Radically questioning the principle of the least restrictive alternative: a reply to Nir Eyal

Abstract

In his insightful editorial, Nir Eyal explores the connections between nudging and shaming. One upshot of his argument is that we should question the principle of the least restrictive alternative in public health and health policy. In this commentary, I maintain that Eyal’s argument undermines only a rather implausible version of the principle of the least restrictive alternative and I sketch two reasons for rejecting the mainstream and more plausible version of this principle.

In his insightful editorial (1), Nir Eyal explores the connections between nudging and shaming. He makes four claims:

• Some nudges deliberately harness the power of shame.

• Other nudges generate shame as an unintended side effect.

• When we can predictably anticipate that a nudge will generate shame as an unintended side effect, it is preferable to select a more restrictive alternative than to unnecessarily expose individuals to shame.

• The Principle of the Least Restrictive Alternative (hereafter, PLRA) is objectionable.

I could take issue with the first claim, which depends on how nudges and shame are respectively defined. Although Socrates made himself quite a name discussing definitions, I here refrain from my professional taste for hair splitting. Instead, here, I take issue with Eyal’s last point. Eyal’s argument undermines only a rather implausible version of PLRA. But there are reasons for rejecting the mainstream version of this principle that I sketch in this commentary.

Public health and health policy issues are rarely due to a single tractable and alterable cause and are notoriously difficult to address. Even partial solutions often require using the law, making regulations, mobilizing complex social institutions, and changing the behavior of a plurality of agents (from individuals to corporations). If proven efficacious and effective, nudges are part of the regulatory toolbox, nothing more, nothing less. As Eyal notes, nudges have been defined as interventions based on insights from behavioral economics that “influence decisions while preserving freedom of choice” (2). The claim that nudges preserve freedom of choice plays a crucial role in the argument in favor of preferring nudges to interventions more restrictive of individual freedoms. Eyal implicitly attacks the following argument:

Premise 1: If two interventions can both efficaciously and effectively address a public health or health policy issue, the intervention least restrictive of personal liberties ought to be preferred (by PLRA).

Premise 2: Health nudges are efficacious and effective, and preserve freedom of choice.

Conclusion: Health nudges ought to be preferred to more restrictiveinterventions.

Eyal denies the conclusion because he rejects Premise 1.

1. According to him Premise 1 is vulnerable to several counterexamples. Let me rehearse only one of them. This counterexample discusses institutional solutions to the problem of the allocation of trainees to patients in university hospitals. As the success rate of trainees is lower than that of trained medical personnel, some patients will be predictably worse-off than they would have been had they been treated by trained personnel. He considers two solutions to this problem:

• An opt-out (nudge) approach: “The default is such that most patients will predictably forego ensuring that only trained personnel attends to them (partly because many patients and their families are too embarrassed to insist), but all maintain their freedom to receive services from trained personnel, because all would get them if they insisted enough” (1).

• An impartial (non-nudge) approach: Patients would be allocated either to trainees or trained personnel according to an impartial rule, such as random lottery. “Patients who insist on refusing them will have to move elsewhere for treatment and perhaps pay steep fees” (1).

A proponent of Premise 1 would conclude that the opt-out (nudge) approach is preferable to the impartial (non-nudge) approach because it is less restrictive of personal liberties. In contrast, according to Eyal, the impartial approach is preferable to the opt-out approach because at least another morally relevant consideration tips the balance the other way, namely, the propensity of an intervention to elicit shame (minimally understood as embarrassment).

Eyal argues that if A and B are two equally efficacious and effective interventions, and if A elicits unintended and undesirable shame, then B is preferable to A, although A might be less restrictive than B. His argument relies on two assumptions: the experience of shame is sometimes a morally undesirable outcome; the opt-out approach counts as an instance of undesirable shaming. Although both claims are disputable and hotly disputed (3), let us grant them for the sake of argument.

