Forecasting earthquakes implies that there are time-varying processes, which depend on the changing conditions deep in the Earth's crust prior to major seismic activity. These processes may be linearly or non-linearly correlated. In seismology, the research has traditionally been centered on mechanical variables, including precursory ground deformation (revealing the build-up of stress deep below) and on prior seismic events (past earthquakes may be related to or even trigger future earthquakes). Since the results have been less than convincing, there is a general consensus in the geoscience community that earthquake forecasting on time scales comparable to meteorological forecasts are still quite far in the future, if ever attainable.The starting point of the present review is to acknowledge that there are innumerable reports of other types of precursory phenomena ranging from emission of electromagnetic waves from ultralow frequency (ULF) to visible (VIS) and near-infrared (NIR) light, electric field and magnetic field anomalies of various kinds (see below), all the way to unusual animal behavior, which has been reported again and again. These precursory signals are intermittent and seem not to occur systematically before every major earthquake and reports on pre-earthquake signals are not widely accepted by the geoscience community at large because no one could explain their origins. In addition, the diversity of the signals makes them look disparate and unrelated, hampering any progress.We review a credible, unifying theory for a solid-state mechanism, which is based on decades of research bridging semi-conductor physics, chemistry and rock physics. This theory, which we refer to as the "peroxy defect theory", is capable of providing explanations for the multitude of reported pre-earthquake phenomena. A synthesis has emerged that all pre-earthquake phenomena could trace back to one fundamental physical process: the activation of electronic charges (electrons and positive holes) in rocks subjected to ever-increasing tectonic stresses prior to any major seismic activity, via the rupture of peroxy bonds. The holes are unusual
I-Introduction 1.1 The cost of earthquakesLarge earthquakes are, by far, the deadliest of all natural disasters, claiming an average of 60,000 lives a year, featuring gigantic fluctuations (e.g. 80,000 victims from 1994 to 2004, and 780,000 from 2001 to 2010; see e.g. Knopoff and Sornette, 1995, for the size distribution of death tolls related to seismic events), which partly mirrors the highly intermittent distribution of seismicity in space, time, and magnitude. On more economic grounds, such disasters are also causing colossal property and industrial damage, with that of the 1989 Loma Prieta earthquake in California alone estimated at $6 billions (over €4 billions), the 1995 Kobe event in Japan estimated at $200 billions (€150 billions), while the 2011 Tohoku earthquake followed by its great tsunami already stands with much higher losses, with costs continuing to rise with the on-going m...