We might prefer the image of Mill as a robust, principled liberal, but that might not be his view.

Piers Norris Turner, 2014

Mill’s views are controversial, and require further analysis, clarification, and debate. This is as he would have wished. For him, no opinion is worthy of acceptance by thinking beings unless it can stand up to a free and critical scrutiny from all sides.

C.L. Ten, 2009

1 Introduction

The current debate on paternalism and behavioral economics has rekindled interest in the political and moral philosophy of John Stuart Mill (1806–1873), in particular, his works On Liberty (1859) and Utilitarianism (1861). While Mill is known to have defended liberty against all kinds of intrusions, a careful reading invites a more nuanced interpretation that suggests he supported paternalistic interventions in certain contexts (Dworkin 1971; Young 2009). Mill, in fact, curbed his Liberty Principle with many exceptions that pave the way for interference with individual actions leading some commentators to describe him as an elastic paternalist (Cohen-Almagor 2012) or that, in his work, weak paternalism is defensible (Brink 1992, p. 91).

Like other great philosophers and economists, a selective reading of his work can offer simultaneous support to two (or more) opposing normative points of views.Footnote 1 There is, however, according to Hamburger (1999), a conventional interpretation of Mill that avoids or dismisses discussions of crucial passages in which he advocates interference with self-regarding conduct. Our view is that this inaccurate interpretation of Mill (as a robust anti-paternalist) has framed the debate on the policy implications of behavioral economics.

We can categorize the contributors to the policy implications of behavioral economics as either coercive paternalists (Conly 2013), soft paternalists (Sunstein 2014), or anti-paternalists (Sugden 2018). Yet, while the coercive and soft paternalists agree that their stance cannot be reconciled with Mill, anti-paternalists argue the opposite, namely, that their views mesh with his defense of liberty. The position taken in this paper is that both miss important nuances in Mill’s texts, nuances that undermine how they think their own ideas should relate to his.

Dworkin’s (1997, p. 62) often cited definition of paternalism—as an “interference with a person’s liberty of action justified by reasons referring exclusively to the welfare, good, happiness, needs, interests or values of the person being coerced”—may be slightly modified to accommodate the policy implications of behavioral economics. Le Grand (2020, p. 2; emphasis added) thus defines a “state intervention as paternalistic with respect to an individual if it is intended to address a failure of judgment by that individual and if it is intended to further that individual’s own good”. This definition should be further modified to accommodate the so-called “soft paternalisms” arising from behavioral economics. These new paternalisms, as Dworkin (2020) labels them, are arguably of a different sort in so far as they are not constrained to judgment failures (and may include the provision of missing information; Hausman 2018, p. 61). They also preserve freedom by imposing little or no cost on choosers. Sunstein (2014, p. 56–57) offers a specimen when he argues that.

“Some varieties of paternalism are highly aggressive, or “hard,” while others are “soft.” Soft paternalism is weaker and essentially libertarian, in the crucial sense that it preserves freedom of choice […] It might be most sensible to understand paternalistic interventions in terms of a continuum from hardest to softest, with the points marked in accordance with the magnitudes of the costs (of whatever kind) imposed on choosers by choice architects […] We should agree that there is a big difference between, say, a severe criminal ban on smoking marijuana and a nominal fine, and between a prison sentence for failing to buckle your seatbelt and a graphic educational warning offering vivid warnings. Under this approach, a statement that paternalism is “hard” would mean that choice architects are imposing large costs on choosers, whereas a statement that paternalism is “soft” would mean that the costs are small […] Nudges would count as soft paternalism as long as they impose no or very small costs on choosers. If their goal is to protect people from their own mistakes, they would run afoul of the Harm Principle—but as we shall see, they might nonetheless be justified”.

It should be noted that, in the existing taxonomy of paternalisms, soft paternalism has two circulating distinct definitions. Among economists and other social scientists, soft paternalistic interventions seek the prudential goals—welfare, well-being, or good—of an individual with policy instruments that do not impose high costs and thus do not cross out choices (Sunstein 2014). Sunstein’s understanding of “soft paternalism” is not an idiosyncrasy of his and others, before him, have used the term in this manner (Glaeser 2006; McQuillin and Sugden 2012; Qizilbash 2012). However, in the philosophical literature, following Feinberg (1971, p. 113), a distinction is made between soft (sometimes also weak) paternalism which permits interference with self-regarding acts if they are non-voluntary targeting children and non-competent adults and hard paternalism which allows interference with self-regarding voluntary acts targeting competent adults (Arneson 1997).Footnote 2 Dworkin (2020) further distinguishes between soft and weak paternalism pointing that the former is paternalism applied to non-voluntary acts while the latter is paternalism applied to interventions that target “the means agents choose to achieve their ends, if those means are likely to defeat those ends”. We prefer to call Dworkins’s “weak paternalism” just “means paternalism” to avoid a cleft between weak and soft which have been used interchangeably in the literature (e.g., Young 2009). Our principal concern is also not “ends paternalism” because the findings and policy interventions advocated by behavioral economists have not sought to modify the ends or goals of agents.Footnote 3 Although Dworkin (2020) questions the extent to which nudges should be considered paternalistic (in so far as many of them do not infringe on autonomy), he agrees that nudges mostly target means not ends. It should be clear, at any rate that, in what follows, soft paternalism involves prudential interventions targeting welfare, well-being, or the good of an individual without imposing high costs or restricting freedom of choice.

There are two considerations that render Mill a worthwhile interlocutor from the nineteenth century. First, Mill was aware that humans have psychological weaknesses, that is instances where they are not operating under full mental capacities. If so, his Liberty Principle should withstand the findings of behavioral economics. In other words, Mill derived his Liberty Principle with some awareness of human psychological limitations and added specific caveats that restrict its application when some psychological conditions are not satisfied. Second, the Liberty Principle allows interference, including aids to judgments, even when an act is potentially harmful to no one but the individual. Similar interventions, it turns out, are also advocated by the soft paternalists. Although Mill does not use the term paternalism in On Liberty,Footnote 4 the interventions which present-day soft paternalists derive from the findings of behavioral economics would not count as unjustified threats to freedom under his own Liberty Principle. These tenets form our working hypothesis in this paper, and our objective in what follows will be to provide sufficient evidence from Mill’s own writings on its plausibility.

