What Could Cause a Big Crunch in the Expansion of the Universe?

[Chalnoth]

No evidence of what?

I am --- or was --- a trial lawyer. I really like evidence, but it must be evidence of something.

I believe the context made the point clear in the post where I made the statement. I assume you're talking about the text that apeiron quoted above.

 

[marcus]

All Apeiron has done, that I can see, is raise the issue of fine-tuning. But tuning for what?

While rejecting god style answers, fine tuning would be a legitmate kind of evidence here. ...<snip>...

But what follows from that? You can either go the god route - look for an intentional creator. Which is obviously unsatisfactory for so many reasons.

Or you can instead look for creating processes. ...

I would agree that the parameters of cosmology (e.g. governing its expansion history) and the parameters of particle physics (e.g. governing nuclear and chemical reactions) appear finely adjusted so as to produce a lot of stars.

If anybody here wants to point to parameter fine-tuning to produce an abundance of stars, I would acknowledge that and admit that it is a legitimate problem, can we explain it?

===
If U had started off expanding much faster, then stars and galaxies would never have had the chance to condense---it would just be an increasingly dilute expanding gas. Expansion must not overwhelm gravity, if we are to have stars. On the other hand if gravity had been too strong or expansion too slow the universe would have collapsed prematurely.

Stable nuclei of elements heavier than hydrogen are necessary for fusion to occur. Supernovae promote star formation (Chalnoth will know about the role of density waves in starforming regions.) Stable chemical elements such as carbon and oxygen faciliate the condensation of overdense regions by radiating off excess energy.

A small percentage change in several of the basic cosmo and particle physics parameters would apparently make stars less abundant.

I am not suggesting that we have a global max, only a local optimum. The parameters do not necessarily represent global optimality, they are merely well enough tuned so that any small change you can imagine would take us down hill in the star-making business, or in any case would at least not improve things (wouldn't have made stars more abundant.)

I see lots of stars, but few signs of intelligence. Indeed few signs of any form or level of life at all. The straightforward conclusion is that if there's any tuning, it's for stars.

And maybe that's wrong. It's just a conjecture. Maybe someone can find a counterexample.

If you grant this apparent fine-tuning, then we could argue that it requires us to look for a natural mechanism to explain it. What naturalistic unintentional process could have caused our parameters to be favorable to stars?

According to the well-worn established traditions of science, we ought first to exhaustively examine any natural unintentional mechanistic explanations---which hopefully will be empirically falsifiable or in some way adequately testable by observation.

 

[LtDan]

Chalnoth, when you use the phrase, "such a hypothesis doesn't help us uncover the process by which our universe came about," you at least hint at the idea of creation. It makes one tend to believe that you believe that "our universe" is not infinitely old and that we are not wasting our time trying to find out how the "coming about" occurred.

As far as finding out how our solar system or galaxy were created, I think we are now in the process of finding those things out, and it doesn't appear, so far, that anything or anyone created them on purpose.

 

[Chalnoth]

Chalnoth, when you use the phrase, "such a hypothesis doesn't help us uncover the process by which our universe came about," you at least hint at the idea of creation. It makes one tend to believe that you believe that "our universe" is not infinitely old and that we are not wasting our time trying to find out how the "coming about" occurred.

Well, the region of the universe which we inhabit is certainly not infinitely old. It's 13.7 billion years old (plus or minus a couple hundred million). That's what I was referring to when I put the qualifier "our" before universe.

As far as finding out how our solar system or galaxy were created, I think we are now in the process of finding those things out, and it doesn't appear, so far, that anything or anyone created them on purpose.

Well, that's good. Because the evidence is manifestly in the other direction.

 

[apeiron]

All Apeiron has done, that I can see, is raise the issue of fine-tuning. But tuning for what?
.

There would be fine-tuning for neutrons and CP asymmetry and much else too. Even to find just three spatial dimensions - out of a potential infinity, that put things in a small corner of probability space.

So a satisfactory answer I feel would be a very general constraints principle, not some particular story to explain some chosen level of organisation (like stars). But of course, consideration of particular examples would help in the formulation of those general principles.

My point was that this kind of work has already been done in an at least handwaving way. It is called systems science.

Well, my first point was that people like Chalnoth should not simply dismiss the religious as though conventional bottom-up science already has all the answers. There are things like purpose, selection, meaning that bottom-up approaches in principle cannot explain. God is a mystical way of providing a matching top-down causality. But science can hope to provide a proper top-down model of causality. That is the correct way to remove gods and other mystical ideas from the intellectual stage.

Note that it is just as mystical when a "scientist" says finetuning "just is", or human consciousness and intentional bios "just is". I realize this has become enshrined in physics as the doctrine of positivism. Thereof what we cannot speak, we must remain silent, etc. Thankfully not all scientists actually stick to that narrow-minded doctrine.

 

[marcus]

Apeiron thank you for answering my first main question. Later I will remind you of the second, which I put in bold. But for now you just replied to the first:

All Apeiron has done, that I can see, is raise the issue of fine-tuning. But tuning for what?
..

There would be fine-tuning for neutrons and CP asymmetry and much else too.
... a very general constraints principle, not some particular story to explain some chosen level of organisation (like stars). ...
... has already been done in an at least handwaving way. It is called systems science.
...That is the correct way to remove gods and other mystical ideas from the intellectual stage.
...

