Paternalists want you to do what is good for you, and they support policies that promote your doing that. On the face of it, this seems like an uncontroversial goal: the successful paternalist enables people to be better off, and individually, being better off seems to be what we ourselves want. Given this, it may hard to see why there should be any conflict between the paternalist and the people whose welfare he or she promotes. In fact, however, many people see paternalism as invidious. The objection is that paternalism at least sometimes interferes with personal liberty, and that interfering with personal liberty cannot be good.
The first problem is that we typically call a policy paternalistic only if it makes you behave differently than you would otherwise have done. As Cass Sunstein writes, we see paternalistic policies being introduced when “a public institution does not believe that people's choices will promote their welfare and is taking steps to influence or alter people's choices for what it considers to be their good” (Sunstein, 2015, p. 129). Merely making it easier for you to do something that you were planning to do anyway, however beneficent in intent, is not usually referred to as paternalistic: if you were planning to visit the museum on Sunday and the government planner institutes a Sunday free admission policy, making this action even more attractive, it is not generally seen as paternalism. Paternalism is thus generally construed as altering preferences. If a paternalist policy increases the cost for participating in an activity one initially preferred, one may still like the idea of engaging in that activity, but one's all-things-considered preference changes – given the new costs of the activity, one prefers not to engage in it. Where you had wanted to do one thing, now that something in the situation has changed, you will choose to do something else. It is this intervention – this concerted attempt to change people's behaviour – that sometimes arouses strong opposition.
Much of the controversy arises, however, from a second issue: the way the change in behaviour is initiated. In some cases, paternalist interference may be positively welcomed. Say you want to go to the museum, but you feel that you cannot afford it and thus decide not to go: all things considered, you prefer staying away from the museum to going. The paternalist's free admission Sunday policy, designed to benefit the population by increasing museum attendance, changes your all-things-considered preference and with it your behaviour, and so you happily embark on an excursion to see the Elgin Marbles. This, to most people, is not objectionable.
Other ways of changing behaviour are perhaps less welcome, but to most people still not positively offensive: education, for example, can be provided with a specific policy objective in mind, but consumers do not seem to mind this. Warning signs on cigarette packs probably are not the first way anyone learns that smoking may cause cancer, but they may serve as reminders of that fact at a crucial time – when someone is about to smoke – and people outside of the cigarette industry seem to think that this is fair enough. Similarly, required calorie postings in restaurants, even when they elicit shock and horror by revealing that a mocha Frappuccino ‘coffee’ contains 390 calories, do not seem to be something that consumers themselves object to, even though they realise that the calorie information has been provided by a paternalist who hopes that it that will encourage less caloric choices.
Other policies, though, have drawn much more negative attention. In the USA, the Obama Administration's attempt to require all citizens to buy health insurance has caused a huge popular and political backlash, even though everyone recognises that it is all but impossible to pay hospital bills in the USA unless one has insurance and that the resulting failure to get medical treatment will be bad for health. The objection seems to be that people are being forced to act – or refrain from acting – in ways they would not otherwise have done, and even if that benefits them in the long run, coercing a person in this way can only be justified when what he or she wants to do is something that is harmful to others. We do not mind the free entrance museum policy because no one who does not want to go to the museum has to go; individual consumers do not object to education about calories because they can still order the 390-calorie coffee. Some policies leave you free to make whatever choice you want, and others do not.
Why, though, is this loss of freedom bad? One might object, first, that coercive paternalist policies force you to do things that you do not want to do, and this will naturally result in a state of dissatisfaction. People are forced, in that they are sanctioned, when, for example, they do not buy insurance, and the sanction is severe enough that a reasonable person would feel he or she must avoid it. That it is not something they want to do is evidenced (on this account) by the fact that they had not already done it. The individual knows best what he or she wants and knows whether he or she prefers to spend or prefers to gamble with his or her health, and those who would prefer to run the risk of not being able to afford health care are wronged when they are forced to make a purchase that might be good for their future health, but is not what they themselves want to spend their money on. They lose the enjoyment of their preferred choice, and furthermore suffer the frustration of knowing that they could have had it if it were not for outside interference.