Powers et al. (2012) pursue a similar objective applied to public health ethics. First, they argue Mill deserves a prominent place in current debates on public health ethics. Second, they reject the familiar “blanket presumption in favor of liberty” (Powers et al 2012, p. 13) arguing that, for Mill, not all liberties are equally important. In their interpretation, the Liberty Principle protects liberties essential to self-determination, an ingredient of individual well-being. Third, they contend that Mill’s Free Trade doctrine balances the interests of society with respect to economic efficiency and public welfare. Thus, while the Liberty Principle protects “major life-shaping choices … of deep moral significance” the Free Trade doctrine “takes maters of public welfare outside the domain of the Principle of Liberty and assumes that the centrally defining considerations to be balanced are economic efficiency and public welfare protection” (Powers et al 2012, p. 10). Their nuanced and more complex reading of Mill provides a liberal framework to public health policies where “irksome, inefficient, ill-advised” daily life interferences do not weaken individuals’ ability to make consequential life choices (such as whether and whom to marry, admissions to professions and careers without restriction based on gender, control of property and personal assets, control over one’s own body and mind, etc.). Such claims have substantial overlaps with our paper which focuses on how multiple issues raised in Mill’s texts On Liberty and Utilitarianism offer precursors to the current debate on the policy implications of behavioral economics.

The outline of the paper is as follows. Section 2 sketches how the novel alliance between behavioral economics and the new soft paternalists positions itself with respect to Mill’s defense of liberty. Section 3 provides our interpretation of On Liberty and the various caveats Mill implanted to constrain its reach. Section 4 explains how the Utility Principle—Mill’s first ethical principle—rules over the Liberty Principle disabling the tension between individual liberty and paternalism. In Sect. 5, I review Mill’s discussion of the Liberty Principle and the Free Trade doctrine to further evaluate the ramifications of his likely advocacy of soft paternalistic interventions. Section 6 concludes.

2 Behavioral economics and the new paternalisms

The argument we shall pursue is that Mill is misclassified as an unrepentant opponent of soft (and coercive) paternalism by those involved in debating the policy implications of behavioral economics. Mill is, thus, described as “the most significant opponent of paternalism” (Conly 2013, p. 18, 47), a libertarian with impeccable credentials and a source of inspiration for other libertarians (Rebonato 2012, p. 34, 93), or that “contrary to Mill’s classic statement on the illegitimacy of government paternalistic intervention to promote an individual’s own good, there are cases where paternalistic interventions are justified” (Le Grand 2020, p. 1). Hausman (2018, p. 55), more subtly, notes that “Mill is not concerned about whether there might be paternalistic actions that do not limit liberty”. Because it is concerned with the tyranny of the majority, Mill’s essay On Liberty (1859/2008) is a major source for such claims in that it aims to erect general principles to protect individuality from the collective mediocrity of the masses (On Liberty, p. 73). More generally, the view that Mill opposes paternalism in his Liberty Principle is adopted by philosophers (Arneson 1980; Hausman 2012, p. 80) as well as those who seek a bridge between contractarian political philosophy and behavioral economics (Sugden 2018).

We may consider two known contributors who agree that Mill’s Liberty Principle is anti-paternalist yet derive radically different policy implications from behavioral economics, namely, Cass Sunstein and Robert Sugden. For Sunstein (2013, 2014), Mill’s Harm Principle (which we have referred to as the Liberty Principle) that interference by the State is only justified if there is harm to others points in the wrong direction. While Sunstein (2014, p. 143) came close to agreeing with us that “soft paternalism might be compatible with the Harm Principle” (p. 19), he considers that Mill’s principle generates problems for the soft paternalism he defends and his aim is to cast doubt on it (p. 163). Sunstein argues that Mill would reject soft paternalism and his objective is to “challenge the Harm Principle on the ground that in certain contexts people are prone to error and paternalistic interventions would make their lives go better” (Sunstein 2014, p. 4–5). While Sunstein (2014, p. 4) agrees with Mill that “individuals often know best”, that “adults should not be treated like children”, or that he “still has a great deal to teach us”, however, he argues that “since his time we have learned a lot that he [Mill] did not and could not know, especially about human error and what we have learned cuts at some of the foundations of the Harm Principle”. For Sugden (2018, p. 3), likewise, “it is integral to Mill’s concept of ‘advantage’ that each person is the ultimate judge of what counts as advantage to him or her. His On Liberty is a passionate defense of individual liberty against paternalism”. There is no escape from the ambiguity and complexity of Mill’s writings on the matter and our objective is to clarify by adding nuance to this current agreement between intellectual opponents. The objective of the present paper is accordingly to provide guidance on Mill’s two works and their implications for policies derived from the findings of behavioral economics.

Do the findings of behavioral economics justify paternalism? Behavioral policy interventions use the findings of behavioral economics—mostly, but not only, the biases and heuristics school of Tversky and Kahneman (1974) and Kahneman & Tversky (2000)—to coerce, aid, guide, nudge, or influence the behavior of either individuals or groups towards a set of goals which may, but need not, be their own ends. Paternalistic behavioral interventions, specifically, use the findings of behavioral economics to improve the well-being of individuals (Le Grand and New 2015). Behavioral policy interventions that target an externality (pollution reduction) or the financial strength of a country (aggregate savings) are not part of our subject matter since they should not be deemed paternalistic. Finally, soft paternalistic interventions encompass libertarian paternalism (Sunstein and Thaler 2003), asymmetric paternalism (Camerer et al 2003), behavioral welfare economics (Bernheim and Rangel 2007, 2009), and light paternalism (Loewenstein and Haisley 2008). Although the interventions advocated by the soft paternalists tend to be associated with judgment failures, they are paternalistic because they promote individual prudential goals (Hausman 2018) and soft because they impose little to no cost on choice (Sunstein 2014).

While Mill claimed utility is the highest moral principle, a view to be elaborated on below, he also valued liberty as one of its most important components.Footnote 5 The parallels to Sunstein and Thaler ‘s (2003, p. 3) libertarian paternalism are apparent since its rules are chosen “with the explicit goal of improving the welfare of the people affected by them” while maintaining their freedom to opt out and seek alternative arrangements. This most popular approach to soft paternalism, and certainly the one which has had the most significant policy impact, is described by Sunstein and Thaler (2003) as a weak, soft, and non-intrusive type of paternalism because choices are not blocked and individuals are free to opt out. Thaler and Sunstein (2008, p. 7) start from the premise that the anti-paternalist assumes people are good decision-makers, they are more like homo economicus and can “think like Einstein, store as much information as IBM’s Big Blue, and exercise the willpower of Mahatma Gandhi”. However, Thaler and Sunstein consider most people are not like that—they are homo sapiens—since they “make pretty bad decisions—decisions they would not have made if they had full attention and possessed complete information, unlimited cognitive abilities, and complete self-control” (Thaler and Sunstein 2008, p. 5–6). In their bestseller Nudge, Thaler and Sunstein (2008, p. 5) contend that libertarian paternalism either “maintains or increases freedom of choice”; that it does not “burden those who want to exercise their freedom” (Thaler and Sunstein 2008, p. 5); and that “it tries to influence choices in a way that will make choosers better-off, as judged by themselves” (Thaler and Sunstein 2008, p. 5, emphasis in original).Footnote 6

