Abstract: In the late nineteenth century, James Bradley Thayer urged that an act of Congress should not be struck down unless the constitutional violation “is so clear as to leave no room for reasonable doubt.” Thayer’s beyond-a-reasonable-doubt test helped define constitutional understandings for more than a half-century; Oliver Wendell Holmes, Louis Brandeis, Learned Hand, Benjamin Cardozo, and Felix Frankfurter were practicing Thayerians. Thayerism provided crucial orientation for Alexander Bickel’s conception of judicial review and his embrace of “the passive virtues,” and also for John Hart Ely’s democracy-reinforcing approach to constitutional law. But Thayerism seems to have dropped out of contemporary constitutional law. One reason is that as a matter of simple psychology, it is extremely difficult for any judge consistently to embrace it. Another reason is that Thayer’s defense of Thayerism was very thin; for the most part, he purported to be describing longstanding practice, rather than to be justifying it. But if we make certain judgments about the likely capacities and performance of judges, legislators, and others, Thayerism would make a great deal of sense. If we make contrary judgments, Thayerism would be preposterous. Selective Thayerism, of the sort defended by Bickel or Ely, might follow from yet another set of judgments. The broader lesson is that no approach to constitutional law can be adopted or rejected in the absence of an answer to the question whether it would make our constitutional order better rather than worse, which requires in turn a set of judgments about the likely behavior of various institutions. We might also understand Thayerism as a kind of arms control agreement: I will adopt a Thayerian approach if you will as well. More particularly, left-of-center judges might be willing to be Thayerian if and only if right-of-center judges are willing to be Thayerian as well. The problem, of course, is that unless a strong norm is in place, both sides will be tempted to defect. And that is, in fact, what we observe.