We present an alternative to Topologically Massive Gravity (TMG) with the same "minimal" bulk properties; i.e. a single local degree of freedom that is realized as a massive graviton in linearization about an anti-de Sitter (AdS) vacuum. However, in contrast to TMG, the new "minimal massive gravity" has both a positive energy graviton and positive central charges for the asymptotic AdS-boundary conformal algebra.

A wide class of three-dimensional gravity models can be put into "Chern-Simons-like" form. We perform a Hamiltonian analysis of the general model and then specialise to Einstein-Cartan Gravity, General Massive Gravity, the recently proposed Zwei-Dreibein Gravity and a further parity-violating generalisation combining the latter two.

The "Minimal Massive Gravity" (MMG) model of massive gravity in three spacetime dimensions (which has the same anti-de Sitter (AdS) bulk properties as "Topologically Massive Gravity" but improved boundary properties) is coupled to matter. Consistency requires a particular matter source tensor, which is quadratic in the stress tensor. The consequences are explored for an ideal fluid in the context of asymptotically de-Sitter (dS) cosmological solutions, which bounce smoothly from contraction to expansion. Various vacuum solutions are also found, including warped (A)dS, and (for special values of parameters) static black holes and an (A)dS 2 × S 1 vacuum.

We present a "Chern-Simons-like" action for the "general massive gravity" model propagating two spin-2 modes with independent masses in three spacetime dimensions (3D), and we use it to find a simple Hamiltonian form of this model. The number of local degrees of freedom, determined by the dimension of the physical phase space, agrees with a linearized analysis except in some limits, in particular that yielding "new topologically massive gravity", which therefore suffers from a linearization instability.Comment: 25 pages, minor corrections plus extended discussion in v

We construct O(D, D) invariant actions for the bosonic string and RNS superstring, using Hamiltonian methods and ideas from double field theory. In this framework the doubled coordinates of double field theory appear as coordinates on phase space and T-duality becomes a canonical transformation. Requiring the algebra of constraints to close leads to the section condition, which splits the phase space coordinates into spacetime coordinates and momenta.