L. Susskind -- The "Megaverse"

[Reality Check]

A preponderance of red-heads would only be a problem if you had prior knowledge of the distribution of red-headed people.

That is not Perpetual Student's analogy, Fudbucker.
It is everyone who is red-headed.
The problem is that the visitor has speculated about (or maybe knows of or maybe is even blond!) the existence of non-red-headed people. So the question is "why are all of the people here red-headed" and Perpetual Student gave a list of possible reasons.

 

[Fudbucker]

That is not Perpetual Student's analogy, Fudbucker.
It is everyone who is red-headed.
The problem is that the visitor has speculated about (or maybe knows of or maybe is even blond!) the existence of non-red-headed people. So the question is "why are all of the people here red-headed" and Perpetual Student gave a list of possible reasons.

The visitor, in Perpetual's analogy, is a visitor with the background knowledge that there are a lot of different hair colors.

Wrt to fine-tuning, we lack that background knowledge. It's looking likelier that the values of the physical constants are arbitrary, but it's still highly speculative.

For the analogy to work, you need someone who really doesn't know what kind of variance of hair color there is, although they've made some educational guesses (which have all turned out to be way off the mark).

 

[Kwalish Kid]

I deliberately didn't read Kwalish's response. I'm curious if he will bring up a similar point. LOL, I read it and can't tell what he's saying.

Let me try again:

1. People claim other universes (of some sort) exist with some specific properties.
2. The reason: fine-tuning.
3. We know that fine-tuning exists because other universes exist...

 

[Perpetual Student]

The visitor, in Perpetual's analogy, is a visitor with the background knowledge that there are a lot of different hair colors.

Wrt to fine-tuning, we lack that background knowledge. It's looking likelier that the values of the physical constants are arbitrary, but it's still highly speculative.

For the analogy to work, you need someone who really doesn't know what kind of variance of hair color there is, although they've made some educational guesses (which have all turned out to be way off the mark).

Look, ripping apart an analogy misses the purpose of the analogy, which is to provide a frame for making a point. Let the visitor to the town of redheads be an alien with no knowledge of human traits. The point is that he might still speculate about the red hair color he sees (let's assume he has human-like vision) without any knowledge of human hair color. That's all. I'm merely pointing out that we do not require prior knowledge of the probability distribution of some phenomenon to speculate about its likelihood. Making rules about when we can and cannot speculate is a bit bizarre.

 

[Perpetual Student]

Let me try again:

1. People claim other universes (of some sort) exist with some specific properties.
2. The reason: fine-tuning.
3. We know that fine-tuning exists because other universes exist...

This is a classic straw man! Can you provide an example of someone reasoning this way?

 

[sol invictus]

Let me try again:

1. People claim other universes (of some sort) exist with some specific properties.
2. The reason: fine-tuning.
3. We know that fine-tuning exists because other universes exist...

That certainly doesn't make much sense, but it's neither the way the question evolved historically nor the way people reason about it now.

Historically, the fine-tuning question has existed for at least decades in various forms. In some cases during the development of the standard model of particle physics, tunings did turn out to have an explanation that was discovered later (not a multiverse, but a dynamical mechanism of the sort I referred to in my third option). So this way of thinking does have a track record of success.

The current situation is roughly this - among other tunings, observation indicates that the cosmological constant is not zero, but has a value that is smaller than its "natural" value by 123 orders of magnitude. Here "natural" refers to the value one predicts using standard quantum field theory with generic values for the parameters (for example, all dimensionless coupling constants chosen with a uniform distribution between 0 and their strong coupling value). So why is the cosmological constant so small? I think that's very clearly a scientific question, and as I said before there are two possibilites:

(1) The value is unique (in which case, why?), or

(2) The value is not unique and varies, in which case it probably takes the value it does in our part of the universe because larger values are incompatible with life.

The value of the CC was predicted by Steven Weinberg well before it was measured, using the logic of (2). To repeat: this logic led to a falsifiable prediction - one that by the way essentially no one at the time believed or took seriously, despite the scientific eminence of its author - that was later confirmed by observation. Luck? Maybe, but to dismiss it out of personal prejudice is not scientific.

 

[Fudbucker]

Look, ripping apart an analogy misses the purpose of the analogy, which is to provide a frame for making a point. Let the visitor to the town of redheads be an alien with no knowledge of human traits. The point is that he might still speculate about the red hair color he sees (let's assume he has human-like vision) without any knowledge of human hair color. That's all. I'm merely pointing out that we do not require prior knowledge of the probability distribution of some phenomenon to speculate about its likelihood. Making rules about when we can and cannot speculate is a bit bizarre.

I'm not "ripping it apart". I thought you made a good point. However...

Aliens would definitely speculate about all these read heads, that seems right. But the presence of all these red-heads wouldn't constitute a problem without more information on which to base a probability judgement.

