Monday, January 13, 2020 ... Deutsch/Español/Related posts from blogosphere

Cosmologists neglecting the motion of the Solar System are crackpots

For a year or two, all of us were aware of discrepancies in the measurement of Hubble's constant.

Several updated, comparably plausible methods yield values of \(H_0\) that differ by as much as 10%. I think that without the elimination of some of these determinations, we are forced to admit that very many cosmological parameters, perhaps including the age of the Universe, are only known with the precision of 10% or so and the "high-precision cosmology" started in the late 1990s was an illusion, at least at this moment it should be treated as one.

It is possible that the resolution to these strange results will be mundane – some error in some of the methods. It may be very deep, connected with some of the deepest facts about string theory that we still don't understand – some deep insights that will completely change our views about the landscape, swampland, and the cosmological constant in string theory. There exist various strategies by which the people try to wrestle with this new Hubble confusion.

I am getting Richard Mitnick's e-mails which are scientific articles copied from various places and I just received a one-week-old Phys.ORG article with an ambitious title,

New evidence shows that the key assumption made in the discovery of dark energy is in error
Cool, so five Koreans' one-month-old paper claims that the type Ia supernovae aren't as well-behaved "standard candles" as previously believed. There is a statistically significant dependence of their luminosity on the age, the authors claim, and this dependence may have the same effect that is normally attributed to the accelerating expansion (i.e. the dark energy i.e. the cosmological constant).

When this dependence is acknowledged to exist, the whole evidence for the accelerated expansion from the supernovae must be considered spurious. A big claim, indeed.

At this superficial level, the possibility makes complete sense. The real question is whether this new "age dependence of luminosity" is real. This is exactly the kind of an alternative, mundane explanation that must have been checked many times by the "supernova dark energy people", right? At this sociological level, it surely seems more likely that it is the Koreans who have made a mistake, not the supernovae people who have discovered the dark energy. But sociological arguments aren't strong.

Fine, who is right?

After some reading, you may find out that the five Koreans stand on the same side of "revolutionaries" as Colin et al., authors of an August 2018 paper
Evidence for anisotropy of cosmic acceleration
OK, these folks said that the cosmic acceleration isn't quite universal. It seems to depend on the direction in the sky. So it has a dipole component and they claim that they have "found" that the dipole is\[

q_0 = q_\mathrm{m} -8.03 \cdot \hat{n}\exp(-38.2 z).

\] Cool. So they have found that the acceleration depends on the redshift \(z\) in this utterly implausible way – the exponential of 38 times the redshift. I laughed out loud when I read this sentence carefully for the first time. You would need extraordinary evidence to support such an utterly implausible, mathematically ugly and artificial relationship. The statement that they have "found" this relationship is clearly a lie, hogwash. At most, they have made a fit (which is just an ad hoc way to emulate a discontinuous function) and showed the compatibility of this formula with some extremely limited subset of the data about the Cosmos. But the claim is still utterly implausible and couldn't have been "found" or "demonstrated".

Are the five Koreans or Colin et al. (the most famous guy in the group is almost certainly Subir Sarkar of Oxford) right?

In mid December, the Quanta Magazine's Wolchover wrote an article
No Dark Energy? No Chance, Cosmologists Contend
that makes it rather clear that other cosmologists don't believe these statements at all. And I think that they don't believe it. One December paper written by David Rubin and Jessica Heitlauf is quoted as the truly damning criticism of those revolutionaries, especially of the Colin et al. paper above and of Nielsen+Guffanti+Sarkar 2015 (who made the first recent claim that the zero acceleration was fine).

Rubin and Heitlauf chose the title
Is the expansion of the universe accelerating? All signs still point to yes
making it very clear what the two sides argue about. Rubin and Heitlauf argued that the "revolutionaries" have incorrectly treated the distribution of light-curve parameters of the type Ia supernovae. Distributions are hard. You know, the supernovae are surely not zero-parameter, totally repeatable "standard candles". They still have distributions and you need to compare "analogous" type Ia supernovae at various places i.e. deal with distributions which isn't easy.

A much more damning criticism is the finding by Rubin and Heitlauf that Colin et al. have failed to subtract the motion of the solar system relatively to the cosmic microwave background which can actually explain all of the famous Colin-et-al. "age dependence of luminosity". Wow. If true, this is a spectacularly elementary mistake, indeed. The solar system is moving by some 370 km/s, over 0.1% of the speed of light, and it may be divided to the motion within the Milky Way and the motion of the Milky Way etc.

This motion of our solar system – and the telescopes – is clearly one of the most elementary effects that are fully understood and that you need to take into account (or correct for) when you are trying to find the truth about the deeper questions that are closer to the fundamental laws of the Universe. Have Colin et al. indeed made this amazingly embarrassing error?