Second, in addition to causing subjective dissatisfaction, such interference may also be seen as disrespectful. The objection is that the government forces people to do what it thinks is good for them, rather than what they themselves want. The paternalist fails to respect the individual's values, and substitutes his or her own ends for those that the individual has chosen to pursue. This substitution of judgement is demeaning, since it deprecates the personal judgement of the individual and exalts the judgement of others who think that they know what is good for us. This division into classes of those whose judgement should prevail and those whose judgement should be disregarded is disrespectful and fundamentally undemocratic, and therefore, critics will argue, should be condemned.
However, as we know if we are familiar with the paternalist literature, these particular objections rest on a misconstruction of contemporary paternalist theory. Most paternalists now embrace a subjective view of welfare (Thaler & Sunstein, 2009 Conly, 2013; Le Grand & New, 2015). That is, their goal is not to impose their own values, but to endorse the values of the individual who is subject to the regulation. Thus, the paternalist's goal is not to substitute his or her own judgement for your own, but to help you to do what you yourself most want to do. In certain areas, we are all prone to error, and paternalist policies help us to avoid the poor reasoning that leads us astray from the satisfaction of our own goals. The judgement that the paternalist endorses is the judgement of the individual, and it is the individual's own long-term goals that the paternalist aims to promote. So, successful paternalism allows for more long-term felt satisfaction, since we will be able to enjoy the states we actually prefer – good health, financial solvency, or whatever. Furthermore, individuals’ values are respected, since the aim of the policy is to help individuals fulfil their own goals, rather than leaving them to languish in failure.
Even if we all agree that welfare should be construed in terms of what the individual himself wants for his or her life, though, objections to paternalism remain. While no one wants to go flying through the windshield in the case of a car crash, seatbelt laws were not popular when introduced into the USA in 1968, not only because car companies objected to the cost of implementing them, but also because some people felt that they were being forced to wear them. Even seatbelt laws designed to protect children, whom we often force to do what is good for them on the grounds that their own judgement is poor, have been seen as objectionable because they place coercive pressure on the parent or driver. “I don't need my mother or my government telling me I must have seatbelts on my kids” is not an atypical response, even among those who accept that seatbelts are in fact a good idea.
Even those who think we should wear seatbelts, who want to wear seatbelts themselves and who want their children to wear seatbelts may object to the policy that forces them to wear seatbelts.
Why? Why is a regulation that requires wearing seatbelts not like creating a free entrance day to the museum – with each making it easier to do what you wanted to do anyway? Why is this not seen as a delightful enhancement to your life, making it easier to pursue your own goals? Free entrance days make doing what we wanted painless; calorie postings let us choose the less caloric option, if that is what we care about; and bans on trans fats, prohibitive taxes on soda and required health care again help us stick to the course we already wanted to steer. What is the problem?
Even if both sorts of policies are intended to help you get what you most want, they operate in different ways. Having things made easier to do and having your attention drawn to certain facts still leave it open to you to choose other than what the policy promotes. Your options remain the same, even if some look more attractive under the paternalist policy. Coercive policies, on the other hand, take options away. It is the nature of coercion that even if it does not make an action literally impossible to perform, it increases the costs of that action enough that one cannot reasonably avoid it.
How much of a burden is required for an action to be one that cannot reasonably be taken is a topic of debate, and I will not attempt to resolve this issue, but on most accounts there is agreement as to some particulars. When the thief threatens to take your money or your life, that is coercive, even though you presumably could choose to have your life taken rather than your money. (Let us assume that this is a principled thief who takes ‘your money or your life’ as an exclusive disjunction and is willing to leave your wallet on your dead body in accordance with your choice.)
The idea, then, is that liberty consists of having certain options, and coercion by its very nature deprives us of options. While all paternalist policies attempt to push us in certain directions, some forms of paternalism do this without foreclosing the possibility of us choosing the way the paternalist hopes we will not. The individual may refrain from going to the museum, or may choose the pastry with trans fats. Thus, those policies may be thought not to interfere with liberty, even if they are manipulative. Cass Sunstein has championed libertarian paternalism in part for this reason: while the libertarian paternalist changes the ‘choice architecture’ to make some actions more attractive and some less attractive, he or she does not actually remove any options (Thaler & Sunstein, 2009; Sunstein, 2015). Thus, ideally, liberty is preserved even while welfare is promoted.
