Is Paul Steinhardt's Statement "Rather Pathetic"? Why?

[marcus]

It might be worthwhile for anyone who can't remember how the discussion started to review post #1 to see what we are talking about. Do we have to give up on the traditional reductionist program of explanation and resort to anthropics, or not?

The issue arose concisely in this post:
https://www.physicsforums.com/showthread.php?p=3725098#post3725098

I linked to Steinhardt's statement and Chalnoth replied that it was pathetic.

Here's the post we began the thread with:
===Originally Posted by Chalnoth===

==Originally Posted by marcus==
http://edge.org/response-detail/805/what-do-you-believe-is-true-even-though-you-cannot-prove-it

or google "Steinhardt annual question 2005"
==endquote==​

I've always found that response to anthropic arguments to be rather pathetic.
===endquote===

In fact Steinhardt's position has largely prevailed, physics has moved on and Chalnoth's complaint is out of date.
In scientific discussion (i.e. outside popular media) one does not hear much anthropics talk these days.
I think that's great and I'm grateful to Steinhardt for leading the attack on it.

There is a lot of obfuscation about this issue and natural confusion as well so I urge anybody who is not familiar with it to take a close look at post #1.

 

[Chalnoth]

Totally agree. But simplicity is a statistical nicety; not necessarily a physical one. The universe simply might not work that way. Multiverses simply might not exist.

That's extremely unlikely.

With a bit of hindsight, now that we know that the appearance of collapse was built into the wavefunction dynamics from the start, it is patently obvious that the entire process of attempting to explain wavefunction collapse was always an unnecessary enterprise. We were trying to explain something that was already explained by the wavefunction dynamics.

The whole enterprise of searching for another interpretation of quantum mechanics is akin to having Newtonian gravity before you and attempting to come up with some rule by which planets can have elliptical orbits. The very idea is nonsensical: Newtonian gravity explains it just fine. Why look for anything extra?

 

[Fuzzy Logic]

As I said in an earlier post, the only problem I have with Steinhardt's statement is his reason in that the universe was not accidental. Of course it was. Saying otherwise infers design. That answers nothing and complicates things immensely.

 

[Chalnoth]

In fact Steinhardt's position has largely prevailed, physics has moved on and Chalnoth's complaint is out of date.

Um, I guarantee you that nearly all theoretical physicists would find that a complete and utter surprise.

In scientific discussion (i.e. outside popular media) one does not hear much anthropics talk these days.

Only because these ideas are very difficult to nail down, and there are other areas where we have lots of good new experimental data to examine. As I said before, we can expect them to resurface once the LHC becomes mature and precision cosmology enters a lull in new data.

I think that's great and I'm grateful to Steinhardt for leading the attack on it.

I guarantee you he had nothing to do with it. Bad, irrational arguments like the one you quoted aren't likely to have had any impact on the scientific community.

 

[marcus]

As I said in an earlier post, the only problem I have with Steinhardt's statement is his reason in that the universe was not accidental. Of course it was. Saying otherwise infers design. That answers nothing and complicates things immensely.

That's great. Then you agree with the main body of the statement (we don't have to give up on the effort to find natural explanations) and just disagree verbally with how he chose to phrase the headline.

I'm sure you agree that Steinhardt was not arguing for "design"!
His basic message is there is no indication that we are through explaining yet. There are further layers of the onion.
We have not yet reached the end of physics where we have found the deepest explanation and where beyond that it "just is the way it is."

Notice how Steinhardt uses the word "desperation" in analyzing the motives of those who resort to anthropics. The pernicious thing is the cop-out: there are lots of different big bangs and this one just is the way it is, so don't ask why. He attributes it to an desperate attempt to excuse some theorist's failure to come up with a unique fundamental theory (by fundamental I mean good up to Planck scale.) A multiplicity of big bangs let's them off the hook.

Since 2006 it seems to me as an onlooker that the community has pretty much decided that theorists are NOT going to be let off the hook. The anthropic excuse for failure has been discarded or has lapsed into disuse.

 

[Fuzzy Logic]

That's great. Then you agree with the main body of the statement (we don't have to give up on the effort to find natural explanations) and just disagree verbally with how he chose to phrase the headline.

You miss the point. While our current model may be sufficient to explain what we already know, it doesn't explain everything. His statement is fine by scientific standards and I agree that the anthropic principle is not enough reason to justify a multiverse theory. I agree that the anthropic principle has no merit.
Would I say it's rather pathetic? No, but it is not a valid argument against multiverse theories, only the anthropic principle.

