1The title of my paper, perhaps a bit oddly, makes a connection between John Stuart Mill and two newly coined terms that Mill himself did not use: soft paternalism and mental capacity. There can be no doubt that Mill is among the first authors to whom modern social and political philosophers turn on questions of state intervention and paternalism. This paper is an attempt to reconstruct and partly challenge the view that is generally attributed to Mill when it comes to the justification of paternalistic interventions. This view, which I call the ‘orthodox view’, characterizes Mill as a ‘soft antipaternalist’, i.e. as someone who rejects paternalism with certain exceptions, the most notable being the possibility to exercise paternalism over individuals who are, for some reason, unable to make decisions for themselves. The following section is dedicated entirely to the reconstruction of the orthodox view and the demonstration of the hypothesis that this reading – although certainly finds support in Mill’s writings – essentially stems from the apparently uncontroversial nature of soft paternalism that is accepted by major liberal authors as well. Creating the orthodox reading could be seen as an attempt, as Feinberg states, ‘to reconcile somehow our general repugnance for paternalism with the apparent necessity, or at least reasonableness, of some paternalistic regulations’.1 The third section will cast doubt on the uncontroversial nature of soft paternalism by discussing some problems that can be raised in relation to the concepts of voluntariness, autonomy and mental capacity. Finally, the last section returns to Mill and shows that on a closer reading his ideas about autonomy and mental capacity imply that in some cases he would have taken an issue with the proposition that people making non-autonomous choices can be indiscriminately paternalized.
- 2 Mill (1975) 10.
- 3 Skinner (1969) 16.
- 4 Skinner (1969) 18-19. See also Brown (2009).
- 5 Mabsout (2021) 4.
2Much ink has been spilled on reconstructing Mill’s view on paternalism. It is quite possible that Mill had no coherent stance towards the justification of paternalistic interventions and it was merely a desire for coherence that led many scholars to attribute – to use Mill’s own words – ‘one very simple principle’ to him when designating the limits of state intervention.2 Quentin Skinner points out that the history of ethical and political philosophy is particularly pervaded by the ‘mythology of coherence’,3 i.e. by the desire to find consistency in the theories of classic writers, even at the cost of discounting statements, passages or even whole works that seem to violate the real or imaginary coherence of an author’s theory.4 Mill himself did not use the term paternalism,5 and his position can only be reconstructed ‘indirectly’ by considering different and sometimes inconsistent passages scattered throughout his oeuvre. To be sure, the starting point remains the following, oft-quoted passage from On Liberty:
That 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.6
3Read on its own, this section, usually referred to as the harm principle, seems to impose an absolute ban on paternalistic interventions. Yet, considering other parts of On Liberty, it is noticeable that Mill does not reject paternalism in every situation. It is possible to identify at least three considerations that seem to relax the absolute ban on paternalism apparent in the previously quoted passage. Without going into a detailed analysis, the following paragraphs give an overview of these exceptions which, taken together with Mill’s main rule on antipaternalism, constitute what I have dubbed as the ‘orthodox view’ on Mill’s approach to paternalism in On Liberty.
- 7 Mill (1975) 10. See also Mabsout (2021) 11; Riley (1998) 93.
- 8 Mill (1975) 71.
- 9 Mill (1975) 10. Cf. Kultgen (1995) 239.
4The first exception to absolute anti-paternalism appears immediately after the formulation of the harm principle. Mill states that only coercive forms of paternalism are prohibited, and it is acceptable, for example, to try to persuade someone to make better decisions.7 Later, in the beginning of Chapter IV, he adds that ‘disinterested benevolence can find other instruments to persuade people to their good, than whips and scourges’.8 Thus, ‘softer’, non-coercive forms of paternalism seem to be acceptable to Mill, although what he means by coercion remains somewhat vague in On Liberty. At one point, Mill seems to extend the concept when stating that compulsion might not only entail ‘physical force in the form of legal penalties’ but also ‘the moral coercion of public opinion’.9
- 10 According to C. L. Ten, the term ‘self-regarding’ is Mill’s own invention but he does not use the e (...)
- 11 Mill (1975) 75.
- 12 Stephen (1991) 66.
- 13 In such cases, the rationale for intervention is ‘mixed’ and it is hard to tell if they are cases o (...)
- 14 Mill (1975) 75.
- 15 Mill (1975) 76.
