Trial and Error

I’m wondering how trial and error might be done.  Professor Woodhouse mentions nuclear reactors, saying “The industry was inflexible, in part because billions of dollars had to be expended before a reactor generated a single kilowatt of electricity.”  Given that, how could more trial and error have been done?

My instinct here is to say that it could not, but I realize that that is legacy thinking and that people instinctively defend the way things already are.  So, how could the industry be more flexible and allow for trial and error?

I’m not sure.  Maybe smaller reactors?  Woodhouse mentions this.  Or what about making one reactor, and testing it?  Perhaps there should have been tests done to see what happens when there is no electricity, with many people on hand to deal with the damage.  The issue with that is that the test itself would have been very dangerous, but I think that only betrays how dangerous the reactors are.  Maybe there could have been simulations.  But honestly, I’m not sure that the problem with the Fukoshima reactors was that there hadn’t been effective trial and error.  I think the problems could have been predicted, but people didn’t think through things well enough.  It seems to me that people rarely do.

I think Woodhouse makes a good point about flexibility.  When a technology is inflexible, it’s difficult to change things when we discover that things are going badly.

A concern I have about the idea of trial-and-error is that I don’t think we necessarily know how something is going to go until we do it.  How can we fix this?  Instead of going to something large-scale right away, we can increase the scale little by little, testing to see whether it’s working.

I suppose that Woodhouse would say that if we can’t effectively use trial and error for something, we should err on the side of caution and not do it.  I do agree with the precautionary principle, but I don’t know if it should apply here.  Are there situations in which it makes sense to do something even when we can’t effectively try it out (or we can’t try it out enough to get a lot of information)?  I think there might, but I think those situations should be the exception, not the rule.  In general, I think there should be more trial and error done.  However, I think a lot of problems are caused not by a lack of knowledge about the harmful effects of something, but about people not caring (or not caring enough to fix them).  Would trial and error have made a difference in the case of the Ford Pinto?  Ford already knew what the problems were, and did not care.  Of course, if they had cared, trial and error would still have saved lives.