There are many compelling reasons to consider multiverse theories.
Most importantly, what happened at T=0.

 

[Chalnoth]

I agree that the anthropic principle has no merit.

That is taking things way way too far. The anthropic principle is, first of all, necessarily true. Because of its necessary truth, it manifests itself as a selection effect that must be applied whenever considering any theory for the birth of a new region of space-time. Failure to apply the selection effect that is the anthropic principle in such a situation will always lead you to incorrect conclusions.

Now, the anthropic principle can be used badly, as can most anything, but that's no reason to disregard it.

 

[Fuzzy Logic]

That is taking things way way too far. The anthropic principle is, first of all, necessarily true. Because of its necessary truth, it manifests itself as a selection effect that must be applied whenever considering any theory for the birth of a new region of space-time. Failure to apply the selection effect that is the anthropic principle in such a situation will always lead you to incorrect conclusions.

Now, the anthropic principle can be used badly, as can most anything, but that's no reason to disregard it.

No, I say it has no merit, because we don't know what the conditions for life really are. Life could just as easily prosper in a universe with completely different physics from our own. The fact that life is so prolific on Earth contradicts the idea of a finely tuned universe. The fact that we can't find life outside of our own planet only proves that the universe is vast, not that life is unique.

If you want to say that it's a necessary truth, then it is no different than saying "I think, therefore I am".

 

[marcus]

His statement is fine by scientific standards and I agree that the anthropic principle is not enough reason to justify a multiverse theory. I agree that the anthropic principle has no merit.

I agree. I think Steinhardt in 2005 led the attack that established that the anthropic principle has no place in fundamental physics. His statement was effective and influential. It was echoed and anthropics lost out e.g. as reflected in subsequent StringsXXXX conferences. Ridiculous to call it pathetic.

Would I say it's rather pathetic? No, but it is not a valid argument against multiverse theories, only the anthropic principle.

That was my point at the outset, specifically regarding the anthropic cop-out.
===Originally Posted by Chalnoth===
==Originally Posted by marcus==
http://edge.org/response-detail/805/...annot-prove-it

or google "Steinhardt annual question 2005"
==endquote==
I've always found that response to anthropic arguments to be rather pathetic.
===endquote===

BTW don't mind predictive testable theories that explain stuff involving subsequent big bangs as a byproduct. If a particular model of our big bang (as long as it is testable) just has to produce baby big bangs/reheats somewhere down the road, that is OK with me.
I just get disgusted when I see arguments like "it's the way it is because otherwise we wouldn't be here and we can't explain any more".

There are many compelling reasons to consider multiverse theories.
Most importantly, what happened at T=0.

Please explain, what do you think happened at T=0 that makes this or that compelling. Are you thinking of socalled rogue regions where inflation did not stop and is still going on---or where reheating occurred later? That was discussed in an interesting thread last summer. You might be intrigued by some of what Ben Crowell had to say regarding a conference at Perimeter about problems with the prevailing early universe pictures and related stuff
https://www.physicsforums.com/showthread.php?p=3403876#post3403876

 

[Fuzzy Logic]

Please explain, what do you think happened at T=0 that makes this or that compelling. Are you thinking of socalled rogue regions where inflation did not stop and is still going on---or where reheating occurred later? That was discussed in an interesting thread last summer. You might be intrigued by some of what Ben Crowell had to say regarding a conference at Perimeter about problems with the prevailing early universe pictures and related stuff
https://www.physicsforums.com/showthread.php?p=3403876#post3403876

I have no opinion on simultaneous multiverse theories, though because of intuition and the singularity problems I am inclined to a big-bounce theory. I don't think that we should be extrapolating back to exactly 0. I expect that the universe bounced slightly before. What we say is T=0 in the big-bang model is more like Planck time. What happened before that was the end of the collapse of the previous universe when it reached critical density and entropy surpassed gravity. If I say too much I'll surely paint myself into a corner. That is just my layman interpretation of what I think happens. I don't pretend to know enough details to argue for it.

I'll check out that thread, thanks.