Here is my second question, if you would be so kind as to address it:

I see lots of stars, but few signs of intelligence. Indeed few signs of any form or level of life at all. The straightforward conclusion is that if there's any tuning, it's for stars.

If you grant this apparent fine-tuning, then we could argue that it requires us to look for a natural mechanism to explain it. What naturalistic unintentional process could have caused our parameters to be favorable to stars?

I think you very likely know one famous answer that has been given to this question---which has still not been refuted. One simply remarks that many stars end up forming black holes. One could equally well formulate the hypothesis that our point in parameter space is a local optimum for producing black holes (rather than stars).

This local optimality would evolve naturally if a black hole sufficiently often causes a fresh expanding region with nearly the same cosmo and particle parameters---the system will evolve towards greater reproductive fitness.

This was the answer Smolin gave in 1992 and it makes some testable predictions. The main overall prediction is that no one will be able to exhibit a small change in any of the fundamental parameters which would have resulted in a universe with more abundant black holes.
The argument was spelled out in detail as to why it was plausible that no one would ever be able to point to a suboptimality of any of the parameters. I recapitulated some of the main points in the argument here:

If U had started off expanding much faster, then stars and galaxies would never have had the chance to condense---it would just be an increasingly dilute expanding gas. Expansion must not overwhelm gravity, if we are to have stars. On the other hand if gravity had been too strong or expansion too slow the universe would have collapsed prematurely.
Stable nuclei of elements heavier than hydrogen are necessary for fusion to occur. Supernovae promote star formation (... the role of density waves in starforming regions.) Stable chemical elements such as carbon and oxygen faciliate the condensation of overdense regions by radiating off excess energy.

Setting aside for a moment the controversy you are having with Chalnoth, maybe you could answer my question. What do you think of this reproductive hypothesis as a natural evolutionary mechanism which would lead to regions of the universe having parameters which appear to be fine-tuned for the abundant production of stars?

This is a simple idea and the local optimum conjecture is falsifiable and has survived scrutiny (Alex Vilenkin tried unsuccessfully to refute it.) It has been treated in several books, and another one (by Smolin and Unger) is in preparation. Maybe this idea deserves to be on your system science agenda.

 

[Chalnoth]

There would be fine-tuning for neutrons and CP asymmetry and much else too. Even to find just three spatial dimensions - out of a potential infinity, that put things in a small corner of probability space.

This isn't true, though. We don't know the probability space.

So a satisfactory answer I feel would be a very general constraints principle, not some particular story to explain some chosen level of organisation (like stars). But of course, consideration of particular examples would help in the formulation of those general principles.

I think it's unreasonable to expect that any but a tiny minority of the regions of the universe end up being conducive to life like ours is.

My point was that this kind of work has already been done in an at least handwaving way. It is called systems science.

As near as I can tell, systems science is about the behavior of complex interconnected networks, and has nothing to do with the nature of fundamental law.

Well, my first point was that people like Chalnoth should not simply dismiss the religious as though conventional bottom-up science already has all the answers.

It is completely ridiculous to claim that science has all the answers. If it did, nobody would be doing science any more! But the fact is that science is the only method we have to discover any answers to questions about the nature of reality.

There are things like purpose, selection, meaning that bottom-up approaches in principle cannot explain. God is a mystical way of providing a matching top-down causality. But science can hope to provide a proper top-down model of causality. That is the correct way to remove gods and other mystical ideas from the intellectual stage.

This makes no sense to me.

 

[apeiron]

What do you think of this reproductive hypothesis as a natural evolutionary mechanism which would lead to regions of the universe having parameters which appear to be fine-tuned for the abundant production of stars?

It is certainly a powerful idea that takes its causality from conventional biology and would manage to explain an awful lot about the particular make-up of our universe. The major issue of course is with the plausibility of the black hole spawning mechanism itself - the whiting out "the other side" in infant big bangs. What's the current view on that?

And then to be evolution-like, the black holes would have to spawn the new worlds with constrained variety. Like DNA, it would not be just "reproduction with errors" - the old hopeful monsters view of Darwinism. Instead, like DNA, we might expect variation with a bounded gaussian mean. So enough variety for a species to explore an environment, but not so much that it cripples the next generation of organism.

I'm not sure anyone has ideas of how black/white holes would spawn in this constrained fashion. It would seem that they would either spawn clones (no variety) or spawn properly random variants (which would obey unconstrained powerlaw statistics rather than constrained gaussian variation - damped exponentially so the new generation is not thrown too far off path).

But such caveats aside, I would agree that if it could be shown the universe is indeed the result of an evolution towards black-hole producing regimes, then that would account for stars as you say. And if you have stars, then planets and life come along with that, needing no further special explaining.

Neither life nor mind are troublesome features of a sufficiently complex universe in my view (and neuroscience/mind science was my speciality). So getting to universes which create supernovas gets you also far enough for planets, life and mind to be considered merely further evolutionary complication.

In modern theoretical biology, the second law of thermodynamics, the maximum entropy production principle, and the theory of dissipative structures, would be enough to explain the existence of life and mind - their intentionality - in natural terms. As opposed to supernatural. Order arises naturally to dissipate entropy gradients. And we know, for example, that conscious humans are champion planetary degraders! Just look at the speed we will blow the world's fossil fuels.

However, granting you that stars and bios would happily follow from a multiverse capable of constrained (fine-tuned!) black hole spawning, so dealing with a lot of the parameter finetuning issues, there would still remain the bigger questions about how the whole shebbang itself arises.

