In Harm’s Way

To the Editors:

In his recent review [“It’s For Your Own Good!” NYR, March 7], Cass Sunstein claims that recent empirical studies undermine John Stuart Mill’s Principle of Harm, also known as his Principle of Liberty. He is wrong about this. Mill’s principle says in part:

The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right.

This is not an empirical principle; it makes no empirical prediction and hence none that can be falsified by empirical findings; nor does it presuppose that people always or even generally make wise choices about their lives.

As Sunstein points out, one of Mill’s arguments for his principle is that individuals are in a better position than government officials or other outsiders to know what is good for them. This is not a good argument. Even apart from exceptions that Mill allowed for, involving small children and severely mentally ill people, sometimes spouses, friends, doctors, and even government officials may know better than an individual about how best to achieve the person’s goals or even whether the goals themselves are wrong. The failure of Mill’s argument, however, does nothing to disprove his Principle of Liberty.

Many people act against their own interests by eating the wrong foods or failing to think clearly. Some would be better off if they changed their diet or took a course in logic, but should they be coerced to do either of these things? Mill’s principle implies that they should not be coerced if they are not harming others. Someone who disagrees can argue the opposite; in the words of the subtitle of the book Sunstein reviewed, one might argue for “coercive paternalism.” But this requires a moral argument, not merely an appeal to empirical findings about people making unwise choices.

Ed Erwin
Department of Philosophy
University of Miami
Miami, Florida

Cass R. Sunstein replies:

Ed Erwin is right to say that a moral argument is required to undermine Mill’s Harm Principle. Mill was a utilitarian, and one of his central (moral) arguments was unmistakably utilitarian in character. In Mill’s account, the individual “is the person most interested in his own well-being,” and the “ordinary man or woman has means of knowledge immeasurably surpassing those that can be possessed by any one else.” When society seeks to overrule the individual’s judgment, Mill wrote, it does so on the basis of “general presumptions,” and these “may be altogether wrong, and even if right, are as likely as not to be misapplied to individual cases.”

Behavioral economists have demonstrated that in at least some contexts, ordinary men or women make systematic mistakes about what will promote their own well-being. To that extent, they have raised a significant challenge to utilitarian supporters of the Harm Principle. Their objection is a moral one. If our goal is to increase people’s utility, the Harm Principle—suggesting that a person’s actions should be subject to coercion only if they will harm others—may prevent us from achieving that goal. In some cases, “nudges” that preserve choices—by means of education and warnings, for example—will be required. (Mayor Bloomberg’s recent proposal to restrict the display of cigarettes counts as a nudge.) In other cases, coercion might be necessary on utilitarian grounds (for example, to protect people from running excessive risks to their health and well-being).

There are nonutilitarian defenses of the Harm Principle. Maybe people have a right to choose freely, even if their choices get them into terrible trouble. Erwin is right to suggest that for those who accept this view, behavioral findings about human error provide no defense of coercion. Even so, such findings help to justify nudges, which allow people to go their own way. And the same findings might well prompt defenders of the Harm Principle to do a bit more thinking about the limits of that principle. Consider, for example, apparently paternalistic regulations designed to protect workplace safety, to increase the fuel efficiency of cars, and to make refrigerators and other appliances more energy-efficient than consumers are demanding. Those regulations have (and deserve) widespread support, but it is far from clear that they can be squared with the Harm Principle.