Radiation risks are real. But no cause for radiophobia.

Editor’s note: This story was originally published by Undark. It appears here as part of the Climate Desk collaboration.

For Andrew Maidment, the chief of physics for the Department of Radiology at the University of Pennsylvania, it was one of the oddest phone calls he’d ever received. “A woman called me up once,” he remembers, “and I hear this child screaming in the background, yelling and crying, crying. She starts off with, can I hug my son again?”

Part of Maidment’s job is consulting with patients and occasionally answering strange questions, but this was a new one: “She explains that her son had broken his leg three weeks earlier, and he’d had an X-ray for his leg. And was he still radioactive? She’s so afraid of radiation she won’t hug her kid.” Maidment reassured the woman that any radiation her son had received vanished as soon as the X-ray machine was turned off, and that her child wasn’t radioactive. “She says, ‘so I could have hugged my son right away?’ And I say, ‘yeah.’”

It was an example of a phenomenon that scientists and other experts have dubbed radiophobia — the fear of ionizing radiation. Every horror filmmaker worth her fake blood and monster makeup knows that nothing is more terrifying than the unseen: the creak on the stairs, the shadow on the curtains, the hint of fatal evil.

And radiation — a mysterious presence that can’t be felt by our senses — is just such a terror.

This fear has become the default setting for most people — and plenty of experts argue that’s for good reason. Ionizing radiation strips electrons from atoms and molecules, converting them into ions, or charged particles. And in atoms and molecules comprising living tissue, such as DNA, it can wreak havoc. If the intensity and rate of exposure are too great, the organism sickens and can die. This is called acute radiation syndrome, or ARS, and it killed many people in the aftermath of the atomic bombings of the Japanese cities of Hiroshima and Nagasaki in the 1940s. It also took a toll following the nuclear accident at Chernobyl, which was recently revisited, to critical acclaim, in a haunting HBO miniseries. Fortunately, most living organisms rarely, if ever, encounter radiation at such toxic levels.

Crystallizing events like these have reinforced in the popular mind a technical model that today informs the regulation of nuclear radiation the world over. It’s called the “LNT” model, short for “linear no-threshold,” and it holds that at any level of radiation exposure greater than zero, there is some damage to human DNA. The higher the level, the greater the harm. The “linear” part of the LNT refers to the shape of a line graph plotting cancer risk against radiation dose: not a curve, but a straight line beginning at zero and proceeding upward to infinity. According to the LNT model, all radiation doses, however small, are dangerous, and the harm is cumulative over time. Thus, any exposure must be held to what nuclear regulators would call an “as low as reasonably achievable” dose, or ALARA.

To most scientists, regulators, and the public at large, this only makes sense. And yet, an increasingly vocal group of nuclear experts suggests that the LNT model and its zero-sum approach to radiation may well be doing more harm than good. Naturally occurring radiation surrounds us all the time, they point out — emanating up from the Earth itself, and raining down on us from the cosmos. And outsized fear of radiation, they say — driven largely by the LNT framing — does much more than frighten people like the mother who queried Maidment. It drives voters and governments, for example, to abandon or avoid nuclear power — even though its fossil-based alternatives pollute the air and warm the planet, arguably killing far more people each year. (Indeed, according to the World Health Organization (WHO), 4.2 million people die every year from outdoor air pollution — orders of magnitude more deaths than those attributable to all civilian nuclear accidents, and the bombings of Nagasaki and Hiroshima, combined.)

Similarly, the partial meltdown of three reactors at Japan’s Fukushima Daiichi Nuclear Power Plant in 2011 led to mass evacuations that many experts now say resulted in more deaths than the earthquake and tsunami that caused the accident. “Overestimating radiation risks using the LNT model may be more detrimental than underestimating them,” argued researchers Jeffry Siegel and James Welsh in a 2015 article in the journal Technology in Cancer Research and Treatment, “as this approach has resulted in unnecessary loss of life due to traumatic forced evacuations, suicides, and unneeded abortions after the Fukushima nuclear accident.”

