Yes, this is philosophy of science and technically an axiom. But it's not an arbitrary one. It's saying that "it exists <=> something would be different about the world if it didn't exist <=> it's in principle observable (though its impact) <=> can be studied by science". Call it my lack of imagination, but I don't believe it's even possible to disagree with this axiom; alternatives don't make any kind of sense.
EDIT: looking at this "axiom" from another angle. Imagine that I claim that atoms are really built out of Zeta faces, that are exactly like quarks, but have smily faces etched on them. I also claim that it's in principle impossible to tell a difference between a Zeta face and a quark (hypothetical quark, after all it's all Zeta faces). They just are there. At this point you could correctly observe that since the universe would look precisely no different whether or not there are smiley faces on quarks - there's not a bit of information distinguishing these two hypotheses - my claim is absolutely meaningless and a waste of time. This generalizes to all kinds of invisible, unmeasurable magic faries under the bed, etc.
(It's probably a well-trodden philosophical ground, but I don't know who to read to find a decent discussion of it.)
>Yes, this is philosophy of science and technically an axiom. But it's not an arbitrary one. It's saying that "it exists <=> something would be different about the world if it didn't exist <=> it's in principle observable (though its impact) <=> can be studied by science". Call it my lack of imagination, but I don't believe it's even possible to disagree with this axiom; alternatives don't make any kind of sense.
There are several logical holes in the axiom.
The first big one is that it implies logic is applicable to the universe, i.e. that the universe must abide by logical necessities (and even more so, by logical necessities the way we perceive them). Setting aside the fact that we cannot prove logic's applicability logically (it's itself an axiom), QM has given us many non-classical-logic compliant observations.
Then, there's the assumption that: "it exists <=> something would be different about the world if it didn't exist". That's again an assumption. I could just as well consider that (the fact of its existence turning from false to true aside) the world could be the same whether something exists or not.
This presupposes when it should instead prove: that existence necessitates difference in the world.
Third, "it's in principle observable (though its impact) <=> can be studied by science" -- whether something can be studied by science or not shouldn't be a criterion for its existence.
Black holes existed for billions of years before they could be "studied by science", or before science even had any understanding of them. Only in retrospect we say that they "can be studied by science". Did the not exist when we couldn't say that, and exist now?
Or they did exist and we only lately came to find out about them, and that we can actually study them? If it's the latter, why wouldn't that -- the possibility of existence- be the case for tons of others things (e.g. sub-plank length stuff)? Because our current theories say so? Or because we thing our current theories are the be-all end-all we'll come up with?
Fourth, one could argue that something could still exist, have impact, and
not be observable. E.g. if it's impossible to make it not exist, or look at that level (so that we have no way to measure its impact). E.g. some entities or energies at the sub-Plank scale level could hold the universe together and be essential substrate. But we just have no way to know about it, because we can't look at that level (and our theories can't even go at that level).
>At this point you could correctly observe that since the universe would look precisely no different whether or not there are smiley faces on quarks - there's not a bit of information distinguishing these two hypotheses - my claim is absolutely meaningless and a waste of time.
The claim would be a waste of time to try to verify.
It could still very well be ontologically true -- which is what we're discussing.
That something that can't be verified in any way doesn't exist is a logical jump. What we can say for certain is that it could or could not exist, but we can't verify it.
We can't even say that it has no impact to us, because without being able to verify it, how can we be sure it has no impact? At best we can say "it has no impact as far as we can tell".
But it could still have a huge impact -- the same way the Quarks you've mentioned have a huge impact, had a huge impact even before we knew existed and before we could measure them, and would have have a huge impact even if we remained forever at the 15th century level of physical knowledge.
> that the universe must abide by logical necessities (and even more so, by logical necessities the way we perceive them)
This axiom can be defined better, to account for the fact that we only perceive the universe through our minds, and this opens a whole can of philosophical worms - but in practice, this axiom does hold, and if it didn't, we could only pack our bags and go back to caves, as nothing would make sense and there would be no point to thinking.
