SCIENCE VS. SCIENTISTS
If there were an annual international prize given for the most interesting peer-reviewed science journal (periodicals that publish detailed, rigorous reports of research within the scientific community - not articles for nonscientists), Health Physics, which is the official journal of
the Health Physics Society (published by Williams and Wilkins 351 West Camden Street, Baltimore, MD 21201-2436), would be, in my opinion, a certain winner - even though the vast majority of its articles are likely to put even the most enthusiastic of young physicists to sleep. Month after month, this journal inches forward, filled to the brim with research papers documenting the most accurate and detailed experiments possible with modern technology concerning radiation protection and exposure. Some of these articles describe rare cases of accidental high radiation doses and resultant health effects, but the vast majority deal with the distribution and detection of very small amounts of radioisotopes and radiation. Since instrumentation for these purposes has become extraordinarily accurate and the underlying science is relatively simple, research results reported in Health Physics tend to be detailed, exact, and complete with very rigorous mathematical analysis.
The underlying assumption supporting most of this work is that radioisotopes and radiation are very dangerous in even the tiniest quantities and must, therefore, be rigorously tracked down and quantitatively measured wherever they are found. It is evident that nearly all of these scientists blindly accept this assumption, which has, after all, been accepted dogma in their profession since the 1950s - although most of their articles do not discuss this. They just report their measurements.
The editors of Health Physics are apparently excellent classical scientists who are willing to publish any submitted paper that survives peer review of its experimental and theoretical rigor and its relevance to the journal's purposes. [You thought, perhaps, that this is the way all scientific research journals are operated? Welcome to the age of political correctness. Even the prestigious Proceedings of the United States National Academy of Sciences has formed a special medical censorship committee, which prevents publication of any paper with medical implications that are contrary to current medical dogma - regardless of the quality of the research or the importance of the conclusions.] It is this willingness of the editors of Health Physics to adhere to unbiased fairness in review that is causing a most startling form of chaos to break out within their publication. While the steady march of apparently endless radiation measurement articles continues apace like tin soldiers from a toy factory, interleaved among them are articles reporting research on health effects of low radiation doses - results that completely invalidate the underlying assumption of radiation dangers.
In two recent issues, for example, are "Setting Standards for Radiation Protection: The Process Appraised'' by H. Wade Patterson, pp 450-457, and an exchange of letters by D. J. Strom and B. Cohen, pp 488-490, Health Physics 72, No. 3, March 1997, and "Questionnaire Study of the Lung Cancer Risk from Radon in Homes'' by B. Cohen, pp 615-622, and "Problems in the Radon vs. Lung Cancer Test of the Linear No-Threshold Theory and a Procedure for Resolving Them'' by B. Cohen, pp 623-628, Health Physics 72, No. 4, April 1997.
The Patterson article points out that the no-threshold linear hypothesis originated during the 1950s to estimate the health effects of radioactive fallout from nuclear weapons tests. This hypothesis had two advantages. First, it was simple, since it involved only a linear extrapolation to zero from health effects at very high doses. Second, it gave results that were politically desirable to those who wished to ban atomic testing. Citing this hypothesis as fact, they claimed world-wide damage to human health from tiny increases in background radiation.
People who certainly were good enough scientists to know better, yet determined to ban testing regardless of the truth, claimed to the public that this hypothesis was established scientific fact - the most infamous example being that of Linus Pauling in his debates with Ed-ward Teller. Teller was working to preserve American nuclear defenses against Soviet technology. From this activity Pauling wangled a Nobel Peace Prize and a Lenin Prize, the latter being well-deserved.
The primary disadvantage of the no-threshold linear hypothesis of radiation damage to health is that there is not a shred of experimental evidence to support it. The predicted health effects are so low that such experiments were exceptionally difficult to carry out. Politically, of course, proponents multiplied the tiny hypothetical effects by the total number of people on the earth and then cried crocodile tears over the calculated human suffering - none of which was ever observed.
Well, time moves on. Pauling is dead, and Edward Teller is revered both personally and as a proxy for the many patriotic scientists whose work preserved American freedom during the Cold War - and kept Western technology far enough ahead of the Soviets that the use of nuclear weapons has, so far, been deterred. In addition, experimental methods have improved. It is now possible to test the no-threshold linear hypothesis. This hypothesis has spectacularly failed all experimental tests. In every instance, low-level radiation has either been found to have no health effects within the experimental errors of the research, or has been found to have beneficial effects - radiation hormesis.
Access to Energy has discussed some of these experiments - most recently in the February 1997 issue. Figures 1 and 2 show additional experiments used as illustrations in the Patterson Health Physics paper.
Figure 1, adapted from "Radium in Man'' by R. D. Evans, Health Physics 27, p 504 (1974), gives tumor incidence versus radiation dose to the bone. Figure 2, adapted from "Further Observations on Environmental Radiation and Cancer in India'' by K. S. Nambi, S. D. So-man, and S. D. Further, Health Physics 59, p 543 (1990), gives cancer incidences as a function of environmental doses.
