… is from page 29 of the 1981 edition of George Reisman’s translation of Ludwig von Mises’s 1933 tract, Epistemological Problems of Economics:
Human action always confronts experience as a complex phenomenon that first must be analyzed and interpreted by a theory before it can even be set in the context of an hypothesis that could be proved or disproved; hence the vexatious impasse created when supporters of conflicting doctrines point to the same historical data as evidence of their correctness. The statement that statistics can prove anything is a popular recognition of this truth. No political or economic program, no matter how absurd, can, in the eyes of its supporters, be contradicted by experience. Whoever is convinced a priori of the correctness of his doctrine can always point out that some condition essential for success according to his theory has not been met.
DBx: Yep. This reality, however, does not render history or statistics useless. Instead, it points, first, to the importance of interpreting data and experience with sound rather than unsound theories, and, second, to the need to be both careful and humble in the handling of history and statistics.
Ludwig von Mises was born on this date in 1881.