- 16 Brink (2022). See also Szerletics (2015) 97.
5Distinguishing between conducts that only affect the agent and conducts that affect others seems to be crucial for the justification of social intervention in Mill’s framework.10 He readily admits that ‘the mischief which a person does to himself may seriously affect, both through their sympathies and their interests, those nearly connected with him and, in a minor degree, society at large’.11 The idea that self-regarding actions may ‘indirectly’ harm others and, as a result, can be a reason for intervention appears regularly among Mill’s critics, including his contemporary, James Fitzjames Stephen, who points out that ‘by far the most important part of our conduct regards both ourselves and others’.12 There is no doubt that Mill allows paternalistic measures if a self-harming action causes harm to others as well.13 Thus, the central problem at the heart of our second exception is not really about the possibility of distinguishing between self-regarding and other-regarding actions but rather about the notion of harm, i.e. when does a primarily self-regarding ‘mischief’ affecting other people cause actual harm to them. Even though Mill does not provide a detailed answer to this question, he states that if a person, through his or her ‘mischief’ violates a ‘distinct and assignable obligation’ to others, ‘the case is taken out of the self-regarding class’.14 A little later, he adds that ‘whenever […] there is a definite damage, or a definite risk of damage, either to an individual or to the public, the case is taken out of the province of liberty, and placed in that of morality or law’.15 Regardless of how someone understands the concept of harm, it is important to note that for Mill the presence of harm seems to be a necessary, but not sufficient condition for coercion. From the perspective of paternalism, this means that a self-harming action that causes harm to others at the same time can, but not necessarily will be limited – societal intervention ultimately seems to depend on consequentialist (utilitarian) considerations.16
- 17 Feinberg uses the expressions ‘weak’ and ‘strong’ paternalism in his early paper. Later, he adopts (...)
- 18 Feinberg (1971) 113.
- 19 Danny Scoccia mentions e.g. Joel Feinberg, Gerald Dworkin, Richard Arneson, Alvin Goldman, Donald V (...)
- 20 It is an open question whether Mill accepted the primacy of autonomy on utilitarian or deontologica (...)
6 The last exception that leaves room for paternalistic interventions in Mill’s framework is based on the distinction between hard and soft paternalism. Mill did not use these terms himself. The distinction was introduced much later by Joel Feinberg in his seminal paper on paternalism and it is based primarily on the voluntariness of the subject’s decisions.17 Hard paternalism aims to protect competent adults from their voluntary self-harming choices. Soft paternalism allows protection from self-regarding harmful conduct, if ‘the conduct is substantially nonvoluntary, or when temporary intervention is necessary to establish if it is voluntary or not’.18 Feinberg opposes hard paternalism on the basis that it violates the autonomy of competent individuals; other liberal theorists follow suit and tend to reject paternalistic interventions that do not respect personal autonomy.19 These authors regard soft paternalism as relatively uncontroversial because it only interferes with non-autonomous self-harming decisions, the negative consequences of which a vulnerable person can or even needs to be protected from. The orthodox liberal reading of Mill, I believe, is rooted in the assumption that the normative conclusion drawn from the distinction between hard and soft paternalism is essentially right and, as a result, it is best to characterize Mill as a ‘moderate’ anti-paternalist placing the hard-soft distinction and, relatedly, the value of autonomy at the heart of his theory in On Liberty.20 To be sure, there are passages in On Liberty that support this reading. Take the well-known example of the unsafe bridge in chapter 5:
‘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 […]’21
7Lack of information can render a person’s choices ‘substantially nonvoluntary’ and make him or her a subject of legitimate paternalistic intervention in Mill’s framework. The same applies to ‘a child, or someone who is delirious, or in some state of excitement or absorption incompatible with the full use of the reflecting faculty’.22 Another clear indication of Mill adopting soft paternalism can be found in the first chapter of the essay, just after the formulation of the harm principle:
‘It is, perhaps, hardly necessary to say that this doctrine is meant to apply only to human beings in the maturity of their faculties. We are not speaking of children […]. Those who are still in a state to require being taken care of by others, must be protected against their own actions as well as against external injury.’23
8These passages provide a solid basis for the orthodox interpretation of Mill’s approach to paternalism which can be roughly described as a form of autonomy-based moderate antipaternalism.
- 24 Riley mentions Danny Scoccia, Sarah Conly, Cass Sunstein and Jason Hanna as representatives of this (...)
