Hello Prof. Bodansky,
I’m a PhD student in geophysics, and I just finished reading your book, Nuclear Energy. I appreciate the trouble you went to in the book to remain effectively neutral as to whether we ought to be pursuing the development of nuclear power. While I can’t say that the book made me into a nuclear advocate, I am less opposed to it in principle now, and believe that it does represent a potential long term energy solution, albeit one with non-trivial caveats. Then again, that seems to be the case with all of our options at this point.
In particular, I liked the discussion of long term, intergenerational ethics which you brought up in connection with waste disposal, because it is applicable to so much more than just nuclear waste, and because while you did point out the huge disparity between our assessment of the risks associated with radiological hazards vs. other forms of pollution, environmental degradation, or more general societal sources of human suffering, you did not use that disparity to argue that we are being too cautious in our approach to waste disposal (or nuclear power in general). I appreciate that because I think it is probably more likely in the long run that we are for some reason being appropriately cautious with respect to nuclear hazards, and inappropriately cavalier in many other contexts.
The only significant criticism I can think to direct at the book is in regards to the methods described for assessing the probability of accidents. I would submit that serious releases of radiation are most likely not to be the product of several independent low probability events taking place, or even of “weak link” or “common mode” failures as described in your book. Chernobyl was, by your own description, not really that kind of accident, as many of the reactor’s safety systems had been disabled for the purposes of the test which was underway. I suspect that there are significant irreducible risks of human error or malicious intent. In the event that we choose as a civilization to pursue nuclear energy as a significant primary power source, with terawatts of installed generating capacity, the right standard in my mind is how safe the reactors are when they are located in a war zone, and operated by an occupying army who has no idea what they’re doing, or who might actively try to precipitate the release of radioactive material. In the fullness of time, with thousands or tens of thousands of nuclear power plants installed globally, surely such a scenario would take place eventually, and potentially more frequently than would be deemed acceptable by those assessing the risk of truly accidental accidents. I am encouraged that many of the reactor designs currently under consideration incorporate passive safety features, and hope that in time we may arrive at a design so simple and inherently incapable of uncontrolled criticality or overheating that these human risks are no longer significant.
A lot of my discomfort with low probability high consequence events comes from having some familiarity with inverse modeling problems, as described for a non-technical audience by Nassim Taleb (“Fooled by Randomness” and “The Black Swan“) and for a technical audience by Albert Tarantola (at the Institut de Physique du Globe de Paris). If you are not already familiar with them, and have the time and inclination, I think both writers are worth attention.
Thank you again for your work,