And here we would have to pull back from merely evolutionary mechanism to talk about the more general causality of developmental processes.

 

[Chalnoth]

Setting aside for a moment the controversy you are having with Chalnoth, maybe you could answer my question. What do you think of this reproductive hypothesis as a natural evolutionary mechanism which would lead to regions of the universe having parameters which appear to be fine-tuned for the abundant production of stars?

Personally let me just say that I think it rather unreasonable to expect that there is any fundamental feature of the universe that would be such that life-bearing regions of the universe like our own should be common.

It therefore seems to be unlikely that any research into physical process that might increase the frequency of regions which are conducive to life is highly unlikely to be fruitful. This is further complicated by the weak anthropic principle, which demonstrates that we can't actually make statements about the nature of things we can't observe without first noting the simple fact that observers can only possibly observe environments conducive to their development and survival, indicating that it is impossible to deduce anything whatsoever from the mere fact that our region of the universe is conducive to our survival, other than the plainly-obvious fact that the probability of generating such a region must be greater than zero.

 

[apeiron]

This isn't true, though. We don't know the probability space.

.

Unless you know good a priori reasons to constrain the number of dimensions then we can start with the observation that 3 is an awful lot less than an infinite number.

As near as I can tell, systems science is about the behavior of complex interconnected networks, and has nothing to do with the nature of fundamental law.

.

Where did you learn this from?

It is completely ridiculous to claim that science has all the answers. If it did, nobody would be doing science any more! But the fact is that science is the only method we have to discover any answers to questions about the nature of reality.
.

I said bottom-up science. As opposed to science that includes downward causality.

This makes no sense to me.

I've gathered that.

 

[Chalnoth]

Unless you know good a priori reasons to constrain the number of dimensions then we can start with the observation that 3 is an awful lot less than an infinite number.

The point is that we don't know the nature of the fundamental laws of the universe, and therefore cannot make any statements as to the probability of things like the probability distribution of the observed number of dimensions.

Where did you learn this from?

"Systems science" was an entirely new term to me, so I looked it up on Wikipedia. Unless you're talking about something completely and utterly different from what is described there, then systems science has nothing to do with the early universe or the nature of fundamental law.

I said bottom-up science. As opposed to science that includes downward causality.

And what do you mean by those two phrases?

 

[apeiron]

The point is that we don't know the nature of the fundamental laws of the universe, and therefore cannot make any statements as to the probability of things like the probability distribution of the observed number of dimensions.

But we can count the number of dimensions we see. And we have notions like zero and infinity. So of course we have a starting point for talking about the improbability of just three spatial dimensions as something that actually exists. It becomes another example of fine-tuning. And many have accepted it as such.

"Systems science" was an entirely new term to me, so I looked it up on Wikipedia.

Well if you've read a whole wiki page on something...

Here is a decent starter on downward causation from Paul Davies...
http://www.ctnsstars.org/conferences/papers/The%20physics%20of%20downward%20causation.pdf

Here is a not too scary introductory site on systems approaches...
http://pespmc1.vub.ac.be/CYBSWHAT.html

And its page on downward causation...
http://pespmc1.vub.ac.be/DOWNCAUS.HTML

And it is worth thinking about the condensed physics lobby (who talk about the bottom up emergence of organising laws, which again is a less scary way of talking about top-down constraints)...
http://waitaki.otago.ac.nz/~martin/Documents/Theory of Everythihng - NYTimes-Dec01.pdf

 

[LtDan]

If the universe is indeed infinite --- and I believe it is --- then probability essentially becomes meaningless.

 

[apeiron]

If the universe is indeed infinite --- and I believe it is --- then probability essentially becomes meaningless.

The issue is not about the infinite extent/duration of our 4D universe but the fact that 4D is a remarkably particular subset of infinite-D, the set of all possible dimensionalities.

What are the chances that there would be so few if we were to pick some random number between 0 and infinity? Even if we just take a Bayesian approach, we can see how this is a fine-tuning question.

Another way to look at it is why is our universe so homogeneously 4D. Why does it not break out into patches of 670002-D here, 8887-D there. Or fractional dimensions even.

Something must be stabilising the outcome. Or the outcome must be self-stable for "emergent" reasons (emergence being code for downward causation).

 

[marcus]

Chalnoth your post #79 appears addressed to me, thanks for the reply. However you didn't get the point of my posts #72 and #76. I was not talking about parameters being fine-tuned so as to be conducive to life. I said nothing about life. You mentioned the "weak anthropic principle" but nothing in my posts was intended to relate to that. So you seem to have missed the point and might want to read those two posts again. I would appreciate your informed comment!

==========

Apeiron, thanks for your response, and for coming to grips with the main points I was making! There are quite a few papers on black hole bounce. Various quantum gravity models of black hole collapse, in some of which there results another tract of spacetime but in other cases things go wrong and a wrong dimensional kind of crippled spacetime occurs. This work is in preliminary stages, has been going on since about 2006.
Names are Leonardo Modesto, Dah-wei Chiou, Kevin Vandersloot, Christian Boehmer. In case anyone wants to do an arxiv search for black hole bounce papers.
But it is almost not worth it. Too new. However there was a workshop on LQG black hole models and related stuff this spring.

It is certainly a powerful idea that takes its causality from conventional biology and would manage to explain an awful lot about the particular make-up of our universe. The major issue of course is with the plausibility of the black hole spawning mechanism itself - the whiting out "the other side" in infant big bangs. What's the current view on that?

