Hume introduced an enormous problem with how we reason from past or present to the future that still plagues philosophers of science and epistemologists today. He pointed out that no matter how comprehensive our past experience, it is never a logical contradiction to deny that the same thing which happened in the past will happen in the future. Take the idea that the Sun will rise tomorrow. Although we have always known it to rise every day, it is not a contradiction to say it won’t rise tomorrow. If one objects that past experience gives us regularities between events like those that occur today and the Sun rising thereafter, Hume’s response would be that we do not know that those regularities will occur in the future. To take another example, oxygen, friction, and combustible material have always resulted in fire, but maybe in the future that very combination will not be followed by fire.

Hume’s problem of induction goes beyond saying that we never know enough to predict the future. His claim is that even when we do know enough to predict the future, where that knowledge has been proven in past experience, we do not know that the patterns of our experience in the future will resemble the patterns of the past. Of course, he did not disregard probability or prudence. His attack was on the notion that we can be certain about the future.