Although liquefaction of soils during earthquakes has been researched intensively for more than 20 years, it has remained a confusing problem - owing to seemingly divergent viewpoints as to the fundamental nature of the problem. During the past several years there has been a clarifying and coming together of these viewpoints, and hence a much clearer framework of understanding has been established. This new perspective is presented and related to previously developed methods of investigation and analysis, and remaining problems are identified and discussed. Several recent advances re parts of the problems - prediction of limited permanent deformations and delayed failure - are also summarized.