Does multiverse really explain fine-tuning?

[Garth]

Steinhardt I think contradicts himself because he says you can discriminate between his ekpyrotic model and inflation by observing the gravitational wave spectrum. but if that's true then how does he justify that inflation can't be tested?

The Ekpyrotic universe model predicts no primordial gravitational waves detectable by BICEP2, whereas Inflation in many guises did. If such g-w's had been found that would have falsified Steinhardt's model.

In fact they weren't discovered as we know, BICEP2 picked up a signal from dust and whether there is a weaker g-w signal hidden in there has yet to be determined.

So it might seem that the Ekpyrotic model has won over the Inflationary one, however Inflation is so flexible that one can simply constrain the parameters fed into the theory and produce a weaker or zero prediction of the g-w signal and thus save the theory. Actually we all do this with our theories - it's just that inflation is more flexible than most.

So i think we have good scientific reasons to believe in a multiverse. But clearly that belief should not be at the same level that we believe in black holes. The whole argument is based on a false notion that science should give us definitive answers to questions. I wish the world worked that way, but alas it doesnt.

Indeed, especially when we are scratching away at the boundaries of what is observable.

Garth

 

[Chalnoth]

That statement itself is quite an assumption!

Why do you think this?

If you take the standard model at face-value using the minimal assumptions required to explain the available evidence, then you have a model which unambiguously predicts a multiverse. You can, of course, add assumptions to get rid of the multiverse, but then you're just making unwarranted assumptions without evidence to support them. Maybe you can get rid of the multiverse (in this case, different low-energy physical laws in different locations), but I find that prospect highly unlikely.

Fine tuning that is required for any observers to exist is in itself powerful evidence of fine tuning. The picture here is that if we imagine the full possible configuration space of the universe as given by our current state of knowledge as a sheet of paper, and we then draw a dot to represent the part of that configuration space where life is possible. Those that want to get rid of the multiverse tend to argue that the rest of the paper isn't real: it's just an illusion due to our lack of knowledge, and the real space of possibilities aren't much larger than that small dot.

But the question that I ask is: why should that be the case? Why should the total space of possibilities tailor itself perfectly for life to exist?

I think most theorists expect that a significant fraction of that space of possibilities is impossible, and that knowledge of the correct GUT will demonstrate this. But I don't think there's any reasonable argument to be had that it won't remain much, much larger than the part of the space where life is possible. The cosmological constant is good example here: independent of our current observations, it is impossible for any structure formation if the magnitude of the cosmological constant were more than about $10^{-120}$ in natural units. For a long time, most theorists believed that there must, therefore, be some kind of symmetry that sets the cosmological constant to zero. But that has proven very difficult (for discussion, Sean Carroll described some of the difficulties in his book here).

Granted, the fact that it's hard to write down theories which predict a small or zero cosmological constant doesn't necessarily mean that it can't be true. But there's just no reason to prefer a unique universe as a default assumption. The simplest assumptions, given current knowledge, lead to some kind of multiverse.

 

[palmer eldtrich]

The Ekpyrotic universe model predicts no primordial gravitational waves detectable by BICEP2, whereas Inflation in many guises did. If such g-w's had been found that would have falsified Steinhardt's model.

In fact they weren't discovered as we know, BICEP2 picked up a signal from dust and whether there is a weaker g-w signal hidden in there has yet to be determined.

So it might seem that the Ekpyrotic model has won over the Inflationary one, however Inflation is so flexible that one can simply constrain the parameters fed into the theory and produce a weaker or zero prediction of the g-w signal and thus save the theory. Actually we all do this with our theories - it's just that inflation is more flexible than most.Indeed, especially when we are scratching away at the boundaries of what is observable.

Garth

That's like saying Mars rovers have shown theories that life is unique on Earth have won over theories that imply life in the universe is not unique to Earth. Technically true but hardly impressive. It's hard to find life outside of Earth even if its there. Its hard to find gravity waves even if they are there. I spoke to a number of people that worked on Planck and they said they had very little expectation of finding the b mode ( although they hoped they might get lucky ) so not finding it was not really a big shock and I don't think you should read too much into it.

 

[Chronos]

The multiverse hypothesis is filled with paradoxes, as deep and varied as the number of metastable vacuum states.

 

[Chronos]

Under a multiverse scenario, such as that proposed by Tegmark, you cannot forbid causal connections between different 'universes'. So there must surely be other universes that are observationally accessible to us. So, to quote Fermi, where are they?