This last paragraph raises the following issues which must be commented on. The first is to guard against assuming that all those working on the policy implications of behavioral economics interpret the findings as mistakes in decision-making. Sugden (2008, 2018), for example, is a contractarian who rejects paternalism as a necessary policy offshoot of behavioral economics.Footnote 7 The second issue relates to our principal argument in this paper, namely, that Mill would have approved at least a part of the soft paternalist prescriptions because of its emphasis on the preservation or even promotion of choice alongside its claim that its policies aim to improve the welfare of individuals as judged by themselves (Ten 2009, p. 9). The third issue forestalls a potential critique, namely, that there is a fundamental tension between Mill and the soft paternalists over the importance of deliberation and practicing choice. According to Brink (1992, p. 79), Mill’s conception of happiness “consist in large part in the exercise of those higher capacities that distinguish us from other animals. Our higher capacities include our rational capacities, especially our capacities for practical deliberation”. Mill believed that the human faculties of choice need to be exercised and, like all faculties, when they are not, they will wither away as individuals settle on customs (On Liberty, p. 77–82). But some policy tools advocated by the soft paternalists by-pass the exercise of choice, specifically, automatic or “non-educative” nudges such as defaults (Sunstein 2018; Thaler and Sunstein 2008). This will not, however, undermine our argument. First, because individuals can lack experience or, if they are already overburdened with decisions, they may wish to delegate at least part of their choices to those in the know including waiters (in restaurants), bankers, doctors, and engineers (Sunstein 2015, 2017). Second, the range of policy tools advocated by the soft paternalists is quite diverse and not all are automatic—some are considered self-consciously educative because they provide boosts that harness decision-making competences (Hertwig 2017; Sunstein 2014, 2016).Footnote 8 The recognition that not all interventions advocated by the soft paternalists by-pass deliberation and active choosing, and that some of them reinforce it, should suffice for now to render the arguments by soft paternalists potentially interesting to someone who approves Mill’s defense of liberty, in particular, his claim on the importance of practicing choice for individual development and personal growth.

3 How anti-paternalist is Mill’s Liberty Principle?

Mill’s two essays of concern us—On Liberty (1859/2008) and Utilitarianism (1861/2008)—were published only 2 years apart but were most likely both conceived between 1854 and 6 (Gray 1991). While these works have been considered irreconcilable (Berlin 1959/1991; Conly 2013; Gray 1989; Himmelfarb 1973; Lyons 1997), Mill insists that utility should be the “ultimate appeal on all ethical questions” (On Liberty, p. 15).Footnote 9

In this section I first lay out elements of Mill’s On Liberty to subsequently tease out five exceptions he identified. In On Liberty Mill argues that the tyranny of the majority, which tends to be expressed in the organs of representative government, is now an evil society should protect itself from (On Liberty, p. 8). His focus is on “the nature and limits of the power which can be legitimately exercised by society over the individual” (On Liberty, p. 5) and claims that unless the State is dealing with children or incapacitated adults,Footnote 10 the only interventions that should be allowed are towards acts that harm others. The State should not ban acts that cause self–harm because such acts are necessary to the formation of spontaneous acts and individuality both of which are necessary for human development. In fact, any intervention that interferes with acts that only affect the good of the individual could backfire and generate resistance (On Liberty, p.92). His most famous statement in defense of individual liberty is the Liberty Principle:

“That [simple] principle is, that the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number, is self-protection. 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 opinion of others to do so would be wise, or even right” (On Liberty, p.14).

Later in the text he restates it: “When, by conduct of this sort, a person is led to violate a distinct and assignable obligation to any other person or persons, the case is taken out of the self-regarding class, and becomes amenable to moral disapprobation” (On Liberty, p. 90) or, perhaps more broadly, “[w]henever, in short, there is a definite damage, or definite risk, either to an individual or to the public, the case is taken out of the province of liberty, and placed in that of morality and law” (On Liberty, p. 91). Accordingly, if an act harms others without their consent, it is no longer self-regarding and “becomes amenable to moral disapprobation” (On Liberty, p. 90). For a society to be considered free it should, nevertheless, not interfere with individuals’ “inward domain of consciousness”, their “liberty of tastes and pursuits; of framing the plan of our life to suit our character”, and the “freedom [of combination] to unite, for any purpose not involving harm to others” (On Liberty, p. 16). When there is no harm to others, individuals’ liberty to pursue and develop their own character, taste, and plans of life—even if these are considered foolish, perverse, or wrong by society—should be promoted without impediments (On Liberty, p. 17).

While Mill acknowledged that most of our acts affect others,Footnote 11 what did he mean by “harm to others”? There is admittedly multiple statements in his text on what harm involves. Consequently, multiple interpretations have emerged. In the narrowest, harm entails a violation of rights (Brink 1992; Donner 2009) or interests (which cover, but are not limited, to constitutional rights; Gray 1989; Rees 1991).Footnote 12 In the broadest, harm is perceptible damage (albeit excluding mere dislike and emotional distress; Riley 1991, pp. 22–23, 2009) or a catchall term for bad consequences (including causing emotional distress in others; Turner 2014). Given the utilitarian underpinnings of the Liberty Principle, ultimately, whether harm to others without their consent should be reprimanded is contingent on contrasting aggregate net social utility that would arise from the distinct available courses of action.

It should be noted that Mill is not just concerned with negative freedom, or freedom from interference, since he also considers the Liberty Principle necessary for human development.Footnote 13 The title of Sect. 3 of On Liberty “Of Individuality, as one of the elements of well-being” is suggestive. His argument is that individualityFootnote 14 is an important constituent of not just personal well-being but also of social progress which comes to a halt without individuality (On Liberty, p. 63, 79). For Mill, what explains European progress is not intrinsic to Europe but its tolerance and promotion of diversity in culture and character which have nurtured its many sided development (On Liberty, p. 80). Humans are not machines but more like “a tree which requires to grow and develop itself on all sides according to the tendency of the inward forces which make it a living thing” (On Liberty, p. 66). Merely following customs blunts the development of human faculties such as perception, judgment, mental activity, and moral preference which are trained with active choice and necessary for discerning what is best for the individual (On Liberty, p. 65). While he contends the forces of individuality and spontaneity used to be excessive, the opposite is true at the time of his writing as a dearth of personal impulses and eccentricity form one of the most significant dangers of his time (On Liberty, p. 68, 75). To counter such trends, society ought to cultivate diversity, countering intolerance against eccentricity and individuality (On Liberty, p. 77). Thus, for Mill, freedom to pursue one’s own inclinations and the ensuing variety of situations which results are two necessary conditions for human development.

