This is an interesting paper, which investigates whether elastic (solid) dark energy could give rise to large scale anisotropies in the observed CMB. An interesting feture is that coupling between scalar, vector and tensor modes is generated by breaking rotational invariance via the anisotropic elasticity tensor.
The model is in some tension with other (e.g. supernova) data because it requires a dark energy equation of state w=2/3.
Their general Lagrangian is specified so that it is only a function of the metric. Since the action should be gauge invariant, and the gauge invariant function of the metric tensor is the Riemann tensor, does this mean that their model is equivalent to a generalized gravity theory with some f(R_{abcd}) action?
[astroph/0602377] Anisotropic perturbations due to dark energy
Authors:  Richard A. Battye, Adam Moss 
Abstract:  A variety of observational tests seem to suggest that the universe is anisotropic. This is incompatible with the standard dogma based on adiabatic, rotationally invariant perturbations. We point out that this is a consequence of the standard decomposition of the stressenergy tensor for the cosmological fluids, and that rotational invariance need not be assumed, if there is elastic rigidity in the dark energy. The dark energy required to achieve this might be provided by point symmetric domain wall network with $P/\rho=2/3$, although the concept is more general. We illustrate this with reference to a model with cubic symmetry and discuss various aspects of the model. 
[PDF] [PS] [BibTex] [Bookmark] 

 Posts: 1522
 Joined: September 23 2004
 Affiliation: University of Sussex
 Contact: