Monday, April 21, 2008

Inflation Sinking


In the 1970's as today, rising prices were on everyone's mind. Inflation was proposed to explain uniformity in the cosmic microwave background (which shows that the speed of light has slowed) and other puzzles. Physicist Alan Guth and others proposed that the early Universe expanded at warp speed, many times faster than light. Inflation would violate both the First Law of Thermodynamics and Relativity's stipulation that nothing can travel faster than light. Theorists invoked mysterious "scalar fields" and "inflatons" to explain their idea. None of these inferrences has ever been observed in nature. Inflation theories have multiplied like epicycles, providing comfortable careers for Guth and others.

April is also the 96th anniversary of the Titanic sinking! Back in 2006 (when St. Louis reached 1000 viewings of this blog) we saw The Puncturing of Inflation and Inflation Leaking. The leak in Titanic didn't look bad at first, but ship's designer Mr. Ismay did some quick calculations and realised they were sunk. Like the first boats from a sinking ship, leading scientists are abandoning the inflationary paradigm. At the beginning of this New Year physicist Paul Steinhardt publicly abandoned inflation, and is now working on cyclic universes. Read the links for more.

As with the string enterprise, inflation theorists are increasingly desperate to prove their idea. 5 years ago they prevailed upon the WMAP team to interpret the data their way. More recently they have claimed that gravitational waves would be a "smoking gun" proof. Too bad that gravitational waves have never been detected. Monday Physicist Lawrence Krauss, who has been arguing against inflation since 1992, revealed to us what he has been up to. In the new issue of Physical Review Letters; Krauss, grad student Katherine Jones-Smith and associate professor Harsh Mathur show that these waves, if detected at all, could come from an entirely different source. So much for smoking guns.

Inflation predicts that the Universe is geometrically flat, like the Earth. The WMAP angular correlation function is from Glenn Starkman and Dominik Schwarz, Scientific American August 2005. Inflation predicts that density fluctuations are the same at all scales, leading to the red prediction curve. In fact fluctuations are nearly zero for angles greater than 60 degrees. Note the error bars--inflation's prediction is ruled out by both WMAP and COBE. As surely as a ship's sails disappearing over the horizon, this shows that the Universe is curved with the radius predicted.

Though it has been useful to cosmologists, the inflationary paradigm can not, repeat can not be proven. Humans can not time-travel to the first 10^{-33} seconds, and no human experiment can approach the titanic energies and densities near the Big Bang. A real Theory would offer a falsifiable prediction: that the speed of light has changed. No experiment can prove that c is constant, for a more accurate experiment can always prove that foolish. It is very possible to prove the Theory. Type Ia supernovae, the "Faint Young Sun" and Lunar Laser Ranging can all be interpreted to show not just that c has changed, but that it has changed in exactly the amounts predicted.

Here's why we are winning: An independent mind is far more agile than a group mind. Independent thought allows new ideas to be considered, checked, discarded or advanced at the speed of thought. Group minds are slowed by speed of communication, time for ideas to be digested by members of the group, organizational rot and reluctance to part with familiar notions. Old paradigms will last until their proponents go to their grave, then a better Theory takes over..

Labels: ,

16 Comments:

Blogger nige said...

Thanks for this excellent post. The correlation in the diagram you provide is extremely weak, indicating that the universe is very flat (only very slight curvature).

When the cosmic background radiation was emitted at 300,000 years after the big bang, the universe was just becoming transparent (it had previously been ionized and opaque to radiation, but after that time the temperature dropped below 3000 K so electrons combined with protons to produce non-ionized hydrogen gas, which is transparent to radiation unlike the ionized form).

Hence, the ripples in the cosmic background radiation as a function of azimuth angle correspond to density fluctuations in the universe (mainly hydrogen gas) which existed at 300,000 years.

These ripples as a function of angle of sky were observed to be about 10,000 times smaller than predicted. Hence, at 300,000 years after the big bang, the universe was 10,000 times flatter than the amount of curvature predicted by general relativity.

As you point out, the inflation bangwaggon is much like that of string theory. People come up with vague ideas to 'test' such things like superpartners of unknown energy or magnetic monopoles, but these aren't really tests because if you don't see them, the theory happily survives to branewash everyone that the unobserved particles 'must be' of higher energy than previously hoped.

One thing I don't grasp from your post, Louise, is why you think:

"Note the error bars--inflation's prediction is ruled out by both WMAP and COBE. As surely as a ship's sails disappearing over the horizon, this shows that the Universe is curved with the radius predicted."

This is a bit misleading. The observed curvature at 300,000 years as shown by the plotted cosmic background radiation temperature fluctuations in your diagram, is extremely small, it's tens of thousands of times smaller than what classical theory (general relativity with constant c and G in it) predicts.

The curve marked theory in the diagram is the result from inflation theory, not from classical general relativity (which would show massive deviation, much bigger!). The inflation theory is actually not a bad fit to the observed data, compared to the predicted curvature from general relativity which is far greater.