 

[ShayanJ]

One of the technical issues here is called the "measure problem".
https://www.quantamagazine.org/20141103-in-a-multiverse-what-are-the-odds/

I don't understand this part:

Proponents of the multiverse idea must show that, among the rare universes that support life, ours is statistically typical.

Why? I don't think that's necessary. Our universe doesn't need to be typical. In an infinite number of universes with all possible values of the constants, its enough to have one that supports life and we can be in that one. Why does it need to be typical?

 

[ShayanJ]

And about accepting or rejecting the multiverse hypothesis. Its not that there can be two types of physicists:proponents of multiverse, opponents of multiverse. There are actually three: proponents of multiverse that accept it as a final answer, proponents of multiverse that this hypothesis makes sense to them but still don't stop searching, opponents of multiverse.
I think the serious objection(from Garth and others) is to the first group. And chalnoth's objection should be toward the third group. But I think most people are actually of the second group, which seems a very reasonable viewpoint to me. I mean, the multiverse hypothesis makes sense, but that doesn't mean we should stop searching. We continue our search till either there is good experimental evidence for multiverse or we find another solution.

 

[Garth]

Nope. The multiverse is the absence of an assumption.

That statement itself is quite an assumption!

Why do you think this?

Because we see just one universe and a sky full of a microwave background that we try to explain with incomplete physics.

When we have a quantum gravity theory and know what the Inflaton, the Dark Matter particle and Dark Energy are then we may know what we are talking about.

Until then we hypothesise and make untested (and perhaps untestable) assumptions.

Garth

 

[newjerseyrunner]

Under a multiverse scenario, such as that proposed by Tegmark, you cannot forbid causal connections between different 'universes'. So there must surely be other universes that are observationally accessible to us. So, to quote Fermi, where are they?

Perhaps in the microwave background radiation, perhaps gravity is weak because it seeps into other universes, maybe dark energy appears to be coming from no where because it's leaking from another universe. Until data contradicts a possibility, that possibility must be considered valid, no matter how unlikely.

Aren't there many different types of multiverse theory? I've heard theories where our universe being a small pocket where the laws of physics are what they are, but as you get further away the just sort of change, we just happen to be in a pocket where everything looks the same. I've also heard theories that our universe may exist within a black hole of a bigger universe and fine tuning is explained by evolution, natural selection favors universes that make a lot of black holes and evolve through recursion. Along lines of the recursion, I've heard ideas in my own field that natural selection prefers universes that produce intelligent life and life is somehow responsible for making more universes (simulation hypothesis.)

 

[Chalnoth]

That's like saying Mars rovers have shown theories that life is unique on Earth have won over theories that imply life in the universe is not unique to Earth. Technically true but hardly impressive. It's hard to find life outside of Earth even if its there. Its hard to find gravity waves even if they are there. I spoke to a number of people that worked on Planck and they said they had very little expectation of finding the b mode ( although they hoped they might get lucky ) so not finding it was not really a big shock and I don't think you should read too much into it.

Yup. Simulations put Planck's 2-sigma detection limit right at the edge of the feasible range for the B-mode signal, so that even if Planck had detected it, it likely wouldn't have been a very significant detection.

 

[Chalnoth]

Perhaps in the microwave background radiation, perhaps gravity is weak because it seeps into other universes, maybe dark energy appears to be coming from no where because it's leaking from another universe. Until data contradicts a possibility, that possibility must be considered valid, no matter how unlikely.

I don't think so. There are an infinite number of possibilities to investigate, so it's very worthwhile to have some ideas of the kinds of things to investigate even before there is any evidence to guide the way.

A pretty good rule of thumb that is frequently used is Occam's Razor: examine the number of postulates of the theories, and the one with fewer postulates should be investigated first. This is one reason why string theory remains popular among a number of theorists: it has very few degrees of freedom in the base theory, and it predicts quantum gravity. It's hard to say how likely it is that this theory describes the world around us, but this is solid ground to at least consider the theory worth investigation.

On the other end of things, there are a number of theoretical ideas that need a number of complicated mechanisms added to them in order for them to work out. For example, if you have a theory that includes tachyons, then any interaction at all between those tachyons and normal matter results in instability (basically, the tachyons cause the universe to explode and never stop exploding).

