Concept formulated by Princeton philosopher Gilbert Harman in his paper ‘The Inference to the Best Explanation’, Philosophical Review 74 (1965). Arguably the idea goes back to Peirce, although Peirce's views are not altogether easy to excavate. The idea is that when we have a best explanation of some phenomenon, we are entitled to repose confidence in it simply on that account. Sometimes thought to be the lynchpin of scientific method, the principle is not easy to formulate and has come under attack, notably since our best current explanation of something may be only the best of a bad lot. There exist cases in which the best explanation is still not all that convincing, so other desiderata than pure explanatory success seem to play a role. See also constructive empiricism.