JaredJames said:
Thank you Dmytry.
As a note, I was referring more to Joe's specific claims when requesting sources.
ahh, also, regarding LNT. The EPA page is kind of out of date. The microbeam studies are additional evidence in support of LNT.
The reason it is not possible to show effects of low levels of radiation is that you need large sample sizes to eliminate noise, i.e. random fluctuations. The statistical noise is proportional to square root of sample size. Meaning that if it takes e.g. 100 people to conclusively show dose effect of 1 sievert - which causes excess cancer rate of 10% on background of 40%, it will take 100 million for 1 millisievert (and another hundred million for control).
It is not possible to control for healtcare (rate of failure to diagnose), age, smoking, race, etc. when big populations are involved. It is theoretically impossible to directly show that radiation effects continue at low doses - there is a threshold to sensitivity of population studies. However, theoretical considerations - and single cell single track studies - lead to conclusion that effects are linear.
Generally, in science, the continuation is adopted as null hypothesis, in absence of proof of non-continuation.
For example, how much money would you bet that 1 gram of matter does not attract 1 gram gravitationally over distance of 2 meters? Such attraction would be EXTREMELY difficult to show, but surely we aren't going to bet our money it isn't true, as simple logic shows there must be some very complicated effect to make gravity not work on 1 gram, but work on 1000 pieces each of 1 gram.
Would you bet human lives on such an assertion? I can't show that 1 gram attracts 1 gram over distance of 2 meters directly, sorry, all i have is theory that it does, based on evidence with larger masses or smaller distances and the perceived complexity of a theory which would fit the experimental data but would not have 1 gram attract 1 gram over 2 meters distance.
It's a simple matter of occam's razor - and occam's razor is very much in favour of LNT.