Are they right? Has the no-amount-is-safe approach to radiation — a paradigm that most of us consider so axiomatic that it informs both law and no shortage of fictionalized movie-house nightmares — actually made us less safe? Or, while perhaps technically tenuous, does it nonetheless provide a rational posture, given the potential risks? Jan Beyea, a nuclear physicist and the founder of Consulting in the Public Interest, a firm offering advice and expertise in the biological and physical sciences to nonprofits, universities, law firms, and other groups, suggests those are heated questions indeed. In a May 2012 article in Bulletin of the Atomic Scientists, Beyea wrote: “The debates can be brutal — so much so that, at times, they make the spats between William Jennings Bryan and Clarence Darrow look tame.”

The divisions run deep: “Underlying the [existing] risk models is a large body of epidemiological and radiobiological data,” the Environmental Protection Agency opined in support of LNT in 2011. “In general, results from both lines of research are consistent with a linear, no-threshold dose (LNT) response model.” That assessment was echoed by the Congressionally chartered National Council on Radiation Protection and Measurements (NCRP) in April 2018. Reviewing more than two-dozen recent studies of the effects of low-dose radiation, the group concluded that the results added “substantial weight to the judgment on the use of the [existing] model for radiation protection.”

But Carol Marcus, a radiobiologist and nuclear medicine physician at UCLA who has recently petitioned the U.S. Nuclear Regulatory Commission (NRC) to relax some regulation of low-dose radiation exposures, disagrees. “The LNT,” she declared in her petition, “is based on hogwash.”

Ask most people how many were killed by radiation from the reactor accidents at Chernobyl, Fukushima, or Three Mile Island, and you’d probably get a wide range of answers. The 1986 Chernobyl disaster would no doubt be on top, with estimates ranging from a few thousand to hundreds of thousands. Fukushima, the first nuclear accident to occur during the era of 24-hour cable news and internet coverage, would probably draw estimates anywhere from a hundred up to thousands. And although most of us are aware that a serious disaster never occurred at Three Mile Island in 1979, some would probably guess that at least a few fatalities resulted, with leukemia or thyroid cancer cases manifesting themselves years later.

The actual figures are surprising. Most official sources give a figure of a few dozen deaths caused directly by radiation from the Chernobyl accident, though estimates of projected future deaths from cancer vary widely. That’s partly due to the unreliability and incompleteness of data from the former Soviet Union, as well as the impossibility of distinguishing cancers caused by Chernobyl from those from other causes. Projections range from around 4,000 (from scientific groups affiliated with international agencies such as the WHO and the United Nations) to 100,000 (from environmental organizations like Greenpeace). And while some research suggests that the rate of birth defects increased in areas contaminated by radiation following the meltdown, a lack of data on pregnant women’s actual exposure — along with other confounding factors like alcohol consumption and diet — make it impossible to draw definitive conclusions.

Fukushima is a considerably easier case. At the moment, only one radiation-related death has been reported — a Fukushima plant worker who died of lung cancer first diagnosed in 2016. As for Three Mile Island, there have been no official deaths from radiation.

This chasm between public perception and actual casualty data is a testament to decades of misconceptions, misinformation, and misapprehensions about radiation, reinforced by the relentlessly baleful portrayals in popular culture and the dark shadow of the mushroom cloud. From the earliest post-Hiroshima B-movies, comic books, and pulp novels, the underlying message has been the same: Radiation is bad. It can make humans and other creatures grow or shrink; give them special abilities; or and mutate them into monsters.

Radiation, of course, does none of those things, though it certainly does have devastating effects at high doses. It’s the over-regulation of even low-dose exposures, however, that LNT critics say have done more harm than good — and that have given rise to what they characterize as irrational fears. The major alternative to the LNT model posits the existence of a threshold, or a measurable limit, below which there’s no danger. This reflects the reality that in our daily lives, radiation can never be eliminated, only limited. “Radiation is ubiquitous,” says Chary Rangacharyulu, a physicist at the University of Saskatchewan. Every day, we are bathed in a natural low-level radiation background from various sources: cosmic rays, radon gas, radioactive elements in the Earth, even our own bodies.

Millions of years of evolution have provided earthly life with intricate genetic and cellular mechanisms that, in most cases, can either repair or neutralize damage by causing overly radiated cells to self-destruct. “The basic premise of LNT that any amount of radiation, however small, is harmful, is flawed,” says Rangacharyulu.

