> However, I dispute calling nuclear "one of the cheapest and safest power sources humankind has ever developed".
I'd rewrite it as
> one of the safest power sources humankind has ever developed[0]
because this is difficult to dispute. Cost is a more complicated matter and depends what you're counting and how you're comparing. Are you including tragedy of the commons? I.e. carbon emissions, mining, storage (batteries, waste, etc), land usage, etc. If any of these are included in the costs the price changes quite a bit and I'd argue it is not in good faith to compare solar and wind to hydro, coal, nuclear, etc if you don't include battery storage/costs because intermittent energy isn't equal to continuous (unless you over allocate).
To continue along the line of the gp the worst case estimates of nuclear deaths (including future deaths) is still lower than 8 years of of coal in the US using the best case stats. If we use more reliable data (the consensus from the UN and other independent studies) even including future deaths are lower than a single year of coal's best case in the US. So how much do you value human life? Because we're talking about 81 people[1] (4k if we assume we don't get better at treating cancer over the next 20 years) vs 39k PER YEAR (low 7.5k high 42k per year US coal particulate related deaths). Let that sink in again: 31 people died immediately from Chernobyl, an additional 50 due to Chernobyl, 0 from 3 Mile Island, and 0 Fukushima. Those are the deaths of nuclear power.
> The order of magnitude of Fukushima ...
> Still on safety: nuclear enthusiasts often equate that reliable under nominal conditions implies reliable in practice. As if negligence and stupidity weren't factors in the real world.
I'm often frustrated by Fukushima being brought up and people implying that it came down to human stupidity. Yes, that was a factor, but there was a much larger and significantly more important factor. The earthquake that caused the tsunami was the largest in recorded (by actual measurement) history (for Japan) and the 4th largest EVER recorded. The second largest recorded was an 8.5 in 1896 and the second largest theorized was an 8.9 in the year 869. Remember this is not linear growth. Based on this the likelihood of an earthquake and tsunami like that happening within the lifetime of the powerplant was incredibly low. Were the earthquake an 8.5 Fukushima wouldn't have happened. It is an absurd statement to chock this up to human fault. We're talking about an absurdly rare and freak event, and no one died.
As to the cost of the cleanup (which includes tsunami cleanup as well as radiation fwiw), well we could make it a lot cheaper if we reduced the acceptable levels of early death by radiation. The hard part is that there may or may not be small radioactive particulate on the ground and this could get introduced into water or food (getting inside the body). Letting people move back would still likely result in less deaths than coal does per year. I'm not advocating for this, I think that'd be insane, but why is this insane? Why is pumping out more coal not more insane? What's the difference? We're talking about human lives and livelihood here. When it is more humane to send people back to live in a radioactive zone than keeping coal plants online, maybe we should re-question our priors. What issue are we trying to solve? Because if it is saving human lives we're doing a pretty bang up job. If it is cost, ditto. If it is climate, ditto.
The arguments being made here are often absurd. There is no question that nuclear is clean and safe. The question is about cost, how much you value human life, how much you value carbon, and how much you value pollution. But if you think lives are cheap, climate change isn't real, and (not or) pollution isn't a big deal, then yeah, continue fighting against nuclear because it is costly.
[0] https://www.statista.com/statistics/494425/death-rate-worldw...
[1] https://www.bbc.com/future/article/20190725-will-we-ever-kno...