Sounds like a contradiction - light being faster than itself. However, by FTL I mean "Faster Than the speed of Light in a vacuum".
Recently my attention has been drawn to some reports of light pulses in an optical fibre travelling FTL. They tend to be quite sensational - making much of the great but unspecified implications for communication. I thought I'd check it out.
There are a bunch of papers about this, like this one. So it's real! Next I had to actually read them, and find people in the field to help me understand them - one of the advantages of having a scientific community. Gradually understanding dawns and, as suspected, it's a trick.
I was shown a nice animation that illustrates the effect quite well.
When the articles say "FTL light pulse" what we imagine is that someone points a torch (or something) down a fiber, turns it on, turns it off. The middle of the pulse (the "group velocity") is FTL. But that's not quite what is going on here.
If you watch the animation, nothing at all physical is going FTL ... the individual photons are all STL. Light moving Slower Than Light is not a problem, lightspeed in a medium is less than in a vacuum, it's how optical fibers work. What happens is that each of the colors travel at a different STL speed, and so they interfere with each other. The colors still add up to white light, but the brightest part of the beam (not pulse) changes position quickly - FTL. Those are the big wiggles in the animation. It is the bright bump in the beam that the authors of the papers are calling the "superluminal pulse" and the sensational articles are "forgetting" that this pulse is just part of an overall beam.
And the great consequences for communication?
The superluminal pulses cannot be used for FTL communication. The other effects, slowing down the long-wavelength colors and all that, those are the bits with the potential use. Greatly slowing light pulses would be useful in making optical computers for example. Light is just too darn fast.