You don't need to study too much because a few days later, a month ago, Colin et al. published their
A response to Rubin & Heitlauf: "Is the expansion of the universe accelerating? All signs still point to yes"
where they very clearly admit that they haven't subtracted the motion of the solar system. In fact, they not only admit the mistake. They give two pompous names to their embarrassing mistake. To transform the data to the CMB frame is called an example of "arbitrary corrections" and those illustrate the "cosmological fitting problem" (which they capitalize, to make their arrogance look even more breathtaking, but I refuse to capitalize it) which says (George Ellis and W. Stoeger 1987) that it is immoral to correct for the motion of the Sun and similar things because our Universe is surely inhomogeneous and to produce corrected data is a blasphemy and cherry-picking.

This argumentation removes the authors from the list of trustworthy scientists. I will never seriously consider anything written by them again because they explicitly say that experimenters should be sloppy and they even invent and promote a whole ideology that is supposed to imply that it's right for experimenters to suck and be sloppy.

It's the main reason why we celebrate Copernicus – he discovered heliocentrism i.e. the fact that the observations done from the Earth aren't necessarily the most fundamental ones. Instead, they have to be transformed to more natural frames, the solar (or CMB) frame, to get more fundamental data which aren't distorted by relative trivialities. To deny (or even stigmatize) the basic and undoubtedly correct Copernicus' lesson in 2020 is just breathtakingly idiotic.

To remove simple, well-understood mundane effects from the observations when you want to find something "deeper" belongs among the first steps that a great, good, or even fair experimental scientist does at the very beginning. For example, when you observe something through a telescope with bluish glasses, you subtract the bluishness and you're careful not to make far-reaching conclusions about the objects' being blue. When something looks distorted in the telescope, a good observer asks whether it's due to the well-known unusual shape of his lenses or something like that.

When your apparatus evaporates a lot of water, you first check whether all the water has actually evaporated, instead of making claims – like Andrea Rossi – that you have discovered cold fusion.

When we see the CMB to be hotter in one direction than in the opposite direction, it may have various sources but the most mundane one is the Doppler shift. Its contribution is almost surely nonzero, regardless of any deeper assumptions about cosmology, and your interpretation of the data simply cannot afford to pretend that it should be zero. There may also be a dipole that has a more fundamental origin but that is not enough to neglect the fact that the Sun is moving by hundreds of miles per second relatively to the Milky Way, the CMB, and other frames that are clearly more fundamental and more natural in cosmology than the Sun's frame.

There is nothing "arbitrary" about the correction that you have to make to avoid far-reaching interpretations of effects that actually have a mundane explanation. Also, the need for this correction – e.g. one needed to transform our observations from the terrestrial frame to the CMB-at-rest frame – must be done regardless of the more fundamental assumptions about the homogeneity and isotropy of the Universe. Even in an inhomogeneous or anisotropic Universe, planets, stars, and galaxies still move and it's still a catastrophic mistake for a decent scientist to overlook these effects.

To claim that this absolutely elementary and unavoidable correction is "arbitrary" is just breath-taking. Indeed, it's the very same talking point that we may sometimes hear from scientifically unrefined climate skeptics who always explode whenever they hear "a correction". Corrections for mundane, large enough effects must always be made when deeper and less obvious effects are being looked for! When someone abuses the process of corrections to introduce biases, it's bad but the fact that some people are dishonest doesn't strip a good and fair scientist from taking mundane effects into account and correcting them!

For them to defend the indefensible overlooking of the motion of the solar system by big words such as "the cosmological fitting problem" is an example of a highly arrogant excuse. They basically invent a whole ideology and propaganda with its own terminology that is meant to justify their being lousy experimenters. In reality, their being sloppy has nothing to do with the inhomogeneous Universe: a scientist shouldn't be sloppy in any kind of the Universe and he should correct for all the sufficiently large effects that are understood.

It's really a universal fact about the scientific method that is true regardless of any technical questions about the laws of the Universe: some of the effects are easier and understood earlier, others are harder or deeper or waiting to be understood, and one must always properly account for all the former (easier or old) effects when or before he is discovering the latter (harder, deeper, or new) effects! To misunderstand the previous method means to misunderstand the very basics of the scientific method.

I am sorry but such a fundamental flaw in the methodology is enough for a sensible cosmologist not to take any claim interpreting some observations from Colin, Mohayaee, Rameez, and Sarkar seriously again. Ever. And to conclude that there is zero scientific evidence for a dipole in the acceleration and other claims that they have made.

These people have basically admitted that they're sloppy and they're proud about it; they may only be praised by the cesspools on the Internet, not by genuine scientists. Even if some paper had some point or found something nontrivial, it is extremely likely that the effect is actually due to some mundane effect that was overlooked, and probably deliberately overlooked because these people think that it is cool to overlook mundane effects and naively interpret uncorrected data as direct measurements of the fundamental phenomena in the Universe. Why would you care about the result of such a work that isn't done well? It's much better to ignore it and do your own analysis if you need the answers to similar questions.

Well, it's not cool to overlook or hide mundane effects. It's an admission of one's (or four's) systematic sloppiness that kills his credibility among all good scientists.

Eppur si muove and it matters. And that's the memo.

Add to Digg this Add to reddit

snail feedback (0) :

(function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//','ga'); ga('create', 'UA-1828728-1', 'auto'); ga('send', 'pageview');