Or, as Siegel, Charles W. Pennington, and Bill Sacks wrote in a 2017 article in The Journal of Nuclear Medicine: “The primary LNTH [linear no-threshold hypothesis] fallacy is it excludes this evolutionary biology, ignoring the body’s differing responses to high versus low radiation doses. Low doses stimulate protective responses; high doses overwhelm and inhibit such protections.” In another article, Siegel and Pennington contend that “LNTH-derived policy is as unsafe a practice as shouting fire in a crowded theater.”

If the LNT model is as flawed as these critics argue, how did it become the foundation of our regulatory framework, not to mention our cultural attitudes, regarding radiation? The first strong evidence of a connection between radiation and genetic damage came from experiments in 1926 and 1927 by the American geneticist Hermann Joseph Muller. Muller exposed groups of fruit flies to high doses of X-rays at high rates, reporting what looked to be definitive evidence of mutagenesis, or the production of genetic mutations. He eventually won a Nobel Prize in 1946 for this work, claiming in his Nobel lecture that there was “no escape from the conclusion that there is no threshold dose.”

Further research in the interim since Muller’s early experiments, however, suggested that in fact there was a threshold at low doses or exposure rates. It remains unclear whether Muller merely dismissed these new findings as incorrect, or purposely ignored them. Edward J. Calabrese, a toxicologist at the University of Massachusetts Amherst, is the most vocal proponent of the latter view. He argues that close examination of the historical and scientific record proves that Muller was being intentionally “misleading and deceptive,” and tried to “blunt criticism of [his] Nobel Prize lecture and its impact on his reputation.”

But others have responded that Muller was an extraordinarily conscientious researcher and his unequivocal support of the LNT was based on his honest, if imperfect, appraisal of the scientific evidence of the time. In any event, Muller’s Nobel speech helped cement the scientific acceptance of the LNT model, and the 1956 recommendations of the National Academy of Sciences’ Committee on the Biological Effects of Atomic Radiation (BEAR I) made it official. Subsequent NAS committees combined the work of Muller and others on radiation mutagenesis with preliminary analyses of data for Hiroshima and Nagasaki to extend the application of the LNT to carcinogenesis.

This, the LNT’s opponents argue, was a major mistake. The possibility of huge doses of radiation causing genetic mutations in future generations, which was the core of Muller’s work, is one thing; causing cancer in living organisms with low radiation doses over time is quite different. In other words, a high dose of radiation all at once isn’t the same thing as small amounts over a long time.

Siegel, Pennington, Sacks, and Welsh have pointed out that “the difference between the lowest experimental dose [used by Muller] and zero dose is huge,” arguing that it’s “unacceptable to simply assume, without further observation,” that there’s some lasting damage all the way down to the lowest dose. As far back as the 16th century, it was recognized that while a substance might be harmless or even beneficial in small amounts, there is a threshold above which it becomes harmful. According to the Renaissance physician Paracelsus, known as the father of toxicology, “Poison is in everything, and no thing is without poison. The dosage makes it either a poison or a remedy.”

Nonetheless, as the New York Times science reporter Gina Kolata noted nearly two decades ago, “the idea that radiation’s effects were directly proportional to its dose caught hold and soon was being used to predict cancer cases.”

In radiation-induced cancers, the issue is the definition of “low dose.” Cellular mutations happen constantly in your body, most often caused not by radiation but by oxygen, and are usually handled quite efficiently by natural cellular repair mechanisms. Yet sometimes damaged cells can elude repair and reproduce imperfectly, perhaps becoming cancerous over time. So at what point does that happen? According to the LNT scenario, once a cell is hit by radiation it’s probably damaged for good, and every future dose of radiation just ups the total bodily harm, possibly resulting in cancer. But if in fact a threshold exists, then it’s only when it is exceeded by too much radiation that there’s a problem.

With most toxic substances, the key question is the dosage or amount of exposure. Some may behave differently at lower doses than at higher ones, but a DDT molecule, for example, remains a DDT molecule, whether in isolation or in the company of a million others. The effects of ionizing radiation, however, are complicated by the fact that there’s more than one type, with each varying in the way it inflicts damage. Radiation comes in several varieties: alpha and beta consist of subatomic particles and can be stopped in their tracks by a sheet of paper or a layer of skin, while gamma radiation and X-rays consist of highly energetic and penetrating photons. All have the ability to ionize atoms, sometimes generating further radiation in the process, making things even more complicated.