And then to be evolution-like, the black holes would have to spawn the new worlds with constrained variety. Like DNA, it would not be just "reproduction with errors" - the old hopeful monsters view of Darwinism. Instead, like DNA, we might expect variation with a bounded gaussian mean. So enough variety for a species to explore an environment, but not so much that it cripples the next generation of organism.
...
...

But such caveats aside, I would agree that if it could be shown the universe is indeed the result of an evolution towards black-hole producing regimes, then that would account for stars as you say. ...

Yes! I have highlighted one of your sentences Apeiron. I am glad you agree.

I think a most important point to make is that what we now have is a testable empirical hypothesis that has nothing to do with plausibility of black hole bounce.

One can simply ask are the parameters tuned for black hole abundance? As a local optimum.

The hypothesis challenges you to find a small adjustment in the parameters of the standard particle model, or of cosmology, which if it could have been made would have resulted in a greater BH abundance. If you can think of one, this refutes the hypothsis of local BH optimality.

This has nothing to do with a possible explanation. The question is simply "are the numbers a local optimum, or not?"

If, after appropriate scrutiny and testing, the hypothesis is found tenable, then the question is how to explain it. And then one could talk about the plausibility of various reproductive mechanisms and various ideas of evolution. But first there is the bare hypothesis of local optimality. That a small percentage change would not have increased abundance.

Ways of deriving predictions from the hypothesis, and testing it by astronomical observation are discussed here.
http://arxiv.org/abs/hep-th/0612185
"The status of cosmological natural selection"
This may be too condensed and abstract, there may be earlier papers that explain the testing strategies better. I will see if I can suggest some later.
(Apeiron, I suspect you may already be familiar with this, but we should put some links in case anybody else reading this thread is curious to know more.)

 

[Chalnoth]

True, you didn't, Marcus. But my point was that any such discussion of fine tuning requires consideration of the weak anthropic principle. I suppose I just assumed that's why you were talking about the apparent fine tuning of the universe to produce stars.

My point is that stars appear to be necessary for the formation of life. As such, we can't honestly be surprised by the existence of stars, because without them there would be nobody here to observe the universe. For the same reason that we can't be surprised we don't find ourselves on the surface of Mercury rather than Earth.

 

[marcus]

But my point was that any such discussion of fine tuning requires consideration of the weak anthropic principle.

I disagree. The whole point of Smolin's discussion is that it does NOT refer at any point to life.

I suppose I just assumed that's why you were talking about the apparent fine tuning of the universe to produce stars.

No, that was not a correct assumption. What I said didn't have anything to do with life.

My point is that stars appear to be necessary for the formation of life.

I don't want to talk about that because it is irrelevant to the logic of what I'm saying, as will become clear to you if you think about it. I don't know what life is. I don't worry about it. And I view it as an accidental side effect or byproduct of whatever processes caused the universe to be the way it is.

We need to separate off the life issue and ignore it.
Then we can focus on the core issue which is whether or not the 30-some numbers which are the parameters of cosmology and the standard particle model are well-adapted for black hole abundance. (Not star abundance per se, black hole abundance.) If they are then that is very interesting. If not, too bad---we just forget about it.

From the paper I linked, if interested you can get some of the discussion, predictions, and "postdictions"---regarding tuning for fusion, stable elements, heavy element chemistry, and the collapse of sufficiently massive neutron stars to form black holes.

The hypothesis of black hole optimality was put forth around 1992 together with a falsifiable prediction concerning neutron star masses, which has not yet been falsified.

 

[apeiron]

And I view it as an accidental side effect or byproduct of whatever processes caused the universe to be the way it is.

We need to separate off the life issue and ignore it.
Then we can focus on the core issue which is whether or not the 30-some numbers which are the parameters of cosmology and the standard particle model are well-adapted for black hole abundance. (Not star abundance per se, black hole abundance.) If they are then that is very interesting. If not, too bad---we just forget about it.

.

There would be a strong body of opinion now that life and stars can both be seen as consequences of the second law of thermodynamics - more particularly, as entropy degraders or dissipative structures. So both would be equally natural to the Universe in this respect.

I understand what you mean here though. If black hole production is the attractor for universe production, then everything else can be viewed as entrained to this evolutionary process. It controls the standard model parameters (which physicists want to explain) and higher levels of dissipative structure, like life, arise incidentally. They are part of the picture, but their existence is not necessary.

Stars then become the right scale of focus because that is the scale of the black holes that get produced.

So this is a consistent stance. It passes the "not even wrong" test as something that could be wrong.

Personally I am not optimistic because I cannot see how black holes can spawn infant big bangs. My way of looking at spacetime fabric does not suggest that it could "rip" locally in such a fashion.

I also cannot see an easy way to get white hole spawning with gaussian (constrained) variation. In biology, that of course was the big issue eventually solved by DNA. So far, this only appears to be covered by friendly handwaving by Smolin and his crew. But it has been a few years since I ready any relevant work on this.

A third issue is that - taking up the thermodynamic/dissipative structure approach to explaining biology - the whole spawning multiverse would probably have to be sliding down some entropy gradient. Smolin's treatment assumed a closed system perspective (each bang wipes the slate and has all the negentropy/energy needed to create another whole universe).

I would take this as a suspect assumption. And perhaps the most difficult hurdle of all. We already discussed this in regard of big bounce cosmologies.