Edit: sorry I didn't actually address your question. Why is it compelling.
I think it's a compelling reason because we can't explain the singularity event. Not only can we not explain what happened in the first moments, we can't explain how the universe came into existence at all. If I have to choose between saying the the universe just popped into existence and time began or saying that the universe has existed forever, without any direct evidence in favour of either argument the circumstantial evidence would suggest that the life cycle of the universe is closed system, just like the rest of nature. Nowhere do we have any evidence of spontaneous existence, but everywhere we can see examples of cyclic evolution.

 

[marcus]

Part of what causes confusion is that some people talk as if we have a settled theory of the mechanism behind inflation.
We don't. Smart people are still arguing about whether inflation even occurred (see Ben Crowell's condensed digest of the Perimeter conference last summer)
and there are several ideas of how it might have worked, if it did occur.
Here's a relevant Crowell post:
https://www.physicsforums.com/showthread.php?p=3404021#post3404021

Just to illustrate with an example, here is a recent paper by Perimeter's Laurent Freidel and others that proposes quite a different mechanism.

http://arxiv.org/abs/1201.5423
Dirac fields and Barbero-Immirzi parameter in Cosmology
G. de Berredo-Peixoto, L. Freidel, I.L. Shapiro, C.A. de Souza
(Submitted on 26 Jan 2012)
We consider cosmological solution for Einstein gravity with massive fermions with a four-fermion coupling, which emerges from the Holst action and is related to the Barbero-Immirzi (BI) parameter. This gravitational action is an important object of investigation in a non-perturbative formalism of quantum gravity. We study the equation of motion for for the Dirac field within the standard Friedman-Robertson-Walker (FRW) metric. Finally, we show the theory with BI parameter and minimally coupling Dirac field, in the zero mass limit, is equivalent to an additional term which looks like a perfect fluid with the equation of state p = wρ, with w = 1 which is independent of the BI parameter. The existence of mass imposes a variable w, which creates either an inflationary phase with w=-1, or assumes an ultra hard equation of states w = 1 for very early universe. Both phases relax to a pressureless fluid w = 0 for late universe (corresponding to the limit m→∞).
16 pages

From the conclusions section on page 15: "... the fermionic matter behaves effectively as a cosmological constant and creates an inflationary phase which is relaxed at late time into a pressureless fluid."

So here's another possible inflation mechanism just now proposed, which will quite possibly be worked on. At this point it isn't clear that it would for instance involve "rogue regions" where inflation does not turn off and which continue inflating. The idea needs to be explored and one does not know which problems it would or would not share with inflation mechanisms which people have speculated about earlier.

I guess the moral (which bears repeating) is "don't assume you know what you don't know and draw draw conclusions from it."

 

[salvestrom]

My thoughts on the subject solidified last night and rapidly reached two seemingly diametric dead ends.

If the expression of all possible outcomes is an intrinsic property of the universe then the first event was simply this: existence or non-existance (possibly even 'are all possile outcomes expressable or not). This automatically creates a universe and a non-universe. From here questions abound about the specific makeup of our early universe, requiring the acceptance of a trillion other alternatives with less or more of one thing or another. It gets messy quickly, but seems to offer an answer to the most fundamental question.

Here's the opposite:

All events today are the result of all the events over the last 13.5 billion years and so, far from being many posible outcomes, there is only one possible outcome for any event: that which happens. Although Fatalistic, the reality is we cannot know the future because we cannot know all prior actions in order to calculate 100% any future ones. Oddly enough, the line "the expression of all possible outcomes is an intrinsic property of the universe" still applies, it's just there's only ever one. This idea offers little revelation in the explanation of the existence of the universe, other than to accept that it couldn't have ever been anything else.

 

[Fuzzy Logic]

I guess the moral (which bears repeating) is "don't assume you know what you don't know and draw draw conclusions from it."

If I assumed, I wouldn't bother looking for answers. Satisfaction in ignorance is no better. The pursuit has and always will be fact, whether that coincides with my own expectations or not. I did not make a conclusion, I postulated an idea. Is it evidently wrong?

 

[marcus]

If I assumed, I wouldn't bother looking for answers. Satisfaction in ignorance is no better. The pursuit has and always will be fact, whether that coincides with my own expectations or not. I did not make a conclusion, I postulated an idea. Is it evidently wrong?

Fuzzy, I don't see the connection between that general "moral" and your posts. It wasn't directed towards them or reflecting on them.

But I would like to amplify and explain some, if I've got time.

All "inflation" means is exponential expansion e^Ht with a a high nearly steady Hubble rate H. The Hubble "constant" is a frequency, a reciprocal time.

