L. Susskind -- The "Megaverse"

[Perpetual Student]

Yes, but the question of "fine-tuning" is one that arises if, and only if, a set of conditions hold for a theory of the cause of universal constants. We do not have such a theory.

No.

Certainly, whether there is a mechanism that produces various constants is an interesting scientific question, but until such a mechanism has been identified, fine-tuning isn't an issue.

No.

 

[TubbaBlubba]

No.


No.

Oh, okay.

 

[Fudbucker]

Precisely. How can one then say that the "tuning" of the constants is "improbable"?

The results coming from the LHC suggest the universe may be unnatural (i.e., there's no underlying principle that would allow us to derive the values of the physical constants). If the universe is unnatural, the fact that we are here is incredibly improbable. So much so, that positing a macroverse of nearly infinite universes is becoming an accepted alternative. Even though there is, for all intents and purposes, no evidence that any other universes exist.

No, I am saying that _I do not understand_ how people with intellectual honesty can understand it to be a problem.

This is my layman's understanding: it's been a problem for decades, but the prevailing view was that the universe is ultimately natural; there is some underlying framework that would explain all the values. Now unnaturalness is looking more and more likely, so fine-tuning is looking more and more like a major problem.

 

[TubbaBlubba]

The results coming from the LHC suggest the universe may be unnatural

Do you or anyone else have a good article on this to recommend?

(i.e., there's no underlying principle that would allow us to derive the values of the physical constants). If the universe is unnatural, the fact that we are here is incredibly improbable. So much so, that positing a macroverse of nearly infinite universes is becoming an accepted alternative. Even though there is, for all intents and purposes, no evidence that any other universes exist.

If there is no underlying principle, how could one arrive at a probability density for our existence?

 

[Kwalish Kid]

No.


No.

Hmmm. It seems that you have no idea what you are talking about.

TubbaBubba is exactly right that we cannot make any sense of the idea that the values of the physical constants are improbable unless there is some assignation of probability to the values that they constants can take.

Those who take the fine-tuning problem seriously do attempt to produce some way to account for the origin of values of the physical constants and the probability that the values will end up in a certain range. So far, this seems in the very speculative stage.

 

[Perpetual Student]

Hmmm. It seems that you have no idea what you are talking about.

TubbaBubba is exactly right that we cannot make any sense of the idea that the values of the physical constants are improbable unless there is some assignation of probability to the values that they constants can take.

Those who take the fine-tuning problem seriously do attempt to produce some way to account for the origin of values of the physical constants and the probability that the values will end up in a certain range. So far, this seems in the very speculative stage.

Well, sorry, it's not the first time. I wonder if Susskind, Hawking and thousands of other physicists around the world who regard fine-tuning as a problem are aware that they too "have no idea what (they) are talking about"?

 

[TubbaBlubba]

Well, sorry, it's not the first time. I wonder if Susskind, Hawking and thousands of other physicists around the world who regard fine-tuning as a problem are aware that they too "have no idea what (they) are talking about"?

Are they perhaps making un- or seldom stated assumptions?

 

[Fudbucker]

Do you or anyone else have a good article on this to recommend?

I posted this once before, I think. Can't remember.
http://www.scientificamerican.com/a...cations-lend-support-to-multiverse-hypothesis

If there is no underlying principle, how could one arrive at a probability density for our existence?

Very good question. Maybe one of our resident physicists could answer it?

ETA: my crack at it is, the expected values of the constants were so far off what was actually observed, we assume there is a huge range of possible values.

 

[TubbaBlubba]

ETA: my crack at it is, the expected values of the constants were so far off what was actually observed, we assume there is a huge range of possible values.

I don't think I can fully appreciate the consequences of the following:

However, in order for the Higgs boson to make sense with the mass (or equivalent energy) it was determined to have, the LHC needed to find a swarm of other particles, too. None turned up.

With the discovery of only one particle, the LHC experiments deepened a profound problem in physics that had been brewing for decades. Modern equations seem to capture reality with breathtaking accuracy, correctly predicting the values of many constants of nature and the existence of particles like the Higgs. Yet a few constants — including the mass of the Higgs boson — are exponentially different from what these trusted laws indicate they should be, in ways that would rule out any chance of life, unless the universe is shaped by inexplicable fine-tunings and cancellations.

Still - if our universe is some sort of random fluctuation in space-time, never mind its precise nature, there's no reason to assume there's only one. So I guess the multiverse hypothesis is a natural consequence of that.

I don't see how fine-tuning would be the central question in this, though. It might possibly be a supporting argument for multiple universes, however - but again any speculation about the birth of the universe requires many assumptions for it to even arise.

 

[Perpetual Student]

Roughly speaking a model for selecting values over a probability distribution.

