harley used the "gummed film" measurements in his attempt to discredit my correlation of nuclear testing with the rises in infant mortality, writing that "fallout before 1954 was exactly the opposite of what was stated by Dr. Sternglass."

A few days after the paper on infant mortality had been returned by Science, the paper on the leukemia rise in the Troy area was also returned with a note of rejection. It was the same story again. Two of the three enclosed reviews were clearly by public-health physicians and statisticians and were quite favorable. One of these reviewers, in fact, stated that "the comparability between past fallout and past irradiated cases is `impressive,'" while the other reviewer began with the statement: "The conclusion of this paper, if correct, is clearly a most important one." The third reviewer, however, was completely negative, and, exactly as in the case of the paper on infant mortality, there could be no doubt to which organization this third reviewer belonged. Almost word for word and point for point, the third review resembled a critique that had been sent to me a few months earlier by John Conway, Chief Counsel for the Joint Committee of Atomic Energy. And this critique had been sent to him by the AEC's Division of Biology and Medicine.

Perhaps the most remarkable point made by the third reviewer, in view of my experience with the Health Department of the State of New York, was the following: After arguing that the data on Albany-Troy were "incomplete," the reviewer asserted that "Sternglass could obtain the missing data."

And then, quite unexpectedly, less than a week after the two papers had been returned by Science, the following letter arrived:

Bulletin of the Atomic Scientists
A Journal of Science and Public Affairs
Eugene Rabinowitch, Editor

Dear Dr. Sternglass:

The drawings which we have for some of the figures in your article on Infant and Fetal Mortality Increase in the U.S. are not dark enough to be printed. Could you send us the original drawings -- or very clear, dark copies -- for figures 1 and 3.

Thank you for your assistance.

Sincerely yours,

Merry Selk
Editorial Assistant

I gradually realized that the Bulletin must have decided to publish the report I had sent to Dr. Rabinowitch merely for his information.

As I later learned from the managing editor, Richard S. Lewis, in the face of strong reviewers' opinions both pro and con, it had been decided that the grave issues raised by my findings should be publicized and discussed as widely as possible, both by the scientific community and the general public. This was indeed good news. For although Science had a far wider circulation, the Bulletin reached an important group of physical and political scientists in university and government circles around the world. Furthermore, like Science, it was carefully read behind the Iron Curtain, hundreds of copies of each issue being sent to the Soviet Union, Poland, East Germany, Czechoslovakia, and Rumania.

Subsequently, I presented all the new evidence before a meeting of the National Council of the Federation of American Scientists, and the Council voted to set up a special committee to look into the evidence in detail. Dr. John T. Edsall, a noted biologist at Harvard University, agreed to head up the study committee, and after many months of investigations during which he consulted a number of specialists, he indicated in a letter that he would personally be willing to urge the editor of Science to reconsider his decision not to publish my findings.

Meanwhile, at the suggestion of a mutual acquaintance, I sent copies of my data to Dr. Luis Alvarez, then president of the Physical Society, who was now heading a physics research group in Berkeley, California. I included the maps showing increases in infant mortality downwind from the Trinity test site in New Mexico, for it had been Alvarez who watched the mushroom cloud from the first atomic explosion drift off across the United States. Thus, he was one of the few individuals in the world who had firsthand knowledge of the way the cloud had broken up and the directions in which it had drifted.

In the first paragraph of the reply I soon received from him, he stated that he had found the evidence "very impressive, particularly the map of the United States with the percent excess in mortality showing an effect only downwind of the Trinity site." He added that "in view of the enormous statistical significance of the results you plot on your map of the United States it is difficult to question your findings."

My article appeared in the April 1969 issue of the Bulletin. Interestingly, the managing editor, Richard Lewis, later told me that pressure had come both before and after publication in the form of long-distance calls from Washington from individuals who claimed to be long-term government friends of the journal. They said it was a grave mistake for the Bulletin to publish my article. When Lewis asked their names, they refused to identify themselves.

The April issue also carried an article by Dr. Freeman J. Dyson, a theoretical physicist at the Institute for Advanced Studies in Princeton, entitled "A Case for Missile Defense." It was clear that the basic premise of Dyson's argument in favor of an antiballistic missile (ABM) system would not be valid if my conclusions on the vulnerability of the developing infant to radiation were correct. His basic assumption was that a defensive system, once installed, regardless of how really effective it might be, would force an attacker to concentrate many of his missiles on a few defended cities, thereby reducing the number of cities that could be attacked with a given number of missiles and saving those cities that could not be attacked. But in the process, the "saved" cities would be inundated with intensive fallout.

I wrote a short note in rebuttal to Dyson's article, hoping that it would be published in the Bulletin. A few weeks later, a letter arrived from the Bulletin containing galley proofs of my letter and a reply by Dyson that began as follows:

I welcome this chance to call attention to Ernest Sternglass' article "Infant Mortality and Nuclear Tests" in the April Bulletin. I urge everyone to read it. Compared with the issues Sternglass has raised, my arguments about missile defense are quite insignificant.

Sternglass displays evidence that the effect of fallout in killing babies is about a hundred times greater than has been generally supposed. The evidence is not sufficient to prove Sternglass is right. The essential point is that Sternglass may be right. The margin of uncertainty in the effects of worldwide fallout is so large that we have no justification for dismissing Sternglass' numbers as fantastic.

If Sternglass' numbers are right, as I believe they well may be, then he has a good argument against missile defense. . . .

Thus it appeared that once the evidence on the dangers of worldwide fallout was allowed to reach the scientific community at large, responsible scientists would be willing to reconsider their past judgments.

Next | ToC | Prev