When I was being trained for work in the nuke industry (1972), we were given a tour of a laboratory where nuclear experiments of the non-explosive variety were carried out. One apparatus we examined was a vault were items were placed to measure their potential emission, tiny amounts of radiation (in soil and water samples, etc). The device was heavily shielded so that radiation from the environment (not just the lab, but in the world itself) would not affect the measurements.

The man giving the tour happened to mention that the shielding for the vault was shaved from the armor plating of a mothballed battleship that had been built before 1945, that is, the steel had been forged prior to the first nuclear explosion in WWII.

The idea was that the shielding, instruments inside the test chamber, as well as the containers that held the samples, could be measured separately so readings from any radioactive materials incorporated into their manufacture could be subtracted from the sample. Radiation still emitted by contamination from the shielding itself had to be from unavoidable natural background sources, because the steel was crystallized prior to the first Yucca Flats detonation. The instruments were so sensitive that radiation from decades of weapons testing, could easily contaminate and overwhelm the readings, and even the tiny amounts of background radiation from cosmic or geological sources was detectable.

This does not imply that nuclear weapons testing residual fallout was necessarily dangerous, just that it was measurable, and that as we know now, we had absolutely no idea how dangerous it might be. The levels of ionizing radiation, as well as the types of radiation due to the nucleides created by fission and neutron bombardment, of the planet had been altered on a world-whie, and permanent, basis; and that contamination was already easily measurable by the primitive technology avaialble almost a half-century ago.

It wasn't immediately obvious that just because animals didn't die immediately after exposure didn't mean that level of radiation was safe. In all toxicology, the death rate is plotted against the exposure at high concentrations of toxicity, and a lethality as a function of exposure curve is plotted. For low levels of exposure, the deaths might not be immediately obvious unless we waited a long time, and looked at enormous populations. Would a million extra deaths a year worldwide, each resembling an ordinary cancer, even be noticeable? And if it could be confirmed, would it be acceptable? And who makes that decision, who does the risk/benefit analysis, and how much do they have a stake in the outcome?

And of course, it's not just radiation. Ever single industrial chemical we release into the environment, particularly the complex synthetics that have never been present in nature, has the potential to have some hidden risk, either in its own right, or synergystically when combined with other substances which might be benign by themselves.

Most are probably harmless, true, but it is almost certain that millions die every year because of chemicals we mass-produce to make some small improvement to our quality of life or the profitablity or marketability of our products. Only if the effect becomes obvious, is anything done about it. And even if the effect is obvious, there are those who have everything to gain by making sure nothing gets done about it. They usually succeed, too.

I personally, am willing to accept a small amount of risk for these obvious benefits. But I sure wish I knew just what the numbers were.