- 25 Mill (1975) 95. See also Arneson (1980) 472-473; Feinberg (1986) 79; Archard (1990); Fuchs (2001) 2 (...)
- 26 Arneson (1980) 486-487.
- 27 Cohen-Almagor (2012) 568.
- 28 Claeys (2013) 10, 80.
- 29 Claeys (2013) 10 and references there.
- 30 Claeys (2013). It is possible to argue that Mill was a perfectionist liberal, i.e. he saw autonomy (...)
- 31 See Cohen-Almagor (2012) 565-573.
- 32 See Claeys (2013) 11-12 and references there.
- 33 Claeys (2013) 14.
- 34 Claeys (2013) 14.
9The dominance of the orthodox interpretation can – and to a considerable extent, has been – challenged from two directions. (1) The first direction is based on the normative assumption that hard paternalism is not necessarily wrong and that an adequate theory of paternalism does not categorically exclude the possibility of interfering with autonomous self-harming choices.24 It seems unnecessary to present the theoretical arguments in support of this assumption here; I would just like to indicate that there are there are some ‘troubling’ examples of hard paternalism in Mill’s own work that cast doubt on the correctness of the orthodox interpretation. One such ‘puzzling’ example is the invalidity of voluntary slavery contracts discussed in the fifth chapter of On Liberty.25 In order to preserve the coherence of the orthodox interpretation, one could argue to disregard this passage,26 or take it as soft paternalism because anyone who decides to become a slave is ‘not rational enough to have full responsibility of one’s future life’.27 However, it seems almost impossible to maintain the orthodox interpretation of Mill’s position on paternalism if we consider his entire oeuvre, including such major works as Principles of Political Economy and Considerations on Representative Government.28 Claeys points out that ‘critics have […] discerned markedly paternalistic trends in Mill’s thought’.29 Social justice and equality, feminism and Malthusian population control are three central, increasingly important ideals for Mill that can justify various – sometimes hard paternalistic – interventions in individual liberty,30 for example compulsory education, regulation of perpetual contracts, birth control, regulation of social offences and indecent conduct.31 In fact, the level of inconsistency between On Liberty and other works is such that some have concluded that there are actually ‘two Mills’: a moderate antipaternalist and an interventionist one.32 As Claeys points out, this apparent incoherence can be traced back to the presumption that On Liberty presents Mill’s complete statement of the issue of liberty.33 This presumption is far from being true; focusing on this text as the single authoritative source on matters of liberty and paternalism has ‘produced a largely unhistorical portrait of Mill’s intentions’.34
10(2) The second approach capable of challenging the orthodox interpretation of Mill’s position to paternalism starts off from a slightly different premise than the previous one. While the previous approach made the distinction between hard and soft paternalism somewhat irrelevant by accepting the possibility of hard paternalism, this one builds on the idea that the distinction is untenable, partly because it is impossible to draw a non-arbitrary line between autonomous and non-autonomous decisions. The rest of my paper is devoted to exploring this second approach. The next section shows that there are compelling arguments in favour of abandoning the soft-hard distinction, but this should in no way be taken as an argument for hard paternalism. It rather implies that if one resorts to paternalism, people might have to be paternalized in the same way regardless of their status with respect to autonomy. Section 4 picks up the thread on Mill again and tries to show that, even though Mill did endorse paternalism in the cases of ‘lunatics’ and children, some of the problems presented in section 3 come up in Mill’s works as well. Thus, the orthodox reading of Mill’s paternalism can be challenged, and not only on the basis that he was a hard paternalist but also because he could be read as having a more chiselled approach to those who are generally subjected to ‘soft’ paternalism without any second thought.
11For most people, soft paternalism seems relatively uncontroversial: if a self-harming decision is ‘less than voluntary’, it is not considered to be the person’s own decision and, as a result, he or she can be paternalistically treated. The main goal of this section is to question the uncontroversial nature of soft paternalism which seems to inform the orthodox reading of Mill. This will be carried out in three steps. The first subsection briefly presents the most important objections to Feinberg’s voluntariness standard, while the second one critically examines a more comprehensive criterion, i.e. the notion of personal autonomy. Finally, we look at one specific aspect of autonomy, mental capacity and some of the objections that can be associated with the acceptance of (soft) paternalism in the case of those who are said to lack mental capacity. Given the complexity of these issues, an in-depth analysis is hardly possible here. It seems sufficient to give an overview of these problems; our ultimate goal in section 4 will be to show that reservations about soft paternalism caused by some of these problems are implicitly present in Mill’s thinking.
