Lee Smolin and Daughter universes

[vincentm]

Lee Smolin and Daughter universes...

He postulates that daughter universes can arise from black holes. How is this so?

 

[marcus]

He postulates that daughter universes can arise from black holes. How is this so?

Come on Vincent, he explicitly does NOT postulate that. Please read the articles themselves. I will get some links.

Smolin CNS conjecture is simply that our universe's fundamental constants are OPTIMIZED FOR MAKING BLACK HOLES.

If this is wrong it should be easy to falsify it, because it makes some clear unequivocal predictions which are subject to test by observations currently in progress. In fact it has stood up to testing for around 14 years already.

 

[marcus]

In science, empirical science, the Baconian method in case anyone has forgotten, you are not supposed to propose a theory unless it can be falsified by some future experiment. It has to bet its life on some prediction of some possible observation that COULD go against it.

It can't be infinitely adaptible so that it would accommodate any possible outcome of any future observation. If it is that accomodating it is MUSH having no predictive or explanatory value----i.e. not science.

Like saying the universe is compatible with the existence of conscious life. Because what observation could ever show that it was not compatible with life?
A lot of "multiverse" talk is like that and is not science.

The CNS conjecture challenges you to find some possible modification of the fundamental constants which would increase astrophysical black hole abundance.

It PREDICTS, in effect, that you cannot find any slight tweaking of the constants which, if done, would make the universe have more black holes

This prediction has been standing since 1993 when Smolin published it, and people have TRIED and they have so far FAILED to falsify the CNS conjecture. You can try. See if you can think of something. If you can you will have refuted CNS and it will be shot down.

One way would be if an astronomer found a neutron star with more than 1.6 solar mass, because that would show that tweaking the mass of the top quark so as to make that neutron star collapse to hole would be possible.

If someone finds a 1.7 solar mass neutron star that will prove that our universe is NOT OPTIMIZED FOR HOLES.

 

[marcus]

As far as daughter universes growing from black holes, John Archibald Wheeler already had that idea before Smolin.

It is consistent with the LQC model of the big bang which eliminated the singularity and goes back to a prior gravitational collapse. But that is at present just a theory which needs observational testing. LQC fits all the existing cosmological data because it has standard FRW cosmology as its classical limit, it differs only right close to the moment of the "bang" which it says was a bounce. So ways need to be found to TEST LQC. That is another issue.

For now all one can say is that there is no scientific reason to reject Wheeler's idea that the universe might have emerged from a black hole gravitational collapse.

But that is not what we are talking about. Speculating about that is putting the cart before the horse. It certainly is POSSIBLE that an astrophysical black hole leads to a bounce and another region of expanding spacetime.

WHAT WE SHOULD BE ASKING IS DIFFERENT THAN SPECULATION---WE SHOULD BE ASKING WHETHER OR NOT THE UNIVERSE IS OPTIMIZED FOR HOLES. Because either it is or it isn't and that is something that can be tested . If it turns out that the parameters of physics ARE optimal for hole abundance, then that would be very interesting and one would want to ask why.

The universe is NOT in any obvious sense optimized for the abundance of LIFE, in fact as far as we can see life is rather scarce and a lot of places seem darn inhospitable. But suppose after a lot of checking we determine that it IS optimized for holes. Then we need to construct some scenarios that could explain that. Smolin has one that the LQC research I mentioned has recently made more plausible. Universes may begin with bounces, black holes may end with bounces. The bounce phenomenon is being studied using computer models, and attempts made to derive testable prediction (something to check against the CMB).

But my point is that all that explanatory scenario business is SECONDARY to testing what Smolin calls the MAIN CNS PREDICTION which challenges you to find some adjustment of fundamental constants causing more holes, and predicts that you will not be able to do that.

 

[vincentm]

Ok, postulate was a bad word for the purpose of this discussion, thanks for the information i'll pour over it soon as get home from work.

 

[marcus]

Our posts crossed. I didnt see your last one. thanks for your respose!
You can ignore the following which I wrote before I saw your post:
daughter universes is not a postulate

there is no indication that Smolin himself believes it AFAICS and no indication that he wants to persuade anyone else to believe it, a scientists business is not to believe but to construct and test theories

the daughter universes idea is a possible scenario that could explain why the main CNS conjecture is true, if the main conjecture continues to be confirmed empirically.

the main CNS conjecture is hole-optimality of the fundamental numbers determining the physics of our universe. this is what the testable predictions derive from (predictions having to do with neutron star mass, among other things)

A recent statement of the main CNS conjecture, or "master conjecture" was on page 6 of this paper

http://arxiv.org/hep-th/0612185
The status of cosmological natural selection

==quote==
Hence the theory predicts that a randomly chosen member of the ensemble will have the following property, which we may call the master prediction of CNS

M: Almost every small change in p from its present value either leads f(p) unchanged or leads to a decrease in f(p).

Since our universe can be assumed to be randomly chosen we conclude that if the hypotheses made above are true, almost no change in the parameters of the standard model from the present values will increase the numbers of black holes produced.

==endquote==
what he is denoting by f(p) is the abundance of black holes, as a function of parameter of the standard model----fundamental constants of physics.

 

[marcus]

Ok, postulate was a bad word for the purpose of this discussion, thanks for the information i'll pour over it soon as get home from work.

Hi vincent sorry about my impatience.

I will try to find some more readable statements. I find that recent paper I gave you link to rather indigestible.

There is a 2004 paper that has parts that are readable.
And a critical discussion by a philosophy of science guy named Rudiger Vaas.
I personally don't know any piece of writing that is clearly and simply written on this. Maybe someone else has some suggestions.