A bewildering lexicon has proliferated over the past century to describe and quantify radiation, from “roentgen,” “becquerel,” and “rem” (roentgen equivalent man), to “rad,” “gray,” and “sievert.” Some terms are widely used, while others are obsolete or appear only in certain circumstances. Each has its own definition, based on various factors. Currently, the sievert (Sv) is the unit most often used when considering the health effects of radiation dosages. The inescapable natural background radiation we receive from the environment is about 3 to 10 mSv (millisieverts, or a thousandth of a sievert) a year; a chest X-ray gives about 0.1 mSv; a mammogram, about 0.4 mSv; a CT scan 2 mSv or more, depending on the type.

Most experts place the threshold for cell mutation or cancer at roughly 100 mSv, for a dose from all sources above natural background radiation. The Health Physics Society, an organization of radiation safety experts, says in an official position statement that “radiogenic health effects (primarily cancer) have been demonstrated in humans through epidemiological studies only at doses exceeding 50-100 mSv at high-dose rates. Below this, estimation of adverse health effect remains speculative.”

And yet, the Health Physics Society, along with a large number of other experts, suggests that establishing the precise threshold below which we ought not worry is a slippery business indeed — a fact that argues in favor of maintaining the LNT framing. “The Health Physics Society also recognizes the practical advantages of the linear no-threshold hypothesis to the practice of radiation protection,” the group’s statement continued — adding: “Nonetheless, risk assessment at low doses should focus on establishing a range of health outcomes in the dose range of interest and acknowledge the possibility of zero effects.”

The International Commission on Radiological Protection (ICRP), an independent international scientific body, concurred, stating in a 2005 report that while the “existence of a low-dose threshold does not seem unlikely for radiation-related cancers of certain tissues, and cannot be ruled out for all cancers as a group, the evidence as a whole does not favor the existence of a universal threshold, and there seems to be no particular reason to factor the possibility of a threshold into risk calculations for purposes of radiation protection.”

The National Council on Radiation Protection and Measurements (NCRP), an organization of scientists and other nuclear experts that disseminates information and research data about radiation exposure and protection guidelines, agrees. While not affirming that the LNT is necessarily correct scientifically, the group maintained in a 2018 report that “recent epidemiologic studies support the continued use of the LNT model for radiation protection.”

In a response to emailed questions from Undark, NCRP’s director of science John Boice and Roy Shore, an emeritus member of the organization who formerly worked for the Radiation Effects Research Foundation, or RERF (a joint U.S.-Japan organization focused on studying the effects of radiation on atomic bomb survivors) wrote that “protection from any toxic agent needs to err, if anything, on the prudent side to assure protection, because the available risk data are inevitably too limited to give precise risk values. Therefore, in order for the dose threshold model, which discounts low-dose risk, to possibly be prudent, virtually all the epidemiologic and animal data would have to show no excess risk at reasonably low doses.”

The pair added: “The question is what model(s) approximately reflect the knowledge base available (which is constantly increasing) and also are sufficiently conservative so as to protect human health.

“The LNT model,” they wrote, “meets those criteria reasonably well.”

Since 1956, the BEAR (later renamed BEIR, for Biologic Effects of Ionizing Radiation) committees have met several times to reevaluate the current state of scientific knowledge on radiation. The most recent report, BEIR VII, came out in 2006 and it continues to support the LNT, while hedging its bets. “It is unlikely that there is a threshold below which cancers are not induced, but at low doses the number of radiation-induced cancers will be small,” says a summary report that also acknowledges that competing hypotheses exist, including ones holding that “risks are smaller than predicted by the linear no-threshold model.”

Apart from carcinogenesis, the BEIR VII committee also examined the latest results of RERF’s Life Span Study, which tracked tens of thousands of survivors of the Hiroshima and Nagasaki bombings, as well as their children. They found “a lack of significant adverse genetic effects,” also stating that the risk of inheritable radiation-induced mutations “is sufficiently small that it has not been detected in humans, even in thoroughly studied irradiated populations such as those of Hiroshima and Nagasaki.”

