Relativity

[davieddy]

To sum it up "there are no preferred reference frames".
OTOH Cosmic background radiation makes the idea of a "fixed" one plausible.

Which newly jumped-up Supermod saw fit to delete my post to this effect?

Er... LET ME GUESS

[xilman]

Originally Posted by davieddy

To sum it up "there are no preferred reference frames".
OTOH Cosmic background radiation makes the idea of a "fixed" one plausible.

Which newly jumped-up Supermod saw fit to delete my post to this effect?

Er... LET ME GUESS

Not me guv.

If you recall, I suggested, in the gateway to anywhere thread, that the frame in which the CMB has zero dipole moment is a prime candidate for a preferred reference frame.

Upon further consideration, there is a potential problem which may be impossible to solve. What if the universe is rotating? If that is the case, the frame in which the CMB has a static appearance may not be inertial --- a body which is at rest in such a frame would have an angular momentum which could be detected by local measurements such as centripetal/centrifugal forces, Coriolis forces or relativistic frame dragging.


Paul

[ewmayer]

Originally Posted by davieddy

To sum it up "there are no preferred reference frames".
OTOH Cosmic background radiation makes the idea of a "fixed" one plausible.

Which newly jumped-up Supermod saw fit to delete my post to this effect?

I haven't been on the forum in the last nearly 24 hours - Did the post in question end up in the "Useless" collection thread?

-----------------------

Originally Posted by xilman

If you recall, I suggested, in the gateway to anywhere thread, that the frame in which the CMB has zero dipole moment is a prime candidate for a preferred reference frame.

Haven't been following the GTA thread, but IIRC "zero dipole moment" only fixes things w.r.to rotation. In a homogeneous spherically-symmetric (on large enough scales) linear-power-law expansion - the only kind which permits isotropy, interestingly enough (in other words the assumption of isotropy implies Hubble's law, up to the expansion-rate constant) - how would zero dipole moment fix things in the rectilinear-motion sense?

[xilman]

Originally Posted by ewmayer

I haven't been on the forum in the last nearly 24 hours - Did the post in question end up in the "Useless" collection thread?

-----------------------


Haven't been following the GTA thread, but IIRC "zero dipole moment" only fixes things w.r.to rotation. In a homogeneous spherically-symmetric (on large enough scales) linear-power-law expansion - the only kind which permits isotropy, interestingly enough (in other words the assumption of isotropy implies Hubble's law, up to the expansion-rate constant) - how would zero dipole moment fix things in the rectilinear-motion sense?

Not as I understand it.

The CMB has an observed dipole moment. It is hotter in one direction on average than the other. Equivalently, the CMB is bluer in one direction and redder in the opposite. If the observer is made to move with an appropriate velocity the observed average red shift can be reduuced to zero. That is, the movement compensates for the rectilinear motion.

Assuming that has been done, three of the six degrees of freedom have been removed. The remaining three, the angular momentum vector, are still undetermined. If the universe is rotating you can choose either to rotate with the universe and cancel out the centrifugal (etc) forces or you can keep the CMB stationary and observe the universal rotation. Personally, I'd choose the former option but would still feel uncomfortable about the choice.

In addition, I see no a priori reason for a homogeneous spherically symmetrical universe. Actually, that's not a strict requirement for Hubble's law to hold if you allow the topology to be sufficiently interesting and/or the universe to be sufficiently large. All that's required is that our local region to have that symmetry. There are plausible theoretical reasons why that symmetry shouldn't hold. IIRC some cosmologies predict very large voids (as appears to exist in the direction of Eridanus) and a large scale flow of matter in a particular direction, for which some evidence exists in the form of the "Great Attractor".

That said, my understanding of cosmology is somewhat shaky for anything proposed in the last 25 years or so.


Paul

[ewmayer]

Originally Posted by xilman

Not as I understand it.

The CMB has an observed dipole moment. It is hotter in one direction on average than the other. Equivalently, the CMB is bluer in one direction and redder in the opposite. If the observer is made to move with an appropriate velocity the observed average red shift can be reduuced to zero. That is, the movement compensates for the rectilinear motion.