His strongest argument for non-interference with self-regarding acts is that the knowledge each individual possesses of her concerns cannot be matched by anyone else (On Liberty, p. 85).Footnote 15 Mill claims any external interference in the personal, self-regarding, sphere of thought and action is more likely to target the wrong person, at the wrong time, and the wrong place (On Liberty, p. 92).Footnote 16 Mill, nevertheless, considers that when harm to others arising from individual action is involved, that is on questions of social morality, the “opinion of the public” should be acted on “because on such questions they [the public] are only required to judge of their own interests; of the manner in which some mode of conduct, if allowed to be practiced, would affect themselves” (On Liberty, pp. 92–93). If public opinion or State sponsored policies cannot match a person’s capacity at identifying what is in her best interest, in the sphere of duties owed to others, though carrying a high risk of being wrong, such interventions are perfectly legitimate.

3.1 Exceptions to the Liberty Principle

As Mill defines and sets the contours of the Liberty Principle, he also lists exceptions to its application. First, he recalls that the good of the individual,

“either physical or moral, is not a sufficient warrant [for interference]. He cannot rightfully be compelled to do or forebear because it will be better for him to do so, because it will make him happier, because in the opinion of others to do so would be wise or even right (On Liberty, p.14).

So far so good. But then he adds,

“there are good reasons for remonstrating with him, or reasoning with him, or persuading him, or entreating him, but not for compelling him or visiting him with an evil in case he do otherwise” (On Liberty, p.14).

In a subsequent passage he claims

“considerations to aid his judgment, exhortations to strengthen his will may be offered to him, even obtruded on him, by others; but he himself is the final judge. All errors which he is likely to commit against advice and warning are far outweighed by the evil of allowing others to constrain him to what they deem good” (On Liberty, p.85, emphasis added).

And towards the end of the essay he recalls that “advice, instruction, persuasion, and avoidance by other people” (On Liberty, p. 104) are the only measures that can be used to express disapprobation. These passages are important because they show how Mill is willing to allow some interference as long as the final decision remains up to the individual. He is also arguing against compulsions (or bans) but not other interventions such as remonstrating, persuasion, reasoning, aid in judgment etc. Mill, like the soft paternalists, is here advocating for policies that are choice preserving. Such passages provide evidence that Mill, before the dawn of behavioral economics and soft paternalism, explored similar ideas. Indeed, the measures Mill proposes do not dispute some of their current policy prescriptions. Thus, in so far as they preserve freedom of choice, the interventions of soft paternalists accord with the measures Mill allows in concomitance to his Liberty Principle.

Second, Mill argues that the good of one person is often connected to the good of others since

“no person is an entirely isolated being [...] If he injures his property, he does harm to those who directly or indirectly derived support from it, and usually diminishes, by a greater or less amount, the general resources of the community. If he deteriorates his bodily or mental faculties, he not only brings evil upon all who depended upon him for any portion of their happiness, but disqualifies himself for rendering the services which he owes to his fellow creatures generally, perhaps becomes a burden on their affection or benevolence” (On Liberty, p. 88–89).

Mill concludes that if intemperance or extravagance in individual behavior generate negative spillovers on the well-being of others (ability to pay debt, providing for the family), the individual may not just be reprobated but should also be punished (On Liberty, p. 89). Such constraints limit the application of the Liberty Principle not because of the extravagance per se but for the breach of moral duties to others.Footnote 17

The third caveat to the Liberty Principle is that it is not applicable in societies that have not developed channels for free and equal discussion to have an impact on the world thus ruling out its application in societies which cannot self-improve through persuasion (On Liberty, p. 15). Mill further restricts the application of the Liberty Principle to adults with fully developed mental capacities. If an individual is under “some state of excitement or absorption incompatible with the full use of reflective faculty” (On Liberty, p. 107) she should not just be warned of the danger but the Liberty Principle is no longer applicable.

Behavioral economics has expanded our understanding of the number of ways in which decisions are mediated by heuristics and emotions. For example, Ariely and Loewenstein (2006) have shown how individuals under the heat of the moment, specifically if sexually aroused, will take risks they would not have taken otherwise (the so-called hot–cold empathy gap). Mill would have approved restricting the Liberty Principle in similar situations where adults, under the heat of the moment, have their full use of reflective faculties compromised. Again, such discoveries may further constrain the range of applicability of the Liberty Principle.

Related to the third caveat, Conly (2013, p. 55) argues that Mill considers the number of “occasions upon which people need to be saved are relatively few: left to their own devices, they will typically choose for themselves that course which is best for them given their particular desires”. Mill does argue that no one “is warranted in saying to another human creature of ripe years, that he shall not do with his life for his own benefit what he chooses to do with it. He is the person most interested in his own well-being” (On Liberty, p. 84). Here, however, Mill is not affirming that he thinks individuals will succeed in serving their own best interest but instead that they are the ones most interested in realizing what is in their own best interest. Whether individuals, left to their own device, are the most successful in achieving what is in their best interest is not elaborated on. This point is restated in the Principles of Political Economy (PPE 1848/2008) where Mill argues that while the consumer may be the best judge of the commodity to be consumed, this is not universally true and there are numerous exceptions (PPE, p. 338). Specifically, Mill thought that even if it is possible to keep someone out of harm’s way or on a good path, this would take away from her the chance to gain experience in the faculties of observation, reasoning and judgment to foresee, discrimination to decide, firmness in self-control etc. What matters is not the effectiveness with which ends are achieved, but the process and agency used to attain them (On Liberty, p. 66). There is, at any rate, textual support that Mill did not consider individuals very effective in pursuing their own well-being. For example, he claims that across all social classes, people conflate what is in their interest with the customary (On Liberty, p. 68). Specifically, he argues, “I do not mean that they choose what is customary, in preference to what suits their own inclinations. [But] it does not occur to them to have any inclinations, except what is customary” (On Liberty, p. 68). The claim that most people will pursue what is good for them if given the freedom to do so is also difficult to reconcile with Mill’s numerous statements on the moderate intellect, or to use a modern term, the bounded rationality, of the average of mankind (On Liberty, p. 73–4, 77).Footnote 18

The fourth caveat allows interference with the Liberty Principle when an individual is likely to be harmed by her own acts yet if she knew her act would cause harm she would have taken another course of action. This goes some way in supporting the view that Mill favors an informed desire stance on well-being (Crisp 1997, p. 29; Qizilbash 2012; West 2009). This hypothesis is further corroborated in Utilitarianism where he claims pleasures of the intellect are qualitatively superior to others. In a famous passage in On Liberty, Mill illustrates with an example which Cohen-Almagor (2012, p. 574) describes as consistent with soft paternalism:

“if either a public officer or anyone else saw a person attempting to cross a bridge which had been ascertained to be unsafe, and there were no time to warn him of his danger, they might seize him and turn him back, without any real infringement of his liberty; for liberty consists in doing what one desires, and he does not desire to fall into the river” (On Liberty, pp. 106–7).