I don't see how you can explain the magnitude of the observed deviations using a varying speed of light unless you go in for inflation as Guth suggests. You need some way to explain why the fluctuations and thus the curvature of the universe is so small at 300,000 years after the big bang.

My calculation from a quantum gravity idea shows that G is directly proportional to time since the big bang. This predicts the flatness of the universe at 300,000 years very simply because G was 300,000/13,700,000,000 = 1/46,000 times today's value.

This explains very simply why gravitation was so weak, giving a small amount of curvature at early times when galaxies were being seeded by density fluctuations in the expanding cloud of gas.

The actual mechanism that I'm dealing with is that gravitation effects are the shadowing of the exchange of gravitons between masses. Since the outward Hubble recession v = HR gives a = dv/dt = Hv = RH^2, the effective outward force of the receding universe (mass m) around us is on the order of F = ma = mRH^2. By Newton's 3rd law there's a reaction force, and this is carried by gravitons. When the details are worked out, this reaction force predicted by Hubble's and Newton's laws is predicts gravitational strength G correctly to within experimental error (10% or so) by a simple Le Sage shadowing geometric mechanism.

Traditionally the argument against varying G is Teller's argument that star radiance is a sensitive function of G and this would upset evolution during the age of the earth by affecting temperature. However, Teller ignored the fact that there are great similarities between electromagnetic and gravitational mechanisms (both are inverse square law, infinite range forces, whose coupling constants are accurately predictable by this mechanism of exchange radiation). Hence, it is clear that both gravitation and electromagnetism are different aspects of the same unified force and they both vary with time in the same way. This masks the effect on solar radiating power, because the variation in Coulomb potential with time then offsets the variation in fusion rate that teller naively predicted would result from a variation in gravitational compression in the star (Coulomb forces between protons tend to resist fusion).

So the graph you produce is explained by quantum gravity with G varying instead of c varying?

10:13 AM  
Anonymous Anonymous said...

Oh yea! What? Huh? I like apples.


-Steve

11:35 AM  
Blogger Matti Pitkanen said...

I dare to believe that quantum criticality and hierarchy of Planck constants (gigantic for flux tubes mediating gravitational interactions) will be the natural follower of the inflation paradigm.

Mere criticality together with imbeddability fixes the cosmology apart from the duration of the critical phase: also super-critical periods are possible but also last also only finite period before change to criticality followed by subcriticality.

Quantum criticality implies that no scales are present except those associated with space-time sheets themselves and naturally explaining the lack of angular correlations at very large angle separations.

I have tried to communicate the message about quantum criticality as a natural replacement of inflationary scenario for at least 17 years now. Without any intelligent response.

Dare I hope that time is becoming ripe now? Or is the hierarchy of Planck constants and macroscopic quantum gravitation too much for the string hegenomy? Do colleagues want still to keep the cherished belief that quantum gravity is testable only at Planck scale or are they finally ready to see the reality and start to develop and test TGD?

5:48 PM  
Blogger Kea said...

Matti, in order to see reality, they must abandon most of their cherished ideas about the particle spectrum, let alone cosmology. Although they are getting awfully close, especially on the mathematical side of M theory, it might still take a lifetime for another generation to go to its grave.

Thanks Louise, for a fantastic post.

9:24 PM  
Anonymous Anonymous said...

peanut butter is good with grape jelly...did I add to the debate?

1:46 PM  
Anonymous Parvulus said...

"Though it has been useful to cosmologists, the inflationary paradigm can not, repeat can not be proven. Humans can not time-travel to the first 10^{-33} seconds"

Just as we can say:

"Though it has been useful to cosmologists, the Copernican Principle can not, repeat can not be proven. Humans can not travel outside the Local Group of Galaxies."

Very interesting posts, Louise. Keep it up.

1:53 PM  
Blogger L. Riofrio said...

Thank you all for your such informed comments. This is a long fight, as Krauss has been arguing against inflation since 1992. Perhaps as old ideas fade the ideas of Kea, Matti, and Nige will recieve consideration.

For nige: Theory predicts a curvature of radius R = ct, which produces a spectrum close to COBE and WMAP. I will post the prediction graph again soon. There is room for your varying G to be considered too. Teller completely ignored that, according to astrophysics, life shouldn't have evolved on Earth because solar luminosity has been increasing. The Faint Young Sun Paradox can be explained by varying c or G.

For parvulus: True, but at least we can prove that the Sun doesn't circle the Earth.

7:08 PM  
Anonymous Anonymous said...

Hey babe, can I be your dumb trophy hubby/arm candy?


-Steve

11:25 AM  
Anonymous Ethan Siegel said...

Louise,

Your post is interesting, but extremely (and unfairly) harsh on inflation. There is, of course, no direct way to test inflation, as the energies required to make the Universe inflate again are ridiculous, nearly string-theory like. (See this post for details.)

But there are many different models of inflation, and these models can predict many different angular correlation functions at large angles. Inflation really isn't as predictive in that regime as you attribute it to be.