Fine-tuning isn't just speculation, it's a problem that's got more and more people thinking about multiverse theory. Part of the problem is that our predictions for these values are way off, so we do have some idea what they should be.

 

[Fudbucker]

Look, ripping apart an analogy misses the purpose of the analogy, which is to provide a frame for making a point. Let the visitor to the town of redheads be an alien with no knowledge of human traits. The point is that he might still speculate about the red hair color he sees (let's assume he has human-like vision) without any knowledge of human hair color. That's all. I'm merely pointing out that we do not require prior knowledge of the probability distribution of some phenomenon to speculate about its likelihood. Making rules about when we can and cannot speculate is a bit bizarre.

You do need to know something about the phenomenon. "Problems" regarding theories often arise due to surprising results. The only way to have a surprising result is to already have an expectation that a piece of evidence will be like X. That requires some amount of prior knowledge. When the evidence turns out to be completely not like X, then you have a problem.

There's two surprises that make up fine-tuning:
1. We didn't expect the universe to be so balanced wrt to all these values. Just a small change in one of like 20 different values, and complex structure would have been impossible.
2. The values of some (all?) of these constants were way off what we expected them to be.

 

[Fudbucker]

That certainly doesn't make much sense, but it's neither the way the question evolved historically nor the way people reason about it now.

Historically, the fine-tuning question has existed for at least decades in various forms. In some cases during the development of the standard model of particle physics, tunings did turn out to have an explanation that was discovered later (not a multiverse, but a dynamical mechanism of the sort I referred to in my third option). So this way of thinking does have a track record of success.

The current situation is roughly this - among other tunings, observation indicates that the cosmological constant is not zero, but has a value that is smaller than its "natural" value by 123 orders of magnitude. Here "natural" refers to the value one predicts using standard quantum field theory with generic values for the parameters (for example, all dimensionless coupling constants chosen with a uniform distribution between 0 and their strong coupling value). So why is the cosmological constant so small? I think that's very clearly a scientific question, and as I said before there are two possibilites:

(1) The value is unique (in which case, why?), or

(2) The value is not unique and varies, in which case it probably takes the value it does in our part of the universe because larger values are incompatible with life.

The value of the CC was predicted by Steven Weinberg well before it was measured, using the logic of (2). To repeat: this logic led to a falsifiable prediction - one that by the way essentially no one at the time believed or took seriously, despite the scientific eminence of its author - that was later confirmed by observation. Luck? Maybe, but to dismiss it out of personal prejudice is not scientific.

It helps having a physicist in the discussion. I hadn't heard about Weinberg's prediction. How did he predict it? He just figured, what's the value needed for someone like me to be around to observe it?

 

[Perpetual Student]

OK. In that case, I fully agree. It certainly is meaningless to talk about the values of dimensionful constants unless you've specified a set of units. But units are just a way of measuring one dimensionful quantity relative to another - in other words, taking ratios of two dimensionful quantities to produce a dimensionless quantity.

So it's only dimensionless quantities that have any physical meaning.

Are not all coupling constants dimensionless by definition?

 

[mijopaalmc]

It helps having a physicist in the discussion. I hadn't heard about Weinberg's prediction. How did he predict it? He just figured, what's the value needed for someone like me to be around to observe it?

Anthropic Bound on the Cosmological Constant*

In recent cosmological models, there is an "anthropic" upper bound on the cosmological constant Λ. It is argued here that in universes that do not recollapse, the only such bound on Λ is that it should not be so large as to prevent the formation of gravitationally bound states. It turns out that the bound is quite large. A cosmological constant that is within 1 or 2 orders of magnitude of its upper bound would help with the missing-mass and age problems, but may be ruled out by galaxy number counts. If so, we may conclude that anthropic considerations do not explain the smallness of the cosmological constant.

(emphasis added)

*free full-text pdf

 

[Fudbucker]

Thanks, Mijo.

 

[Perpetual Student]

I would like to try another analogy -- please indulge me.
Suppose you entered a town you had never visited before and found that everyone you saw had red hair. In every shop, in every home and everyone you saw in the street walking along had red hair. Now, might you not speculate how this can about? Was the town founded centuries ago by some redheads? Does everyone here dye their hair? Do only red-haired people move to this town? Is there something in the water causing then to become redheads?
Well, you can freely speculate why everyone you see in this town has red hair without knowing anything about the probability distribution of red hair. Of course, knowing the probability distribution will tell you what the chances are of this being a result of chance, but that's not necessary to speculate.
I know this analogy is not exact since we know what possible colors of hair humans have but we do not know anything about what alternative values (if any)the constants might have. My point here is to provide some framework as to why it is reasonable to speculate about fine-tuning with no knowledge of a probability distribution.
Now you can tell me why this is all wrong!

