Hard cases present a dilemma at the heart of constitutional law. Courts have a duty to decide them—to vindicate rights, to clarify law—but doing so leads to errors (judges do not know the “right answer”) and strains the credibility of courts as impartial decisionmakers. Theories of constitutional adjudication tend to embrace one horn of this dilemma. We explore a principle for deciding hard cases that appreciates both. We argue that courts should decide hard cases against the party who could have more easily avoided the conflict in the first place. This is the conflict-avoidance principle. The principle builds on and systematizes “least cost avoidance” in private law and myriad constitutional doctrines. We apply the principle to several cases, generating insights into discrimination, affirmative action, religion, and so on. The principle represents a form of common-law constitutionalism, and it reveals connections between rights, markets, and state power. It also invites objections, to which we respond. Conflict avoidance is not “value-neutral,” and it cannot resolve every hard case. But it can resolve many in a practical way.