In this hypothetical scenario Mill is assuming imminent danger to a potential bridge crosser and that, if there is not enough time to warn him, anyone present can seize him. While Mill assumes most people do not wish to fall in the river, according to one interpretation, the individual should not be seized if there is enough time to warn him and an adult with full mental capacities must be allowed to attempt and cross if he so insists (Feinberg 1971, p. 112). In the latter case, said individual has his own reason to cross the unsafe bridge which no one can scrutinize better than himself and, if no harm to others is involved, he should be allowed to cross. If, however, there is only risk of self-harm to said person, if authorities, say, are not sure whether the bridge is unsafe, Mill claims the walker should not be seized and may, at most, be warned (On Liberty, p. 107).Footnote 19 I shall return to the bridge example below when the interaction between the Liberty Principle and the Free Trade doctrine is discussed. It should be said, at any rate, that Mill uses the bridge example to recommend warnings such as labels on potentially dangerous drugs. Such labels, Mill insists, do not infringe on liberty as they help people achieve their goals in so far as no one wishes to ignore, or be misinformed about, their potential risks.

The provision of information such as warnings is part and parcel of the soft paternalist toolbox. Sunstein (2016, p. 26, 2017) lists the following policy instruments available to the soft paternalist: simplifications, disclosure of factual information, warnings, and reminders. Mill’s informed desires view is also aligned with libertarian paternalism which presumes interventions ought to “make the choosers better off as judged by themselves” (Thaler and Sunstein 2008, p. 5, emphasis in original). Given that most bridge crossers do not wish to fall in the river, if time allows, soft paternalistic interventions in the form of warnings are justified and ought to make potential bridge crossers better off as judged by themselves.

The fifth and final constraint we consider on the Liberty Principle appears in the final chapter of On Liberty, where Mill argues nonreversible self-regarding acts that constrain one’s own future freedom—such as selling oneself into slavery—may be interfered with:

“the ground for thus limiting his power of voluntarily disposing of his own lot in life, is apparent, and is very clearly seen in this extreme case. The reason for not interfering, unless for the sake of others, with a person’s voluntary acts, is considerations of his liberty. His voluntary choice is evidence that what he so chooses is desirable, or at least endurable, to him, and his good is on the whole best provided for by allowing him to take his own mean of pursuing it. But by selling himself for a slave, he abdicates his liberty; he forgoes any future use of it beyond that single act. He therefore defeats, in his own case, the very purpose which is the justification of allowing him to dispose of himself. He is no longer free; […] The principle of freedom cannot require that he should be free not to be free. It is not freedom, to be allowed to alienate his freedom” (On Liberty, p.113-114).

This passage has been interpreted to offer grounds for a wide range of paternalisms (Dworkin 1997, p. 73). Indeed, like selling oneself into slavery, suicide and substance abuse—which could lead to irreversible destructive psychological and physiological effects making “it impossible to continue to make reasoned choices in the future” (Dworkin 1997, p. 78)—can justify interfering with one’s self-regarding actions in so far as they wipe out, or critically undermine, one’s future freedom.

We listed and discussed five constraints on Mill’s Liberty Principle. These are, to recap, (1) the permission to aid in judgment, to remonstrate, to reason with, and to persuade an individual in what only concerns himself; (2) reprobation and punishments of individual acts, including intemperance and extravagance, that spillover and harm others who depend on the individual (for example, next of keen or members of the community); (3) the Liberty Principle is inapplicable in societies where free and equal discussion and dissuasion have not developed or for individuals who are not in possession of their full reflective capabilities; (4) interference with self-regarding acts is allowed, including seizing someone, when it is suspected that the individual is not fully informed about the consequences of his actions and there is not enough time to warn him; (5) acts that diminish one’s future freedom, including selling oneself to a slaveholder, can be interfered with. Together, these constraints entail the Liberty Principle is not an exceptionless rule. Furthermore, they are sensible constraints if the Liberty Principle is a secondary ethical principle guided by the higher-level ethical principle of utilitarianism.

4 Utility as a first ethical principle

Among other commentators, Conly (2013, p. 49) finds that Mill’s Liberty Principle is at odds with his utilitarian philosophy because if there are acts that undercut one’s own happiness they should be restricted instead of banning the restriction as Mill commends. In this section we shall see how On Liberty and Utilitarianism can be reconciled. Although published after On Liberty, Utilitarianism provides foundations for and underpins his other works, including On Liberty. The connection between On Liberty and Utilitarianism, specifically, operates through Mill’s contention that there are first and secondary ethical principles: “Whatever we adopt as the fundamental principle of morality, we require subordinate principles to apply it by” (Utilitarianism, p. 157). The first principle he vouches for is, of course, the utilitarian doctrine that one’s action should produce the greatest happiness overall. Happiness is the only desirable thing as an end (Utilitarianism, p. 168), “the sole end of human action, and the promotion of it the test by which to judge all human conduct” (Utilitarianism, p. 172). Liberty, like justice and virtue, is a valuable ingredient to happiness, it is part of that end and not merely instrumental to it (Utilitarianism, p. 170). But nothing can be desirable in itself without also being an ingredient of happiness: “there is in reality nothing desired except happiness. Whatever is desired otherwise than as a means to some end beyond itself, and ultimately to happiness, is desired as itself a part of happiness, and is not desired for itself until it has become so” (Utilitarianism, p. 172, emphasis added). The first principle of utility, accordingly, supports secondary principles which include customary morality where the cumulative wisdom of humanity lies. Secondary ethical principles encompass the prevailing principles of justice, the Liberty Principle, the Free Trade doctrine, among others.Footnote 20

Mill considers that utility, or happiness, is made of pleasures and that there are different grades of pleasures—there is discontinuity between lower and higher ones. In Chapter 2 of Utilitarianism, Mill introduces this distinction noting that “utilitarian writers in general have placed the superiority of mental over bodily pleasures chiefly in the greater permanency, safety, uncostliness, etc. of the former—that is in their circumstantial advantages rather than their intrinsic nature” (Utilitarianism, p. 138). But for Mill, some pleasures, specifically pleasures of the mind or the higher faculties, are qualitatively superior to pleasures of the body. Mill argues any person who experienced both pleasures will tell them apart.Footnote 21 He lists multiple explanations for these qualitative differences though in the end he claims dignity is the most appealing distinguishing feature. Mill still admits that, while someone acquainted with both pleasures may at times choose the lower one, he does so because of men’s “infirmity of character, [which makes] their election for the nearer good, though they know it is to be the less valuable” (Utilitarianism, pp. 140–1).

