WHEN SHOULD WE INVOKE ULTERIOR MOTIVES IN A SCIENTIFIC CONTROVERSY?
Edward Calabrese brought a charge against H J Muller [1890-1967] reviving a Cold War claim that exposure to radiation in low doses is either harmless or good for the population exposed to it. A reporter for the Chronicle of Education did an article on it and tried to be balanced in his coverage of the charge by Calabrese that his field of hormesis rules out harmful effects of any low doses of radiations, chemical mutagens, carcinogens, additives, pollutants, and wastes whether in our drinking war, foods, our workplace, or the air we breathe. He claims such noxious exposures stimulate the immune system and give us resistance to these agents or they compensate for the exposure in some unknown way. To me that reeks of self-interest and wishful thinking because the weight of scientific evidence favors, in the peer reviewed published record, harm done to organisms given attenuated doses of radiation. Small exposures to radium by watchmakers in factories during the 1920s resulted in bone cancers to the women who worked in these assembly lines. Physicians and dentists, in the days before there were dosimeters, used to invoke a threshold dose below which radiation did no harm, The test: put your hand under the x-ray machine and if reddening of the skin occurs you have reached the threshold dose. Please don’t try that! There is no dispute by Calabrese and Cold War critics of Muller that high doses of exposure are harmful. How could one deny Hiroshima and Nagasaki or those exposed during accidents to massive doses of radiation? Nor do they deny that it is linear from roughly 100 R to 12,000 R in fruit flies and other systems where such doses could be applied to a set of organisms and sufficient numbers of their progeny be counted for a controlled experiment. But is radiation sickness at high doses the only harm done by exposure to ionizing radiation? What is difficult to do for fruit flies or mice are experiments involving lower doses of 25 R. Why are such experiments difficult? The lower the dose the larger the number of flies or mice you have to involve in an experiment at low doses because those doses may only increase the incidence of mutation by 2 or 3 fold. If the spontaneous rate in a fruit fly experiment is 1 X-linked lethal per two thousand sperm, how many individuals would you use for the treated and controlled? The statistics for small numbers would argue a range of 1-12 per 2000 would be a control rate in any one given experiment. So you would need to examine a much larger number of flies to show more mutations occurred in the low dose and these two results would be non-overlapping. Let us say you did such an experiment and in the controls you sampled 2000 and got 3 mutations and in the exposed (let us say 25 R) you got 12 mutations. You would think, if you didn’t know the statistical issues involved, that 25 R produced those extra mutations. But let us say you repeated the experiment and this time you got 8 controls mutations and 2 at the 25 R range. You would either say radiation did not induce mutations at that dose or you would say radiation is good for you at low doses because you get less mutations by being exposed. Actually in both these experiments you get inclusive results because you haven’t examined a sufficiently large number of vials of flies (each vial represents one exposed or one control sperm). When you have to use tens of thousands of vials you run into a lot of problems because that is as very big experiment. The larger such experiments are, the more variables you introduce and unless the effects are dramatic statistically, it is hard to know how carefully these alternative factors played a role (light, temperature, food batch, being in a tray under other trays or being in the top of several layers of trays in one or two rooms with different conditions. This why attenuated doses are more frequently used to test the presence or absence of threshold effects. The best known was done by Muller’s student in Edinburgh, S P Ray-Chaudhury in 1939. His thesis wasn’t published until 1944 because of the war. Later he went back to India and repeated his technique of attenuated doses once for measuring chromosomes breaks (which he found also to be linear) and once to look again at gene mutations.
In science there is a tradition of challenging the work of one’s peers and repeating the work of one’s peers. Out of this healthy debate emerges more careful experiments and more confidence in the findings found in a first report of something new. For most of tens of thousands of such scientific findings the disputes never leave the professional journals. The issues are strictly scientific. The persons who decide the validity and care of an experiment are other scientists in the field. To get published there is peer review. No one calls in a scientist from another field or a reporter to evaluate the claims of the accuracy of the two experimental groups claiming contradictory findings. Such things happen when the nature of the findings have serious human implications. We have controversy over agents that cause embryo damage (e.g., the thalidomide controversy), over stem cell research (a form of identical twinning from embryonic cells before they have implanted into a uterus), over potential carcinogens (like the butter yellow used to color margarines in the 1940s when the dairy industry forced margarine companies to put a pellet of butter yellow for the purchaser to knead into the white lard-like fat so that customers would not think they are buying real butter). Butter yellow was a liver carcinogen and pulled off the market by our FDA regulatory commission. Frequently in these controversies there are vested interests who favor a view that what they are doing, what they are using, and what they are selling is perfectly safe and that the claims of mutations, cancers, or toxicity are either bad science or politically inspired, or ideological by irrational environmentalists or health nuts. Also in these debates are scientifically not well informed environmentalists, health nuts, and others who without knowledge of the science involved believe everything that is not organic, natural, or familiar should be treated as potentially damaging to the public. It works both ways to our disadvantage because most people involved in the debate are not the scientists involved in the appropriate fields to do so.
I am frustrated when a reporter calls and asks for my “side” of a debate. How do I spend an hour or more trying to show all the variables involved just in the science involved? Should I mention that the accuser of a position I hold is funded by agencies or industries that favor a view that what they do is without harm and no government regulations are needed in an industry which knows best what is harmful and what is harmless? Of course it is irrelevant to the real issue of whether, in this case, low doses are harmful, beneficial, or inactive with respect to the mutation process. But is the scientist on the side of that position a geneticist? No. Should I respect his judgment nevertheless when he has not immersed himself in the field the way Muller did for his entire adult life? Do thousands of hours doing genetics get neutralized by someone doing virtually no genetics and coming in a field of toxicology from a department of entomology for his PhD? Does it matter? Whose job is it to look into this? Is it mine as a geneticist or is it a reporter’s with very little, if any experience in doctorate level genetic experimentation? This forces reporters to look more at motivation, conflict of interest, ideology, scientific method (self-deception, constructed realities, unexamined bias) rather than at the detailed scientific experiments and how they were done and what was inadequate in their experimental design. When science gets politicized by this type of publicized controversy it frustrates everyone involved. I wish I knew a remedy for this. I don’t.