The way blame gets assigned and risk gets assessed by human society is often quite pathological. In a typical setting, there are options A & B. As an example,

The way blame gets assigned and risk gets assessed by human society is often quite pathological. In a typical setting, there are options A & B. As an example, let A be the option to allow young kids to ride independently on a bus and B the option to legislate against letting young kids ride independently on a bus. The cost of A seems high to human emotions and the lazy thinker, typically involving rare but high-impact events (something really bad happening to the well-trained kid taking a bus) that get cherry-picked as important to consider. The cost of B seems low to human emotions and the lazy thinker, typically involving diverse continual low-impact events (all kids affected by legislation losing out on chance to learn independence, environmental impact of using a car instead of public transport) that get neglected because of their low emotional appeal and more complicated accounting. (It may be hard to prove that lots of low-impact things add up to a big thing -- there are too many things and causal relationships to point to.) In many cases, the cost of A is actually objectively lower than the cost of B, so A is the better option. In such cases, whereas all options and their consequences should be considered together with their probabilities, in reality, option B is often deemed to be the better choice, and A a bad choice, because the cost of A appeals more strongly to emotional. Consequently, someone who picked A (law-maker or bureaucrat) gets blamed but someone who picked B doesn't get blamed and is harder to blame, including in court. To protect themselves, they pick the inferior option. I think this sort of pathological risk assessment is the root of the cover-your-ass culture that permeates society.