Whether or not Muller’s hugely irradiated fruit flies actually did pass down mutations to subsequent generations, the scientific consensus has been that inheritable genetic damage from radiation has not been definitively detected in higher organisms like human beings. A BEIR VIII report is said to be in the works, though it’s uncertain when it will appear.

In the swirl swirl of all this complexity, it’s easy to understand the appeal of a one-size-fits-all theory like the linear no-threshold model. It not only seems reasonable, but is more politically convenient, allowing legislators and regulators to claim they’re doing everything possible to protect people while avoiding criticism that they’re exposing them to any potential risks. The NCRP came to roughly the same conclusion in its meta-analysis last year, stating that use of the LNT model “is in accord with judgments by other national and international scientific committees, based on somewhat older data, that no alternative dose-response relationship appears more pragmatic or prudent for radiation protection purposes than the LNT model.”

So what’s the problem? Critics of LNT argue that “playing it safe” has turned out to be not so safe after all. The side effects of radiophobia manifest themselves not only with large-scale events such as Fukushima, but also with routine doctor-patient interactions. For physicians and other health professionals, it can be a considerable challenge. “People think that the risk is additive, and it’s not,” Maidment said. “Radiation is like buying a lottery ticket. You buy a ticket and you have a chance of winning. And then the next week, if you want to try again, you can buy a ticket and you have a chance of winning. People think that the more they play the lottery, the better their chances are of winning. But it’s not true.”

He uses a CT scan as an example. “Let’s imagine the risk is one in a thousand. So people think that, well, if I have two CT scans, my risk is now one in 500. If I have 10, my risk is now one in 10. But it’s not like that. The first CT scan had a risk of one in a thousand. You either got cancer from that CT scan or you didn’t.”

“The other thing that people never put into this equation is the benefit,” he adds. “Risks should be considered in light of benefits.”

He notes that in the days before CT scans, exploratory surgery was the only means to get a detailed view of many internal injuries and disease conditions. “They’re cutting you open to look inside you to make sure everything was O.K. You could die of anesthesia. You could die of infection. You had a scar forever. Now you stick yourself into this donut for 15 minutes and we have a relatively definitive answer. That’s an incredible benefit for a theoretical risk years in the future. And if you get a cancer 30 years from now, we can detect it and we can probably treat it with medicine that’s 30 years better than it is today.”

The NCRP’s Boice and Shore observe in their emailed replies that “small doses evoke fears that are disproportionately higher than the potential risks from the low doses. Some scientists or health workers of questionable legitimacy stoke those fears by exaggerating the effects of low-level radiation in public forums and writings.”

On the other hand, they add, many radiation scientists recognize that any excess risk of cancer at very low doses would be so small as to be undetectable. “People encounter much larger immediate risks (e.g., from auto accidents) in everyday life,” the pair wrote.

There’s no way to know how many people have suffered or died needlessly as a result of refusing diagnostic imaging tests or radiation therapy, but the details attending events like Fukushima are somewhat easier to evaluate. As a 2018 study on nuclear power from the MIT Energy Initiative remarked, “the evacuation experiences following the Fukushima nuclear accident have shown that current emergency planning and evacuation requirements in all likelihood do not correctly balance the risks of large-scale evacuations with the health risks of low-dose radiation exposure.” Despite the lack of a large number of radiation-induced casualties at Fukushima, more than 1,600 evacuation-related deaths were recorded. “Causes of indirect deaths include physical and mental stress stemming from long stays at shelters, a lack of initial care as a result of hospitals being disabled by the disaster, and suicides,” according to The Japan Times.

The 2015 report on Fukushima by the director general of the International Atomic Energy Agency stated that “no early radiation induced health effects were observed among workers or members of the public that could be attributed to the accident.” The United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) found that “no discernible increased incidence of radiation-related health effects are expected among exposed members of the public and their descendants,” and concluded that the true public health damage of Fukushima was psychological: “The most important health effect is on mental and social well-being, related to the enormous impact of the earthquake, tsunami, and nuclear accident, and the fear and stigma related to the perceived risk of exposure to ionizing radiation. Effects such as depression and post-traumatic stress symptoms have already been reported.”

