Debate on "background independence" in theoretical physics

On Oct 10, 2010, at 4:43 PM, JACK SARFATTI wrote:
I have not had time yet to properly address James's more recent messages regarding my conjecture on the amplification of gravity in a yet-to-be-made large area plate made of stacked 1-atom thick graphene sheets separated by some other kind of material in which the composite is a higher-than-room temperature superconducting meta-material - whose several properties are tunable by nano-engineering techniques.
Remember my position is that James may prove correct in the end that such gravity amplification is impossible, but not for most of the detailed reasons he gives.
If he is correct in the end, then it will be a sad day for Man, because we will be stuck here and it will be impossible for real distant ETs to get to us, or us to get to them in single life-times. It would also mean that all claims of actual material contact with alien ETs are mistaken.
On Oct 10, 2010, at 3:47 PM, Paul Zielinski wrote:
I think what James has written below exposes some basic misconceptions, not only
about bimetric theories of gravity, but also with regard to the concept of "background
independence" itself.
String/membrane theory is "bimetric," i.e. G = GNewton(1 + ae^-r/b) in the static limit.
Background independence is a Red Herring. It is simply, for the special case of Einstein's 1916 GR, the universal gauge invariance of the global dynamical action of all matter fields and their local field equations from the action principle &S = 0 with respect to the localized 4-parameter Abelian translation group T4(x) whose Lie algebra elements are the global total linear 4-momenta of the matter fields.
To get beyond Einstein's 1916 GR extend to the Poincare and the conformal groups.
Vacuum/Ground state solutions of the field equations may spontaneously break that symmetry, similarly to the way a superconductor breaks the U1 internal EM symmetry.  Here the original quanta are replaced by elementary excitations (quasi-particles & collective modes) in the long-wave IR limit, e.g. the massless quasi-electrons in 2D flat graphene sheets (presently limited to tiny flakes) whose effective coupling to light is 300xvacuum fine structure "constant". Similarly for gravity in special superconducting meta-materials, possibly graphene-based, I suggest as a Popper-falsifiable hypothesis.
On Wed, Oct 6, 2010 at 10:23 PM, james f woodward wrote:
I think you'll find that my comments about background independence are
not incorrect as you claim.  I suspect that this claim may be the result
of confusion of what background independence actually is -- a confusion
easily understood because background independence means the opposite of
what one might intuitively think. 
A confusion that is easily understood since no one seems to be able to give a clear definition
of "background independence".
I gave a very clear definition formally as local T4(x) gauge transformations. Physically as transformations between locally coincident small detectors each in arbitrary possibly off-geodesic & rotating motions including applied torques.
 
The historical origins of all this go
back to the '50s and '60s and the Rosen [the same as the EPR and "bridge"
Rosen] "bi-metric" theory of gravity where the usual Einstein metric
theory is supplemented by a global Minkowski "background" spacetime.
Actually they go back to the devastating 1917 Kretschmann argument which trivialized general
covariance from the physical standpoint. Kretschmann's argument clearly de-linked general
relativity from general covariance. A way had to be found to rescue Einstein's notion of "general
relativity", and the conceptual vacuum was filled (as it were) by Anderson's "principle of reciprocity"
and his "no absolute objects" rule -- leading to the contemporary principle of active diffeomorphism
invariance currently promoted by Rovelli, Smolin, and others.
I agree a lot of excess baggage here. I think I have gotten this down to the essential physics in a Zen like, more with less operational way.
It's true any physical theory can be made "background independent" - indeed in my T4(x) definition the equivalence principle means that all physical theories must be background independent.
Background independence, in my opinion, is simply the particular application of the local gauge principle to the universal symmetry groups that define the spacetime fabric.
 
After considerable debate, folks agreed that the Minkowski background
metric added nothing whatsoever of physical meaning to Einstein's GR,
Rovelli in "Quantum Gravity" says global Minkowski spacetime is simply an unstable vacuum solution to Einstein T4(x) covariant field equations - with invariant global action integrals S over all 4D spacetime.
The "folks" who agreed on this, agreed on this. Others did not, and still don't.
 
and
the metric of GR describe the actual physical state of spacetime. 
You need to use ds^2 = guvdx^udx^v = gu'v'dx^u'dx^v' as the actual physical state, not the detector-dependent representations guv or gu'v'.
In any bimetric version of GR you have a deformable physical metric in addition to a Minkowski
reference metric. The physical metric is determined by the matter distribution just as in plain
vanilla GR.
Formally, the Einstein Equivalence Principle (EEP) simply means a class of coincident LIF can be realized in which the metric is approximately Minkowski over a scale small compared to the locally variable radii of curvature.
So this is an obvious red herring. Straw man.
 
That
is why GRT is called background independent.
That is not why GRT is called background independent.
OK I think I see your point about the "confusion". :-)
 
 A background-dependent theory is one that presupposes the existence of a global Minkowski metric spacetime into which physical stuff is loaded.  The loading stuff (fields and particles and the like) has no effect on the metric at all.
This is simply false.
This is all related to the "relationalist" vs. "substantivalist" debate regarding the physical
vacuum of GR. The relationalist idea is that if all gravitating matter is removed, then
there is no metric and no spacetime -- that the spacetime of GR represents *only*
relations between material bodies. The substantivalist position is that after all matter is
removed you still have a physical spacetime with a Minkowski metric (aside from
gravitational  waves).
De Sitter famously showed that Einstein's field equations have vacuum solutions, and
this along with Kretschmann's earlier 1917 critique upset Einstein's earlier "Machian"
relationalist position.
 
