> The real questions are: Does our current model describe all observations? If not, can we find a model that does? And, even better, can that new model make pre

> The real questions are: Does our current model describe all observations? If not, can we find a model that does? And, even better, can that new model make predictions that we can verify? But every prediction your model can make comes from a "why" question that the model answers.