 

[marcus]

And perhaps the most difficult hurdle of all. We already discussed this in regard of big bounce cosmologies.

In that respect you agree with Roger Penrose. He too considers the Second Law to be the main impediment, either to big-bounce or to "little-bounce" cosmologies like what's being considered.

Indeed we did discuss whether it is meaningful to claim that the Second Law holds across a bounce (where presumably ordinary 3D space ceases to exist and there is no clear distinction between geometry and matter) given the discontinuity of observation.

As you know I consider the jury still out on that one: I'm not sure the Second Law applies or is even meaningful in the so-called "quantum regime" at near-Planck density. But people continue to study that kind of thing (both big and now increasingly the little bounce cases)

Today I happened to come across a talk about a toy model where the universe was represented by a graph or network (governed by a Hamiltonian of some simple form) where the idea of temperature was implemented and the qualitative properties of the graph changed with temperature. There was no bounce in this simple case, only an initial condition of starting at high temperature. You might be casually interested. It was at the Emergent Gravity conference at Vancouver British Columbia. I will get a link.


This is the fourth conference on Emergent Gravity (called EG4):
http://www.emergentgravity.org/index.php?main=main_EGIV_about.php


Here is the EG4 Programme:
http://www.emergentgravity.org/index.php?main=main_EGIV_programme.php
Notice the "Networks" topic on Monday afternoon:
14:30 Loll
14:45 Conrady
15:00 Bilson-Thomson
15:15 Rideout

This is the sub-approach where geometry and matter is supposed to emerge from networks. Those are the basic underlying objects and one must guess at a dynamic according to which they evolve. In this case what caught my attention was the talk by Florian Conrady, who has often collaborated with Laurent Freidel in spinfoam work. Here he doing something quite different and, I think, speculative and risky:

"Conrady
Simplicial complexes from condensation
I will talk about work in progress on a statistical model of graphs. The configurations are arbitrary graphs with a fixed number of links. The Hamiltonian is a simple function of the graph that favors the formation of 2d simplicial complexes. Monte Carlo simulations show that at low temperatures the system is in an ordered phase, where the links condense to 2d simplicial complexes. At high temperatures one has a disordered phase and the links form graphs of high connectivity."

Apeiron, as I conceive of this, there is a Hamiltonian that governs the evolution of some graphs and at the "big bang" time when it is very hot there is no regular idea of space because of "high connectivity" meaning almost every point is connected to almost every other--a totally crumpled up chaos, no real space and no real matter. It depicts what Ashtekar calls the "quantum regime" of existence immediately around a bounce.
Then as it cools down it evolves a regular dimensionality, and it looks like a simplicial complex (a more familiar idea of geometry). That seems like an interesting toy model of the emergence of (maybe not gravity but at least) geometry. I offer it only in case you find it amusing or intriguing, not as an answer to your Second Law objection, which at this point cannot I think be satisfactorily answered.

 

[apeiron]

Marcus, here is what Paul Davies said about Smolin's multiverse. A good summary.

Note that Davies recently wrote The Goldilock's Enigma about fine-tuning. And his group of colleages includes Charlie Lineweaver and Tamara Davis. They all broadly take what I would call the systems view of cosmology and are also solid in their science.

Davies from http://arxiv.org/ftp/astro-ph/papers/0403/0403047.pdf

"Another multiverse model has been discussed by Smolin14. He proposes that “baby”
universes can sprout from existing ones via the mechanism of gravitational collapse.
According to the classical picture, when a star implodes to form a black hole, a spacetime
singularity results in the interior of the hole. Smolin suggests that a quantum treatment
would lead instead to the nucleation of a tiny new region of inflating space, connected to
our space via a wormhole. Subsequent evaporation of the black hole by the Hawking
process severs the wormhole, thereby spatially disconnecting the baby universe from
ours. Furthermore, following Wheeler15, Smolin proposes that the violence of
gravitational collapse might ‘reprocess’ the laws of physics randomly, producing small
changes in values of parameters such as particle masses and coupling constants. Thus the
baby universe will inherit the physics of its parent, but with small random variations,
similar to genetic drift in biological evolution. This process could continue ad infinitum,
with baby universes going on to produce their own progeny. It would also imply that our
universe is the product of an earlier gravitational collapse episode in another universe.
Those universes whose physical parameters favoured black hole production, for example
by encouraging the formation of large stars, would produce more progeny, implying that
among the ensemble of universes with all possible variations of the laws of physics, those
universes with prolific black hole production would represent the largest volume of
space."

As a further amusing aside, this black holes story is rather like flu pandemic logic.

Viruses swiftly evolve away from lethality so that they are contagious but not self-eliminating. The first few weeks of the spanish flu saw high mortality, then it followed a flattening curve (which is why so much effort has gone into delaying the spread through the population - even a few weeks can make an evolutionary difference as the virus finetunes its lethality!)

 

[Chalnoth]

I disagree. The whole point of Smolin's discussion is that it does NOT refer at any point to life.

Then that's a serious problem. Because anything about our universe that is required for the existence of intelligent observers is not something we have a right to be surprised about. And stars certainly appear to be a requirement. So we have no right to be surprised about their existence, which makes an attempt to solve the question of why they exist in the first place to be solving a problem that doesn't exist.