- 35 Feinberg (1986) 104.
- 36 Feinberg (1986) 117-123 ; Hannah (2018) 25 ; Szerletics (2015) 51 ; Hannah (2018) 180-181.
- 37 Arneson (2005) 268-269.
- 38 Arneson (2005) 268.
12I would like to mention two reasons for criticizing Feinberg’s voluntariness standard. Firstly, voluntariness – just like autonomy or mental capacity – is not a binary, all-or-nothing concept but it is a matter of degree.35 Thus, it is necessary to set a threshold for voluntariness, above which self-harming decisions should be respected, but below which individuals can be legitimately paternalized. However, this threshold can hardly be objectively determined. Feinberg himself speaks of a ‘sliding scale’ of voluntariness and treats voluntariness as a variable concept: certain factors (e.g. magnitude of harm, risk of harm) can affect the level of voluntariness required to make a given decision.36 On a more fundamental level, it is possible to reject voluntariness as the basis for differentiating between soft and hard paternalism, because voluntariness is only one element of autonomous decision-making and acting voluntarily does not per se guarantee that the person acts autonomously as well.37 After all, it can be plausibly argued that prima facie voluntary self-harming choices determined by inauthentic preferences (i.e. by preferences that are not ‘really’ one’s own, e.g. because they are the result of social indoctrination) are legitimate objects of paternalistic interventions. Thus, one can agree with Arneson that a soft antipaternalist focusing on voluntariness, like Feinberg, ‘only defends personal sovereignty lite’.38
- 39 Authors that conceptualize soft paternalism similarly include Scoccia (2008) 351; Fateh-Moghadam an (...)
- 40 Ashley (2012) 13.
- 41 Ashley (20212) 14.
- 42 See e.g. Christman (2014) 371-375.
- 43 Oshana (2006) 101. Cf. Szerletics (2015) 156.
13Moving beyond the notion of voluntariness, one can conceptualize soft paternalism as a form of paternalism that interferes with the self-harming or imprudent decisions of those considered to lack autonomy. Soft paternalism, even if it involves some form of coercion or limitation of liberty, does not violate autonomy because the paternalized person is essentially non-autonomous, the background assumption being that it is impossible to violate something that does not exist.39 Perhaps the most serious problem with this approach stems from the conceptual diversity of personal autonomy, which means that the range of persons who can be legitimately paternalized will ultimately depend on how one understands the concept of personal autonomy. There are many different theories of personal autonomy: without going into a detailed examination, it is worth mentioning the divisions between internalist and externalist, and proceduralist and substantivist models of autonomy.40 Procedural approaches emphasize the ‘unimpaired’ nature of the decision-making process, while substantive approaches focus on the content of decisions. The internalist – externalist division focuses on the conditions of autonomy. Internalist models emphasize the agent’s internal reflective and evaluative capacities, whereas externalist models tend to put the spotlight on ‘external enabling conditions that pertain to the environment in which the agent is operating’.41 The recent notion of relational autonomy further complicates the theoretical landscape: the basic idea behind relational autonomy is that enabling social relations can help someone, ‘normally’ considered to lack capacity, make autonomous decisions and conversely, oppressive social relations can prevent someone from leading a truly authentic life.42 An oft-cited example is the case of the Taliban woman who seems to ‘willingly’ accept her subordinated role in her family but her acceptance is most likely the result of continuing social repression and conditioning.43 This study does not intend to choose between rival conceptions of autonomy; the point here is that someone can be autonomous according to a specific account of personal autonomy, while the same person can be non-autonomous according to a different one. Thus, the soft-hard distinction of paternalism seems to be contingent upon the specific notion of autonomy adopted.
- 44 See e.g. Fuchs (2001) 231-232; Szerletics (2015) 100-104.
