In terms of experiment and observation, astronomers have gathered data from phenomena like quasars and pulsars, and concluded that gravity is not instantaneous. Yet its speed is calculated by many of these astronomers as being significantly faster than the speed of light, a conclusion which, if accepted, would challenge many propositions taken as axiomatic by physicists.
While Newton took gravity to take effect instantaneously, he also considered any instantaneous effect at a distance as problematic. Many Newtonian thinkers after him solved the problem, or side-stepped it, by saying that the speed of gravity was so fast that it should simply be calculated as instantaneous.
We can conceptualize the question in a thought-experiment:
Let us imagine a universe devoid of matter except for three simple objects - e.g., three solid metal spheres, like ball bearings. These three are in motion, traveling in a plane. Although we are imagining a standard four-dimensional universe, i.e., the three Euclidean dimensions and time, the objects are moving only on a plane. They are in inertial states of motion as follows: two of them are on parallel paths, such that their routes would never intersect, save for the forces of gravity, and they will never leave the plane; the third is moving on a path such that it will collide with one of the other two. At any point in time prior to the collision, we can nicely predict their routes and project them into the future. The forces and their mathematics are well-known, and we have eliminated complications like friction or gravitational fields from other objects. The only quantities to be considered would be the masses of the objects, their speed and direction (i.e., velocities), and the forces of gravity between them. The two objects on parallel paths will presumably be drawn toward each other by gravity so that their paths would cease to be parallel. Let us stipulate that they will not collide until after the third object first collides with one of them, if ever. Up to this point, the situation is unremarkable, not controversial, and easily predictable by Newtonian means. At the point in time at which the third object collides with one of the first two, the two objects in the collision will change velocity. Let us stipulate that the collision does not destroy the objects; imagine billiard balls colliding on the table. Given that the collision and the change in velocity are either instantaneous or nearly so, we will regard this as a single point in time. The force of gravity upon the third object, the one not in the collision, will change also at this point in time. Given the change in the force of gravity upon that object, and the stipulated absence of moderating factors like friction, the third object’s path will change as the force of gravity upon it changes. Our question is then this: is the point in time at which the third object’s path changes different than the point in time at which the collision took place? If yes, then gravity is not instantaneous, and travels at a speed. If no, the gravity is instantaneous.
Philosophers are well aware that attempts to answer questions, whether successful or not, usually yield more questions and meta-questions. The present case is no exception. Is our thought experiment well-formed? Does it have any bearing on physical reality? It certainly fails to yield an answer to the question of whether gravity is instantaneous.
Any attempt to translate this thought experiment into observational natural science will be fraught with complications. Rather than working with a stipulated idealized universe, we would have to work with the real one. Attempting to observe and measure, e.g., tiny variations among asteroids, which travel through space with inertial motion and occasionally collide with one another, would be affected by numerous other gravitational fields in the area - fields emanating from other objects. Calculating these would be impossible or nearly so. The small changes in path and the precision needed to measure such small bits of time would exceed our technological grasp.
If there is any value attributable to this thought experiment, it might be this: that it presents the question in an isolated form. This is the procedure normally adopted by physics textbooks when presenting less controversial matter.
The question about the speed of gravity touches upon questions about the mechanism of gravity, and about the cause or source of gravity. Newton famously violated his own dictum that he would form no hypotheses. Commenting upon this benign self-contradiction, Professor Lawrence Sklar writes:
His hypotheses about gravity, for example, often have a very Cartesian flavor to them, as they postulate “ethers” that fill the universe with various fluid properties of pressure and resistance, and whose relation to matter (perhaps of lower pressure where matter is present, resulting in a “push” that moves matter toward matter) might, possibly, explain the law-like behavior of gravitational attraction. Such “mechanisms” might also remove from gravity the taint of action at a distance. It is worth noting here that the elements that later function to suggest the replacement of “action at a distance” theories by theories that propose an ontology of “fields” intermediate between the interacting objects, that is to say the time lapse in inter-particle actions and the violation in conservation of energy that results if one is not very careful in framing an “action at a distance” theory, play no role in the controversies embroiling Cartesians and Newtonians in Newton’s time.
Given that there are still unanswered questions and significant controversies about gravity, Newton’s suggestions, even if some of them are ultimately rejected, are still worth examination. Sklar continues:
Some of Newton’s hypotheses remain only curiosities in the history of science. Others, such as his particle theory of light, remain, if not really correct, important contributions to the development of later science. Still others, such as his hypothesis expressed in the “Queries” to the Opticks that there might be other forces along with that of gravity by which matter influences matter, and that these other forces might account for such things as the structure and behavior of materials, are prophetic insights into what became large components of the future growth of scientific understanding.
In post-Newtonian physics, questions about the speed of gravity take a different form. They might be posed as questions about the speed at which spacetime can change shape, i.e., curve or become curved. Or they might be asked about the speed of gravity waves, instead of the speed of gravity.