Such psychological scars can impact lives as surely as physical effects. “Precisely because damage from very low-level radiation cannot be detected, people exposed to it are left in anguished uncertainty,” reporter Andrew Revkin observed in a post on The New York Times website. “Many believe they have been fundamentally contaminated for life. They may refuse to have children for fear of birth defects. They may be shunned by others who fear a sort of mysterious contagion.”

Alarmed by sometimes inaccurate media coverage of Fukushima — a group of scientists and medical professionals created a website and eventually an independent group called Scientists for Accurate Radiation Information (SARI) to counter the LNT model. In 2015, SARI member Carol Marcus formally petitioned the Nuclear Regulatory Commission for changes in the federal radiation protection standards currently based on the LNT model, including abolition of the “as low as reasonably achievable,” or ALARA dosage standard.

The petition argues that thousands of scientific papers show that the LNT is in error, claiming “the attitude of today’s regulators is reminiscent of the Catholic Church at the time of Galileo.” And that, the petition argues, has real-world consequences. “Use of the LNT assumption enables regulators to feel justified in ratcheting down permissible worker and public radiation levels, either through actual dose limits or use of the ‘as low as reasonably achievable’ (ALARA) principle, giving the illusion that they are making everyone safer (and creating ever-increasing workload for themselves and their licensees).” In other words, the LNT theory imposes undue regulatory burden — and costs — on all things involving potential radiation exposure, from medical equipment handlers and nuclear power plant operators to crews working to clean up sites contaminated by nuclear materials.

“The LNT model is more like a religion than anything else. It certainly isn’t science,” Marcus’s petition continues. “Imposing it upon the citizens of the United States must stop.”

Four years later,  the NRC has yet to respond to Marcus’s petition, but the response to her and other members of SARI from elsewhere has sometimes been withering. One of the most active opponents of LNT critics has been Beyea, the founder of Consulting in the Public Interest. He has sharply criticized Edward Calabrese’s charges of Muller’s scientific misconduct, and pushed back against the anti-LNT contingent in various other forums.

“My position is that there is uncertainty at those low doses, and it can go up or down,” he says. “We do not know really what’s going on down at low doses. There’s a lot of uncertainty and I believe that policy should be based on that uncertainty.” In his May 2012 article in Bulletin of the Atomic Scientists, he argued against the anti-LNT notion of a threshold below which there’s no effect: “A completely zero response is unlikely, given the heterogeneity of human populations, including differing immune and repair systems among people. However, a quasi-threshold, with a dose response below 0.1 Sv, is a standard theoretical possibility … demonstration of a quasi-threshold would be unlikely to assuage those who abhor radiation-producing technology on existential grounds, but it might eventually affect regulations and overall opinion.”

One point on which both hardcore LNT supporters and LNT critics seem to agree is the maddening uncertainty that creeps in at low doses of radiation. As Beyea says, “When you look at the population, you’re averaging over a wide difference of genetic makeup within the population. So that’s what radiation protection has to deal with. We don’t know all the genetic factors and who responds which way. Some people are hypersensitive, some people are insensitive. We have to look at the average over the population.”

Some have also noted that some SARI members are guided by the view that not only are low levels of radiation harmless, but may in fact be beneficial — a controversial and competing theory of radiation exposure impacts known as “radiation hormesis.” Others characterize SARI and its membership as “fringe” advocates, many of whom have somewhat troubling connections to industry. Relaxing radiation exposure standards would, after all, be a substantial boon for nuclear power and other radiation-related industries, which would no longer have to take costly steps to limit the exposure to workers and the public.

Such were the implications of a lengthy investigation published by the nonprofit Center for Public Integrity (CPI) in February, which noted that anti-LNT advocates had been gaining influence and audience within the administration of President Donald J. Trump. Success on this front, the CPI report argued, could mean “vast savings for radiation-reliant industries, allowing them to install new medical equipment, deploy new types of nuclear reactors, open or enlarge facilities for making key nuclear weapons components, and continue the clean-up of contaminated nuclear sites without undertaking such expensive precautions to limit radiation exposure to workers, patients, and the public.”