 

[oldman]

...Davies from http://arxiv.org/ftp/astro-ph/papers/0403/0403047.pdf

"Another multiverse model has been discussed by Smolin14. He proposes that “baby”
universes can sprout from existing ones via the mechanism of gravitational collapse.
According to the classical picture, when a star implodes to form a black hole, a spacetime
singularity results in the interior of the hole. Smolin suggests that a quantum treatment
would lead instead to the nucleation of a tiny new region of inflating space, connected to
our space via a wormhole. Subsequent evaporation of the black hole by the Hawking
process severs the wormhole, thereby spatially disconnecting the baby universe from
ours. Furthermore, following Wheeler15, Smolin proposes that the violence of
gravitational collapse might ‘reprocess’ the laws of physics randomly, producing small
changes in values of parameters such as particle masses and coupling constants. Thus the
baby universe will inherit the physics of its parent, but with small random variations,
similar to genetic drift in biological evolution. This process could continue ad infinitum,
with baby universes going on to produce their own progeny. It would also imply that our
universe is the product of an earlier gravitational collapse episode in another universe.
Those universes whose physical parameters favoured black hole production, for example
by encouraging the formation of large stars, would produce more progeny, implying that
among the ensemble of universes with all possible variations of the laws of physics, those
universes with prolific black hole production would represent the largest volume of
space." ...

I'm aware of Smolin's ideas about cosmic evolution via bouncing black holes. I've never taken them very seriously. Too much Hat and not enough Cattle, as they say in Texas. But in this thread it seems that there are people who do take such stuff seriously. And the Davies quote you kindly contributed puts Smolin's views succintly.

Perhaps one of the posters in this thread could clarify the following difficulty I have:

The bounce of a old black hole into a new universe is described above specifically as a process A process is something that takes time. But just whose "time" is this?

That of observers in the old universe outside the black hole? Can't be. For them any process involving the actual formation or growth of a black hole; the actual tranfser of mass through its horizon, takes an infinite time (although they don't find the external gravitational field of infalling matter to be static).

In fact it is hard to see how any such observers can claim that any black holes "exist", since existence involves being present "now". And "now" is not a universal instant when there are black holes bouncing around.

That of observers inside an horizon? Lucky them if their infalling is accompanied by enough matter to render tidal forces innocuous!

God's time? Then physics has morphed into theology --- more Hat!

Isn't Smolin (or Davies) being rather sloppy here --- or perhaps just too anthro'centric?

 

[Chalnoth]

Perhaps one of the posters in this thread could clarify the following difficulty I have:

The bounce of a old black hole into a new universe is described above specifically as a process A process is something that takes time. But just whose "time" is this?

I don't think that's a valid complaint. The word "process" in physics just denotes the behavior of some physical system or other.

That of observers in the old universe outside the black hole? Can't be. For them any process involving the actual formation or growth of a black hole; the actual tranfser of mass through its horizon, takes an infinite time (although they don't find the external gravitational field of infalling matter to be static).

In fact it is hard to see how any such observers can claim that any black holes "exist", since existence involves being present "now". And "now" is not a universal instant when there are black holes bouncing around.

The thing about general relativity is that it demonstrates that there is no such thing as a global "now". "Now" is only explicitly-defined at a singular point, and observers at one point can't make any definitive statements about what is happening "now" far away. That's not to say you can't define a "now" far away, rather that the definition is arbitrary.

That of observers inside an horizon? Lucky them if their infalling is accompanied by enough matter to render tidal forces innocuous!

I don't quite see why this is important. The time coordinate is going to get quite messy here anyway, and you really shouldn't expect naive analyses to capture the nature of how time applies to this hypothetical process.

 

[oldman]

I don't think that's a valid complaint. The word "process" in physics just denotes the behavior of some physical system or other.

But "behaviour" also implies finite duration --- can you give me an example of behaviour without such duration? Perhaps you are thinking along the lines of a "block universe" such as envisaged by Julian Barbour?

The thing about general relativity is that it demonstrates that there is no such thing as a global "now". "Now" is only explicitly-defined at a singular point, and observers at one point can't make any definitive statements about what is happening "now" far away. That's not to say you can't define a "now" far away, rather that the definition is arbitrary.

Yes. But when you ordinarily talk about something that "is" or "exists", say a black hole at the centre of our galaxy, you're implying that it does so "now". You can define "now" by choosing a section of spacetime in which you define events to be simultaneous, according to some agreed protocol. E.g. in a FLRW universe you could say "now" everywhere is when the CMB has the same temperature as that you find. Is this what you mean by an arbitrary definition?

What do you do when a Schwarzschild event horizon --- which dilates duration to infinity for external observers -- separates the universe into different domains? How can you then even talk about evolution, such as "black hole formation", or "black hole growth", in ordinary terms which imply duration? As Davies does.

I don't quite see why this is important. The time coordinate is going to get quite messy here anyway, and you really shouldn't expect naive analyses to capture the nature of how time applies to this hypothetical process.

Sounds as if it is too tricky to describe clearly. But thanks.

 

[Chalnoth]

But "behaviour" also implies finite duration --- can you give me an example of behaviour without such duration? Perhaps you are thinking along the lines of a "block universe" such as envisaged by Julian Barbour?

That's only a failure of our language. It's easy enough to describe the behavior of physical systems even if you wish to consider a hypothetical system without any time coordinate.

