The Open SystemPhysics, unsettled

The Speed That Refuses to Add Up

Walk forward on a moving train and the ground sees your speed add up. Do the same with a torch beam and it does not. This single refusal reorganised physics.

Galileo's principle of relativity says that the laws of physics look the same in every uniformly moving frame. Newtonian mechanics obeys it, and so does our daily experience: throw a ball forward on a moving train and its speed relative to the ground is the sum of the two speeds. Velocities add, simply and reliably, as long as the speeds involved are small.

Light is the exception. Measure its speed from a moving train, from a station platform, from a satellite, in any direction and at any time of year, and you get the same number: about 299,792,458 metres per second in vacuum. Not approximately the same after correction for the observer's motion — the same. This is not a statement about the technology of measurement. It is the postulate that the theory is built around, and it has been verified to remarkable precision.1

How the contradiction dissolves

There is a price, and Einstein paid it. If the speed of light does not change for any observer, then something else has to give: simultaneity, length, and the flow of time. Two events that are simultaneous in one frame are not simultaneous in another. A moving rod is measured shorter along its direction of motion. A moving clock is observed to tick slowly. These are not philosophical claims — they are the operating rules of the universe at high speed, and they have been confirmed with particle beams, supersonic clocks, and satellite systems.

Velocities do not add the naive way. At ordinary speeds the correction for adding two velocities is smaller than a rounding error, which is why the classical rule feels exact. At speeds close to light the correction is what keeps the total below the limit. An object travelling at 0.9 c relative to you that launches something forward at 0.9 c still does not exceed c. The arithmetic never reaches the ceiling; it approaches it.

Two facts worth keeping apart

Theoretical. No physical object with mass can be accelerated to the speed of light; the energy required grows without bound.

Observational. A speed limit does not mean nothing is fast. Cosmic rays and accelerator particles routinely approach c to within a few parts per billion.

What we actually measure

First, the speed of light is no longer a measured constant in the everyday sense. Since 1983 the metre has been defined in terms of it, so the number is exact by convention and it is the metre that is measured.2 Second, the claim that the speed of light is independent of the observer's motion is tested in many ways. Laser ranging to the Moon, the tracking of planetary spacecraft, and the positions of synchrotron sources observed after rotating the apparatus all produce results consistent with a single frame-independent speed of light. One independent constraint comes from observations of gamma-ray bursts: differences between high-energy and low-energy photons from the same distant explosion are consistent with zero, sharply limiting exotic theories that would make the speed of light energy-dependent.3

The kinetic consequences are routinely used. In particle physics, energies and momenta are computed relativistically; in any other way, accelerator experiments would mispredict by enormous factors.

The honest limits

  • confirmed The speed of light in vacuum is the same for all inertial observers.
  • confirmed Massless particles travel at c; gravitational waves do as well.
  • open Whether the maximum propagation speed of information is exactly identical for all signals, and whether new physics appears near the Planck scale, is not settled.

The speed limit is not a traffic law imposed on nature from outside. It is a structural feature of spacetime: causally connected events are connected by the geometry itself. Once you accept that, the strange parts of relativity stop being strange and become arithmetic.

Further reading (external sources)

  1. Einstein, A., “Zur Elektrodynamik bewegter Körper”, Annalen der Physik 322, 891 (1905) — doi:10.1002/andp.19053221004.
  2. BIPM, SI Brochure, definition of the metre (1983) — bipm.org.
  3. Fermi Gamma-ray Space Telescope, constraints on Lorentz invariance violation from GRB observations — science.nasa.gov/mission/fermi.

This article is an original explanatory text written for The Open System; it is not a translation or a reproduction of any source above. Sources are listed for verification and further reading.