There are extremely few things (if any) we can know for certain. Thinking in terms of boolean logic like this appears very primitive to me. In my opinion, the

There are extremely few things (if any) we can know for certain. Thinking in terms of boolean logic like this appears very primitive to me. In my opinion, the only rigorous way to improve our knowledge, is to employ Bayes Theorem, and Bayesian thinking in general. You and I may have different prior on a topic. We may see the same evidence, but end up believing different things because of the prior. As long as we state the prior, that is good and fair. If the evidence contradicts my prior, I should lower the confidence I have in my belief, even if the prior was big enough that I still believe the same thing. Unfortunately, what often happens, is that people lock themselves into believing things absolutely. If you lock some prior probability to 1, Bayes' Theorem has n way to let you update that belief. This is especially common when it comes to beliefs related to religion and identity. In fact, this is one way to define religious belief. Anything you believe that you will NOT stop believing regardless of evidence, could be defined as religious. If you believe in God, and we discover that Heaven is not on the other side of the starts, the believer just modifies non-core parts, and claims that "God is everywhere and nowhere". If belief in racist oppression is what defines you, and you find that most individuals are not individually racist anymore, you can just introduce "Structural Racism" that claims that hard work, staying married and maths are racist.