Yes. But when you ordinarily talk about something that "is" or "exists", say a black hole at the centre of our galaxy, you're implying that it does so "now". You can define "now" by choosing a section of spacetime in which you define events to be simultaneous, according to some agreed protocol. E.g. in a FLRW universe you could say "now" everywhere is when the CMB has the same temperature as that you find. Is this what you mean by an arbitrary definition?

What do you do when a Schwarzschild event horizon --- which dilates duration to infinity for external observers -- separates the universe into different domains? How can you then even talk about evolution, such as "black hole formation", or "black hole growth", in ordinary terms which imply duration? As Davies does.

Well, you just have to recognize that reality is a bit more subtle.

Sounds as if it is too tricky to describe clearly. But thanks.

I wouldn't say that. I'd say it's too tricky to describe simply. Any truly accurate description is going to be messy and complex and include lots of mathematics. Now, I suppose it is possible that one could find a contradiction in the math there somewhere (I haven't looked at it in detail myself), but finding an apparent contradiction in a simplified description isn't going to invalidate the hypothesis. It just means the simplified description is likely missing some important details of the physics which it is trying to describe.

 

[Rymer]

We can never see the photons that leave it after a certain time (not simply given by its recession velocity, but instead by the future expansion history of the universe). But this doesn't mean that we cease to see it: we see its after-image forever. It just gets dimmer and dimmer. And, as near as we can tell time slows and slows for this image as time goes forward, and the apparent age of the object in our after image asymptotically approaches the age at which the object crossed our horizon.

Note, however, that this is only true in an accelerating universe. If the universe were not accelerating, or stopped accelerating at some point in the future, then there would be no future horizon, and, given infinite time, we would be able to see the full history of all objects in the universe.

Acceleration has nothing to do with it.

 

[sylas]

Acceleration has nothing to do with it.

Boggle. Of course it does. A non-accelerating expansion does not have this horizon. You only get the horizon with an accelerating expansion.

 

[marcus]

Marcus, here is what Paul Davies said about Smolin's multiverse. A good summary.
...

Apeiron, I think Davies account misses the main point and is not at all a summary, much less a good one. One problem is simply that Davies article is old (March 2004) and references nothing of Smolin's but a popular 1997 book.

Evidently the discussion has changed over the course of time---maybe your reference to the evolution of viruses is pertinent.

What Smolin has emphasized in recent talks about this, and what interests me about it, is what I consider to be a hard empirical hypothesis. There are thirty-some parameters that go into either the current cosmo model or the standard particle model. Perhaps fewer, but call them the Thirty, for short. Are the Thirty optimal for black hole abundance?

This has no claim in it about the existence of a multiverse, or about what happens down a black hole. It represents an hypothesis to be tested. If it is false then it should be possible to find a parameter which, if it had been slightly different, would have resulted in greater hole abundance.

I have to go, back later today.

 

[Chalnoth]

Ah, okay. This is an idea I've heard before. I find it very interesting to think about. Now, I also find it highly speculative and almost certainly wrong. But it is very interesting.

One thing I've been thinking about is that in order to falsify the hypothesis, you don't only need to show that tweaking the parameters in a certain direction would create more black holes, but you also need to show that there is still at least the possibility of life in that direction (a difficult thing, but at least we can put outer bounds on it).

What we expect to see from this hypothesis is that the parameters are right on the outer bounds of what's possible for life, with black hole number being maximized given the constraint of life existing. Not because life has to exist, but because living beings like ourselves can only observe regions where it does: the hypothesis likely predicts that by far most of these "baby universes" will not be conducive to life.

 

[marcus]

BTW my nickname for the 30-odd essential parameters that rule the standard models of cosmo and particle physics is a classical joke
In Greek history the Thirty were a group of oligarchs which governed Athens for a brief period around 404 BC.

The hypothesis is very simple and says nothing about Life or "Multiverses". It is simply the falsifiable conjecture that the Thirty are optimal for black hole abundance.

We are being challenged simply to falsify this. That means to find a small change in the parameters which would have led to more black holes.

This challenge was published in 1992. People (like Alex Vilenkin) have tried to shoot it down, but as yet it has not been shot down. The longer it survives the more likely people are to see it as something requiring explanation.

 

[Rymer]

BTW my nickname for the 30-odd essential parameters that rule the standard models of cosmo and particle physics is a classical joke
In Greek history the Thirty were a group of oligarchs which governed Athens for a brief period around 404 BC.

The hypothesis is very simple and says nothing about Life or "Multiverses". It is simply the falsifiable conjecture that the Thirty are optimal for black hole abundance.

We are being challenged simply to falsify this. That means to find a small change in the parameters which would have led to more black holes.

This challenge was published in 1992. People (like Alex Vilenkin) have tried to shoot it down, but as yet it has not been shot down. The longer it survives the more likely people are to see it as something requiring explanation.

I like your Thirty.

Question all this seems to be assuming there are really 'black holes'. I'm not sure anymore what the term means. At one time it seemed to mean that NOTHING could escape. But my current understanding is that it is not the case. There are 'quantum' level events that do escape (at least these and maybe more).

The reason I bring up the question is that, is this the prediction 'black holes' or just very super-compacted matter objects? And how would the difference effect such postulated tests?

Not clear what is being 'counted'.

 

[russ_watters]

Boggle. Of course it does. A non-accelerating expansion does not have this horizon. You only get the horizon with an accelerating expansion.

A non-accelerating expansion most certainly does have a horizon, as does a non-expanding universe! All that is required to have a horizon for the "observable" universe is for the universe to be larger in light years than it is old in years. Ie, a universe that is 1 year old and not expanding will have an observable size of 1 ly, a universe 14 billion years old will have an observable size of 14 billion light years.