There are various possible mechanisms that could cause inflation. But some people seem to have a fixed notion so when you say "inflation" all they can think of is a fixed range of notions mostly dating back 20 or 30 years which have been drummed into them.

That's the kind of thing I meant by "assuming you know what you don't, and drawing conclusions from it".

I'm short on time. I'll try to add some more clarification to this when I can.

BTW an interesting side aspect to all this is the "superinflation" that automatically occurs (and automatically ends) in the LQC bounce. It is faster than ordinary inflation and involves a rapidly INCREASING Hubble frequency H, so that you get faster than exponential growth. You just solve the bounce equations and this is what you get.

H is negative during the contraction phase, then crosses zero (that defines the moment of the bounce) and increases rapidly to something like Planck frequency, and then slacks off. This is not due to some imagined "quantum fluctuation" or anything not under the model's control. It is built in. Superinflation is a deterministic feature of the loop cosmology bounce by which the singularity is resolved and it happens only then. And it terminates deterministically and very quickly at that.

This is not to say LQC is right, that is something that will eventually be tested by observation. It may or may not be shown false. It fits observation so far---see Rinaldi's recent review of all the QC options. http://arxiv.org/abs/1201.4543

So i don't say anything about right/wrong. this is just an example to illustrate that the mechanisms underlying a brief period of exponential expansion don't have to correspond with anyone's (e.g. Chally's) preconceptions.
So drawing elaborate conclusions is really really premature.

 

[Chalnoth]

No, I say it has no merit, because we don't know what the conditions for life really are. Life could just as easily prosper in a universe with completely different physics from our own. The fact that life is so prolific on Earth contradicts the idea of a finely tuned universe. The fact that we can't find life outside of our own planet only proves that the universe is vast, not that life is unique.

If you want to say that it's a necessary truth, then it is no different than saying "I think, therefore I am".

It's not nearly so bad as that. We may not know the requirements for life in detail, but it is very easy to place limits based upon general, overall requirements. For example, if you want to have life, you are going to need structure formation. That is, you need galaxies. And simple limits like this are enough to make pretty powerful statements about the possible values of some parameters that any observer can potentially measure.

 

[Fuzzy Logic]

It's not nearly so bad as that. We may not know the requirements for life in detail, but it is very easy to place limits based upon general, overall requirements. For example, if you want to have life, you are going to need structure formation. That is, you need galaxies. And simple limits like this are enough to make pretty powerful statements about the possible values of some parameters that any observer can potentially measure.

It is interesting how fickle life seems to be but it doesn't matter how many times you roll the dice, the odds of rolling a 1 are always the same. You can maximize chance with iterations but not odds. It is just as likely that life emerged on the first iteration or the trillionth.

All of nature is uncanny how it manages to work, not just life. I don't think that uncanny is evidence of anything.

 

[Chalnoth]

It is interesting how fickle life seems to be but it doesn't matter how many times you roll the dice, the odds of rolling a 1 are always the same. You can maximize chance with iterations but not odds. It is just as likely that life emerged on the first iteration or the trillionth.

All of nature is uncanny how it manages to work, not just life. I don't think that uncanny is evidence of anything.

What are you trying to say here? Because as near as I can tell it has nothing to do with anthropic arguments. Anthropic arguments are, at their heart, arguments about what aspects of nature we have a right to be surprised about. If a certain aspect of nature seems, on its face, highly unlikely, but it turns out that something like it is required for life to exist, then we don't have any right to be surprised to see it. The cosmological constant is a good example here. Sure, the number 10^{-120} seems fantastically small, but it can't possibly have been much bigger and still allowed the existence of galaxies. Because a cosmological constant this small is required for us to exist, we can't, by rights, be surprised about it.

 

[MathematicalPhysicist]

I am not sure, my message was really understood.

How do you define "universe"?

If it's:"All existing matter and space considered as a whole; the cosmos."

Then multiverse=universe, multiverse is just another name for "all that exists".

If on the other hand you say that a universe is all that is made of the same matter as we are, then there might be other universes with different type of matter different than ours.

But it depends on your definition.

 

[Chalnoth]

I am not sure, my message was really understood.

How do you define "universe"?

If it's:"All existing matter and space considered as a whole; the cosmos."

Then multiverse=universe, multiverse is just another name for "all that exists".