The importance of options as a criterion of liberty has been stressed by others. Isaiah Berlin famously defended the value of what he called ‘negative liberty’ – the freedom from interference – over accounts that defended such restraints in favour of some internal ‘positive’ liberty in decision-making. Negative liberty is freedom from constraint by others, whereas positive liberty, for Berlin, obtains when I am “moved by reasons, by conscious purposes, which are my own…” (Berlin, 2002a, p. 178). For Berlin, the danger is that stressing positive freedom – where we are free insofar as we do what it is rational for us to do – is that it might allow coercion by outside forces: states might argue that they are only ensuring that we act in the way our reason would direct, if we were being rational. Thus, valuing positive freedom too highly would allow all sorts of coercion, all sorts of infringements on negative freedom, and could be used to justify a virtually totalitarian state As Berlin described negative liberty, it consists of being able to pursue one's chosen activities without the interference of other people, or, as it has been taken, of the ability to do what one wants to do (Berlin, 2002a). A natural response to this, however, is that we may imagine a slave – a prime example of someone who lacks liberty – who might, depending on his psychology, do whatever he wants to do because everything he wants to do aligns with what the master wants him to do. While the contented slave can do what he wants, no one could plausibly call him free (Flikschuh, 2007). Berlin saw this, and amended his idea of liberty to include options to act in certain ways (Berlin, 2002b, pp. 30–32). The slave would not be able to do certain things if he did want to do them, including, presumably, deciding whether he wants to work, where he wants to live, etc. Since he lacks these options, he is not free, even though he can do exactly what he prefers to do.
This idea that options, even if not pursued, have value seems to underlie the defence of the importance of many freedoms. Many people who are legally able to vote do not choose to. In the USA, even the presidential elections find at best about 63% of eligible voters taking advantage of that option (New York Times Editorial Board, 2014). Still, if we were inclined to deprive 37% of the electorate the right to vote, even the very same 37% who have failed to vote, then that would be seen as an outrage. Even if we could somehow predict with complete certainty that the same 37% would never vote if given the option, removing the option of voting would still be seen as a significant loss. It is good that we have the option to vote, even if we do not actually vote; it is good, according to some, that we have the option to eat trans fats, smoke cigarettes and eschew seatbelts, even if we wisely avoid those things in the pursuit of goals that are not consistent with them.
It is true that coercive paternalism deprives people of options. That, typically, is the point of coercion. As noted above, the fact that we are coerced to do something does not necessarily mean that it is literally impossible not to do that thing, but it does mean that it is, for most, prohibitively costly to do it. So, it is fair to say that, for practical purposes, coercion removes an option.
Why and when are options important, however? Most people, including anti-paternalists, would agree that some options are more important than others. Part of the argument against banning trans fats has been that allowing a ban on trans fats would lead to the loss of other, more important freedoms (Resnik, 2010). The loss of trans fats is not particularly significant in itself, for a number of reasons. Most of us do not have lives in which a concern for trans fats plays a big role. No one wants them for their own sake, and while we do often want the pastries and snacks they make more readily available, even those do not typically serve as an axis of endeavour. The ban on trans fats is important (to those for whom it is important) because it is symbolic – a ban would reflect a decision that the government knows better than the individual what is good for that individual, and that because of this the government has the right to interfere in individual decision-making. The concern seems to be that if we allow this, the same sort of justification – that we have poor judgement as to means – could justify our losing the right to vote.
This is why even those who consistently choose to wear seatbelts object to being coerced into wearing seatbelts – today it is the freedom not to wear seatbelts that is under attack, tomorrow it is the Magna Carta and the Bill of Rights.
Why, then, are some options more important? Presumably because some options are options to do more valuable things, and others are options to do less valuable things. Which options are more important may depend on the particular theory that is used. Those who endorse subjective welfare accounts may think it is a function of our subjective goals and desires, how much we want a thing or how much felt satisfaction its achievement will yield. Others may think that the exercise of some options expresses our rationality, in it being the product of a deliberated choice, or may further hone our rationality in making us better at rational deliberation. Whatever the theory, it seems to make sense to think that some options have more value than others, because some are options to engage in more valuable activities than others. Probably most value theories would find the activity of self-government more important than the activity of eating a pastry made of trans fats, or even of eating pastry in general. If, in some imagined world, being able to pick our own pastry prevented us from being able to pick our own president, most would consider that to have been a poor bargain, all things being equal.
