Re: Article: A Century of Einstein
From: Eric Baird (eric_baird_at_compuserve.com)
Date: 10/06/04
- Next message: Eric Baird: "Re: Relativity, Scepticism and Humility in Science"
- Previous message: Eric Baird: "Re: Is the speed of light really constant ?"
- In reply to: Tom Roberts: "Re: Article: A Century of Einstein"
- Next in thread: Tom Roberts: "Re: Article: A Century of Einstein"
- Reply: Tom Roberts: "Re: Article: A Century of Einstein"
- Reply: Ole D. Rughede: "Re: Article: A Century of Einstein"
- Messages sorted by: [ date ] [ thread ]
Date: Wed, 6 Oct 2004 03:29:46 +0000 (UTC)
On Sat, 02 Oct 2004 13:58:47 GMT, Tom Roberts <tjroberts@lucent.com>
wrote:
>Eric Baird wrote:
>> IMO, the usual constraint that gravitational theory has to reduce to
>> SR is "traditional", but does not seem to be required.
>
>What is required is that any theory not be refuted by existing
>experiments. As SR is so well tested, and applies in situations in which
>gravitation is negligible, then in a suitable limit any valid theory of
>gravitation must reduce to either SR or a non-gravitational theory that
>is experimentally indistinguishable from SR.
>
>
>> Well, people in the mainstream seem to think that reduction to SR is
>> compulsory so if it is true that compliance with SR means that a
>> theory can't then obey the principle of relativity in its most extreme
>> form, and the mathematicians are being told that only SR-compatible
>> solutions need be considered, then anyone working to those constraints
>> and trying to produce a "completist" theory of relativity is going to
>> be pretty much doomed to failure, because they are going to be working
>> to an impossible design brief.
>
>See above. This is not an "impossible design brief", this is a basic and
>fundamental property of valid theories of physics: that they are not
>already refuted by existing experiments.
The thing that I was referring to as being an "impossible design
brief" was the task of writing a theory that obeyed the principle of
relativity in its most extreme ("Machain") form, and making the theory
also reduce perfectly to SR.
I thought that you were saying that nobody had managed a purely
Machian theory, so I was responding that since everybody seemed to be
also trying to achieve SR-compliance, this was perhaps not surprising.
Since most of the mainstream community don't seem to have any idea
what a "non-SR" theory of relativity would look like, I'm not sure
that they are equipped to judge whether or not such a theory would be
refuted by, or incompatible with, existing experiments.
>> I think that the whole idea of inertial physics being a flat spacetime
>> problem is an evolutionary leftover.
>
>No. It is not "leftover" at all, it is a consequence of observations
>about our world. <shrug>
Ever heard of a high-energy particle physics experiment where they
repeat the Fizeau-type approach with light shone along a straight-line
section of relativistic particle beam, and conclude that the upstream
and downsstream lightbeam velocities are shown to /not/ change
velocity as the particle beam energy changes, that the moving
particles are therefore shown /not/ to drag light, and that the
experiment therefore proves that lightbeam geometry in the core of a
high-energy particle beam /is/ flat?
I haven't.
If we have observations going back to the 1850s showing that moving
particles do drag light, and we've apparently not managed to use our
beeeg particle accelerators to produce any results that disagree with
that, then I don't see how we can impose the assumption on theory that
particles do NOT drag light, and claim that this is known to be true,
based on observations.
We can observe that SR is quite effective at helping us model the
behaviour of particle accelerators (when we bear certain exceptions in
mind), but to be able to claim that SR's success in this respect is
"significant", we have to know what the alternative predictions might
have been, and people growing up today and being taught physics from
modern textbooks seemingly don't get provided with that sort of
knowledge.
The way that SR testing currently goes seems to be a bit like me
insisting that a certain animal must be a pig, because it has four
legs and a tail, and eats food. All those observations are certainly
/consistent/ with the animal being a pig, but they don't /prove/ that
it's a pig, and they aren't even particularly significant evidence
that it is a pig.
Showing that other animals also have those same properties doesn't
automatically mean that the animal is NOT a pig, but it certainly
undermines one's faith in the supplied proofs, and makes one wonder
whether the existing claims of the animal's "proved" pigginess are
actually based on anything concrete.
>> It's historically understandable,
>> and its its a quick and easy way of getting results, but now that
>> we've learnt a load of additional stuff from GR that we didn't know at
>> the time that SR was developed, and know how to do fun things with
>> curved spacetime, I really dson;t see why we are still sticking with
>> it.
>
>Because it works.
It works fairly effectively if one bears in mind certain exceptions
and sets certain limits by hand according to where experimental data
is expected to diverge too far from the theory's predictions
(and if one is prepared to make the occasional correction when an
experiment generates a "wrong" result), yes.
