1. the principle that causes must temporally precede their effects, and never vice versa. See also false cause; reverse causality. 2. in path analysis and similar statistical procedures, the categorizing of causal variables as either more or less direct.