Julia Galef: Think Rationally via Bayes' Rule (2)

Recently I posted this talk by Julia Galef, along with the following comment on Google+:

I am working to steep myself in Bayes' Theorem or Bayes' Rule. I very much like the fact that it helps weigh our beliefs against perhaps a host of evidence that confirm (or disconfirm) our beliefs to varying degrees. The principles underlying Bayes Theorem help sharpen our thinking and strengthen our reasoning. In an optimal scenario - that is, if we are open and willing - we ought to modify our beliefs depending on the evidence we encounter. In other words, we ought to be think logically or rationally about ourselves, people and the world around us.

But the thing is, we as humans are not entirely logical or rational; we are also intuitive, subjective and instinctive. For example, we may hold on to the belief that all teenagers are undisciplined and irresponsible, even though we meet some who are quite the opposite. So the tenacity with which some may uphold their beliefs, despite evidence to contrary defies Bayes' Theorem. This is the stuff of prejudice and discrimination.

However, this example of defiance doesn't necessarily discredit the usefulness of Bayes' Theorem, in ways that Julia Galef talks about it. Say, we have a friend who's prejudiced or discriminatory against teenagers, we might try to firmly and-or gently persuade him to think otherwise via Bayes' Theorem.

Our tact may first draw on one of the Seven Habits of Highly Effective People, in particular: Seek first to understand, then to be understood. Maybe he's faced an assault or trauma in relation to teenagers; maybe he's a father who weathered a terrible storm with his teenage children. Using empathy - that is, the ability and willingness to walk in others' shoes or to sit in their seats, we might help our friend resolve such trauma or storm, so that perhaps he comes to see that despite all of what he experienced, not all teenagers are undisciplined or irresponsible.

So: empathic understanding + Bayes Theorem underpin not only more effective thinking and reasoning, but also better understanding and relationships between people.

My post generated a bit of a discussion:

+Gerallt G. Franke What you say makes very good sense, and it resonates with my working notion that we approach any thing, any phenomenon or context as it is. That is, we do our best to avoid (or at least set aside) any preconceived idea, preexisting framework, or preset methodology. In short, we approach that thing etc. with a blank slate and an empty "toolkit." Then, we let that thing itself point us to an idea, framework or methodology for knowing it, analyzing it, and solving it.

I believe you're quite right: If, instead, we began with a prior logic, reasoning, inference or deduction - perhaps because we have difficulty tolerating or grasping uncertainty - then there is a good likelihood that logic etc. will be off (at best) or break down (at worst). Alternatively, by examining something as it is, again without preconceived ideas, we are much more likely to arrive at the right logic for that thing itself.