An article calls the radiation safety model a lie. The thread fact-checks it line by line

A beer analogy, radioactive apartments, and a charge of deception

How to lie about radiation, by Alex Chalmers and Ben Southwood for Works in Progress, opens on a deliberately blunt analogy: "Drinking one beer a night for a year is a lot less harmful than drinking 365 beers in one go. The same applies to radiation exposure, but regulation doesn't agree." Their target is the Linear No-Threshold (LNT) model: the regulatory assumption that risk scales linearly with total dose all the way down to zero, with no safe floor.

Their centerpiece is a natural experiment in Taiwan: apartment blocks built in the 1980s with steel rebar contaminated by cobalt-60, giving thousands of residents years of elevated chronic exposure. The cancer rate among them came out lower than the general population, not higher. On Chernobyl, the authors put the thyroid-cancer toll at fifteen deaths so far and perhaps 200 in the long run, and argue most of that was avoidable, since Britain after the 1957 Windscale fire simply discarded contaminated milk for 44 days. The headline word is the thesis: regulators, they say, lie about the danger of small doses.

A Chernobyl tumor sequencer asks where the data is

The sharpest pushback in the thread came from someone who had worked the actual genomes.

"The article is long on emotion, exposition, but very short on the data… nor any major chromosomal aberrations. They don't have the technology to measure DNA damage that might be significant." — a researcher who examined whole-genome sequencing of Chornobyl thyroid tumors

The factual claims drew line-item corrections. The article's assertion that Chernobyl is the only commercial-nuclear accident to fatally irradiate anyone was met with a counterexample:

"The Tokaimura incident (Japan, 1999) comes to mind… three workers were severely irradiated, and two of them subsequently died." — duskwuff

A reader who checked the geography flagged the milk argument as cherry-picked: the fallout plume blew north into Belarus, so picking Ukrainian numbers understates the contaminated population.

"There's some very suspicious cherry-picking going on with the author's choice of Ukrainian milk and raw numbers… which is never how public health researchers describe impact." — an engineer who pulled up the directional fallout maps

Not everyone dismissed it. The founder of the Chernobyl Tissue Bank, Geraldine Thomas, was quoted from a 2011 Guardian piece saying she went from anti-nuclear to pro-nuclear after studying the aftermath, citing "an unwarranted fear of radiation." And a reader pointed to the late physicist Bernie Cohen, whose epidemiology found home radon levels negatively correlated with lung-cancer risk: more radon, lower risk.

The deepest cut wasn't physics but provenance. Two commenters who grew up under the Soviet system argued the entire dataset is poisoned at the source.

"I personally treat any statistics and data from the soviet government to be unreliable to the degree of meaninglessness." — an ex-staff engineer at IQM Quantum Computers

That cuts both ways, and the thread knew it: if Soviet numbers can't be trusted, they can't be trusted to make Chernobyl look mild any more than to make it look catastrophic. One reader reframed the whole argument as a policy question rather than a physics one. Even if low doses are harmless on average, contamination isn't spread evenly, so "what if you happen to ingest traces of the harmlessly radioactive material now surrounding you?" The model under fire is conservative precisely because it doesn't have to know.