Given this, it is wise to think about what we should do when some options undercut others. If the presence of a less valuable option that is very likely to be exercised makes it more likely that a much more valuable option will not be exercised, is it still true that any loss of an option should be taken as a harm? If the availability of unhealthy substances makes it more likely that we will consume them and less likely that we will be as healthy as we would like to be, is it a good thing that we have both the option to eat more healthily and the option to eat less healthily? Granted, the option to eat a healthy diet is there when one has both options – the person is free, in theory, to ignore advertising and the lure of cheap, simple foods – but if this option is likely not to be exercised given the option to do what is unhealthy, are we really better off? Having more options need not be better than having fewer options. Indeed, it is hard to see that there has been a substantive loss if the more valuable option is now more likely to be chosen. The more valuable option may also be more fecund, to use Bentham's term: exercising the options that lead to good health, for example, typically gives options to do many things we value, like pursue relationships and careers more effectively. Granted, poor health brings its own set of options – ‘which medication?’, ‘which hospital?’ – but these are typically options to engage in activities we would prefer to avoid. So, some options are more valuable than others because they are options to engage in more valuable activities. These options to do valuable things become even more valuable as we are more likely to exercise them. If we do not exercise them, we do not engage in the valuable activity. It makes sense to say that an option to do something good that is predictably very unlikely to be chosen has less value than a similar option that is more likely to be chosen. The option to choose healthy behaviour is worth more if conditions are conductive to it being acted on. An option to choose a valuable activity that is unlikely to be chosen has, obviously, less potential to result in a valuable state.
Some might object that this talk of making some options more likely to be chosen is ungrounded. That is, they may say that even if it is conceded that some options are more valuable than others, any option may be chosen, no matter what the surrounding circumstances. That is what it is to be an option, after all – something that it is open to the individual to choose. Nothing in the existence of cigarettes, mocha Frappucinos or trans fats means that you have to smoke, drink high-calorie beverages or consume compounds that contribute to heart disease. You are entirely able to avoid these, but if you do not, that is your choice. Some may be willing to concede that ignorance interferes with option selection – it is hard to avoid the carcinogen if you do not know it is a carcinogen – but as long as we are well informed and legally permitted either to choose or not to choose something, we have equal options to choose or not to choose. All doors are open to us, and we may go through any one we want.
Let us put aside general questions about free will and determinism, given the length of this essay (and the intractable nature of the subject). We may concede for the purposes of argument that in some sense it is true that the individual who knowingly makes an unhealthy choice could have chosen otherwise. What follows from this, though? It does not follow that in making the unhealthy choice he or she chose what he or she wanted most overall, as discussed above. It does not follow that he or she is blameworthy for choosing poorly, so that we may somehow dismiss his or her error as evidence of culpable character. ‘Blameworthy’ suggests that he or she has made an error that the average person could have avoided, and we know that these errors are such that only an extraordinary person could consistently avoid them. If one action is much easier to take and the alternative is much more difficult to take, we are more likely to choose the easier of the two. Whatever the mechanism of the will may be, it is no mystery that cognitive failures and the temptations of both appetite and inertia make it less likely that many individuals will make the choice that would be best for them. My argument is that this makes what would have been a valuable option less valuable, because it will not result in the correlative valuable activity or state. If we consider an option to do a valuable thing that has a 1% likelihood of being exercised and an option to do a valuable thing that has a 99% likelihood of being exercised, the second option is more valuable, even though both of the options are indeed options – things that might be chosen.
If this sounds like a cost/benefit analysis, that is because it is a cost/benefit analysis. When we analyse the value of a choice, we consider not just the net utility it would produce, but also the probability that it would produce that utility. At least, we do if we are rational. If a potential choice is likely to result in something valuable, it is more important that it is an option for us. If a potential choice is not likely to result in something valuable, the option to choose in that way is less valuable.