As I indicated above, perhaps the analogy is better if the visitor to the town is an alien with no information about human hair color. The above analogy was only intended to address the claim that, we must have a probability distribution, otherwise speculation about the likelihood of something is somehow not justified or even "irresponsible." If you were to come across any species of animal that you have never seen before and noticed a certain color pattern to its skin, fur or whatever, it would not be unusual or "irresponsible" to speculate about whether the individual you are observing is typical or if there are many variations of coloring among this species.
I did not intend to address the larger question of fine-tuning and what evidence one might need to extend our speculation to that issue. It seems to me that it is when we observe many fundamental quantities in combination that fine-tuning comes up.
Nevertheless, the comment Susskind makes in the interview about Λ is quite compelling in and of itself.

 

[sol invictus]

It helps having a physicist in the discussion. I hadn't heard about Weinberg's prediction. How did he predict it? He just figured, what's the value needed for someone like me to be around to observe it?

Large positive CC prevents galaxies and other complex structures from forming. Weinberg calculated the maximum possible value that allowed structure formation, and then reasoned the CC should be close to that value (since its natural value is much, much larger, so it's very unlikely to be smaller).

Are not all coupling constants dimensionless by definition?

No - for instance Newton's constant is the coupling constant for gravity, but it has dimensions.

 

[Perpetual Student]

No - for instance Newton's constant is the coupling constant for gravity, but it has dimensions.

It has been my understanding that G is the gravitational constant, not the coupling constant for gravity, which is denoted by α_G and is dimensionless. Is that not the case?

 

[Reality Check]

The visitor, in Perpetual's analogy, is a visitor with the background knowledge that there are a lot of different hair colors.

That is what "the visitor has speculated about (or maybe knows of or maybe is even blond!) the existence of non-red-headed people" means.

So the visitor can say there is there is something distinctive about there only being red-headed people in the town. IOW: The town is tuned for red-haired people . But they do not need to know the distribution of red-headed people in different towns.

 

[Kwalish Kid]

The value of the CC was predicted by Steven Weinberg well before it was measured, using the logic of (2). To repeat: this logic led to a falsifiable prediction - one that by the way essentially no one at the time believed or took seriously, despite the scientific eminence of its author - that was later confirmed by observation. Luck? Maybe, but to dismiss it out of personal prejudice is not scientific.

That is a gross mischaracterization of what Weinberg did. There was no prediction in that paper, merely the setting of an upper bound on the cosmological constant given the structures we see and assumptions about the mass-energy densities of matter. All Weinberg needed for "luck" was to be not wildly off on his assumption of the mass-energy density of matter (he was off, but not wildly).

Additionally, I'm not sure who did not take this limit seriously, especially since it is cited in a number of papers (3501 according to Google scholar) and Weinberg's work on the subject is cited in Carroll, Press and Turner 1992 (Annu. Rev. Astron. Astrophys. 1992.30:499-542), a paradigm example of the literature on attempting to derive the cosmological constant from quantum phenomena.

That people do not flock to anthropic reasoning (on this subject or any others) is because it shouldn't do anything that scientists aren't already doing: taking into account the observable facts.

 

[sol invictus]

It has been my understanding that G is the gravitational constant, not the coupling constant for gravity, which is denoted by α_G and is dimensionless. Is that not the case?

It's not the case, no.

 

[sol invictus]

That is a gross mischaracterization of what Weinberg did. There was no prediction in that paper, merely the setting of an upper bound on the cosmological constant given the structures we see and assumptions about the mass-energy densities of matter. All Weinberg needed for "luck" was to be not wildly off on his assumption of the mass-energy density of matter (he was off, but not wildly).

Sorry, but you're wrong. Yes, Weinberg set an upper bound. But he also predicted that the actual value would be within 1-2 orders of magnitude of that bound - as opposed to zero, which is what nearly everyone else believed. His prediction was confirmed more than ten years later.

Now, if the intrinsic distribution of pV values is
smooth and featureless below the anthropic bound (as
would be expected in the models of Refs. 3-5, 8, and 9 if
the natural scale for pV is set by the Planck mass), then
it seems likely that pV would be within 1 or 2 orders of
magnitude of its upper bound. (This can be made more
precise by calculations like those of Carter in Ref. 2.)
With a lower bound of 550p0 on the anthropic upper
bound on pV, we would then conclude that pV must be
much greater than the present mass density p0.

(p0 is the matter density, pV is the vacuum energy).

We now know that pV is about 3 times larger than p0 - not MUCH greater, but greater, and within 1-2 orders of magnitude as he says (and later, more careful, analyses improved that further).

That people do not flock to anthropic reasoning (on this subject or any others) is because it shouldn't do anything that scientists aren't already doing: taking into account the observable facts.

Again, no. Observations can set bounds, but they cannot make predictions. Weinberg made a prediction.

 

[Kwalish Kid]

Again, no. Observations can set bounds, but they cannot make predictions. Weinberg made a prediction.

This is a very, very generous reading of his paper, especially if you think that anthropic reasoning is anything other than taking observations into account.