Time variations in the gravitational constant and in the speed of light have been tightly constrained so that G ~ t is out; G. Raffelt's paper as of a few years ago concluded that for G ~ t^b, |b| < 0.08, constraining your scenario tightly. Similar constraints (although the info isn't at my fingertips) exist for the time variation of c.

Inflation, generically, predicts a flat Universe with uniform temperature and entropy, devoid of topological defects and ultra-high energy relics, and a spectrum of density fluctuations that agrees remarkably with experiment. I think your gripes are sensationalized, and I don't see the validity to the alternative you are advocating.

Ethan

8:43 AM  
Anonymous Ethan Siegel said...

Louise,

Here you are; I've gone and looked up all of the relevant information constraining the time variation of the speed of light.

You can read about it here: http://startswithabang.com/?p=366

It quite clearly rules out your theory that c is proportional to t^(-1/3). By a lot. The signal your theory predicts is 200 times the experimental limits.

Ethan

2:57 PM  
Blogger L. Riofrio said...

The thoughtful comments from astrophysicist Ethan are appreciated too. His cited wikipedia link: http://en.wikipedia.org/wiki/Fine-structure_constant is useful reading. Webb and Linaweaver are both well-regarded by Australians. Both those experiments measure not c, but the fine-structure constant which depends on product hc. If c is decreasing, then h must increase. This is mentioned long ago in the "Very short paper" of 2004, and has great implications for thermodynamics.

7:11 PM  
Anonymous Ethan Siegel said...

No. While changes of a fraction of a percent over billions of years may be possible for G and c, this is not possible for Planck's constant. Why not? Big Bang Nucleosynthesis. This has been known since the year I was born. See Ira Wasserman and Brecher's paper: http://prola.aps.org/abstract/PRL/v41/i14/p920_1

So let's follow your logic. If c changes as t^(-1/3), Planck's constant changes as t^(1/3) to keep the fine-structure constant relatively unchanged. But Wasserman and Brecher find that h can vary, at most, by a factor of 3, from the time when the Universe was a factor of 10^15 younger than it is now. But your prediction yields a difference of a factor 100,000.

How do you account for that?

8:16 PM  
Blogger L. Riofrio said...

A proper reply will require going to the university library, which may take until after the weekend. From my hazy memory, Wasserman and Brecher's paper took the opposite tack: Considering c fixed and looking at change in h. Changing the speed of light was not even a subject in 1978. I did the BBN calculation back in school and found no dependence on fixed h. Those notes will be dug up soon.

For Ethan: Many things were indeed known when your were born. One can repeat those things through school and possibly land a university position. You should concern yourself with what happens after you are born, and what difference you will have made. The "dark side" can be tempting, but leads to epicycles of irrationality and will never solve cosmological problems.

6:25 AM  
Anonymous Ethan Siegel said...

Again, no. There are different combinations of the constants h, c, and G, but they don't always appear together at the same order. The fine-structure constant tests the combination hc. G. Raffelt's paper tests G on its own. And the Wasserman/Brecher paper tests, depending on which equation you're looking at, h^(5/2)c^(3/2) or h^(2)c^(6).

My point is that there isn't a conspiracy to keep your theory out of publication, there are experiments and observations that contradict your theory. I'm one of the most open-minded theorists you'll find out there, and I'm telling you that the onus is on you to show that your theory can stand up to this scrutiny.

I've read the "very short papers" you have linked to, and they don't stand up to the four constraints I've told you about (time-varying G, time-varying alpha from QSO absorption lines, time-varying alpha from Oklo, and time-varying h from BBN). And they don't stand up to it by orders of magnitude.

You can complain about a theory being ugly and about dark matter and dark energy being epicycles or patches all you want. You've devised a beautiful "cosmographicum mysterium" a la Kepler, which is to say, it's pretty, but it's also invalid. I've done it, too; I've submitted a paper before that had invalid results. It gets rejected, and you move on to the next interesting idea you have. That's all.

8:30 AM  
Blogger L. Riofrio said...

"I've done it, too; I've submitted a paper before that had invalid results."

So sad that it should happen. In school we also learn about "straw man" arguments. There is nothing here about hating inflation, varying G or alpha. (More about BBN soon.)

2:32 PM  
Blogger Stephen said...

The layperson is told that Inflation (the Inflation hypothesis?) was invented because the temperature of the CMBR (cosmic microwave backround radiation) across vast reaches of the Universe are very similar. And, that could only be true if these regions communicated with each other.

But couldn't it be that there was a common process that tended to produce similar results? I mean, if a car company has two plants, and they make cars from the same plans, it's OK to call both sets of cars '08 Mustangs.

Now, perhaps quantum fluctuations don't qualify as a common process. But it seems that not everything is known about how the Universe works. Maybe there's some rule we haven't learned.

As i understand it, Quantum and GR are known to be incompatible. These are really, really solid theories, even though one or both are wrong.

So, in laymen's terms, is there a good reason for Inflation? WMAP evidence? Supernova data? Is Inflation getting stronger, or are competing theories getting ruled out?

Oh jeez. This capta's impossible.

11:06 AM  

Post a Comment

Links to this post:

Create a Link

<< Home

Locations of visitors to this page