>And because it is a property of all valid theories of physics.
WHOAH there! :)
WHY is SR-compliance supposed to be a property of all valid theories
of physcs?
Because authority-figures say so?
That's bunk, before that statement can be treated as fact, there has
to be a logical chain of arguments that show that "non-SR" relativity
can't be right, and those arguments have to be transparent and open to
examination and testing, or else its just assertion.
You can invoke authority, but a Christian fundamentalist can also
invoke authority to claim that the Earth was "really" created about
4000BC.
Their arguments are /also/ consistent with the evidence (if we allow
certain assumptions designed to support their case), but I wouldn't
believe a fundamentalist's claim that no other explanations can exist,
or that no other explanations are valid, unless they had some better
sort of argument to support their case ... and a similar level of
proof ough to apply to relativists claims that a particualr
implementaiton of the principle of relativity is the only possible
answer,and no othe rsolutions can exist, or that no other solutins can
be valid.
The fundamentalists have had maybe a thousand years or so to come up
with a decent supporting argument for the earth being created
recently, and so far they haven;t managed it (okay, there /were/
admittedly one or two good arguments, but those were overturned), so
maybe ... gasp ... it's possible that their claims might not be true
Compared to this, the SR community has only had about a century, but
that century has had many of the finest physics minds that ever ticked
working with the theory, and if none of them have produced a decent
general proof that SR is really unavoidable, then maybe ... claims
that SR is unavoidable might no tbe true.
I think I know the usual reasons why people often /think/ that SR is
unavoidable, and IMO the core logic is usually good (once you strip
away the cloud of surrounding metaphysics), but it seems to be founded
on initial assumnptions that are either unsafe or experimentally
wrong.
I think that if SR really /is/ known to be true, a century should have
been long enough to produce a watertight supporting argument.
If that doesn't yet exist, and people are still making these claims,
then , well, perhaps the scientific method is not being adhered to
quite as strongly in this subject as it could be.
<shrug>
>>>This is, at base, no more than Riemannian geometry -- locally every
>>>manifold looks flat to arbitrary accuracy, in a small enough local
>>>region. In GR, "flat" means essentially "SR applies to within the
>>>measurement accuracy".
>> Okay, here's a distinction between the theory of geometry and "physics
>> theory" that nobody seems to mention:
>>
>> If you zoom in far enough on a "well behaved" smoothly-curved surface,
>> you must end up with a region that is arbitrarily flat ... that's just
>> geometry, and I obviously have to agree that it seems to be an
>> inescapable /geometrical/ result.
>>
>> =BUT=
>>
>> If you have a "well-behaved" smoothly-curved-spacetime relativistic
>> /physics/ and zoom in arbitraily far, you are NOT guaranteed to end up
>> with a reduction to relativistic /physics/ operating as a flat
>> spacetime theory (SR).
>
>But GR does so.
Indeed it (current GR) does.
But to think that this means that all other physical theories have to
reduce in this way to be logically consistent, is wrong.
We can start with GR, and say, casually, " ... and then when we
examine a small enough region of the metric to get an effectively flat
region, we obtain SR", and that can be a good description of how
things are with SR/GR, but the idea that we can take /any/ curved
theory and prove geometrically that it /must/ reduce to the physcs of
special relativity too ... that's disprovable.
> And observations do so.
Nope, apparently not. The lightbeam geometry of the Fizeau experiment
does not sound very flat and isotropic to me.
If the experiment has since been repeated with modern particle
accelerators, and those old light-dragging effects have been
demonstrated to be absent (showing that perhaps Fizeau's result was
wrong, or that perhaps water molecules interact with light according
to different laws of physics to gold ions or helium ions or particuate
observers), then perhaps someone here can point me to it.
If nobody knows of such an experiment, then perhaps experimental
observations do not [yet?] show us that the lightbeam geometry really
is flat between objects with significant "relativistic" relative
velocity.
>And the method by which GR
>reduces to SR is the same as Riemannian geometry reducing to Euclidean
>geometry locally.
Yep, that does seem to be a fair description of how things are under
current GR, agreed.
=Erk= (Eric Baird)
: " The madman is not the man who has lost his reason.
: The madman is the man who has lost everything except his reason. "
: -- G.K. Chesterton
- Next message: Eric Baird: "Re: Relativity, Scepticism and Humility in Science"
- Previous message: Eric Baird: "Re: Is the speed of light really constant ?"
- In reply to: Tom Roberts: "Re: Article: A Century of Einstein"
- Next in thread: Tom Roberts: "Re: Article: A Century of Einstein"
- Reply: Tom Roberts: "Re: Article: A Century of Einstein"
- Reply: Ole D. Rughede: "Re: Article: A Century of Einstein"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
