Earthquake lights were considered to be apocryphal in the lobbies of academia, and largely ignored. They were taken up by the likes of Paul Devereux, on the fringe of science. He wrote two books on the subject – and they are very illuminating. No actual scientist had made a study or witnessed them and therefore they did not exist – the usual pretence. This could be described as the perfect example of consensus science.
It was not until a series of photographs saw the fresh air of modern technology, which showed too clearly that strange lights did occur (during an earthquake in Japan). As a result of this, and due to Devereux too, some scientists began to investigate the phenomenon. See http://blogs.smithsonianmag.com/science/2014/01/why-do-lights-sometimes-…
Science has had a problem in explaining earthquake lights. Various hypothesis have been offered. One theory, popular at one time, is that tectonic movement of rocks that include quartz, is capable of generating a pizoelectric field that produces flashes of light. Others prefer the idea that tectonic stress somehow allows electromagnetic energy to be generated, triggering changes in the magnetic charge of the ionosphere (high in the atmosphere). A more muted version of the latter, which deletes the ionosphere, can be seen in a paper in Seismological Research Letters by Robert Therialt et al. They set out to analyse the geology behind 65 earthquakes that had produced reports of light in order to find out what they might have in common. Earthquake lights occur before or during rather than after an earthquake. Rocks appear to conduct energy up to the surface – but not to the ionosphere (they have decided). Stress causes negatively charged oxygen atoms to break apart and these flow through cracks in the rock towards the surface. At this point, or the theory claims, high energy groups of these charged atoms will ionise pockets of air forming a charged gas – plasma. This emits the lights. It is the stress leading up to the rupture, over an extended period of time, that creates the burst of light (or energy). It is not the actual seismic activity itself.
This is probably not the end of the story as some scientists have elsewhere claimed earthquakes can be predicted by looking at unusual activity in the ionosphere – high above the surface of the earth. Piers Corbyn has also tried to predict earthquakes and volcanoes by looking at activity generated by the solar wind (and ultimately from plasma with an origin on the Sun).