A classical model for the extension of singular spacetime geometries across their singularities is presented. The regularization introduced by this model is based on the following observation. Among the geometries that satisfy Einstein's field equations there is a class of geometries, with certain singularities, where the components of the metric density and their partial derivatives remain finite in the limit where the singularity is approached. Here we exploit this regular behavior of the metric density and elevate its status to that of a fundamental variable -- from which the metric is constructed. We express Einstein's field equations as a set of equations for the metric density, and postulate junction conditions that the metric density satisfies at singularities. Using this model we extend certain geometries across their singularities. The following examples are discussed: radiation dominated Friedmann-Robertson-Walker Universe, Schwarzschild black hole, Reissner-Nordstr\"{o}m black hole, and certain Kasner solutions. For all of the above mentioned examples we obtain a unique extension of the geometry beyond the singularity.

In the Palatini action of general relativity the connection and the metric are treated as independent dynamical variables. Instead of assuming a relation between these quantities, the desired relation between them is derived through the Euler-Lagrange equations of the Palatini action. In this manuscript we construct an extended Palatini action, where we do not assume any a priori relationship between the connection, the covariant metric tensor, and the contravariant metric tensor. Instead we treat these three quantities as independent dynamical variables. We show that this action reproduces the standard Einstein field equations depending on a single metric tensor. We further show that in this formulation the cosmological constant has an additional theoretical significance. Normally the cosmological constant is added to the Einstein field equations for the purpose of having general relativity be consistent with cosmological observations. In the formulation presented here, the nonvanishing cosmological constant also ensures the self-consistency of the theory.

We derive an expression for the luminosity distance as a function of redshift for a flat Robertson-Walker spacetime perturbed by arbitrary scalar perturbations possibly produced by a modified gravity theory with two different scalar perturbation potentials. Measurements of the luminosity distance as function of redshift provide a constraint on a combination of the scalar potentials and so they can complement weak lensing and other measurements in trying to distinguish among the various alternative theories of gravity.

It has been suggested that the accelerated expansion of the Universe is due to backreaction of small scale density perturbations on the large scale spacetime geometry. While evidence against this suggestion has accumulated, it has not yet been definitively ruled out. Many investigations of this issue have focused on the Buchert formalism, which computes spatial averages of quantities in synchronous comoving gauge. We argue that, for the deceleration parameter of this formalism to agree with observations, the spatial average of the three dimensional Ricci scalar (spatial curvature) must be large today, with an $\Omega_k$ in the range of $1 \le \Omega_k \le 1.3$. We argue that this constraint is difficult to reconcile with observations of the location of the first Doppler peak of the CMBR. We illustrate the argument with a simple toy model for the effect of backreaction, which we show is generically incompatible with observations.

In most theories of gravity involving torsion, the source for torsion is the intrinsic spin of matter. Since the spins of fermions are normally randomly oriented in macroscopic bodies, the torsion generated is normally negligible. However, in a recent paper, Mao et al. point out that there is a class of theories in which the angular momentum of macroscopic spinning bodies generates a significant amount of torsion. They argue that by the principle of action equals reaction, one would expect the angular momentum of test bodies to couple to a background torsion field, and therefore the precession of the GPB gyroscopes should be affected in these theories by the torsion generated by the Earth. We show that in fact the principle of action equals reaction does not apply to these theories. We examine in detail a generalization of the Hayashi-Shirafuji theory suggested by Mao et al. called Einstein-Hayashi-Shirafuji theory. There are a variety of different versions of this theory, depending on the precise form of the coupling to matter chosen for the torsion. We show that for any coupling to matter that is compatible with the spin transport equation postulated by Mao et al., the theory has either ghosts or an ill-posed initial value formulation. These theoretical problems can be avoided by specializing the parameters of the theory and in addition choosing the standard minimal coupling to matter of the torsion tensor. This yields a consistent theory, but one in which the action equals reaction principle is violated, and in which the angular momentum of the gyroscopes does not couple to the Earth's torsion field. Thus, the Einstein-Hayashi-Shirafuji theory does not predict a detectable torsion signal for Gravity Probe B. There may be other torsion theories which do.