For a detailed analysis of this work, see the published papers. Any ten-year-old (at least a home-schooled ten-year-old) can see, however, that linear extrapolation to zero dose would be ridiculous for the data in Figure 1 and that the line best fitted to the data in Figure 2 is that labeled "Hormesis Hypothesis.'' Professor Bernard Cohen (who will be speaking at our San Diego meeting in June) has become the worst nemesis of the no-threshold linear hypothesis with his extensive studies of lung cancer in the United States as a function of radon levels. First, as we have described earlier, he has gathered data on 1,729 counties containing 90% of the United States population and including nearly a million deaths from lung cancer and 20 million deaths from other causes, along with mean radon levels for the homes in those counties (see Access to Energy, February 1997 and earlier articles). This work shows an inverse correlation between lung cancer and radon in the low dose levels - the higher the radon, the lower the lung cancer. Since average radon values are used (but specific radon values are not known for each lung cancer victim), these experiments rigorously disprove the no-threshold linear hypothesis but do not rigorously prove the hormesis hypothesis, so Co-hen has been cautious in claiming lung cancer protection by radon.
Considered with data from numerous other experiments, however, his results provide strong support for radiation hormesis.
Now Cohen has driven a new, unique nail into the coffin of the anti-radiation industry. In the first of his two articles in the April 1997 issue of Health Physics, Cohen reports the results of 266,000 individual radon measurements in American homes, which include 4,438 homes where an occupant died of lung cancer and 7,756 homes where a death occurred from some other type of cancer. Here deaths are individually paired with individual home radon levels. He treats basement radon levels separately from living area levels, smokers separately from non-smokers, and both the cancer-free homes and the homes with cancer other than lung cancer as separate control groups. The cancer victims had been living in their houses for an average of 19 years. Lung cancer is estimated to have a five-year latency period between induction and symptoms.
Although the statistical significance of his calculations varies, as expected, with sample sizes, in every instance there is unequivocal rejection of the no-threshold linear hypothesis. Figure 3 shows the lung cancer result when all 266,000 measurements are included - lung cancer and control. The linear regression curves shown are curved by the logarithmic axis (ask your home schooled twelve-year-old to explain this). The line labeled "theory'' (curving toward the top with no data near it) is that predicted by the no-threshold linear hypothesis, and the line labeled "best fit'' is the line resulting from a best mathematical fit of the actual data: "+3SD'' is the 99% confidence level of the theory line, while +1SD is the 67% confidence level of the best fit line.
Sample sizes were too small at the hormetic radon dose levels to prove beneficial effects from radiation, but most of Cohen's calculations of this work do show deviation of the measured averages in the hormetic region that are consistent with radiation hormesis.
Meanwhile, what is happening in Health Physics? (This is a very reputable peer-reviewed journal in a substantially exact science.) Well, the articles on exact measurements of tiny radiation exposures keep marching out of cookie-cutter laboratories, and the articles invalidating their underlying significance are being published right alongside them. This has led to another phenomenon that I have never before observed in a mainstream research journal.
The good guys are beginning to augment their articles with quotations about the scientific method. This is the equivalent of mathematics journals carrying parenthetical reminders that we must not forget that two plus two equals four. Both Patterson and Cohen have taken to quoting Richard Feynman. From Patterson we have: " 'As Richard Feynman said, 'In general we look for a new law by the following process. First we guess it. Then we compute the consequences of the guess to see what would be implied if this law we guessed is right. Then we compare the result of the computation with nature, with experiment or experience, compare it directly with observation, to see if it works. If it disagrees with experiment it is wrong. In that simple statement is the key to science.
" 'It does not make any difference how beautiful your guess is. It does not make any difference how smart you are, who made the guess, or what his name is - if it disagrees with experiment it is wrong. That is all there is to it.
" 'Another thing I must point out is that you cannot prove a vague theory wrong. If the guess you make is poorly expressed and rather vague, and the method you use for figuring out the consequences is a little vague - you are not sure, and you say, "I think everything's all right because it's all due to so and so, and such and such do this and that more or less, and I can sort of explain how this works . . .'', then you see that this theory is good because it cannot be proved wrong! Also if the process of computing the consequences is indefinite, then with a little skill any experimental results can be made to look like the expected consequences. ' '' (Feynman 1965) (This last Feynman paragraph is beautifully applicable to the global warming and ozone scare industries.) In the second April 1997 Health Physics paper, Cohen computes the discrepancy between the no-threshold linear hypothesis and his experimental data from nearly one million cancer deaths and twenty million deaths from other causes to be "20 standard deviations, which has a probability of occurring by chance equal to the inverse of the total number of electrons plus nuclei in the entire [known] universe.''
The no-threshold linear hypothesis is as dead as the hypothesis that the earth is flat, and with it has died the underpinnings of the anti-nuclear energy industry - which spawned, through similarly devious science, most of today's "environmentalist'' movement. Their only last hopes are the claims that nuclear plants might explode (like nuclear weapons) or burn like Soviet plants (with no containment buildings). The explosion claim is inconsistent with basic physical laws - it just cannot happen - and nuclear plants without containment buildings could only be built by communists who, as Petr Beckmann used to say, decided that people were cheaper than concrete.
All we need do with nuclear waste is dilute it to a low radiation level and sprinkle it over the ocean - or even over America after hor-mesis is better understood and verified with respect to more diseases. We leave it as an exercise for the reader to guess how long it will take the failure of the no-threshold hypothesis to reach your television set.