Expense in context
Of course it is going to be expensive to rebuild following the tsunami. However, I fail to see how it helps to have a huge bill to clean up the mess of several severely damaged reactors on top of that. Indeed that surely makes the situation worse.
As for Lewis's line that the reactors exceeded their design target - well, big deal. Yes, the people who designed the reactor did their job, but if the design parameters were inadequate then that's still a major failing. What really matters is how those design targets were arrived at and were shortcuts taken for financial reasons.
To pretend, as Lewis seems to make out that nothing major has gone wrong and there are no serious consequences, well I beg to differ. Three Mile Island took 12 years and about a $1bn at 1980s prices to dismantle the reactor (and the main building remains contaminated and will incur further costs when that is finally removed). Adjusted for inflation, that's more like $1.7bn.
It's clear that the problem at Japan is much worse with several reactors, higher levels of contamination, wrecked buildings, storage pools affected and at least one of the containment vessels seriously damaged. How much is this going to cost to clear up? Who knows, but it's going to be several billion dollars minimum. That's before the costs of replacing the lost generating capacity is considered or the economic costs of power shortages. Whilst one of the reactors was due to be turned off in 18 months or so, some at the site had much longer operational lifetimes. Whether any of the reactors at the site are ever used again, is surely still uncertain. Certainly none of those which have sustained serious core damage will ever operate again.
Some suggest that the cost of the damage to Japan's infrastructure is of the order of $200bn (according to the Economist). Dealing with the consequences of the damaged reactors and replacing lost capacity could easily approach 10% of that. If, as Lewis suggests, some reactors could be restarted behind improved tsunami protection, then that surely points to a major risk assessment failure as such measures would surely have cost a fraction of clearing up the mess and replacing lost generating capacity. At a time
I'll repeat again, I am not anti-nuclear, but the pretence that there are no serious consequences is a bad joke. The idea also that this was an unforseeable event is also highly suspect. Plant like this, where the economic and environmental consequences of severe damage is so high cannot be designed for things like 100 year events (which would give a 40%+ chance of an occurrence), they have to be designed for much rarer ones.
Lewis's argument that we still design roads, railway and bridges despite the dangers is also way off. To a large extent those are essential to modern life - there is no choice, albeit we seek to do them better. Power generation is, of course, similarly essential, but the difference here is that these reactor installations have now proven to be highly vulnerable - and vulnerable in a way that was foreseen, as the US report shows. This is a management and governance failure, not primarily a technical one, but a failure it is.