I will get what links I can, and you can try to see if they work for you.the one thing I am sure of is that he has this main conjecture, or master conjecture, which asserts optimality and which you can test up or down regardless of whatever you believe about what might explain it.

Here is that 2004 paper
http://arxiv.org/abs/hep-th/0407213
Scientific alternatives to the anthropic principle

In this paper the master prediction is on page 29, and from it (starting from page 30 onwards) he derives from it various specific predictions like the neutron star mass one. In this paper the master prediction is stated

==quote==
this is sufficient to lead to
observational tests of these hypothesis, because this implies the prediction that:
If p is changed from the present value in any direction in P the first significant changes
in F(p) encountered must be to decrease F(p).

==endquote==
here the notation is essentially the same. p is the n-tuple of parameters of the standard model (the fundamental constants of physics) and F(p) is black hole abundance. he is saying that p is at a local optimum, so changing it a little will not immediately cause and increase in F(p)

GLOBAL optimality is not asserted, only optimality in a small local neighborhood.
====================

I think there may also be a fairly clear statement of the main prediction in one of Smolin's early 1990s papers on this.

 

[Chronos]

Consider, as well, a great deal depends on how 'soft' the equation of state is for condensed matter [i.e., Brown, et al]. There is a fair amount of evidence to suggest it is difficult to form black holes via stellar collapse, no matter how massive the progenitor star. Enormous amounts of mass are typically expelled during these events - probably enough to prevent the formation of stellar mass black holes. I suspect most, and perhaps all stellar mass balck holes are formed via neutron star mergers in binary systems. In an odd respect, that tends against the CNS conjecture. Core collapse black holes would seem to be preferred in a universe optimized for their production.

 

[heusdens]

In science, empirical science, the Baconian method in case anyone has forgotten, you are not supposed to propose a theory unless it can be falsified by some future experiment. It has to bet its life on some prediction of some possible observation that COULD go against it.

It can't be infinitely adaptible so that it would accommodate any possible outcome of any future observation. If it is that accomodating it is MUSH having no predictive or explanatory value----i.e. not science.

Like saying the universe is compatible with the existence of conscious life. Because what observation could ever show that it was not compatible with life?
A lot of "multiverse" talk is like that and is not science.

The CNS conjecture challenges you to find some possible modification of the fundamental constants which would increase astrophysical black hole abundance.

It PREDICTS, in effect, that you cannot find any slight tweaking of the constants which, if done, would make the universe have more black holes

This prediction has been standing since 1993 when Smolin published it, and people have TRIED and they have so far FAILED to falsify the CNS conjecture. You can try. See if you can think of something. If you can you will have refuted CNS and it will be shot down.

One way would be if an astronomer found a neutron star with more than 1.6 solar mass, because that would show that tweaking the mass of the top quark so as to make that neutron star collapse to hole would be possible.

If someone finds a 1.7 solar mass neutron star that will prove that our universe is NOT OPTIMIZED FOR HOLES.

It is not a testable prediction. We can't change the constants of nature.

 

[marcus]

Hi Heusdens, I think differently from you. The constants of nature i am talking about are the dimensionless parameters that go into the standard models of physics and cosmology

These models are reasonably well confirmed and they tell us what the universe would be like with different values of the paramters----with different inputs to the models.

So although one cannot actually change the proton/electron mass ratio from 1836 to some other number, one can in principle say what atoms would be like if it were different.

Smolin's conjecture, stated as a challenge to find some tweaking of the inputs that results in more holes, needs to be qualified by mentioning that this is according to the standard version of physics/cosmology.

Please add that assumption to what I said and it will be more correct. so if you can find an adjustment of the parameters which would result (according to the standard physics) in more holes then

1. either Smolin's conjecture is wrong
2. or the standard model of physics is wrong.

Or both.
But I would take it as a falsification of Smolin's CNS conjecture. :-)

 

[marcus]

He postulates that daughter universes can arise from black holes. How is this so?

Hi vincentm,
dont know if you are still around but my response didn't adequately address the interesting "How is this so?" question you raised.

I took issue with your saying postulate. I don't think Smolin or anybody postulates that BabyUniverses arise from black holes. That is just one possibility, and AFAIK nobody knows how to check it. At the moment what I think is important is the conjecture that I talked about, which is testable.

What you asked about was just the subject of a discussion at KITP singularities workshop a couple of days ago. Ted Jacobson was at the blackboard a lot of the time. The workshop folks were trying to see if they could rule out babyU within various frameworks---various theoretical contexts.

You might want to download the video and watch some. I will get the link
http://online.kitp.ucsb.edu/online/singular_m07/
look down till you see the 16 January discussion
1/16, 12:00 p.m. Discussion Do Baby Universes Emerge From Black Holes?[Podcast][Aud][Cam]
and click on that to get
http://online.kitp.ucsb.edu/online/singular_m07/babybh/

I would advise skip the first 20 minutes or so---just drag the time pointer. If you have the video downloaded on your desktop you can skip around, watch different parts, don't have to watch the whole hour.
For me the most revealing part was where Steve Shenker asked Ted Jacobson what his basic intuition---his "starting hunch"--- was about what happens when a black hole forms.
This is around minute 42 of the video, like starting at 42:10 and lasting for a minute and a half or so.

Ted says that intuitively evolution has to continue because there's a timelike vector. Steve asked continue to WHAT. Ted: to something we don't know.
And another big bang producing another region of spacetime is one of the possibilities that they talked about.
At one point Ted remarked "Forget complementarity!" The issue is pretty wide open among the experts at this point.