In his 1976 book, A Scientist at the White House, George Kistiakowsky, President Eisenhower's Science Advisor, told what he wrote in his diary in 1960 on being exposed to the idea by the Federal Radiation Council:
It is a rather appalling document which takes 140 pages to state the simple fact that since we know virtually nothing about the dangers of low-intensity radiation, we might as well agree that the average population dose from man-made radiation should be no greater than that which the population already receives from natural causes; and that any individual in that population shouldn't be exposed to more than three times that amount, the latter figure being, of course, totally arbitrary.
Later in the book, Kistiakowsky, who was a nuclear expert and veteran of the Manhattan Project, wrote: "...A linear relation between dose and effect... I still believe is entirely unnecessary for the definition of the current radiation guidelines, since they are pulled out of thin air without any knowledge on which to base them."
Sixty-three years of research on radiation effects have gone by, and Kistiakowsky's critique still holds. The Linear No-Threshold (LNT) Radiation Dose Hypothesis, which surreally influences every regulation and public fear about nuclear power, is based on no knowledge whatever.
At stake is the hundreds of billions spent on meaningless levels of "safety" around nuclear power plants and waste storage, the projected costs of next-generation nuclear plant designs to reduce greenhouse gases worldwide, and the extremely harmful episodes of public panic that accompany rare radiation-release events like Fukushima and Chernobyl. (No birth defects whatever were caused by Chernobyl, but fear of them led to 100,000 panic abortions in the Soviet Union and Europe. What people remember about Fukushima is that nuclear opponents predicted that hundreds or thousands would die or become ill from the radiation. In fact nobody died, nobody became ill, and nobody is expected to.)
The "Linear" part of the LNT is true and well documented. Based on long-term studies of survivors of the atomic bombs in Japan and of nuclear industry workers, the incidence of eventual cancer increases with increasing exposure to radiation at levels above 100 millisieverts/year. The effect is linear. Below 100 millisieverts/year, however, no increased cancer incidence has been detected, either because it doesn't exist or because the numbers are so low that any signal gets lost in the epidemiological noise.
We all die. Nearly a half of us die of cancer (38% of females, 45% of males). If the "No-Threshold" part of the LNT is taken seriously, and an exposed population experiences as much as a 0.5% increase in cancer risk, it simply can not be detected. The LNT operates on the unprovable assumption that the cancer deaths exist, even if the increase is too small to detect, and that therefore "no level of radiation is safe" and every extra millisievert is a public health hazard.
Some evidence against the "No-Threshold" hypothesis draws on studies of background radiation. In the US we are all exposed to 6.2 millisieverts a year on average, but if varies regionally. New England has lower background radiation, Colorado is much higher, yet cancer rates in New England are higher than in Colorado—an inverse effect. Some places in the world such as Ramsar, Iran, have a tenfold higher background radiation, but no higher cancer rates have been discovered there. These results suggest that there is indeed a Threshold below which radiation is not harmful.
Furthermore, recent research at the cell level shows a number of mechanisms for repair of damaged DNA and for ejection of damaged cells up to significant radiation levels. This is not surprising given that life evolved amid high radiation and other threats to DNA. The DNA repair mechanisms that have existed in yeast for 800 million years are also present in humans.
The actual threat of low-dose radiation to humans is so low that the LNT hypothesis can neither be proven true nor proven false, yet it continues to dominate and misguide policies concerning radiation exposure, making them grotesquely conservative and expensive. Once the LNT is explicitly discarded, we can move on to regulations that reflect only discernible, measurable medical effects, and that respond mainly to the much larger considerations of whole-system benefits and harms.
The most crucial decisions about nuclear power are at the category level of world urban prosperity and climate change, not imaginary cancers per millisievert.