14Additional problems emerge if we take into account that autonomy – regardless of its precise conceptualization – is a complex notion that requires several different external and internal conditions to be simultaneously met for its existence. Partly due to this complexity, autonomy, just like voluntariness, seems to be a matter of degree which implies a threshold that can separate legitimate from non-legitimate interventions to someone’s liberty of action. This leads us to probably the most profound question that can be raised with respect to the distinction of soft and hard paternalism, i.e. why does the lack of autonomy make paternalistic interventions legitimate in the first place? In other words, why is the difference between those who have and those who do not have autonomy morally relevant and why does it justify treating these groups differently? After all, it can be plausibly argued that an autonomous self-harming decision is just as bad as a non-autonomous one if they have the same consequences. The answer probably lies in the fact that those who believe that the soft-hard distinction serves as a genuine guideline for paternalistic interventions put autonomy in the first place: they see it as a paramount value and let no other considerations (e.g. individual or collective well-being) influence their views about paternalism. Thus, the adoption of the soft-hard distinction can be associated with deontological ethics and the rejection of consequentialism. It is often argued that Mill’s view on liberty was primarily informed by deontological considerations and that – despite his professed utilitarianism – he abandoned consequentialist principles in On Liberty.44 This seems to be in line with the orthodox reading of Mill as a ‘soft antipaternalist’.
15This section focuses on some problems related to the idea of mental capacity. Mental capacity most often refers to the ability to make decisions about one’s own life and it is seen as constituting an internal element of personal autonomy: most people would prima facie agree with the propositions that mental capacity is primarily associated with an agent’s internal characteristics and that someone who lacks mental capacity can hardly be seen as leading an autonomous life. To be sure, someone who does have mental capacity is not necessarily autonomous either: lack of information, undue pressure and a host of other factors can render one’s actions non-autonomous and make him or her subject to paternalistic interventions (cf. Mill’s example of the unsafe bridge). The problems discussed in the following paragraph are partly the result of challenging the previously mentioned, seemingly uncontentious propositions.
- 45 Section 3 MCA 2005 (England and Wales).
- 46 Arstein-Kerslake (2017) 72.
- 47 The MCA recognizes this idea when it states in section 2 that ‘[…] a person lacks capacity in relat (...)
16(1) Similarly to voluntariness and autonomy, mental capacity is a matter of degree which implies that any threshold for capacity will be, at least to some extent, arbitrarily set. (2) Mental capacity is also a complex notion in the sense that it implies the presence of different cognitive functions that can vary depending on the actual characterisation of the concept; section 3 of the Mental Capacity Act, for example, states that a person is unable to make a decision for himself if he is unable (a) to understand, (b) to retain or (c.) to use or weigh the information relevant to his decision, or if he is unable (d.) to communicate his decision.45 (3) Somewhat similarly to the argument made previously in relation to autonomy, it is possible to assert that there are many different ways to conceptualize mental capacity, which, in turn, might have an impact on what counts as soft paternalism. Even if one accepts that mental capacity is essentially an ‘objective’ medical fact, medical experts will disagree about its proper characterisation and assessment. Those inspired by the social model of disability argue that mental capacity, similarly to disability, is a social construct. Mental disability exists because the world is constructed in a way to accommodate the needs of ‘normal’ people, i.e. those in the majority of society. Thus, mental capacity ‘is a subjective line fashioned by the perspective of an external audience’ that reflects and reinforces the artificially created difference between ‘disabled’ and ‘normal’ people.46 (4) Pursuing this idea further, it is possible to cast some doubt over the exclusively ‘internal’ character of mental capacity in the sense that someone’s ability to make decisions can be significantly altered by external social factors e.g. by empowering social relations that provide support during the decision-making process. This is where the previously mentioned concept of relational autonomy comes in, together with the idea of supported decision-making which is increasingly adopted by jurisdictions all over the world as an alternative to paternalistic substitute decision-making. Additionally, the fact that external factors can affect mental capacity is evident in the recognition that mental capacity is situation-specific: the decision-making ability cannot be abstracted from the specific context in which the decision-making takes place.47 Thus, the presence of mental disability does not, in itself, mean that someone lacks mental capacity; the disabled person might be perfectly able to make certain types of decisions in certain situations, while unable to make other types of decisions in other situations.
17As discussed in section 2, representatives of the ‘orthodox view’ portray Mill as a ‘moderate antipaternalist’ who rejects paternalism with the exception of nonvoluntary self-harming conducts. This section will cast doubt on the correctness of this characterization by arguing that Mill himself was probably aware of some of the difficulties that appear in relation to the distinction between soft and hard paternalism and the possibility of exercising unrestricted paternalism over ‘those who are still in a state to require being taken care of by others’.48
- 49 Mill (2003) 125; Stanley (2017) 54; Macleod (2023) 75-76.
