Autonomy of the Nudged

The concept of autonomy usually refers to individuals’ capacity for self-control and self-determination. It is most often mentioned as a fundamental principle in Kantian deontology, but autonomy can also play a role in consequentialist theories, as it is conducive to well-being.

Nudging has been described as a form of libertarian paternalism by Thaler and Sunstein (2008). It is paternalistic as it influences choices “in a way that will make choosers better off, as judged by themselves” and at the same time libertarian because nudges don’t coerce or restrict choices (p. 16). It is tempting for behavioral science practitioners to rely on this definition to alleviate ethical worries around the nudged’s autonomy: Behavioral architects have your interest in mind and nudges don’t coerce you, so autonomy concerns are relatively unimportant.

Nudge critics would point out that questions of individual autonomy are largely independent of this interpretation, as they deal with the following concerns:

1. Lack of transparency in nudges
2. Nudges take advantage of people’s biases
3. Making nudge-based choices weakens individual agency

Nudges work with and counteract a number of different decision-making problems and human shortcomings, especially uncertainty, inaction, and impatience; thus, nudging may exploit people’s irrationality. Savings defaults, also known as auto-enrollment, for example, take advantage of individuals’ inaction (inertia). Since nudges are thought to operate mainly on a subconscious level, particularly nudges classified as Type 1, critics argue that they lack transparency and are manipulative. They undermine autonomy because they exploit irrationality and influence people ‘behind their backs’. In this sense, Thaler and Sunstein arguably overstate the choice-preserving aspect of nudges. Some contend that only the System 2 route of rational persuasion preserves individuals’ autonomy over their own choices.

It has also been argued that, even if nudges aren’t manipulative, their excessive convenience diminishes active choice. Automatic pension enrollment and other defaults are prime examples of passive choice. In addition, even if there is no real loss of control, people may feel manipulated if there is perceived loss of autonomy.

If choice architects discourage autonomous, reflective decision-making, they may contribute to an infantilization of citizens and consumers: People’s ability to make their own decisions, form their own preferences, and self-regulate might diminish in the long term. Nudging can also deprive individuals of the capacity for making bad choices and thereby erode their ability to take responsibility for their own choices. This is not how people should be treated, especially by governments.