If we restarted the Universe n Times from T=0......

[phinds]

I consider proving something false to be a proof.

Ah, well in that case I do understand your point of view. I don't agree w/ it but you are being logically consistent and making a reasonable argument given your premise. I don't think your point of view is one that is widely shared.

 

[newjerseyrunner]

I had to think about @mfb and @votingmachine's philosophies for a second. Pay attention to the capitalized words.

I agree that you can absolutely never prove that the mathematical description of the universe we've developed will ALWAYS work. It's not possible to do all experiments.

I agree that you can absolutely prove that the mathematical description of the universe we've developed will NOT ALWAYS work. We just need one test to violate it.

I can negate the first and say you can absolutely never prove that a description with ALWAYS NOT work.So I agreed with @mfb at first, but now I agree with @votingmachine. Science is about coming up with hypothesis and eliminating possibilities. Proofs are nice to show that math is consistent, but only way physics can advance is by eliminating possibilities, not proving them. There can be two equally mathematically valid equations for something only by elimination can you move forwards. I agree that proving what is not valid is in itself a form of proof, thought not the same kind that @mfb was originally referring to.

 

[votingmachine]

I thought it was just an oversight in the use of the word "prove". I did not think disagreeing to clarify the point was a deep philosophical point. I know we can't prove a theory true ... as you point out it is not possible to run an infinite variety of tests. And one contradictory test result proves that the model is not ubiquitously true, and is at the least, incomplete.

That really is just a semantic point. I am surprised that the opposite semantics was argued. I thought anyone reflecting on it would also see the proofs on the "disprove" side.

It isn't a big deal ... I thought I had the same semantics as everyone on this. I'll be careful with this word use in the future.

Inconceivable!

 

[Endan Rarity]

If you do explain more later, take care that you are not violating the forum rules on personal theories.

I will take care with that.

 

[Hybrid]

Inconceivable!

You keep using that word. I do not think it means what you think it means.

 

[phinds]

Inconceivable!

Good comeback. Love that movie !

 

[mfb]

@votingmachine: Okay, if you count "this is falsified" as "proof", then it is possible to make a proof. That was not the way I meant it in my post, and I thought that was obvious.

 

[votingmachine]

@votingmachine: Okay, if you count "this is falsified" as "proof", then it is possible to make a proof. That was not the way I meant it in my post, and I thought that was obvious.

I originally typed in my explanation the example that you can prove two masses don't fall at different rates, but I thought it was too much and deleted it. It was probably better to explain, but thought it sounded like I was a bit obsessive ... so then I sound obsessive later ... I understand what you meant, and what you meant is absolutely correct. I'm not actually quite as obsessive-compulsive as that explanation may sound.

 

[mfb]

I originally typed in my explanation the example that you can prove two masses don't fall at different rates

Can you? You can prove your two masses don't fall at different rates now and in your laboratory. What about tomorrow? What about other masses (different chemical composition, or antimatter)? What about repeating the experiment on Moon? You can do thousands of tests, at some point you are highly convinced that all known types of masses and probably unknown types as well fall at the same rate, but you cannot prove that it is true for all types of masses everywhere in space and time.

You can disprove the theory "all masses fall at the same rate" if you find a single counterexample. None was found so far (always assuming vacuum, absence of electromagnetic fields an so on of course).

 

[votingmachine]

Can you? You can prove your two masses don't fall at different rates now and in your laboratory. What about tomorrow? What about other masses (different chemical composition, or antimatter)? What about repeating the experiment on Moon? You can do thousands of tests, at some point you are highly convinced that all known types of masses and probably unknown types as well fall at the same rate, but you cannot prove that it is true for all types of masses everywhere in space and time.

You can disprove the theory "all masses fall at the same rate" if you find a single counterexample. None was found so far (always assuming vacuum, absence of electromagnetic fields an so on of course).

Hmmm. That makes sense. But it also a slight restatement of the "proof". The theory was that heavier masses fell faster. That was disproved. The new theory is that masses fall at the same rate ... it would be a rejection of science if you were to say that re-running the previous experiment exactly might give a different result. But that does not prove the new theory.

Hmmm. I admit the wording is puzzling me now. We can show that two different masses fall at the same rate (controlling all the interfering forces). That experimental result disproves a theory that predicts that heavier masses always fall faster. It does not prove affirmatively the theory that different masses have the same acceleration of gravity.

I'm just confused now. Previously I thought there was a clear proof of the falseness of one theory. But if that is worded as "proof" of the opposite theory ... then it seems wrong.

My model is that we can "prove" some things are false. But not that an open ended generalization is true. We know that the open ended generalization may break down under some conditions. I think that I might need to be more careful in my word selection. I still think there are things we can prove are not true, but those theories need to be quite specifically worded. Otherwise, it becomes an affirmation of the generalization on the other side.

You could be right. It might be better to simply say that nothing can be proven. Things can be known with great certainty, but not proven. I still have reservations about that position though. I'm sure you see the what I am saying ... I don't mean to beat a dead horse.

EDIT: I thought of an example. There was the logical theory that "all swans are white". But it was dis-proven by the discovery of a black swan. Now if you take the new theory to be "all swans are either black or white" ... then it is still not proven. But if the new theory is "swans are not always white", then it is proven (rather useless, but proven). I think there needs to be some caution around the wording of the new theory.

 

[phinds]

... But if the new theory is "swans are not always white", then it is proven (rather useless, but proven). I think there needs to be some caution around the wording of the new theory.

But here again, you are just proving a negative and I at least have already agree with you that you CAN prove a negative. But that is, as you say, pretty useless. It's proving a positive that you cannot do in physics.