Aren't there many different types of multiverse theory? I've heard theories where our universe being a small pocket where the laws of physics are what they are, but as you get further away the just sort of change, we just happen to be in a pocket where everything looks the same. I've also heard theories that our universe may exist within a black hole of a bigger universe and fine tuning is explained by evolution, natural selection favors universes that make a lot of black holes and evolve through recursion. Along lines of the recursion, I've heard ideas in my own field that natural selection prefers universes that produce intelligent life and life is somehow responsible for making more universes (simulation hypothesis.)

Yes. Max Tegmark produced a concise classification, as well as proposing his own new multiverse concept:
http://space.mit.edu/home/tegmark/PDF/multiverse_sciam.pdf

Basically, they boil down to:
1. Different initial conditions.
2. Different low-energy physical laws.
3. Different quantum realizations.
4. Different mathematical structures (this is the one that Tegmark proposed).

 

[mfb]

Under a multiverse scenario, such as that proposed by Tegmark, you cannot forbid causal connections between different 'universes'. So there must surely be other universes that are observationally accessible to us. So, to quote Fermi, where are they?

Why? I cannot forbid that your door bell rings while you are reading this post. Does it imply it has to happen?

 

[Chalnoth]

Why? I cannot forbid that your door bell rings while you are reading this post. Does it imply it has to happen?

It seems to be the case in quantum mechanics that that which is not expressly forbidden necessarily occurs.

However, there are two facts that limit this:
1. With a complex theory, it isn't always clear exactly what is and isn't possible within that theory. It isn't as simple as, "I can't think of a way in which this theory prevents it, therefore it must occur." You have to do more work to show that it really is possible within the theory.

2. Just because something is possible within the theory doesn't mean it's at all likely. If you're going to attempt to tie it to observation, you have to actually show that the effect of interest is expected to be observed.

 

[mfb]

It seems to be the case in quantum mechanics that that which is not expressly forbidden necessarily occurs.

Only with MWI. And not even there in every world.

2. Just because something is possible within the theory doesn't mean it's at all likely. If you're going to attempt to tie it to observation, you have to actually show that the effect of interest is expected to be observed.

Exactly, and that's what I was missing in your post.

 

[Chalnoth]

Only with MWI. And not even there in every world.

I was talking about Gell-Mann's Totalitarian Principle, described here:
https://en.wikipedia.org/wiki/Totalitarian_principle

 

[rootone]

Does this mean that everything that is not expressly forbidden is compulsory?

 

[Chalnoth]

Does this mean that everything that is not expressly forbidden is compulsory?

This rule seems to be accurate when calculating the probability of various interaction outcomes in quantum field theory. Some might argue that this isn't "real", it's just a mathematical tool that is used to calculate those probabilities, but it's at least suggestive.

 

[Chronos]

I'm nor suggesting logic trumps the scientific method, but, I do suggest science without logic is a mess.

 

[Haelfix]

The Ekpyrotic universe model predicts no primordial gravitational waves detectable by BICEP2, whereas Inflation in many guises did. If such g-w's had been found that would have falsified Steinhardt's model.

In fact they weren't discovered as we know, BICEP2 picked up a signal from dust and whether there is a weaker g-w signal hidden in there has yet to be determined.

So it might seem that the Ekpyrotic model has won over the Inflationary one, however Inflation is so flexible that one can simply constrain the parameters fed into the theory and produce a weaker or zero prediction of the g-w signal and thus save the theory. Actually we all do this with our theories - it's just that inflation is more flexible than most.
Garth

I explained last time in the other thread about this subject that this was an attribution fallacy. Essentially comparing apples to oranges. After Bicep2, Neil Turok was busy writing down Ekyrprotic models that fit the new values of r. Of course that if you compare a specific model (the simplest version of the Ekyprotic universe) with an idea (all possible inflationary models), you will find that the former is more predictive than the latter.

What really matters in this business is model independent predictions. Inflationary theory has 5 or 6 nontrivial predictions, and every single one of them (other than a g-w spectrum) has been verified. This includes very specific shapes of the CMB, which everyone in the field thought was crazy talk when it was first proposed. Of course what happened was all new ideas were tempered on these new constraints after experiment picked them out and not surprisingly there is still a large amount of wiggle room and we are still left with a multiplicity of options that have to be tested/constrained and eventually thrown out (and it is conceivable that we might never be able to throw out all possibilities).

