Lecture 3 thoughts

Lecture 3 explored the concept of unintended consequences in depth. At its most basic, the notion of unintended consequences is relatively simple. All actions have consequences, no matter how small. Those consequences (whether negative or positive) may present themselves in unexpected ways, and can propagate on to create even further ripples and consequences of consequences. Given that our goal in this class is to find ways to better utilize science and technology for the well being of humanity, it makes sense that we want to steer technical resources in a way that minimizes negative outcomes. Specifically, I noted that both the frequency and severity of negative consequences were metrics that we seek to minimize. It is helpful to define bad events in this fashion, because both of these factors together do a good job of covering the spectrum of unforeseen disasters. For instance, nuclear power disasters occur extremely infrequently, yet the results of an incident are devastating. Small industrial chemical spills are typically containable and addressable, but the number of incidents and violations surrounding waste dumping laws are very numerous. If we truly want to develop a plan to minimize unforeseen consequences, we need to put safeguards in place to reduce both the magnitude of a potential disaster and the statistical likelihood of it occurring in the first place.

The second part of the lecture that I found particularly interesting surrounded the notion of "normalization" of accidents. In many fields we can both empirically and mathematically prove (to a high degree of accuracy) that accidents will occur with some given frequency. While the specific modes of failure may be unknown or complex to calculate, basic statistics can take a macro view of historic events and condense it into a close estimate. Normalization asks that if we know that an event will happen with a certain frequency (even if we don't know how or precisely when), can we really call it an "accident" when it actually occurs? I would go further, to ask if we should choose to do such a thing. In the event of school shootings, like in our supplementary reading material, normalization appears to make horrific acts of violence commonplace by training students to expect it to happen. On the other hand, we know earthquakes and natural disasters will happen regularly too, and we still feel justified in calling those "accidents". The key seems to lie with our degree of involvement in the accident. We can't cause natural disasters (at least not on a short time scale), and neither can we ever hope to eliminate all incidents and industrial accidents, no matter how carefully we try. I think that the term "accident" is still accurate when the timing of the event cannot be known with any detail, because it still entails an element of surprise. At the same time, however, society needs to stop associating the term "accident" with a freedom of liability or responsibility for the consequences. Just because an accident occurs does not mean we don't have to deal with the fallout, and we need to be vigilant to identify risks that are not worth the benefit.