What do people actually think about nudging and choice architecture? Do they have serious ethical objections? Or do they believe that nudges are acceptable or desirable, or even morally obligatory? Do they distinguish among nudges, and if so, exactly how?

The answers cannot, of course, dispose of the ethical questions. The issue is how to resolve those questions in principle, and empirical findings about people's answers are not decisive. Perhaps those answers are confused, insufficiently considered, or wrong. There is a risk that if people are responding to survey questions, they will not have time or opportunity to reflect, especially if those questions do not offer relevant facts (e.g., about the costs and the benefits of the policies in question). Even if their answers are reflective, perhaps people do not value autonomy or dignity highly enough, or perhaps they do not quite know what those concepts means. Perhaps people pay too little attention to social welfare, or perhaps their judgments about social welfare are off the mark, at least if they are not provided with a great deal of information.

Perhaps different nations, and different groups within the same nation, offer different answers, suggesting an absence of consensus. Behavioral scientists would emphasize a related point: People's answers to ethical questions, or questions about moral approval or disapproval, might well depend on how such questions are framed; slight differences in framing can yield dramatically different answers. Those differences are themselves a nudge; they can have major effects, and they are not easy to avoid.

Here is a small example of how ethical judgments can depend on framing. If people are asked whether they think that young people should be valued more than old people, they will usually say, “certainly not.” They will strenuously resist the idea that government should give a higher value to young lives than to old ones. But suppose that people are asked whether they want either (1) to save 70 people under age of 5 or (2) to save 75 people over the age of 80. It is reasonable to speculate (and evidence confirms) that most people will choose (1), thus demonstrating that they are willing to value a young person more than an old one.