Let's take the simple example, used by Susskind, of the electron. As he points out, there is no fundamental reason for the existence of the electron -- it just is! We can certainly imagine a universe without the electron -- and it would certainly be a very different universe, devoid of the chemistry we know.

Now, there is also no known reason for the electron to have the mass and charge that we know it has. So, what is the probability of the electron having some specific mass? I seems it would be like asking what is the probability of randomly selecting some value on the real line, which, of course, is zero.
So, the probability of the mass of the electron, (assuming there is no theoretical basis for its given value -- and there seems to be none) is zero.
Now, you can extend that logic to every other value of all the fundamental constants we know. Lacking any theoretical basis for the values we have, the probability of these values is zero -- individually and collectively.

 

[TubbaBlubba]

There is no uniform probability distribution over the reals. There CANNOT be, by the axioms of probability.

 

[Perpetual Student]

There is no uniform probability distribution over the reals. There CANNOT be, by the axioms of probability.

The fact that one cannot construct a "uniform" probability is not relevant to my point.

 

[TubbaBlubba]

Yes, it is, because you are implying that the values are uniformly distributed over the real line. This is well known to lead to paradoxes.

 

[Perpetual Student]

Yes, it is, because you are implying that the values are uniformly distributed over the real line. This is well known to lead to paradoxes.

Look at it as analogous to the Dirac delta function, which is vital to QM, but yet is not a rigorously defined function.

If you are arguing that the value of the probability of some value of any fundamental constant is not zero, simply because we cannot define a uniform distribution, you are being pedantic; my point does not depend on "uniformity."
The fine-tuning question, as I have presented it (or as Susskind described it -- if you don't like my description), is quite real and does not require some "underlying principle."

 

[Kwalish Kid]

Well, sorry, it's not the first time. I wonder if Susskind, Hawking and thousands of other physicists around the world who regard fine-tuning as a problem are aware that they too "have no idea what (they) are talking about"?

Yeah, you're being overly defensive.

Sadly, it is true Hawking and thousands of other physicists around the world don't have any idea what they are talking about sometimes. Anyone who attempts to present a fine-tuning problem without giving some account of the probability distribution of the parameters in question is presenting nonsense. If someone gives a bad account of the probability distribution, then they aren't doing a good job.

 

[TubbaBlubba]

Look at it as analogous to the Dirac delta function, which is vital to QM, but yet is not a rigorously defined function.

Correct, but as I understand it, with emphasis on function. It's a mathematical object that can sometimes be thought of as a function.

If you are arguing that the value of the probability of some value of any fundamental constant is not zero, simply because we cannot define a uniform distribution, you are being pedantic; my point does not depend on "uniformity."

Okay, I postulate that the distribution isn't uniform, but is a very but not entirely sharp spike around the current values, for every constant. For a sufficiently well chosen distribution the probability of a universe supporting life approaches 1.

 

[Kwalish Kid]

Now, there is also no known reason for the electron to have the mass and charge that we know it has. So, what is the probability of the electron having some specific mass? I seems it would be like asking what is the probability of randomly selecting some value on the real line, which, of course, is zero.

So, the probability of the mass of the electron, (assuming there is no theoretical basis for its given value -- and there seems to be none) is zero.

Now, you can extend that logic to every other value of all the fundamental constants we know. Lacking any theoretical basis for the values we have, the probability of these values is zero -- individually and collectively.

But this is a ridiculous argument. It is assigning a particular probability distribution in ignorance. There is no reason to prefer the Susskind assignment of probabilities over one that assigns probability of 1 to all the measured values.

 

[Roboramma]

Okay, I postulate that the distribution isn't uniform, but is a very but not entirely sharp spike around the current values, for every constant. For a sufficiently well chosen distribution the probability of a universe supporting life approaches 1.

That still leaves us with the question of why this distribution rather than some other.

 

[Perpetual Student]

...
...
Okay, I postulate that the distribution isn't uniform, but is a very but not entirely sharp spike around the current values, for every constant. For a sufficiently well chosen distribution the probability of a universe supporting life approaches 1.

Well, then why not three spikes or fifteen spikes or 10¹⁵ spikes or no spikes. The point is we do not know, so all these are equally likely. Hence, we have a fine-tuning question.

 

[Perpetual Student]

But this is a ridiculous argument. It is assigning a particular probability distribution in ignorance. There is no reason to prefer the Susskind assignment of probabilities over one that assigns probability of 1 to all the measured values.

But this is a ridiculous argument. We are indeed ignorant of whatever probability distribution might determine the values, so we have no reason to prefer one over another. Assigning a probability of 1 would be like assigning 27 equal spikes (see above) equally distributed by some factorX(1/π). It's not a guessing game; we really don't know!