Something must be said here on Mill’s critique of customs in On Liberty and his contention that following customary morality contributes to the overall good in Utilitarianism. As Donner (2009, pp. 154–5) explains, Mill “draws the general distinction between relying upon the wisdom of accumulated human experience, which he lauds, and habitually and uncritically conforming to custom that is stultifying and impedes cultural improvement, which he condemns”. Mill argued that ignoring the moral customs entails humanity has not learned and will never learn any general “conclusions from its experience of human life” (Utilitarianism, p. 157). Secondary ethical principles, as articulated in customary morality, can contain the wisdom of experience and should not be ignored—there is no need to start writing the rules of proper moral conduct from a blank slate. But Mill also acknowledged that “there exists no moral system under which there do not arise unequivocal cases of conflicting obligations” (Utilitarianism, p. 158). Thus, two secondary principles may offer conflicting guidance over proper conduct and–when they do—the (first) principle of utility is invoked to decide between them. In fact, for Mill, the wide adoption of a secondary ethical principle as a moral custom is neither sufficient nor necessary for its acceptance. As Wolff (2015, p. 127) aptly surmises, “Mill’s pretended adherence to “one simple principle” does not reflect how complicated his beliefs really are”. A finer distinction then must be made between customary morality’s utility promoting secondary principles (such as justice and liberty) from customary morality’s harmful secondary principles (such as patriarchy and racism).Footnote 22

Following our discussion of Utilitarianism we conjecture that for Mill “harm to others” is not a sufficient condition for interference. In other words, even if harm to others is involved interfering with self-regarding acts and thoughts may still not be justified. To see why consider a race or competition where being first “harms” the other competitors yet there is no justification to interfere or modify the rankings because it is not in the general interest to do so. Specifically, Mill distinguishes acts harmful to others because they are ruled out by customary morality (cheating) and which may be interfered with and acts which are harmful to others which do not interfere with customary morality (winning in a competition) and which do not justify interference on the grounds that general happiness is best served by non-interfering. This last distinction clarifies how the Utility Principle supports the Liberty Principle in so far as interference is ruled out even when there is harm to others since undoing the harm would undermine general happiness.

The last two sections have reviewed major themes in Mill’s aforementioned two works with implications for his stance on paternalism. Interpreting Mill’s On Liberty as an anti-paternalistic tract obviates a series of weaknesses and inconsistencies (Young 2009). If, instead, we read On Liberty as providing a defense of liberty while at the same time advocating a utilitarian consequentialist philosophy, then many of these alleged weaknesses and inconsistencies vanish. I have, accordingly, argued that Utilitarianism underpins many of Mill’s claims in On Liberty where the Liberty Principle is a secondary ethical principle subordinate to the first Utility Principle. I have also argued that Mill identifies various exceptions to the Liberty Principle that constrain the range of cases over which it should be applied. The previous section, in particular, has identified five claims in Mill’s On Liberty that may be used to advocate soft, and perhaps even coercive, paternalistic policies, among them, interference that does not eliminate choice (reasoning, advice, remonstrating, persuasion, warnings etc.) for acts that harm no one but the person involved as well as bans for individuals that cannot use their full reflective faculties or individuals about to commit irreversible acts that undermine their freedom in the future.

5 The Free Trade doctrine and the Liberty Principle

The previous sections have shown that isolating On Liberty from Utilitarianism generates interpretational problems for the Liberty Principle, specifically, its consistency with other claims Mill made. The ensuing embedding of the Liberty Principle in Mill’s (indirect or act) utilitarianism also undermines the currently circulating simplistic narrative that he would have objected to all forms of paternalism. In this section, I evaluate how Mill’s Liberty Principle meshes with his defense of, and advocacy for, free trade.

A connection, in fact, was recently made between Mill’s defense of free trade and the Liberty Principle by Sugden (2018, p. 3) who suggests that a core tenet in Mill’s liberalism is that “cooperation for mutual benefit is a governing principle of social life”. Thus, as he quotes the Liberty Principle, Sugden (2018, p. 3) explains that “it is integral to Mill’s concept of ‘advantage’ that each person is the ultimate judge of what counts as advantage to him or her… [N]o one is accountable to others for decisions that only affect only himself”. In the last chapter of On Liberty Mill discusses this seemingly evident connection between the Free Trade doctrine and the Liberty Principle. However, as we shall see, it generates complex, if not perplexing, policy proposals.

Trade for Mill is fundamentally a social act that carries concomitant obligations. He considers that the Free Trade doctrine “rests on grounds different from, though equally solid with, the principle of individual liberty” (On Liberty, p. 105). The doctrine of Free Trade states that “producers and sellers [are] perfectly free, under the sole check of equal freedom to buyers for supplying themselves elsewhere” (On Liberty, p. 105). Mill thinks so because restrictions on trade

“affect only that part of conduct which society is competent to restrain, and are wrong solely because they do not really produce the results which it is desired to produce by them. As the principle of individual liberty is not involved in the doctrine of Free Trade, so neither is it in most of the questions which arise respecting the limits of the doctrine as for example what amount of public control is admissible for the prevention of fraud by adulteration; how far sanitary regulations or arrangements to protect workpeople employed in dangerous occupations, should be enforced on employers” (On Liberty p.105-6).

For Mill then there are two distinct spheres, the social which delineates the scope of free trade and the individual which delineates the scope of personal liberty. Sugden (2010) expresses some surprise as to why a component of the Liberty Principle, specifically, the liberty of combination,Footnote 23 does not cover free trade. However, Sugden (2010) also contends that the Liberty Principle, for Mill, regulates the consumption side of markets while the Free Trade doctrine, which operates in the social sphere, regulates the supply side of markets. Accordingly, Mill contends,

“there are questions relating to interference with trade which are essentially questions of liberty … the prohibition of the importation of opium from China; the restriction on the sale of poison—all cases in short, where the object of the interference is to make it impossible or difficult to obtain a particular commodity. These interferences are objectionable not as infringements on liberty of the producer or seller but on that of the buyer” (On Liberty, p. 106).

If we accept Sugden’s interpretation, the production and supply of commodities pertains to the social sphere where free trade is the guiding doctrine. Furthermore, if no harm to others arises, the Liberty Principle (operating on the self) shields individual consumption from external interference. While the individual and social spheres are regulated by distinct secondary principles, they both fall under the higher order Utility Principle. The Free Trade doctrine, then, is not an autonomous principle but, like the Liberty Principle, it is subordinate to the overall good that is the Utility Principle. According to Mill, the market as a social institution self-regulates, and its participants may decide to constrain their own freedom as they do when they set standards for countering fraud and establishing hygiene and quality benchmarks. Such constraints, of course, will be justified if they are conducive to the overall good.

