We don't "prove" observable things, if you want to be really technical about it. Over time we have seen these observable phenomena and have constructed around them explanations which satisfy the ability to be proven wrong and that adequately predict such events and future events.
Bookmarks