Universe not accidental: Is this Steinhardt statement "rather pathetic"? If so, why?

bapowell

Yeah, but that doesn't constitute observational evidence since you are referring to domains outside the causal Hubble patch. In particular, you are assuming that a sufficiently large universe exists in which the order parameter can take on sufficiently many values to give enough variation on which to base an anthropic argument. This may be true, but it is not, nor ever will be, an empirical argument. But agreed -- it is certainly suggestive.

bapowell

Look, the idea is that we can observe the universe outside the Earth. We know that a larger space exists with other places in it that we could have lived. That alone breaks the analogy. The problem with the multiverse is that it is almost by definition unobservable.

Chalnoth

That's not even necessary, though, as the nature of quantum mechanics guarantees that all of these different symmetry breaking events are realized right here as different branches of the wavefunction.

bapowell

Oh, and you can observe these other branches? They are epistemologically objective?

skydivephil

Yes we know that now, but we did not have any evidence of other planets before the 1990's, so what would have been your view then on the position of the Earth?
Also you didnt reposnd to the claims that it is possible to obtain evidence of the mulitverse. I dont know whether to believe these claims or not. But what I can say is that I do see more papers offering what they claim is an observational signal to the multiverse than papers claiming an observable singature of a final theory mentioned by Steinhardt.

bapowell

I believe I answered this above. I don't think that it matters whether other planets are observed. What matters is that there are other regions of space where we could have existed. We can observe the rest of the universe, and we can conclude that there are indeed many places where we can't likely exist -- like on Mercury or Pluto (and yes, these count!). We can conclude then that there is nothing fundamental about Earth's properties -- we live here simply because we can. We can't say the same about the Universe because we have no evidence for the existence of, and no knowledge about the characteristics of, the multiverse. We don't know whether it exists let alone whether there are vast regions that do not support universes like our own (this is the measure problem that was referred to by someone earlier -- we don't even have the theoretical tools to understand an infinite multiverse). That said...

Sorry, I didn't even see these! Yes, it's certainly true that some models that postulate the existence of other universes might have observationally distinct signatures. And observing these would be suggestive. Of course, it would be necessary to rule out all other degenerate explanations, and of course, the evidence would need to be strong, given what they say about extraordinary claims and all that...

skydivephil

Yes I agree with that, extraordinary claims require extraordinary evidence and we would have to rule all out other less prosaic explanation if such a bubble collision was claimed as observed. It would be easy to fool ourselves thinking we had seen such a pattern becuase we want to . But I belive the authors of the afromentioned papers went to great lengths to exclude such effects, they seemed to be doing science to me.
They may be on a fools errand and chasing something that doenst exist or that does exist but we can never detect, but maybe they will get lucky and make a definitive detection.
I think we should wait and see rather than have a priori assumpotions.

bapowell

Absolutely agreed. I don't mean to suggest that I have an a priori assumption that multiple universes don't exist. I simply don't know, and welcome any evidence that supports their existence. My point was simply that the anthropic principle -- which depends on the existence of a multiverse -- is not a scientific proposal simply because we currently lack evidence of this existence.

marcus

I think that is quite a reasonable position to take and that Steinhardt could agree with you!
He opposes making unscientific assumptions. But he does not, himself, assume that patches of the universe with other values of what we call physical constants do NOT exist. He does not need that assumption.

Just an undirected comment: the reductionist program (call it Baconian if you like, in honor of the Elizabethan gentleman sometimes accused of writing Shakespeare's plays) has never been to explain why existence exists---only why it works the way it does.

To find the simplest natural explanations for as much as you can, for the time being, proposing no theory unless it can be tested, assuming nothing for which there is no evidence. And when that is done, look for the explanation behind THAT explanation.

Skydive, try looking back at the closing paragraphs of Steinhardt's statement and see if they are inconsistent with what Brian Powell just said.

I don't think they are inconsistent! In either case it is the open minded reductionist faith: you don't have to give up asking why, and you don't have to make stuff up.

