The debate over the ontological status of space has been waged over the millennia on many battlefields. In ancient times the major focus was on the possibility of a vacuum, an entity that was both something and yet, in another way, nothing. Newton shifted the scene of action to dynamics, to the question of the ontological commitments needed to give an adequate explanation of observable phenomena. But along with the dynamical effects of Newtonian absolute space came also the spectre of motions unobservable in principle, a notion attacked from Leibniz onward as conceptually absurd or meaningless. So the theater of operation was moved again, especially by verificationists, this time into theories of meaning. Now John Norton and John Earman have opened an entirely new line of assault on the question. They argue that considerations about the possibility of determinism have direct implications for the substantivalist-relationist controversy.