Eyal assumes that proponents of the mainstream version of PLRA would disagree with him and claim that A is preferable to B. Is that plausible assumption? Perhaps not: Premise 1 does not accurately characterize the mainstream version of PLRA. Rather, the mainstream version of PLRA claims that A and B need to be equal in all morally relevant respects for the principle to kick in. Almost nobody claims that the moral evaluation of public health and health policy is restricted to their impact on liberty interests (and effectiveness and efficacy). There is broad consensus that other morally relevant considerations are important, including social justice and respect for the dignity and integrity of persons. Therefore, Eyal’s argument successfully defeats Premise 1 but fails to question the mainstream version of PLRA.

To reflect the views of most public health ethicists (4–7), Premise 1 should be replaced by the following, more accurate, premise:

Premise 1’. If two interventions can both efficaciously and effectively address a public health or health policy issue and are equal in all other morally relevant respects, the intervention least restrictive of personal liberties ought to be preferred (by PLRA).

Not only Eyal’s counterexample does not defeat Premise 1’, but I suspect Eyal actually endorses Premise 1’ and is therefore in agreement with defenders of the mainstream version of PLRA. In contrast, my view is that there are good reasons for rejecting Premise 1’. I suggest two reasons in support of the radical view that the mainstream version of PLRA is mistaken.

The first reason is that if we take the “all things equal” clause seriously, Premise 1’ is of little help. For in the real world, there are almost no situations in which policy-makers have to decide between two interventions that are equal in all morally relevant respects but for their impact on liberty interests, their effectiveness and their efficacy. Eyal’s example of the choice between the opt-out and the impartial approach is a case in point. The opt-out approach favors extraverts and “healthcare system smart” patients who know that if they insist they can have it their way, and puts those who would feel ashamed of asking not to be treated by a trainee at a disadvantage. The impartial approach is egalitarian but perhaps a bit too rigid and might therefore elicit feelings of frustration. Each option affects differently patients’ level of trust in the quality and fairness of the healthcare system; and so on and so forth. PLRA is not problematic because it excludes all moral considerations but those centered on liberty interests, as Eyal assumes. It is problematic because it includes them all. It can therefore rarely assist us in evaluating policy options: it is rarely a deal-breaker.

But suppose partisans of the principle of PLRA could find a way around this objection. I would still argue that it should not be endorsed for a second, more fundamental, reason.

Let me expand on the mainstream version of PLRA. According to its proponents all public health and health policy measures (laws, regulations, policies) have to pass a rigorous test: they are impermissible unless public authorities are in a position to demonstrate they have selected (all things equal) the means least restrictive of personal liberties to achieve their objective (4–7). The mainstream version of PLRA is objectionable because it generalizes a principle of US constitutional law that was originally much more limited in scope—and for good moral and practical reasons. In US constitutional law, PLRA is meant to provide enhanced protection only to fundamental liberties and rights (6). In other words, it says that public authorities ought to be in a position to demonstrate that they have not infringed on fundamental liberties and rights in the pursuit of their public welfare objective without compelling reasons. Because running through all the predictable consequences of feasible alternatives is not an easy task and because the least restrictive alternative is not always the easiest or least costly to implement, the requirement does not and should not apply when the state interferes with nonfundamental liberties. Obviously, a lot relies on the possibility of distinguishing fundamental from nonfundamental liberties. But even if the borders of what counts as a fundamental liberty are fuzzy and debatable, there are clear-cut cases. For example, according to all major traditions in moral and political philosophy, freedom of movement is a fundamental liberty that should only be overridden in exceptional circumstances (e.g. those that standardly justify quarantine as a last resort); selling soft drinks in X-large cups is not a fundamental liberty by any stretch of the term. I discuss these points in greater detail elsewhere (7–9).

In sum, I concur with Eyal’s view that PLRA is questionable, but do not believe he has offered reasons for rejecting the mainstream version of this principle. In this comment, I have sketched such an argument. The hope is that our exchange will open up a much-needed debate over the moral foundations and the scope of application of PLRA.

Ethical issues

Not applicable.

Competing interests

Author declares that he has no competing interests.

Author’s contribution

YS is the single author of the manuscript.

Citation: Saghai Y. Radically questioning the principle of the least restrictive alternative: a reply to Nir Eyal’: Comment on “Nudging by Shaming, Shaming by Nudging”. Int J Health Policy Manag 2014; 3: 349–350. doi: 10.15171/ijhpm.2014.106