The world is not accidental. At every step, there are reasons. You may not be able to answer all the questions at once, or why existence exists, but there is always one further deeper explanation of how it works. We are not yet ready to give up on the reductionist program. There is no evidence that we need to give up yet. I think that's Steinhardt's message.

skydivephil

Yes but surely the point of doing sciencve is to try and find that evidence. What the point in just cofnirming what we already know? Well ok we should always keep testing our theories to destruction, but no one won a Nobel prize for ther next decimal place confirmatrion of an existing thoery.

I agree we currently do not have sufficient evidence to claim the multiverse exist and i think those that say its the only option are wrong. However there seems to be hint of a multiverse in our current scientific picture.
I see this in that
1) inflation has good evidence in favour of it.
2) inflation as descirbed by both its main supporters: Guth, Linde, Vilenkin etc and itsdmain detractors Steinhardt, Turok etc is eternal and hence implies a multiverse.

Now as Marcus has pointed out before, inflation may not be eternal. Furthermore maybe inflaiton will faill at the final hurdle (detection of B mdoe/primordial gravitational waves) and turn out ot be a failed theory. But with the evidence so far in favour of it (inflation) it seems to me that whilst we shouldnt accept the mutliverse as true, we shouldnt equate it to pseudo science as some people have done. It seems to me that in the classification of whats science and what isnt there is a grey area here.

Chalnoth

They are an unambiguous consequence of the nature of quantum mechanics. You can get rid of them, of course, if you assume by fiat that they do not exist. But there is no reason whatsoever to make that assumption.

And the model in which these other branches occur, by the way, makes some very specific predictions about the nature of wave function collapse, predictions that have been tested:
http://vigo.ime.unicamp.br/~jc/p4887_1.pdf

Basically, requesting that we can only test a model in certain, specific ways is irrational and itself extremely unscientific. Any decent model makes a wide variety of predictions, and it is not in any way required that all of those predictions be testable. Demanding that one specific prediction be testable, when there are other tests that can be done instead, is simply refusing to engage in a critical examination of the idea.

bapowell

Sure, these are good candidates (notwithstanding Marcus' reasonable objection that inflation might not be eternal). As is quantum mechanics which Chalnoth mentioned earlier. But we can't observe these other regions of the universe. In order for anthropic reasoning to be successful, we must observe them, and we must learn about the distribution of these spaces -- what their properties are and so forth. Until we do this, all we have are suppositions. These might be based on well-tested theories, and these theories might even require the existence of a multiverse in order to be consistent. This would constitute a strong prediction for the existence of a multiverse. But a prediction requires evidence for its validation. My contention is with this latter point: we don't at present have any evidence for it, and so anthropic arguments are, at present, fundamentally non-scientific.

bapowell

But we're not trying to establish the correctness of a given model, like quantum mechanics. We are trying to understand the nature of one of its predictions. Until you empirically verify the existence of other universes, I'm sorry, they do not correspond to any objective reality. Of course QM makes other predictions, and of course these are sufficient to support the validity of the theory. It sounds like we are arguing two different points. All I'm saying is that it is unscientific to construct a principle based on the details of unobserved phenomena. I'm surprised you won't concede this.

Chalnoth

This attitude of yours is fundamentally anti-science. It is exactly like the creationists complaining that we don't have transitional fossils because we don't have a continuous line of fossils of every lineage. Or that evolution isn't science because it can't be repeated.

Models do not exist in a vacuum. Refusing to believe in a definitive prediction of a model, even though that model has been thoroughly vetted through other means, and even though this other prediction is a natural consequence of the parts of the model that has been vetted, is just plain anti-science.

bapowell

It's not a matter of belief (who sounds like the creationist?) I've stated above (which you didn't read because you were too busy planning your rebuttal) that I am agnostic to the existence of the multiverse because it has not been observed. Once evidence accumulates in favor of it, I will accept it as part of objective reality. <---- THAT is science. To do otherwise is to merely have faith in its existence. Which is precisely what you are doing whether or not you choose to accept it.

Another important point you missed, is that my argument is not so much about the purported existence of the multiverse, but of its character. The anthropic principle presupposes not just its existence, but makes assumptions about its character. Such assumptions I refuse to accept without evidence.

There's a little thing called inductive reasoning. You should google it.

Chalnoth

No, it's a matter of evidence. And refusing to take evidence at face value is the problem here.

bapowell

OK. What evidence, taken at face value or otherwise, can you provide for the existence of mutliple universes?