Well like I said above I'm no gen-rel geek but I do believe one should, rather, discuss the speed of gravitational waves. LIGO's site (Laser Interferometer Gravitational-Wave Observatory) has a discussion about it--
"Gravitational waves are ripples in the fabric of space and time produced by violent events in the distant universe, for example by the collision of two black holes or by the cores of supernova explosions. Gravitational waves are emitted by accelerating masses much as electromagnetic waves are produced by accelerating charges. These ripples in the space-time fabric travel to Earth, bringing with them information about their violent origins and about the nature of gravity."
Well that was actually kind of lame. Lemme see if I can find anything better-- here's something written in '99--
"According to Newton's theory of gravitation, the binary period of two point masses (e.g., two stars) moving in a bound orbit is strictly a constant quantity. However, Einstein's general theory of relativity predicts that two stars revolving around each other in a bound orbit suffer accelerations, and, as a result, gravitational radiation is generated. Gravitational waves carry energy and momentum at the expense of the orbital decay of two stars, thereby causing the stars to gradually spiral towards each other and giving rise to shorter and shorter periods. This anticipated decrease in the orbital period of a binary pulsar was first observed in PSR 1913+16 by Taylor and Weisberg. The observation supported the idea of gravitational radiation first propounded in 1916 by Einstein... [who] showed that the first order contribution to the gravitational radiation must be quadrupolar in a particular coordinate system..."
The "quadrupole moment tensor of the energy density of the source", which is gotten to after you fix and gauge and pick a coordinate system and solve for Einstein's (linearized) equations Dal(h'_mu*nu) = -16PiGT^mu*nu (h' = trace-reversed perturbation, G = Einstein tensor, and I don't know what the hell T is... Tensor shorthand for something...) and fool around with a Fourier transform and take out the spatial part, is given on page 10 and has too many super/subscripts for me to put it here easily...