I disagree about the emphasis on prediction. Science involves predictions, but it does not involve only predictions. Chaos theory, for example, shows that phys

I disagree about the emphasis on prediction. Science involves predictions, but it does not involve only predictions. Chaos theory, for example, shows that physical, deterministic systems can be rigorously understood while prediction is (approximately) impossible. Likewise, you can make predictions without understanding, with sufficiently advanced machine learning (interpretability vs. flexibility).