Teleparallel theories of gravity have a long history. They include a special case referred to as the Teleparallel Equivalent of General Relativity (TEGR, aka GR$_{\|}$). Recently this theory has been generalized to f(T) gravity. Tight constraints from observations suggest that f(T) gravity is not as robust as initially hoped. This might hint at hitherto undiscovered problems at the theoretical level. In this work, we point out that a generic f(T) theory can be expected to have certain problems including superluminal propagating modes, the presence of which can be revealed by using the characteristic equations that govern the dynamics in f(T) gravity and/or the Hamiltonian structure of the theory via Dirac constraint analysis. We use several examples from simpler gauge field theories to explain how such superluminal modes could arise. We also point out problems with the Cauchy development of a constant time hypersurface in FLRW spacetime in f(T) gravity. The time evolution from a FLRW (and as a special case, Minkowski spacetime) initial condition is not unique.

We are investigating the dynamics of a new Poincar\'e gauge theory of gravity model, which has cross coupling between the spin-0$^+$ and spin-0$^-$ modes. To this end we here consider a very appropriate situation---homogeneous-isotropic cosmologies---which is relatively simple, and yet all the modes have non-trivial dynamics which reveals physically interesting and possibly observable results. More specifically we consider manifestly isotropic Bianchi class A cosmologies; for this case we find an effective Lagrangian and Hamiltonian for the dynamical system. The Lagrange equations for these models lead to a set of first order equations that are compatible with those found for the FLRW models and provide a foundation for further investigations. Typical numerical evolution of these equations shows the expected effects of the cross parity coupling.

The Poincar\'e gauge theory of gravity has a metric compatible connection with independent dynamics that is reflected in the torsion and curvature. The theory allows two good propagating spin-0 modes. Dynamical investigations using a simple expanding cosmological model found that the oscillation of the 0$^+$ mode could account for an accelerating expansion similar to that presently observed. The model has been extended to include a $0^{-}$ mode and more recently cross parity couplings. We investigate the dynamics of this model in a situation which is simple, non-trivial, and yet may give physically interesting results that might be observable. We consider homogeneous cosmologies, more specifically, isotropic Bianchi class A models. We find an effective Lagrangian for our dynamical system, a system of first order equations, and present some typical dynamical evolution.

We propose a cosmological model in the framework of the Poincar\'e gauge theory of gravity (PG). The gravitational Lagrangian is quadratic in curvature and torsion. In our specific model, the Lagrangian contains (i) the curvature scalar $R$ and the curvature pseudo-scalar $X$ linearly and quadratically (including an $RX$ term) and (ii) pieces quadratic in the torsion {\it vector} $\cal V$ and the torsion {\it axial} vector $\cal A$ (including a ${\cal V}{\cal A}$ term). We show generally that in quadratic PG models we have nearly the same number of parity conserving terms (`world') and of parity violating terms (`shadow world'). This offers new perspectives in cosmology for the coupling of gravity to matter and antimatter. Our specific model generalizes the fairly realistic `torsion cosmologies' of Shie-Nester-Yo (2008) and Chen et al.\ (2009). With a Friedman type ansatz for an orthonormal coframe and a Lorentz connection, we derive the two field equations of PG in an explicit form and discuss their general structure in detail. In particular, the second field equation can be reduced to first order ordinary differential equations for the curvature pieces $R(t)$ and $X(t)$. Including these along with certain relations obtained from the first field equation and curvature definitions, we present a first order system of equations suitable for numerical evaluation. This is deferred to the second, numerical part of this paper.

Investigations of the dynamic modes of the Poincare gauge theory of gravity found only two good propagating torsion modes; they are effectively a scalar and a pseudoscalar. Cosmology affords a natural situation where one might see observational effects of these modes. Here we consider only the ``scalar torsion'' mode. This mode has certain distinctive and interesting qualities. In particular this type of torsion does not interact directly with any known matter and it allows a critical non-zero value for the affine scalar curvature. Via numerical evolution of the coupled nonlinear equations we show that this mode can contribute an oscillating aspect to the expansion rate of the Universe. From the examination of specific cases of the parameters and initial conditions we show that for suitable ranges of the parameters the dynamic ``scalar torsion'' model can display features similar to those of the presently observed accelerating universe.

An energy for the homogeneous cosmological models is presented. More specifically, using an appropriate natural prescription, we find the energy within any region with any gravitational source for a large class of gravity theories--namely those with a tetrad description--for all 9 Bianchi types. Our energy is given by the value of the Hamiltonian with homogeneous boundary conditions; this value vanishes for all regions in all Bianchi class A models, and it does not vanish for any class B model. This is so not only for Einstein's general relativity but, moreover, for the whole 3-parameter class of tetrad-teleparallel theories. For the physically favored one parameter subclass, which includes the teleparallel equivalent of Einstein's theory as an important special case, the energy for all class B models is, contrary to expectation, negative.

For the Poincare gauge theory of gravity we consider the dynamical scalar torsion mode in a cosmological context. We explore in particular the possibility of using dynamical torsion to explain the current state of the accelerating Universe. With certain suitable sets of chosen parameters, this model can give a (qualitatively) proper description of the current universe without a cosmological constant, and the universe described is oscillating with a period of the Hubble time.