The distinctions between the social and the individual, on the one hand, and the Liberty Principle and the Free Trade doctrine, on the other hand, are used by Mill on boundary cases where the consumption of commodities contains both social and individual elements operating under different, and often times conflicting, principles. Mill examines how such boundary cases justify restrictions on trade which in turn interfere with individual liberty. These cases include potentially harmful consumer commodities (opium, alcohol, poison). He claims these restrictions are objectionable because they interfere with an individual’s liberty to purchase them. Nevertheless, he asks, if interference is justified (1) when a commodity such as poison is sold and used to perpetrate crime or (2) when, outside trade, there is a risk for certain actions leading to accidents. He explains: “If poisons were never bought or used for any purpose except the commission of murder, it would be right to prohibit their manufacture and sale” (On Liberty p.106). While this commendation operates on the supply side—it limits the Free Trade doctrine—it is also a restriction on individual freedom. However, if poison is also desired for useful and non-harmful purposes, such restrictions cannot be imposed in one case but not the other. Thus, when a commodity is potentially dangerous and may at the same time be employed for both beneficial and harmful purposes, public authority cannot bar its sale but should at least warn buyers about its consequences (On Liberty p. 106). Mill here is in favor of soft and liberty preserving interventions by the State to reduce the risk of harms resulting from the purchase of a commodity that can be used to commit crime or produce harm to self.

As Mill pursues his discussion of the freedom to buy and sell poison, he adds, “it is a proper office of public authority to guard against accidents” (On Liberty p. 106) and proceeds to the hypothetical example of a walker heading towards an unsafe bridge. We bring back this example because Mill draws parallels between the sale of poison and the risk of harm to self from accident. Mill thus uses the bridge case to shed light on which constraints should apply on the sale of poison. Recall how for the individual heading towards an unsafe bridge, if there is not enough time to warn him, anyone present should seize him and turn him back without infringing on his freedom. However, if there is only a risk of accident then,

“no one but the person himself can judge the sufficiency of the motive which may prompt him to incur the risk …. [and in this case] he ought to be only warned of the danger; not forcibly prevented from exposing himself to it. Similar considerations, applied to such a question as the sale of poisons, may enable us to decide which among the possible modes of regulation are or are not contrary to principle. Such a precaution, for example, as that of labelling a drug with some words expressive of its dangerous character may be enforced without violation of liberty; the buyer cannot wish not to know that the thing he possesses has poisonous qualities” (On Liberty p.107, emphasis added).

The parallels between sales of poison (to possibly commit crime) and action that can harm oneself generates the following policy implications. In combining the doctrine of Free Trade and the Liberty Principle, three possible interventions for the sale of dangerous goods emerge (and actions that may lead to self-harmful accidents). First, if poisons are only used to commit crime bans are justified (seizures are justified too if a passerby decides to cross an unsafe bridge and there is no time to warn him). Second, if the sold commodity (poison) can be used for both innocent and criminal purposes only a warning should be issued (similarly, if there is only risk of harm to self by crossing the bridge, a warning ought to be issued as well). Third, if the consumer is not aware or is mis-informed about the risks of a commodity, she should also be warned. Whether the sale of a commodity should be banned depends on what it is used for. If it is only used for harm (crime), it should be banned, otherwise, if it has multiple purposes not all of which cause harm, the policy maker can only issue a warning.Footnote 24

There are other instructive cases “which lie on the exact boundary line between two principles, and it is not at once apparent to which of the two it properly belongs” (On Liberty, p. 110). In such cases the separation between the social and the individual is not without ambiguity and Mill uncovers valid arguments in support of both. He first considers giving advice or offering inducements as social acts subject to social control in so far as they affect others. But “if people must be allowed, in whatever concerns only themselves, to act as seems best to themselves, at their own peril, they must be equally free to consult with one another about what is fit to be done; to give opinions and receive suggestions” (On Liberty p. 109). Yet giving advice becomes ethically and legally questionable when someone derives benefits from it or promotes ends the State considers evil (On Liberty p. 109). In this case, we have a group of people whose interests undermine the common good. Mill then asks “ought this be interfered with or not? … [and] should a person be free to be a pimp or to keep a gambling house?” (On Liberty pp. 109–110). In such boundary cases Mill is at pain to determine unambiguously on which side they fall or whether the State should regulate them.

On the one hand, society cannot interfere with what concerns individuals who should be free to dissuade one another. On the other hand, society “cannot be acting wrongly in endeavoring to exclude the influence of solicitations which are not disinterested, of instigators who cannot possibly be impartial—who have a direct personal interest on one side, and that side [is] the one which the state believes to be wrong” (On Liberty, p. 110). Mill thus argues that people should be free to gamble among themselves or as members of a group which may even receive visitors. However, he also claims, “public gambling houses should not be permitted” (On Liberty, p. 110) even though he is skeptical that bans work. Mill’s final thought on the matter is neither here nor there: Gambling houses should be allowed to operate but under a shroud of secrecy and mystery so that society knows as little about them as possible.

Mill’s prescriptions on gambling might not be so unpalatable as gambling policy in Great Britain displays similar features.Footnote 25 Munting’s (1993, p. 304, 307) history of gambling in Great Britain highlights several Millian themes such as how no law succeeded in reducing its popularity (bans do not work) and how—in the nineteenth and twentieth centuries—gambling activities took place on the margin of the law without ever being completely outlawed (gambling is tolerated yet not legal). In fact, even though betting on horse races and dog tracks was legal, it was estimated that about 25% of the adult population betted illegally in the 1950s (Wolff 2020). In 1961, the Betting and Gaming Act legalized betting shops and, from no legal casinos in the 1950s, there were over 1000 by the mid-60 s as reported by Wolff (2020, p. 49) who explains,

“[t]he starting assumption was that, while gambling shouldn’t be encouraged, it cannot be stopped. Hence it is better to allow it to exist under highly regulated conditions. This approach yields the concept of ‘unstimulated demand’. If demand can be shown already to exist, then commercial firms can cater—or should we say ‘pander’?—to it, to prevent people breaking the law. But nothing should be done to stimulate demand. In this respect gambling was treated like sex work in the UK: while the activity is legal, steps to drum up trade are not”.

