D. B. Rubin's model for causal inference in experiments and observational studies is enlarged to analyze the problem of "causes causing causes" and is compared to path analysis and recursive structural equations models. A special quasiexperimental design, the encouragement design, is used to give concreteness to the discussion by focusing on the simplest problem that involves both direct and indirect causation. Encouragement designs involve both randomization and self-selection as well as both direct and indirect causation. Encouragement designs provide a simple yet useful "laboratory" in which the issues of direct and indirect causal relationships can be carefully examined. Rubin's model is shown to extend easily to this situation and to specify conditions under which the parameters of path analysis and recursive structural equations models have causal interpretations. (Contains 8 figures and 44 references.) (Author/SLD)