Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Doug_S. comments on Failing to Learn from History - Less Wrong

38 Post author: Eliezer_Yudkowsky 30 August 2007 08:22PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (11)

Sort By: Old

You are viewing a single comment's thread.

Comment author: Doug_S. 31 August 2007 12:55:09AM 7 points [-]

At some point, one does get down to first principles. Remember Newton's answer to "Why gravity"? It was "I make no hypothesis." If you ask "why" enough times, you eventually run out of answers. The modern explanation of "Why gravity?" is curved space-time, but "why curved space-time" is as good a question as "why gravity." At some point, you run out of justifications, and the only answer you can give becomes "Here is a model that makes accurate predictions."

(Does that make any sense?)

Comment author: DanielLC 27 December 2009 04:26:08AM 17 points [-]

The point isn't to give a reason for everything. The point is to be able to make a model. It's not about "Why do things fall down?" "Gravity.", but "Why do things fall down?" "They have a force exerted on them by every object proportional to the product of their masses and inversely proportional to the square of their distance.". The second one actually lets you make predictions. General Relativity (the exact formulae, not just the general idea of curved spacetime) makes the model a little more accurate. It's not an answer to "Why gravity?" It's just a slightly better answer to the original question.

Comment author: Ronak 26 March 2013 08:03:51AM *  1 point [-]

It does, because you are asking a different 'why' than scientists do; the scientist's question is 'why do I need to postulate this?' And when he/she asks 'why does this phenomenon happen?' he/she is asking 'What model can I use to explain this phenomenon?'

So, the question 'why gravity?' can be answered in the first way, saying that it models falling and revolving objects rather well, or in the second way in which case the answer is, as you say, curved spacetime. But the second question, it can only be answered in the first way, as of now; I'm tempted to say that those concepts which we can only answer in the first way are postulates, but I'm sure non-trivial brainstorming (which I don't feel like doing right now) will show that to be bullshit.

Edit: I think the why you speak of is a weirder version of the second, kinda influenced by words like 'fundamental.'