The safety issues surrounding nuclear power will always be rather complex, because the illnesses that may be caused by less-than-instantly-lethal doses of radiation take years to show up, by which time it's possible to also attribute them to other factors. Even with a disaster on the unimaginable scale of Chernobyl, experts can't seem to agree whether the number of deaths caused directly by the incident was more in the region of 100 or 100,000. With lower-grade accidents of the type we've had in this country, there has also been a wide range of uncertainty about the long-term consequences.
The nuclear lobby seem to imagine this ambiguity plays into their hands - they can point to the lowest figure of deaths for each incident, and fall back on the trusty phrase "there is no evidence to suggest..." to wave away concerns about higher estimates. I'm not so sure that will wash. For one thing, the uncertainty about the consequences of mid-level releases of radiation directly feeds into the precautions that are - perfectly rationally - taken in the event of an accident, as currently seen in Japan with the evacuation of hundreds of thousands of people from their homes in a state of intense fear, and with children and babies having to be tested for radiation exposure. The first question the nuclear lobby will have to answer in the coming days is not about the intangible issue of long-term health risk following an accident, but rather the very down-to-earth query of "do we really want a building ten miles from our homes that will put us through all that?". Wind farms may spoil the view, but they rarely have a practical and psychological impact quite this profound.