(I didn't dig too deep in that can of worms, but I vaguely suspect the justification for such axiom could be found in anthropic reasoning.)
> QM has given us many non-classical-logic compliant observations.
Well, classical logic is pretty limited and it isn't be-all, end-all of logics, and not exactly what we use in daily reasoning anyway. I admit to not having much knowledge on the more complex types of logics, but the problems I recall seeing boil down to misapplying logic as an abstraction. For instance, I just stumbled upon the following example of "classical" vs. "resource-aware" logic:
"suppose you're standing at a vending machine that dispenses candy for a dollar and also dispenses coke for a dollar. You might write that as 1$⇒candy and 1$⇒coke. But then inferring that 1$⇒candy&coke is clearly wrong. The dollar gets "used up"."[0]. The post introduces resource-aware logic as a solution, but to me the problem seems to be modelling vending machines as (classical) logical implications; it's kind of dumb if you think about it.
(I'm also aware that probabilistic models are a better way to describe the observed world in practice than formal logic.)
> Then, there's the assumption that: "it exists <=> something would be different about the world if it didn't exist". That's again an assumption. I could just as well consider that (the fact of its existence turning from false to true aside) the world could be the same whether something exists or not.
I don't see how it could be any other way. Without asserting that there is a non-zero diff between universe with something and universe without something, "existence" is an empty label without meaning.
> This presupposes when it should instead prove: that existence necessitates difference in the world.
"Existence" is a term we define, and again, a different definition seems meaningless.
> whether something can be studied by science or not shouldn't be a criterion for its existence.
I didn't mean that current science defines what exists or not. I meant that existence of something has observable impact on universe. Observable in principle - not right now by us, but theoretically observable with sufficient knowledge and technology level. And if something is observable like this, then it falls into domain of science - it could be studied, and conclusions could be drawn. My point is that the set of things that exist is equal to the set of things that are of interest to science. This is not a time or technology-dependent claim.
> Fourth, one could argue that something could still exist, have impact, and not be observable.
How? "Have impact" is synonymous to "in principle observable".
> E.g. some entities or energies at the sub-Plank scale level could hold the universe together and be essential substrate. But we just have no way to know about it, because we can't look at that level (and our theories can't even go at that level).
I think you could only say such entities exist in a meaningful sense if their non-existence would make the universe look different. This way you could indirectly infer their existence through considering the things you've observed and not observed. If the sub-structure of the universe was completely screened away, so that whatever happens there, no bit of information reaches our universe, then I see no meaning in the question of whether or not that sub-structure, or entities inhabiting it, exists.
Again, by "observable" I at not point meant "observable by us, today". I meant "observable in principle".
> How? "Have impact" is synonymous to "in principle observable" (...) If the sub-structure of the universe was completely screened away, so that whatever happens there, no bit of information reaches our universe, then I see no meaning in the question of whether or not that sub-structure, or entities inhabiting it, exists.
Lemme try to answer those, as this is the core of the current disagreement I think.
Consider the case of parallel universes -- they are actively studied by scientists as a notion, and there's very much a question of whether they exist, while still "no bit of information reaches our universe" from them.
Does that preclude their existence, or makes such an inquiry into the possibility of their existence meaningless? Several physicists and cosmologists (even major ones) don't seem to think so.
EDIT: looking at this "axiom" from another angle. Imagine that I claim that atoms are really built out of Zeta faces, that are exactly like quarks, but have smily faces etched on them. I also claim that it's in principle impossible to tell a difference between a Zeta face and a quark (hypothetical quark, after all it's all Zeta faces). They just are there. At this point you could correctly observe that since the universe would look precisely no different whether or not there are smiley faces on quarks - there's not a bit of information distinguishing these two hypotheses - my claim is absolutely meaningless and a waste of time. This generalizes to all kinds of invisible, unmeasurable magic faries under the bed, etc.
(It's probably a well-trodden philosophical ground, but I don't know who to read to find a decent discussion of it.)