That did not remove the stigma of betting shops which kept dark windows and were not allowed to advertise (Horridge 2016; Lamont 2016).Footnote 26 In 1986, betting shops were allowed to overhaul their interior design, paint their walls, serve drinks, etc. and in 2005 the Gambling Act allowed them to advertise across all media. After WWII, Wolff (2020) argues that there was a gradual move to stimulated demand and a change of view from gambling as undesirable to being one leisure activity among others. With the rise of online gambling in the last 20 years, Horridge (2016) remarked that customer experience inside the shop is becoming a priority. However, the pandemic further fueled the surge in online gambling. As a result, the national online self-exclusion scheme (the blocking software Gamstop) reported a record 21% surge in early 2021 (Davies 2021a). Over time, we thus notice a drift away from Millian policy partly justified by the claim that since only 0.5% of the adult population in Great Britain are problem gamblers, gambling is a fun leisure activity harmless to most (Department for Digital, Culture, Media, and Sport 2020).Footnote 27

Mill admits that since any commodity can be used in excess, alcohol buying and selling should not be prohibited. The question remains whether the State should intervene to regulate the production, distribution, and marketing of alcohol. However, the interest of alcohol dealers to promote excessive consumption is a real evil that “justifies the State in imposing restrictions” that would otherwise infringe on legitimate liberty (On Liberty, p. 111). Mill could also have argued that because alcohol can be used for good (socializing, cultural activities) and bad (alcoholism, vandalism) purposes warnings to the public are perfectly legitimate. While Mill wants to take a strong stand against prohibition, if alcohol consumption becomes a societal problem, if the benefits of consumption are outweighed by its evils, he would approve regulating supply even though this would, as he says, infringe on legitimate liberty.

The final borderline case Mill considers in On Liberty is that of taxation. Mill here argues that taxing some commodities because they are harmful is an undesirable interference with individual liberty. However, since taxation is.

“absolutely inevitable … the state cannot help impose penalties which to some persons may be prohibitory … it is hence the duty of the State to consider in the imposition of taxes which commodities the consumer can best spare … and to select in preferences those of which it deems the use beyond a very moderate quantity to be positively injurious. Taxation of stimulants up to the point which produces the largest amount of revenue is not only admissible, but to be approved of” (On Liberty, p.112).

There is another principle at play here—that of public goods and funding the State apparatus—which trumps the Liberty Principle. This kind of reasoning is sensible if, as we have argued, the Liberty Principle, like other secondary principles, is grounded in the Utility Principle. Consequently, the best state of affairs here is one where individual liberty to consume stimulants is infringed to finance, via taxation, the State and its institutions.

Our reading of Mill’s two texts On Liberty and Utilitarianism took us beyond the simplistic interpretation of Mill as a champion of anti-paternalism emerging in the debate on the policy implications of behavioral economics. With respect to Mill’s defense of free trade, he contends that it stands on equal but distinct grounds from the Liberty Principle. However, both are subordinate to the Utility Principle which is the decisive principle when the former two clash as they do in some of the examples he discusses.

6 Conclusion

Three essential points can be surmised from the above. First, I dealt with the commonplace that in On Liberty Mill allows some exceptions to interfere with individual liberty (Conly 2013, p. 18; Sunstein 2014; White 2013, p. 85). I have argued, however, that this claim stands on thin ground given that the said exceptions are too numerous and significant to be counted as such. If Mill offered a passionate defense of freedom in On Liberty, his substantive list of exceptions is hard to justify unless one accepts that the Liberty Principle is subordinate to a higher order principle, that of (indirect or act) utilitarianism.

Second, the connection between Mill’s Liberty Principle and his Free Trade doctrine is far from seamless. This is in part because Mill considered the Liberty Principle operative at the individual level, covering the purchase and consumption of commodities, whereas the Free Trade doctrine operates at the social level covering the production and supply side of commodities. In the last section of this paper, we have seen that Mill’s reasoning over boundary commodities and services that involve the Free Trade doctrine and the Liberty Principle forced him to formulate ambiguous policies such as covering with a shroud of secrecy the existence of gambling houses.

Third, Mill did not argue that the State, or whoever is conducting policy, has no knowledge whatsoever of what may be the best course of action for some individual. Instead, he emphasized that it is best to leave the individual decide and choose for himself or herself, to develop capacities for autonomous choice. Nevertheless, there are instances, also discussed above, where Mill considers an authority conducting policy may assume (and decide to act on the assumption) that an individual may either be misinformed, an uncritical conformist, or mentally incapacitated and misguided on what is in his best interest. Furthermore, if there is breach of social morality an intervention is justified for Mill irrespective of whether the policy maker has been able to identify the true causes behind the moral grievance. Ultimately, whether the State should intervene is a question of gauging the costs against the benefits of the intervention in terms of utility. While I cautiously presume Mill entertains an agnosticism as to whether the State can in some contexts know the true desires, preferences, or goals of its subjects, the argument laid out in this paper hovers above a precise answer to this thorny question.

I returned, instead, to Mill’s two texts On Liberty and Utilitarianism to show that, in his writings, he argued individuals may not always know what is in their best interest. I discussed five constraints to his Liberty Principle including interference with acts that entangle individual interest and the well-being of others as well as acts where individuals do not control their mental faculties. Interferences with individuals in such contexts do not infringe on the Liberty Principle. A passage from Utilitarianism was cited where Mill acknowledges that, because of infirmity of character, people sometime opt for lower and nearer pleasures. Mill also decries conformity to custom, limited mental abilities, satisfaction with mediocrity, intemperance, and extravagance which everywhere hinder the promotion of self-interest and human advancement. Such claims, I conjecture, signal that Mill would have been receptive to the discoveries of behavioral economists including hyperbolic discounting, framing, and inertia. I, therefore, contest Conly’s (2013, p. 48) claim that Mill “was misled by a relatively unsophisticated view of human psychology” and side with Sugden (2018, p. 5) who states that “many of the factors that Mill thinks influence consumer choice and which neoclassical theory assumed away are ones that modern behavioral economics have investigated”. Mill is favorable to a range of policy interventions which do not infringe on the Liberty Principle even when an act harms no one but the individual. These policy instruments include dissuasion, persuasion, exhortations to strengthen the will, and information labelling. These, I have also shown, are part and parcel of the soft paternalist toolbox.

Three final issues have not received the amount of attention they are due and will be left for future works. First, although Mill’s oeuvre is vast, this paper was restricted to two of his essays, On Liberty and Utilitarianism. Additional research is needed to refute or reconcile the arguments made here with those he made elsewhere. Second, there is the problem of aggregating the net benefits and costs of an intervention using Mill’s higher order Utility Principle. As we saw, the higher order principle is employed when two second order principles clash. If, as Mill maintains, utility is qualitatively heterogeneous, summing costs and benefits could be problematic (but not insurmountable). Third, there is no agreement on the definition of paternalism. We have noted that the adopted definition of soft paternalism is not consensual and that there are other definitions that require, for example, the paternalist to act against the target’s will (Scoccia 2018), to violate her liberty or autonomy (Dworkin 2020), to address a failure of judgment (Le Grand and New 2015; Sugden 2018), or not to be only concerned with individual welfare (Shiffrin 2000). Our aim in this paper was not to adjudicate between these different conceptions of paternalism. Instead, our aim was to show how one of them, the soft paternalistic conception adopted by Sunstein and other behavioral welfare economists, is consistent with Mill’s defense of liberty. Nevertheless, more work is needed on this front specifically on exploring how the Liberty Principle meshes with these other conceptions of paternalism.