Abstract

Here are two claims, both of which (I maintain) are very plausibly true: (1) in the late 1870's, A. A. Michelson measured the speed of light, and his result was only about 150km/s off from the currently accepted value of c (approximately 299,792 km/s); and, (2) Strictly speaking, in special relativity, there is no such thing as the speed of light. These claims are clearly in tension, and this paper resolves that tension. The first step is to defend claim (2), which is, remarkably, controversial even among working physicists and philosophers of physics. In most cases, belief that light has a speed is justified by pointing out that all co-ordinatizations that describe inertial reference frames agree that light has a speed (and that it's c), even if they disagree about other things (like whether two space-like separated events are simultaneous, or whether a given massive body is at rest or in uniform motion in a straight line). This line of reasoning, I argue, comes from commitment to what I call the Co-ordinate Abstraction Principle. I show that this principle, in its most straightforward formulation, is clearly false; even the simplest co-ordinatizations of the most ordinary spaces give rise to fatal counterexamples. Even further, I show that any weakened reformulation of the principle able to avoid these counterexamples will, thereby, be rendered too weak to apply to the speed of light in special relativity. Once the plausible truth of (2) is established, I resolve the tension between (1) and (2). The value, 299,792 km/s, merely describes something about our choices of spatial and temporal units. However, the value of c couldn't be purely conventional, or else it could not have been measured. The reasons our choices of spatial and temporal units give rise to an a posteriori discoverable fact is that the structure of Special Relativity makes an independent unit of spatial distance redundant. I explain why this value is so readily represented as a speed, and how Michelson's measurement could track such a relationship.