Ah, I should have been clearer: I meant rectilinear motion in the sense of establisment of some kind of (hypothetical) absolute position with respect to the expansion. The CMB dipole anisotropy is due to movement of the Earth relative to the "comoving cosmic rest frame", which is a local-relative-movement thing. An observer 10 Bln light-years away would similarly zero out their locally observed dipole moment and see precisely the same remaining CMB. These 2 widely-separated observers would be moving (after correcting for the local-velocities) relative to each other according to the global expansion law, but could conclude nothing about any kind of absolute position as a result, only relative position. Under the assumption of large-scale isotropy of standard BBC, one could leapfrog "outward" as many hops as one likes in this fashion, without ever getting any closer to the "edge" of the universe, because there is no edge.

Or am I misunderstanding your claim?

[xilman]

Originally Posted by ewmayer

Ah, I should have been clearer: I meant rectilinear motion in the sense of establisment of some kind of (hypothetical) absolute position with respect to the expansion. The CMB dipole anisotropy is due to movement of the Earth relative to the "comoving cosmic rest frame", which is a local-relative-movement thing. An observer 10 Bln light-years away would similarly zero out their locally observed dipole moment and see precisely the same remaining CMB. These 2 widely-separated observers would be moving (after correcting for the local-velocities) relative to each other according to the global expansion law, but could conclude nothing about any kind of absolute position as a result, only relative position. Under the assumption of large-scale isotropy of standard BBC, one could leapfrog "outward" as many hops as one likes in this fashion, without ever getting any closer to the "edge" of the universe, because there is no edge.

Or am I misunderstanding your claim?

I think you're understanding my claim and that we were discussing slightly different phenomena. I was starting from the observation that the CMB has a dipole moment, thereby indicating that we are not at rest relative to a (possibly inertial) reference frame which appears to have some fundamental significance. If that frame is not inertial the gravitational forces must be very small because we have no observational evidence at all for any rotational effects.

As for the "no edge" statement it's hard to say what may exist beyond our event horizon because anything out there is unobservable pretty much by definition. There may be an edge in the sense of a topological discontinuity such as a cosmic string or plane within the observable universe but we haven't yet observed one, AFAIK. Personally I wouldn't call a point singularity, aka a black hole, an edge but that might just be dimensional chauvinism.

There could also be an edge in the sense of an intersection of our observable universe and another inflationary bubble, if such exist, but what evidence there is is extremely tenuous and not at all widely accepted. According to some calculations in the context of brane theory two universes could collide and bounce (I'm over simplifying somewhat) with evidence of the bounce remaining in the form of symmetrical distortions in the CMB with a particular range of angular sizes and polarization. There's some evidence of such structures, I believe, but their reality if any and their cause is disputed. I'm not sure that I'd call those "edges" either. As a true intersection is predicted (probably) to expand at the speed of light we (probably) wouldn't observe it unless the laws of physics remain very closely similar before and after the event.


Paul

[davar55]

Some excerpts just to indicate what I'd like to question:

Originally Posted by xilman

If you recall, I suggested, ... that the frame in which the CMB has zero dipole moment is a prime candidate for a preferred reference frame.
...
What if the universe is rotating?
...

Originally Posted by ewmayer

... without ever getting any closer to the "edge" of the universe, because there is no edge.

Originally Posted by xilman

...
because we have no observational evidence at all for any rotational effects.
...
two universes could collide and bounce (I'm over simplifying somewhat)
...

If the Universe has a spatial (3-d) preferred reference frame, then it has
a single center, the origin of the coordinate system. That's silly.

If the Universe has a space-time (4-d) preferred inertial reference frame,
its spatial origin must be moving in some spatial direction at some fixed
speed, but relative to what? This is also "silly".

If the entire Universe is rotating, it must be about some axis, which would
single out some static line throughout space. That's "silly".

If there were some way to define the Universe, which contains all that is,
so that there could be more than one Universe, any additional Universes
would have to be null. Multiple Universes are silly.

If the Universe had an "edge" or boundary or border, it would be a
separation of the something of the Universe from nothing. That's
extremely silly.

I really think current cosmology and cosmological physics needs a reset.

[davieddy]

Originally Posted by xilman

If you recall, I suggested, in the gateway to anywhere thread, that the frame in which the CMB has zero dipole moment is a prime candidate for a preferred reference frame.
Paul

I think most folk here would be more comfortable with "Doppler effect" than "dipole moment".
Those CMB maps you see published are corrected for the velocity of the observer routinely.

What is the dipole moment of the earth's mass?
As long as you take the CofM as the origin, it is zero by definition.

I am fascinated by the geoid.
Kaye & Laby gave a formula with a consistent value for the quadrupole moment contribution. Intriguingly, the octupole moment varied IN SIGN
from one edition to the next!

