What are Griffiths effects in the context of condensed matter physics?
From a cursory examination of the literature I've gathered the following: it seems that ordered systems have a "clean" critical point, at which the system makes a sharp phase transition, and that disordered systems have a "dirty" critical point which is perturbed from the clean critical point due to the randomness of the system. A Griffiths phase then occurs when the control parameter is between the clean and dirty critical points, and is characterized by rare occurrences of local order in an otherwise disordered phase.
Is this correct? What does "rare" really mean? Is this a recently discovered phenomenon?