- 50 Mill (2003) 122.
18Although Mill did not use the term ‘autonomy’, the third chapter of On Liberty is a good starting point for exploring his views on the concept. Mill is often regarded as adopting a largely individualistic conception of autonomy, which focuses on the idea of authentic self-rule, meaning that a ‘person is said to have character’ if he acts on desires and impulses that ‘are his own – are the expression of his own nature’.49 Autonomy implies authenticity, which, in turn, implies individuality, basically meaning that a person’s desires are formed independently of others and not by blindly following social customs and traditions. Mill sees the lack of individuality as a major flaw of the society of his time and warns us that ‘[w]here, not the person’s own character, but the traditions or customs of other people are the rule of conduct,’ there is no individuality, which is ‘one of the principal ingredients of human happiness’ and ‘quite the chief ingredient of individual and social progress’.50
- 51 Macleod (2023) 78.
- 52 Macleod (2023) 78-80.
- 53 Mill (2003) 125. Emphasis added.
- 54 Stanley (2017) 67-72.
- 55 Habibi (2001) 171.
- 56 Mill (1846) 4. See also Browne (2016) 1.
- 57 Mill (1846) 4.
- 58 Mill (1846) 4.
19 Even though ‘one’s own nature’ can hardly be authentic if entirely determined by external factors, it does not follow that it will develop in a state of ‘splendid isolation’.51 In fact, Macleod argues that for Mill, human nature needs to be suitably cultivated, otherwise it would be extremely undesirable.52 Thus, one’s authentic nature is his ‘second nature’ shaped by processes of socialisation and acculturation. Mill accepts that certain external social influences and enabling relations are necessary for autonomy: for him, an autonomous person is someone whose desires and impulses are ‘the expression of his own nature, as it has been developed and modified by his own culture’.53 Some even argue that Mill would have been sympathetic to the idea of relational autonomy – a fairly surprising, yet not entirely implausible argument that can be based, among other things, on Mill’s belief that a positive parent-child relation is one of the most important factors that contributes to the development of a child’s autonomy.54 If such relations are conducive to the autonomy of children, paternalism shall not be ‘automatic and universal’ in their case: a blanket denial of liberty seems to be wrong, partly because children might be able to deliver competent judgments in certain situations, and partly because the lack of opportunities to practice skills that are necessary for autonomous decision-making prevents the development of these skills in the first place.55 The conclusion that soft paternalism is not necessarily right in the case of children might be extended to those who lack capacity due to mental disability: even though the two groups differ in certain basic characteristics (most notably in their potential to develop autonomy), if relations matter to autonomy, ‘good’ relations might be capable of making paternalistic interventions unjustified in this group as well. Speaking of pauper lunatics, Mill states in one of his newspaper writings that ‘there cannot be conceived any human creatures more utterly helpless’.56 To be sure, Mill believes that this helplessness is primarily the result of some debilitating mental illness: ‘they can be compared to nothing but the dumb animals, like those, they cannot even tell of their wrongs’.57 However, he also attributes the vulnerability of ‘lunatics’ partly to the fact that ‘they have no relations capable of giving them the smallest protection.’58
- 59 Browne (2016) 1.
- 60 Mill (2004) 290.
- 61 Mill (2004) 290.
20Since Mill did not devote too much attention to the questions of mental disability and mental capacity, any conclusion one might reach on these points is bound to be speculative to some extent.59 Mill certainly does not question the existence of mental illnesses or the need for ‘lunatic asylums’ and ‘madhouses’. It has already been mentioned in section 2 that he introduced ‘paternalistic’ exceptions to the harm principle in On Liberty. He also states in the Principles of Political Economy that the principle of laisser-faire breaks down in the case of those who may be incapable of judging or acting for themselves, including lunatics, idiots or infants.60 He concludes that ‘[i]nsane persons are everywhere regarded as proper objects of the care of the state’.61
- 62 Mill (2003) 158.
- 63 See e.g. Riley (2018) 157-158.
- 64 Mill (2003) 139.
- 65 Mill (2003) 131. Emphasis added.
- 66 Mill (2003) 132.
- 67 Mill (2003) 133.
- 68 Mill (1858) 4.
