CosmoCoffee

Here we have another paper doing the cosmological constant from back-reaction. This time the authors use the suber-Hubble fluctuations in the scalar field rather than gradient terms of energy momentum tensor used previously, so it is a fundamentally different approach.

In the conclusions the authors address the problem of causality by saying it is easy to refute, because "Our formalism is based entirely on the equations of GR, which are generally covariant...", which still leaves me slighlty nervous (previous approaches were also based on covariant GR and turned out to have many problems), but the proposed mechanism is a very interesting possiblity anyway...

A recent paper by R.Wald [gr-qc/0509108] seems to claim you cannot do this...

I finally got to read through this paper.
At first look, the backreaction jargon of this paper can be simply translated to a quintessence with its initial condition set by inflation.

However, in closer reading, the treatment seems to have more serious problems:

1- [tex]\varphi^4[/tex] theory studied here, undergoes coherent oscillations after the end of inflation, and finally decoheres into a radiation fluid, rather than a uniform field with small inhomogeneities, as assumed here. Eventually, [tex]\varphi^4[/tex] theory can only act as a radiation fluid and not quintessence.

2- [tex]\varphi[/tex] field, which is assumed to be the inflaton, at least partially decays into radiation at reheating, which is not considered at all.

3- The first order terms cannot be simply averaged to zero at the 2nd order level.

4- Finally, G_{0i} equation cannot be used to relate metric to field fluctuations, when the scalar field is sub-dominant, as is the case in the matter era.

All this should be added to the fact that WMAP seems to have ruled out [tex]\varphi^4[/tex] as a possible inflationary scenario.