David

[davieddy]

Originally Posted by davieddy

Which newly jumped-up Supermod saw fit to delete my post to this effect?

Er... LET ME GUESS

Originally Posted by xilman

Not me guv.

Paul

Originally Posted by ewmayer

I haven't been on the forum in the last nearly 24 hours - Did the post in question end up in the "Useless" collection thread?

"No editor ever went bust by underestimating the intelligence of his readership".

I thought one or other of you might recognise a rhetorical question when they saw it.

[xilman]

Originally Posted by davieddy

I think most folk here would be more comfortable with "Doppler effect" than "dipole moment".
Those CMB maps you see published are corrected for the velocity of the observer routinely.

Exactly.

The raw data has a non-zero dipole moment. The frame in which it has zero dipole moment is sufficiently fundamental or, at least, interesting that its rectilinear motion is transformed away for publication.

[xilman]

Originally Posted by davar55

Some excerpts just to indicate what I'd like to question:

If the Universe has a spatial (3-d) preferred reference frame, then it has
a single center, the origin of the coordinate system. That's silly.

If the Universe has a space-time (4-d) preferred inertial reference frame,
its spatial origin must be moving in some spatial direction at some fixed
speed, but relative to what? This is also "silly".

If the entire Universe is rotating, it must be about some axis, which would
single out some static line throughout space. That's "silly".

If there were some way to define the Universe, which contains all that is,
so that there could be more than one Universe, any additional Universes
would have to be null. Multiple Universes are silly.

If the Universe had an "edge" or boundary or border, it would be a
separation of the something of the Universe from nothing. That's
extremely silly.

I really think current cosmology and cosmological physics needs a reset.

Several things there to address, as you note.

If I understand you correctly, I agree with your first observation --- that of a unique 3-d reference frame.

As for the second, it is a matter of observational record that the CMB does indeed have a preferred direction and speed as measured by its differential red shift and the direction in the sky where it is hottest and coldest. If you regard the CMB as being at rest with respect to the universe as a whole, and that seems to be a reasonable assumption to most people who have thought about it, we are demonstrably moving with respect to the universe as a whole.

The third point, that of universal rotation, has very little observational support either way (though I've read a couple of suggestive but disputed papers). The observations are consistent with a non-zero rotation but if it exists the rotation is very small --- on the order of one degree rotation over the last ten gigayears. As to whether the concept is "silly", let us suppose that a very sensitive experiment is performed. The supposed experiment has two masses on the end of a string and the masses are kept at a constant orientation with respect to the rest of the universe --- whether the CMB or the distant galaxies or whatever. Suppose the experiment finds that there is a tension in the string. The tension is maximal when the normal to the mid-point of the string has one particular direction. If the masses are made to rotate (slowly!) with respect to the rest of the universe the tension disappears. In every respect, the masses indicate that they are rotating with respect to the visible universe and everything within it, CMB, galaxies, the whole lot. Now, either everything is rotating with respect to some fundamental axis (the normal to the string) or spacetime is rotating with respect to the contents of the universe. Either scenario could reasonably be described as universal rotation. You are quite correct that in this case there would be a preferred direction in space. You are incorrect, IMO, to dismiss the possibility as "silly" until you have performed the necessary experiments and/or observations.

As for the last, if you define "universe" as everything which may potentially exist, observable or otherwise, we are in agreement. My working definition, in common with many if not all cosmologists is to call "the universe" that which is directly observable if only in principle --- that is, everything which lies within our past light cone. The totality of that "universe" and anything else which may or may not exist, is generally called the multiverse.

There may well be things which are not observable, perhaps because they are so far away that signals from them travelling at c have not yet had chance to reach us. If our spacetime is finite, there may be other disconnected regions of spacetime. (A hoary old two-dimensional example is that of beings confined to the surface of a planet. Other beings may be confined to the surfaces of other planets. Because the beings are confined there is no experiment or observation which could resolve their questions as to whether other planetary surfaces and other beings exist.) Even if our spacetime is infinite, there could still be other disconnected spacetimes, finite or infinite. In the two-d example, suppose we live on an infinite plane. Could other planes parallel to ours exist? How about closed surfaces such as spheres? If another plane is not parallel to ours it would intersect with our universe and, presumably, have observable consequences. As I posted earlier, a theoretical treatment of a particular kind of intersection (a "bounce") has been made and which makes testable predictions. There is a small amount of disputed evidence consistent with the predictions.