The reasoning behind abandoning ALARA isn’t particularly strong; it’s based on pragmatism and economics of risk, not the underlying biology and chemistry of ionizing radiation exposure. Linear No Threshold is still one of the more conservative models that accurately depicts the impact of ionizing radiation on the underlying biology of adducts, double strand breaks, and associated repair mechanisms.
We’re already in a shooting gallery of radiation exposure due to solar and terrestrial sources; controlling some radiation sources won’t be worth the cost, but that’s where reasonable comes in.








LNT does not assume that there isn’t a repair mechanism. It assumes that the repair mechanism has a constant failure rate. That’s a pretty good match for the primary biological process involved.
All of the studies I’m aware of looking at lower dose thresholds for radiation occur on timescales and in situations where the benefits of selection due to mutations overrides the negative effects of mutations on cancers. They aren’t actually measuring lifetime cancer risk on an organism level. (If a study is talking about comparing sieverts instead of doses of a specific kind of radiation, it’s also suspect.)
It’s theoretically possible to have a positive effect on an organism level, but that involves much more complicated biological mechanims like immunity and epigenetics which aren’t as easily modeled or as well understood. Moreover, they also have the capability of being supra-linear at low rates of DNA damage.