Two inimical interpretations of psychotherapy look to many of the same features of empirical research. One camp infers that placebo effects are impossible in principle in psychotherapy; the other camp infers from the same research that psychotherapy is essentially placebo. I examine the crucial discussions and conclude that these opposing evaluations ensue because each group presumes a different baseline from which the significance of the research is gauged. I show how different baselines set different standards of significance and invite different (...) comparisons between therapeutic and non-therapeutic practices. Attention to this point about the baseline of significance puts the placebo debate in sharper relief and frames more clearly the merits and demerits of each position. It also leads to the conclusion that neither group really lays hold of in-principle claims, since both interpretations rest on context-dependent features of the empirical situation. Along the way, I discuss salient features of outcomes research, meta-analysis, and the role of symbols in therapeutic transactions. I close with some remarks that link the proliferation of therapies with cultural splintering generally and suggest how this might bear on the placebo issue. (shrink)