The term “bad-enough” participation refers, in a general sense, to the ways that all analysts inevitably cause their patients to suffer and, more specifically, within each analysis, to the ways in which the analyst's participation confirms some version of the patient's worst fears. These manifestations of “badness,” as long as they are honestly considered and creatively used, are expectable, and even therapeutically essential, aspects of analytic relatedness. Frequently, however, an analyst's bad-enough participation, despite its ubiquity and transformational potential, is attenuated or selectively overlooked. As a result, the therapeutic possibilities of analytic work may be compromised. In this paper the analyst's bad-enough participation is considered conceptually, and an argument is advanced for its therapeutic salience. Two cases are discussed, the first involving a treatment impasse, and the second, the beginning phase of a long-term treatment.