What is true, is that when you analyze in details what's left standing, the existence of something like a multiverse is 'generic'. This is particularly powerful, considering that most of the best (for different phenomenological reasons) and simplest models that are left in this group typically do include multiverses as an outcome.

 

[Garth]

What is true, is that when you analyze in details what's left standing, the existence of something like a multiverse is 'generic'. This is particularly powerful, considering that most of the best (for different phenomenological reasons) and simplest models that are left in this group typically do include multiverses as an outcome.

Except that outcome has arisen from incomplete physics which does not have a quantum gravity theory and that does not know what the Inflaton or Dark Matter, or Dark Energy are.

The extrapolation of theory into the infinite landscape of string and inflation hypotheses may be nothing more than the outcome of our ignorance.

The problem, as I have said several times, is this approach may actually dissuade further investigation into any actual physical causes of the phenomena that may be there to be found - and to be found in this universe, the only one that we have actually observed.

Garth

 

[Chalnoth]

Except that outcome has arisen from incomplete physics which does not have a quantum gravity theory and that does not know what the Inflaton or Dark Matter, or Dark Energy are.

The extrapolation of theory into the infinite landscape of string and inflation hypotheses may be nothing more than the outcome of our ignorance.,

Maybe, but there is absolutely no reason to think this. And how would ruling out a class of models dissuade investigations into physical causes? Are multiverse models somehow unphysical?

 

[Haelfix]

Except that outcome has arisen from incomplete physics which does not have a quantum gravity theory and that does not know what the Inflaton or Dark Matter, or Dark Energy are.
The extrapolation of theory into the infinite landscape of string and inflation hypotheses may be nothing more than the outcome of our ignorance.
Garth

Without a doubt there is much more to be discovered, but note that one of the great theoretical virtues of inflation is that (unlike its competitors) it arises from the physics of weak coupling. In other words we can reliably make many statements (but not all) about what it predicts, without knowing too much about messy unknown details of what happens in the deep UV (it is sensitive to these details, but not so much that we can't still make reliable predictions about certain aspects) This is one of the reasons it is so deeply studied.

Now, of course, if you asked me whether this is the final say on the matter I would tell you probably not, but nevertheless this is what we have, its quite successful so let's go with it and see where it leads, rather than throwing the baby out with the bathwater. As far as I'm concerned the question of whether the multiverse exists or not, is basically a spectator that comes along for the ride of this much deeper question.

 

[Garth]

Maybe, but there is absolutely no reason to think this. And how would ruling out a class of models dissuade investigations into physical causes? Are multiverse models somehow unphysical?

'Physical causes' in the sense that such hypotheses can be tested by experiment.

I'm not ruling out a class of models, I just find explanations unsatisfactory when they explain global properties of this universe by invoking others.

Garth

 

[Chalnoth]

'Physical causes' in the sense that such hypotheses can be tested by experiment.

I'm not ruling out a class of models, I just find explanations unsatisfactory when they explain global properties of this universe by invoking others.

And many multiverse models are testable. Your point?

 

[Garth]

And many multiverse models are testable. Your point?

Testable? Exactly how?

Garth

 

[Chalnoth]

Testable? Exactly how?

Garth

Testable. If anybody was claiming that a specific multiverse model were true, then you might have a point.

 

[Garth]

So are you saying that only specific models are testable and that the 'multiverse paradigm' itself is not?

We have an questioned theory of Inflation that predicts a multiverse.

As I said above while there are crucial and pertinent elements of fundamental physics still unknown such a prediction may well be just the outcome of an extrapolation of our ignorance.

Garth

 

[Chalnoth]

So are you saying that only specific models are testable and that the 'multiverse paradigm' itself is not?

Of course. It's typically not possible to test something that isn't specific, and the term "multiverse" isn't specific.

We have an questioned theory of Inflation that predicts a multiverse.

Even without inflation, if you have a theory of high-energy physics that includes any spontaneous symmetry breaking, you have a multiverse.

As I said above while there are crucial and pertinent elements of fundamental physics still unknown such a prediction may well be just the outcome of an extrapolation of our ignorance.

Maybe, but that's hardly a convincing argument. High-energy physics theories are pushing in the opposite direction, towards more types of spontaneous symmetry breaking. Sure, it's conceivable that this is misleading, that a deeper understanding of high-energy physics will validate notions of the universe that were popular before we started delving into high-energy physics, but why should we expect that to be the case?