An analysis of the control groups in double-blind trials of medicines demonstrates broad variation--from 0 to 100 percent--in placebo effectiveness rates for the same treatment for the same condition. In two cases considered here, drug healing rates covary with placebo healing rates; placebo healing is the ultimate and inescapable "complementary medicine." Several factors can account for the dramatic variation in placebo healing rates, including cultural ones. But because variation differs by illness, large placebo effects for one condition do not necessarily anticipate large placebo effects for other conditions as well. Deeper understanding of the intimate relationship between cultural and biological processes will require close ethnographic scrutiny of the meaningfulness of medical treatment in different societies.