For a constantly expanding universe that started at a big bang, that requires an expansion to always exceed the speed of light.

 

[Chalnoth]

A non-accelerating expansion most certainly does have a horizon, as does a non-expanding universe!

Technically, a non-accelerating universe has a past horizon, meaning there exist objects that have never been in causal contact with us. It doesn't, however, have a future horizon, meaning that eventually everything will be in causal contact with us (or rather, with our position in space-time...we certainly won't be around).

 

[marcus]

Hi Rymer and Chalnoth!

Chalnoth, it occurs to me that Smolin may have confused things by putting the cart before the horse and talking too much about a possible explanation which could be offered, if his very interesting conjecture were established.

The conjecture (concisely stated at the bottom of page 29 of hep-th/0407213) is that the Thirty are optimal for black hole abundance.

Nothing said about Life or Multiverses or Bounce Scenarios :-D. There is a clear challenge to disprove the conjecture by finding a small modification of the Thirty that would have increased the count---say by lowering the collapse threshold so that a less massive star could collapse to hole, and thus a higher percentage of stars could form holes.

Or by making it easier for clouds of gas to condense in the first place so that there would have been more and larger stars to eventually form holes.

It would be great if someone could prove the conjecture false.

On the other hand if it is true, then it is a peculiar instance of fine-tuning and definitely calls for some explanation. But first we should see if it is true or not.

That is my point of view, in any case. Sorry if that wasn't clear at the start.
=======

Rhymer, I'm glad you liked the reference to the Thirty Tyrants who ruled Athens for a brief period. I hope that the reign of this Thirty is also curtailed and that our standard models can be simplified so that the number of essential parameters is reduced. Fewer free parameters means more elegant, and also more predictive.

...The reason I bring up the question is that, is this the prediction 'black holes' or just very super-compacted matter objects? And how would the difference effect such postulated tests?

Not clear what is being 'counted'.

Rhymer black holes are a regularly catalogued astronomical object. No one knows what happens inside them, though there are various models. Astronomers know black holes when they see them, and they have their customary ways of distinguishing them from other compact objects like neutron stars. There is a fairly complete picture of what conditions lead to their formation.

So that is what is being counted.

However you might still be wondering what goes on inside---what the competing models are. One can chuck General Rel out because it suffers from a singularity (stops computing meaningful numbers, blows up and says infinite curvature and other unrealistic stuff.)
But there are other models and you can find them at arxiv.org.
Some names of researchers:
Leonardo Modesto
Kevin Vandersloot
Christian Boehmer
Dah-wei Chiou

These people model black hole collapse using a formalism that does not blow up and does not develop singularities. It's work in progress. The research papers consider gradually more and more general cases. The effort is to gradually get rid of simplifying assumptions like homogeneity and isotropy. What happens if the collapse is slightly lopsided? And so on.
If I'm not mistaken, both Modesto and Chiou delivered papers on this at a conference in Beijing last week.
(Of course Smolin must be happy with this research because in some cases what they found was a bounce resulting in a new region of expanding space, but it is still early days. One should not count the chickens before the eggs hatch---nobody knows how this line of research will go.)

 

[Chalnoth]

I think you're failing to get my point again, so let me see if I can restate it.

If we want to falsify Smolin's hypothesis, we must bear in mind that only a limited parameter space is available for investigation: that part of the parameter space that can be observed.

It is entirely possible that the the optimum point in the parameter space is very, very far outside of the region that can possibly be observed. But that doesn't make the hypothesis false, because there might be tons of stuff outside of what can be observed that is more in line with the hypothesis. Only a small fraction will be in the possibly-observable region of the parameter space, but that's okay. There's nothing that says the maximum must be within the observable region.

So, to reach a solution, we are required to stick to just the possibly-observable region, and see if the region we observe is tilted in the direction within this parameter space we expect, given this hypothesis of the universe being fine-tuned for the production of black holes.

And what defines this possibly-observable region? Well, that's the region in the parameter space where observers can be found: the region where complex life is possible. Nothing else is observable, so it doesn't matter for hypothesis-testing.

------------

With that said, let me make use of the above to argue why I think that Smolin is unlikely to be onto anything of importance here. To do this, I'll simply point out what is required for lots and lots of black holes. You need:

1. A universe that lasts a long time. If it recollapses very quickly, obviously few if any black holes will form.
2. A universe that has structure formation. Stuff has to get clumpy before black holes have a chance to form.
3. At least some of the matter needs to be dissipative. That is, it has to experience friction so that it can collapse more readily.

And I believe that's all of it. Point (1) is met by our universe being nearly flat, and the cosmological constant not being too large. Point (2) is met by a combination of the magnitude of primordial perturbations, the existence of normal + dark matter, and the cosmological constant being very small. Point (3) is met by the existence of the electromagnetic force.

The problem I have with this, however, is that these are all also requirements for life. And if the requirements for life include the requirements for black holes, then clearly any universe with life in it is going to have lots and lots of black holes. So even if there is an appearance of fine tuning, I don't think we can take it seriously.

The only real way to test it would be to demonstrate succinctly that within the limited range of the observable parameter space, this theory predicts that universes on one end of the parameter space are vastly, vastly more likely than universes on the other, such that we should find ourselves strongly to one end of said parameter space.