21 However, it seems quite clear that Mill does not equate ‘lunacy’ with the lack of mental capacity. Someone might be incapable of judging or acting for himself without being a ‘lunatic’ and reversely, it is possible that someone with mental disability can nevertheless make decisions for himself. Although there is less evidence in Mill’s work to support the latter claim, it can be argued that Mill would have rejected a status-based approach to mental capacity. The first indication of this is the fact that he does not link the permissibility of soft paternalism directly to childhood or mental illness. Rather, when it comes to exercising soft paternalism, he talks about those ‘who may be incapable of judging or acting for themselves’, or those, who are ‘delirious, or in some state of excitement or absorption incompatible with the full use of the reflecting faculty’.62 Thus, it seems that Mill utilizes a rudimentary conception of mental capacity, which is, however, far from being clearly formulated.63 He mentions the capacity of ‘being improved by free and equal discussion’, a ‘tolerable amount of common sense and experience’ and an ‘ordinary amount of understanding’.64 The words ‘common sense’, ‘understanding’ and ‘reflecting faculty’ suggest that Mill primarily associates mental capacity with the presence of certain cognitive skills that are necessary for an ‘adequate’ decision-making process. However, he does not equate mental capacity with full rationality: he states that ‘if a person possesses any tolerable amount of common sense and experience, his own mode of laying out his existence is the best, not because it is the best in itself, but because it is his own mode’.65 Thus, people with capacity do not always make the ‘best’ decisions; sometimes they make unwise or foolish decisions which nevertheless deserve respect because they were autonomously made. According to Mill, the fact that someone made an imprudent or unwise decision is not to be taken as an indication of lack of mental capacity. He repeatedly expresses his fear that it is possible to abuse existing legal regulations to declare someone lunatic on the basis that he or she made a decision that seems unconventional or unusual for the majority. He writes in On Liberty that those ‘who can be accused either of doing ‘what nobody does,’ or of not doing ‘what everybody does,’ are ‘in peril of a commission de lunatico, and of having their property taken from them’.66 In the accompanying footnote, he adds that when declaring someone ‘unfit for the management of his affairs’, ‘[a]ll the minute details of his daily life are pried into, and whatever is found which […] bears an appearance unlike absolute commonplace, is laid before the jury as evidence of insanity.’67 Approaching this problem from a different direction, it is possible to focus on the unwise decisions of ‘lunatics’, i.e. those suffering from some kind of mental disability. Once again, Mill does not really have doubts about paternalizing these people. However, it is possible to argue that he is concerned about the potentially discriminatory treatment of those who allegedly belong to this group: if someone is labelled a ‘lunatic’, there is a tendency to automatically treat him as incompetent and to interpret his choices that would ‘normally’ be unwise as irrational. He writes that to ‘the other unfortunates [in the madhouse] the ordinary use of speech is virtually denied; their sober statements of fact, still more their passionate protests against injustice, are held to be so many instances of insane delusion’.68
22The second indication which arguably shows that Mill had reservations about automatically conflating mental illness with the lack of mental capacity is the way he generally describes the treatment of lunatics as sanctioned by the laws of England. Take, for example, the following passage from his newspaper article about poor laws:
‘We may be sure, without being told, that there is and always has been a greater amount of brutal indifference, hardened want of feeling, and wanton and capricious tyranny practised towards insane paupers, than towards any other class of unfortunates’.69
23It is possible to argue that Mill sees the ‘capricious tyranny’ of unrestricted paternalism as being fundamentally wrong, even if it is motivated by benevolence. He might prefer to have a system that takes into account actual mental capacity instead of mental illness when deciding about the permissibility of paternalistic interventions.
24The orthodox reading of Mill is influenced by the presumption that soft paternalism is morally permissible because it does not violate the autonomy of the paternalized person. By challenging this presumption, it is possible to open one’s mind to a different reading of Mill, i.e. to see him as someone who does not necessarily agree with unbridled paternalism exercised over those who lack capacity. The aim of this paper was to show that the considerations which might lead us to reject soft paternalism are present, to some extent, in Mill’s work. There are passages, for example, that can be used to argue that Mill’s view of autonomy is not as individualistic as it is usually made out to be: external factors, socialisation and relationships can all play a role in shaping one’s ‘true nature’. It is possible to speculatively conclude that Mill would have agreed that providing support and strengthening relations that benefit the individual are viable alternatives to soft paternalistic interventions.