Whatever the truth of such accusations, anti-LNT advocates argue that the generalized fear of radiation has had consequences that are too-often overlooked. One example: A quiet momentum in favor of nuclear power as a zero-carbon supplement to renewable energy sources in the fight to mitigate climate change had been slowly building over the past couple of decades, invigorated by the urging of climate scientists like James Hansen and environmentalists like Stewart Brand and James Lovelock. But following Fukushima, nuclear power was all but eliminated from the conversation as a workable alternative energy choice, especially in the United States and other leading carbon-producing nations.

“Without nuclear power,” Hansen has argued, “it will be even harder to mitigate human-caused climate change and air pollution.” Furthermore, as he wrote in The Boston Globe last year, “the notion that renewable energies and batteries alone will provide all needed energy is fantastical. It is also a grotesque idea, because of the staggering environmental pollution from mining and material disposal, if all energy was derived from renewables and batteries.”

Other critics have pondered the potential impacts of public panic over a so-called dirty bomb. Certainly, an actual nuclear detonation in an American city would kill many people and cause billions of dollars in damage, but building a nuclear weapon and smuggling it into a city would be enormously difficult and expensive. All that’s needed for a dirty bomb, on the other hand, are a conventional explosive device and some low-level radioactive material, both far easier to obtain than the highly enriched fuel needed for an actual nuclear bomb.

A dirty bomb might blow up a building or a block, killing several dozen in its initial blast. But because it weaponizes radiophobia, its psychological impacts, some analysts have argued, would far outstrip the physical damage by orders of magnitude. “Should such a weapon go off in a city, much of that city will be shut down, and major areas evacuated,” the risk perception consultant David Ropeik writes in Scientific American. “Tens of millions of people in the wider surrounding region, especially downwind, will be afraid. The economic costs will be vast. So will the health effects — not from radiation, but from the sweeping physical impacts of stress, including cardiovascular risk and weakened immune systems.”

For such reasons, Beyea emphasizes that the social fallout of radiophobia needs to be addressed. “I feel that we are completely dropping the ball on social science research here,” he says. “If we have a terrorist attack, if we had a dirty bomb, you’re going to be face to face with this lack of trust, lack of confidence, lack of knowing what to do, what to say.” Part of the solution, he added, would be depoliticizing the discussion over LNT. “I feel there needs to be more respect between the various proponents,” Beyea said, noting the frustrating lack of definitive answers and the passionate feelings on all sides. “I think that we need to recognize that there are arguments on all sides. I don’t think that politicization is helpful for anybody.”

Perhaps the linear no-threshold model made sense in the 1950s, anti-LNT advocates say — particularly in light of legitimate concerns about fallout from atmospheric nuclear testing and a lack of knowledge about the existence of DNA repair mechanisms. But that’s no longer the case. “Current radiation protection concepts should be based on facts and on concepts consistent with current scientific results and not on opinion,” according to a 2009 paper in the journal Radiology, which cites studies on DNA repair and various other work supporting a revised threshold model.

That might involve reconsidering and revising regulations to better reflect our improved understanding of molecular biology since Muller and BEAR I, while continuing research into the biological effects of low-level radiation. The 2018 MIT study on nuclear energy offers what could be a reasonable middle-ground proposal between the LNT and anti-LNT positions, recommending studying the feasibility of “establishing an annual dose below which radiation exposure is not regulated,” noting that such a dose “is not necessarily a threshold below which no health effects occur; rather it is a dose below which health effects simply cannot be identified in the presence of other health risks.”

For their part, the NCRP’s Boice and Shore suggested that providing guidance for a “below regulatory concern,” or BRC, dose that agencies might be able to incorporate into their rule making would not be a bad thing. “If well crafted, they wrote in their response to Undark, “an approach defining BRC could be beneficial, as it would help alleviate over-regulation of exposures.”

Still, that might be a bridge too far for some people — particularly those who believe in the precautionary principle, which in this case would dictate that where the impacts of low-dose radiation are uncertain, the safest thing to do is assume the worst.

Maidment, who routinely deals with the limitations of the LNT, doesn’t necessarily dismiss that notion out of hand, though he believes that the world nonetheless needs a better biological model of risk. “So that we can say, what’s the risk of the CT scan or the coronary bypass procedure or what have you, and we don’t have those today,” he said. “And so we have to fall back to the best model, and that’s LNT, because we don’t know any better.

“It’s the model that has the least assumptions,” Maidment added, “but it should be used with a one-pound box of salt.”