The approach of pragmatism is to test each notion about these first things by tracing and evaluating its consequences. But the criteria for evaluating those consequences are derived from—what, exactly? Surely not our immediate preferences, for those have to be "nudged" from time to time. In order to define what works, pragmatism must have recourse to some principle outside itself. Unfortunately, too often pragmatists such as Sunstein leave it at "the best results on balance," without defending the implicit assumption about those best results. We are left with platitudes about being empirical, nuanced, and avoiding dogmatism, without a serious discussion of the important questions that are so carelessly glossed over. The platitudes provide cover for ignoring those assumptions, and framing one's position as based on facts, rather than "values" derived from metaphysical reasoning.黑体是我加的。还有：
Ultimately, Nudge relies on the idea that there are "choice architects" with objective knowledge of what is best for us, who should frame choices in order to get us to understand what is in our best interest. But these are the same sort of experts who failed, miserably, to predict the consequences of President Obama's stimulus bill on the rate of unemployment. Sunstein and Thaler cannot get around the knowledge problem: how do we know which nudges will be helpful and which will be harmful? And then there's the knower problem: who out there is qualified to be a choice architect?
Worst of all, libertarian paternalism is confused about what it means to have liberty. Liberty is defined by Sunstein and Thaler as the ability to choose among competing options. But government can deprive someone of liberty even while presenting him with several options; and liberty can exist even when only one option is present. As Friedrich Hayek explained so well in The Constitution of Liberty, "the range of physical possibilities from which a person can choose at a given moment has no direct relevance to freedom. The rock climber on a difficult pitch who sees only one way out to save his life is unquestionably free, though we would hardly say he has any choice." The upshot is clear: choice and liberty are not the same thing. Liberty is possible even where choice does not exist, and choice does not always translate into liberty. Libertarian paternalism turns out to be merely soft paternalism.
Yet Sunstein's ability to expose the tension between conservative traditionalism and originalism does not diminish the profound contradictions in his own arguments. In some books he wants judges to defer to tradition and the wisdom of crowds, and to avoid imposing their rationalist blueprints on society. But in others he wants regulators and judges to be free to form citizens' preferences and establish a "deliberative democracy" where people are influenced to think rightly about questions of importance to them. The meaning of the Constitution or a law is not found in the document itself, he claims, and every interpreter must bring his own preferences to the text. But the meaning of such texts is not indeterminate, he warns, lest we give up on reason altogether. On the one hand, the New Deal regulatory state did significant damage to the Constitution. On the other, the meaning of the Constitution changes over time, and FDR's Second Bill of Rights should be the lodestar of American policy and jurisprudence. Professor Sunstein's own mental constitution is one of many minds.