home *** CD-ROM | disk | FTP | other *** search
- From: weeks@hpscit.sc.hp.com (Greg Weeks)
- Date: Fri, 18 Dec 1992 00:57:00 GMT
- Subject: Axiomatic Quantum Gravity
- Message-ID: <56960005@hpscit.sc.hp.com>
- Organization: Hewlett-Packard, Santa Clara, CA
- Path: sparky!uunet!spool.mu.edu!sdd.hp.com!hpscit.sc.hp.com!hplextra!hpcc05!hpscit!weeks
- Newsgroups: sci.physics
- Lines: 52
-
- Not. There is no axiomatic quantum gravity. General relativity doesn't
- have local observables. I've never seen this asserted, so I'm asserting it
- in this note. The following argument is a bit shaky. My hope is that
- exposure will cure it or kill it. So, here is why I say that general
- relativity doesn't have local observables.
-
-
- In special relativity, the metric g is viewed as a part of space-time. The
- isomorphisms of space-time make up the Poincare group. All observables are
- Poincare-invariant. (It's obvious that no experimental result would be
- changed if the universe were uniformly translated two feet in some
- direction.) And yet the "observables" of special relativistic theories --
- the energy-momentum tensor, the electromagnetic field, the electric
- current, and so on -- are Poincare-covariant, not Poincare-invariant. The
- resolution is that it is possible to establish physical coordinate systems
- using clocks, yardsticks, and such. These coordinate systems do not affect
- the systems being observed, and vice versa. The values of
- Poincare-covariants measured with respect to a physical coordinate system
- are Poincare-invariant.
-
- In general relativity, the metric g is not viewed as a part of space-time.
- The isomorphisms of space-time are the diffeomorphisms. All observables
- are diffeomorphism-invariant. Now, is it possible to establish physical
- coordinate systems? I don't think so. [Here is where things get shaky.]
- If the physical coordinate system is much more massive than the system
- being observed, then it will significantly affect the system being
- observed. If the physical coordinate system is much less massive than the
- system being observed, then it will be significantly affected by the system
- being observed. So, the notion of a physical coordinate system is
- problematic at best. My feeling is that the notion doesn't make sense.
- So, without local diffeomorphism-invariants and without physical coordinate
- systems, there are no local observables.
-
- And without local observables, there is no quantum theory, leaving me
- feeling as sick as mud.
-
-
- Incidentally, it is reasonable to ask: If there are no local observables in
- general relativity, why does it seem as if there are? There is an answer
- to this. In practical circumstances, space-time is very nearly flat. It
- is natural to consider coordinate systems in which the metric g is very
- nearly the same as in special relativity. This restricts the symmetry
- group of space-time to the Poincare group plus "infinitesimal"
- diffeomorphisms. Furthermore, in the systems where gravity is considered,
- massive objects move very slowly (compared to light). And it just so
- happens that, to first order in these velocities, the equations of motion
- of massive bodies are invariant under the infinitesimal diffeomorphisms.
- So, to a good approximation, in practice we have flat space-time with
- Newtonian gravitation. This influences our intuition.
-
-
- Greg Weeks
-