If on the other hand you say that a universe is all that is made of the same matter as we are, then there might be other universes with different type of matter different than ours.

But it depends on your definition.

The only way in which it's an interesting topic of discussion is in regard to the low-energy laws of physics varying from place to place within the universe.

 

[Fuzzy Logic]

Ok in that sense sure, but do we really need a principle to define it? That is just elementary deduction. "I think, therefore I am"

I was referring to the anthropic principle being a justification for a multiverse.
As far as I understand, there is no evidence that any of the fundamental conditions must or even could change, only predictions.

 

[Chalnoth]

Ok in that sense sure, but do we really need a principle to define it? That is just elementary deduction. "I think, therefore I am"

The fact that so many people argue so vociferously against it seems to indicate that yes, yes we do.

I was referring to the anthropic principle being a justification for a multiverse.

I don't think that's an entirely correct way of thinking about it. Rather, as I said earlier, the anthropic principle must be taken into account when considering any law for how new regions of space time are born, or how the low-energy laws of physics might vary.

As far as I understand, there is no evidence that any of the fundamental conditions must or even could change, only predictions.

Any spontaneous symmetry breaking event causes a change in the low-energy laws of physics. The electro-weak symmetry breaking is one we know of. There are probably many more.

 

[marcus]

The only way in which it's an interesting topic of discussion is in regard to the low-energy laws of physics varying from place to place within the universe.

The way I think of it can connect with this. I think of the universe as the unique whole of nature with a unique set of fundamental laws.

Spontaneous symmetry-breaking may have resulted in regional variation in some constants that emerge at lower energy. I think that is extremely interesting and I think it is something we humans may be able to study and understand.

At present I don't see any compelling reason to involve multiple big bangs in our model of the big bang---the start of expansion. Pending evidence to the contrary I expect one start of expansion, operating under one set of fundamental highenergy laws of physics will probably fit the data.

It would be quite interesting if we got some evidence of other big bangs having happened, of course. But absent such evidence *shrug*.

So my view is similar to the one expressed in post #1---the reductionist program is on track, no need to give up on the program of finding ever deeper explanations for what we see in terms of one universe, one start of expansion, one set of fundamental physics laws.

 

[Chalnoth]

At present I don't see any compelling reason to involve multiple big bangs in our model of the big bang---the start of expansion. Pending evidence to the contrary I expect one start of expansion, operating under one set of fundamental highenergy laws of physics will probably fit the data.

Well, it kinda has to. But that same model may unambiguously predict other bangs, other low-energy laws of physics (our current model already predicts other low-energy laws of physics...and other bangs are the natural expectation of any model that produces at least one).

Now, I don't think there is any conceivable way that we will ever be able to obtain direct evidence of universes with different fundamental laws. Though I do think it may be interesting to think about the possibility.

 

[marcus]

Since we just turned a page, I'll recall the post of mine (#112) you were responding to just now

==quote post #112==

The only way in which it's an interesting topic of discussion is in regard to the low-energy laws of physics varying from place to place within the universe.

The way I think of it can connect with this. I think of the universe as the unique whole of nature with a unique set of fundamental laws.

Spontaneous symmetry-breaking may have resulted in regional variation in some constants that emerge at lower energy. I think that is extremely interesting and I think it is something we humans may be able to study and understand.

At present I don't see any compelling reason to involve multiple big bangs in our model of the big bang---the start of expansion. Pending evidence to the contrary I expect one start of expansion, operating under one set of fundamental highenergy laws of physics will probably fit the data.

It would be quite interesting if we got some evidence of other big bangs having happened, of course. But absent such evidence *shrug*.

So my view is similar to the one expressed in post #1---the reductionist program is on track, no need to give up on the program of finding ever deeper explanations for what we see in terms of one universe, one start of expansion, one set of fundamental physics laws.

==endquote==

Then continuing the discussion with your post #113, which think was mainly in response to what I just highlighted blue:

Well, it kinda has to. But that same model may unambiguously predict other bangs, other low-energy laws of physics (our current model already predicts other low-energy laws of physics...and other bangs are the natural expectation of any model that produces at least one).

Now, I don't think there is any conceivable way that we will ever be able to obtain direct evidence of universes with different fundamental laws. Though I do think it may be interesting to think about the possibility.

That seems pretty reasonable to me.

 

[Radrook]

I understood Hawking to mean that the universe is as it is because we can't help but perceive it that way because of what we are-human.