Intuitively, the background metric is independent of the contents of
spacetime. 
Just as the quiescent state of a guitar string (understood as a zero vibrational energy
reference state) is independent of how it is plucked.
 
But, counterintuitively, such theories are called background
dependent because the background spacetime is prior to the contents found
in it, and they depend for their existence on the prior existence of the
background spacetime.
Exactly -- the unperturbed background inertial field (Minkowski spacetime) subsists ontologically
prior to and independently of the existence of gravitating matter. But that does not mean that the
physical metric describing this field  cannot be physically perturbed by matter. That is the point of
the physical spacetime metric in bimetric theories!
 
There are two important points here, for our purposes anyway.  First,
because of the prior nature of spacetime in background dependent theories
(that the spacetime is unaffected by its contents),
This is just wrong. The actual idea is that a flat spacetime can subsist ontologically in the absence of
gravitating matter.  The actual physical metric is of course still perturbed by the presence of gravitating
matter.
This "substantivalist" interpretation is clearly supported by the formalism of GR, since the field equations
do in fact have pure vacuum solutions.
You seem to be confusing the question of ontological priority with the question of gravitational deformation
of the physical metric. No one in the substantivalist or bimetric camps claims that physical spacetime is
always flat regardless of the presence of matter!
 
nothing you put into
the spacetime has any effect at all on the spacetime. 
I think this is called "special relativity". Bimetric theories of matter are not at all the same as special relativity,
contrary to what you seem to be arguing here.
 
That means that no
matter what you do, you cannot warp the background spacetime -- which is
the real spacetime. 
This is Minkowski special relativity. This has nothing to do with "background dependence".
You seem to be seriously confused here.
 
So there are no wormholes or warp drives possible in
such theories. 
Not in special relativity, which is what you appear to be talking about.
 
Second, Hal's PV approximation to GRT is a background
dependent theory.  A global Minkowski spacetime is ASSUMED, and the
propagation of light, when it deviates from striaght lines in the
Minkowski spacetime, is attributed to variations in the index of
refraction of a material medium (virtual e-p pairs in the PV model) in
the background spacetime. 
No, the deformation of the physical metric is simply modeled (for engineering purposes)
as a vacuum polarization effect.
 
By ASSUMPTION, no amount of any stuff placed
into the background Minkowski spacetime can cause any warping of the
spacetime. 
You're talking about Minkowski SR here, right?
This has nothing to do wiith background dependent or bimetric interpretations of GR.
I don't think you have understood the meaning of "background dependence".
 
Should you check with Hal, I expect he will tell you that
this is indeed the case for the PV approximation to GRT.  The PV model is
just a heuristic approximation to GRT intended to simplify otherwise more
complicated calculations.
PV is not relevant here, since it is intended only as a naive approximate engineering model.
It is not offered as an alternative to GR, or even as the basis for an alternative interpretation
of GR, as far as I'm aware.
Should you read the paper in light of the foregoing, you'll find that I
have used "background independence" consistently with the above. 
Consistently incorrect, yes.
 
GRT is
background independent --
I'm sorry James but this is nonsense. GR has background dependent and background independent
interpretations,but even the Andersonian background independent "no absolute objects" interpretations
are only partially valid. GR still has absolute background structure. For example, the Lorentz signature of
the metric is an absolute invariant in any viable interpretation of GR, regardless of how matter is
distributed in spacetime, and regardless of whether there is any gravitating matter at all.
 
which is why it is physically meaningful to
talk about wormholes and warp drives as real physical possibilities.
This is nonsense. Bimetric theories don't allow gravitational deformation of the physical spacetime?
Horse feathers. Poppycock.
 
Other theories like PV and string theory are background dependent.  As
such, they presume the reality of spacetime to be, for example, a global
Minkowski spacetime.  Since that spacetime is not affected by its
material contents, wormholes and warp drives are by assumption
impossible.
It just means that if you remove all gravitating matter from spacetime, there is still a "there"
there. This is fully supported by the formalism of GR, since GR has pure vacuum solutions --
as shown long ago by de Sitter.
If you want to see stargates and warp drives become a reality, keep your
fingers crossed that Einstein was right about the background independence
of GRT, and that reality conforms to GRT.
So e.g. Rosen's bimetric theory doesn't admit a Rosen bridge? I beg to disagree.
I've attached a paper "Who's Afraid of Background Independence" by Rickles, raising a number
of points regarding background dependence vs. background independence that you may
find useful.
Z.
On Wed, 06 Oct 2010 12:14:00 -0700 JACK SARFATTI writes:
> If we can make a low-power high Tc superconducting warp/wormhole
> drive I am sure graphene sheets will be the key substrate.
> BTW a point about James Woodward's incorrect argument about
> background independence being spoiled inside the material.
> By the same false reasoning one can argue that special relativity is
> spoiled inside the material because the speed of light in material
> is clearly not invariant.
> Just remember locality demands that (index of refraction)^4G/c^4 is
> the gravity coupling inside the material because the speed of light
> in vacuum is from scattering off virtual electron-positron pairs and
> in a material we also have real particle charges. It's the same
> thing as far as gravity is concerned because of the equivalence
> principle. The fact that the universe is accelerating is proof
> positive that virtual particles inside the vacuum (anti) gravitate
> equally with real particles outside the vacuum.
>