Yes, and radiation can be serious. But nature knows how to deal with high levels of radiation (see Chernobyl's flourishing ecosystem) after a period of time, same with any other disaster. You as a human could go into some moderately radioactive areas since the civilian limits are set so extremely low below the non-stochastic effects, and maybe not have much more of an elevated cancer risk than if you went to the hospital and got an MRI or PET scan (which is unregulated in terms of legal dosage limits).

Just because radiation causes different constraints on cleanup than oil on a large ocean or arsenic in the water table or issues in a space rocket means it is morally worse? That's the part I fail to understand, so long as the engineering continues to behave ethically behind all the systems in their design and construction and retrofit.

Adding to this - there seems to be a lack of observable damage from nuclear accidents apart from self-imposed evacuations.

Literally no-one died at Fukushima. It is the only energy disaster I know of where no-one ended up dead.

This is strong circumstantial evidence that we are being too safe, because we implicitly accept a few deaths when things go wrong in, eg, coal (pollution & extraction deaths), solar ( mainly in installations not in operating), hydro (big-time risks).

Going from 1 death to 0 deaths on this scale is a huge marginal cost. It almost certainly outweights the benefits.

EDIT: We haven't had a solar disaster yet, but coal & hydro disasters happen and can be very bad indeed.

I agree. Anywhere there's a dam, people will drown in it. In fact, if it's close enough to a town and people like to drink then we'll see many more deaths in the winter since it looks like you an skate on it---but can't.