Some may say that this talk of the value of different options misses the point of having options. That is, they will say that it is not just possible outcomes that give options value, but also the act of choosing between them. That is why those who oppose paternalistic laws that might lower your credit limit before you acquire unpayable debt are still in favour of the individual who cancels his or her own cards so that he or she will not run up unpayable debt. He or she is making his or her own decision, even though that decision forecloses options. Options have extrinsic value, which derives from being the objects of consideration. We are not going to use our evaluative skills to make decisions between options we do not have. We could, presumably, hypothesise about different impossible choices, but most of us do not engage in such a fruitless exercise. Even if we were willing to speculate about impossible actions, if we cannot really make a choice, we cannot get feedback about the outcome. We cannot learn whether a choice we made was good or bad, and so we do not develop the capacity to discern what a good option is and what is not. Our capacity to make decisions will not be exercised, and this capacity is, on some accounts, the trait we have that makes us valuable, and that distinguishes humans from other animals. To that extent, we would live like unreflecting animals. Furthermore, if we do not exercise that capacity, our ability to reflect when we do face choices diminishes. That is, we will get worse at choosing when we have the opportunity to choose (Mill, 2003, pp. 133–134). Thus, even if individual options are unimportant or even destructive, it is necessary that we have the opportunity to choose among them if we are to use and to improve our capacities for evaluation. The possibility of error is the inevitable concomitant of the best part of the human condition.
This is a fair concern, since we certainly do not want to lose our capacity to make good decisions. The mistake is in the belief that having more options contributes to the exercise of our decision-making capacities. The idea comes perhaps from an analogy. If I am taking piano lessons, for example, I need to practice – nothing but the actual activity of hitting the notes will make me better at playing the piano. The more practice, the better, within the boundaries of physical and psychological health. We say, indeed, that practice makes perfect, and while that might not be an option for me in terms of playing the piano, practice certainly makes better. So, one might think that just as practice is necessary for improvement at playing the piano, practice is also necessary for improvement in decision making.
However, the piano practice analogy may be taken the wrong way. It is not enough to go out there and bang randomly on the keys if you want to get better at playing the piano. You have to hit the right notes. If you practice and continuously hit a C# when that should be a D#, all you are getting better at is hitting the wrong note. You will not suddenly hit the D# when it matters (your recital). Hitting the C# repeatedly and saying each time “Drat! That should have been a D#!” is not helpful. Practice involves avoiding errors if it is to result in improvement. We may hone our skills by making good decisions, and we do not hone them by making bad decisions. We know, after all, that when it comes to public health, the fact that one has made a few poor choices (to smoke a few cigarettes, to eat a few unhealthy meals, etc.) does not mean that one will then see the error of one's ways and shape up. If that were true, we would not find two-thirds of Americans to be overweight. Bankers and credit card companies would joyfully extend credit to those who have gone bankrupt, sure that by now they have learned the error of their ways and will be an especially good credit risk. Too often, bad choices result in bad habits. Part of a bad habit is a failure even to exercise the evaluative capacity the champions of free choice celebrate. What we do habitually we do with very little reflection. The unfettered ability to make bad choices is no strategy for improvement in the exercise of the evaluative capacities. Making good choices involves a number of things, including in many cases the impossibility of making the worst choices that turn into the worst habits.
This is not to say that we never learn from bad decisions, and never change our behaviour in light of that learning. Coercive paternalism would not be suitable in every area of choice, to say the least, and one of the many reasons for this is that, in some cases, making choices among options has the effect that the champions of choice would like it to have – we rethink our bad decisions and we avoid them the next time. In some areas, though, this does not seem to happen, or happens not often enough, and these are often very significant areas when it comes to living a successful life. This is particularly true when we are looking at harms that accrue from incremental damage, where education as to likely outcomes is just not enough to dissuade us from taking each small, bad step. Realising after one's diagnosis of cancer that the choice to smoke was a bad one is a bit late when it comes to making better choices in the future, since too often there will not be much of a future.
All of this may make it sound as if liberty is only valuable as a means, rather than as end in itself – as if its value lies in what it brings us, rather than simply in what it is. If it sounds this way, that is because this is in fact the most plausible view of political liberty. Different options have different values because they lead to more or less valuable states of affairs, or because they contribute or do not contribute to the development of an important capacity.