- Quality of explanations is measured by:
- rigidity: Degree to which it can be varied without losing explanatory value. Less is better. Good Explanations are hard to vary. Being able to vary an explanation does not lead to contradictions, but unnecessary restricts our freedom to act in accordance to our values.
- reach: “The ability of some explanations to solve problems beyond those that they were created to solve.” or “Degree to which it can be applied in other circumstances under which it was discovered.” More is better. Effectively, this means that explanations with more reach do not need all premises under which they were discovered. Contradictions are a sign for one of the premises an explanation makes is not being met. There are no contradictions. Check your premises
- Finding good explanations is hard but good explanations are often obvious. In the same vein, creation is often hard and destruction is trivial.
- Progress only happens when people put in the work. Namely We create better explanations through conjecture, criticism and testing.
- It’s tempting to judge explanations by the level the argument is made on. Like “quantum physics encompasses explains all things that happen in the multiverse”. This is called reductionism. But High-level simplicity can emerge from low-level complexity. The other way around like “love is the answer” is an error of the same category which is called holism. The quality of an explanation can not be deduced by the level it is on
- It’s tempting to judge decisions on the result they deliver. This is called resulting. However, because we live in a world with uncertainty we can’t escape, we need to make decisions with what we have right now (explanations!). Thus, The quality of a decision is completely explained by the quality of the explanation that went into making the decision.
- All explanations are built upon abstractions. We can’t perceive reality as is. We leave things out to focus on other things. If what we leave out is needed for a better explanations, we are using a bad abstraction. Bad abstractions are illusions.
- Good explanations build on top of things that we know about the world. For example, the Jobs to be Done Theory allows us to form explanations around product design decisions, that we can evaluate if we see the Theory as true. Deciding which Theory to base our explanation on is itself up for discussion/dependent on good explanations.