Skip to content

The Language of Freedom in Public Health: the Case of the Smoking Ban

Alberto Giubilini


Enough manipulation of the definition of man, and freedom can be made to mean whatever the manipulator wishes

(Isaiah Berlin, Two Concepts of Liberty, 1958)


The UK Prime Minister has announced his plan to ban the sale of tobacco products to young generations in England. Smoking will be phased out by progressively increasing the legal age for buying tobacco every year. Assuming the plan is effective and does not simply open the door to a black market, young generations in England will be prevented from starting to smoke. According to the Prime Minister, “this measure will be the single biggest intervention in public health in a generation.”

It is hardly necessary to provide figures about the risks of smoking. Lighting up that first cigarette is one of the most unhealthy choices one could ever make. In fact, it is a decision many regret later in life. The question is: to what extent is a government justified in preventing competent individuals from making unhealthy decisions for themselves?

Appeals to public goods provide some support for the ban. For instance, if the policy turns out to be effective, there might well be positive implications for the NHS. On some views, this might justify the cost in terms of freedom. However, pressed by a BBC journalist with the question whether the smoking ban would not excessively infringe upon individual freedom of choice, the Chief Medical Officer for England chose a different argumentative strategy. (video interview at this link)

He implied that in fact there is no conflict between individual freedom and health. On his view, preventing people from smoking means helping them make genuinely free choices.

We have heard this before. As I have discussed in a recent article, some of the argumentsused to justify restrictive measures during the recent pandemic contained a similar narrative: by restricting individuals’ freedom in one sense (say, with lockdowns or vaccine mandates), restrictive Government policies promote individuals’ freedom in some other, allegedly more important sense (say, freedom from disease).

In the case of the smoking ban, the Chief Medical Officer replied:

smoking is an issue based on addiction. Most people who smoke wish they had never taken it up. They try to stop and they can’t. And that’s the point: their choice has been taken away from them. As a doctor I’ve seen many people in hospital desperate to stop smoking because it’s killing them and yet they cannot – their choice has been removed”. (see video interview at the link above)

What to make of this statement?


An element of paternalism

Let me start by pointing out the obvious element of paternalism involved in that statement. As Jonathan Pugh explained in his recent blogpost about the smoking ban, paternalism can be understood as “as an interference with another person, against their will, in order to promote that individual’s own interests”. In this case, an interest in one’s own health.  If it is in one person’s interest not to take up smoking, and assuming a smoking ban is effective at doing that, then a smoking ban is paternalistic. This does not say whether the ban is, all things considered, ethically justifiable or consistent with liberal values. On some views, whether it is largely depends on what type of paternalism we are talking about.

The Chief Medical Officer’s statement suggests that the ban would represent a form of  ‘soft paternalism’, often considered less problematic and consistent with liberal values. As explained in the same blogpost, soft paternalism “only permits interference with an individual’s involuntary choices”, while hard paternalism “permits interference with an individual’s voluntary choices”. The Chief Medical Officer’s words suggest that the really voluntary choice is the one of not taking up smoking or of quitting smoking, as evidenced by all the cases of people regretting having started smoking and who cannot get out of their addiction.

One assumption here is that often what we choose is not what we really want or what we would choose if we were in the condition to make really rational, or really autonomous choices. In the philosophical literature, different pairs of terms have been used to draw this type of distinction  – such as first-order vs second-order desiresor experiential vs critical interests3. These pairs do not precisely overlap, but they both imply a broadly Kantian understanding of autonomy according to which there are (at least?) two levels of decision making. One (second order, critical) is more reflective, rational, or autonomous and the other (first-order, experiential) is more spontaneous, irreflective, and inconsiderate. Only the former represents the level of our genuinely voluntarily choices. Presumably, then, the choice to take up smoking is often if not always an instance of the latter.

Thus, a smoking ban would help most people make the choice they would autonomously make, if only they were in the condition to make autonomous choices.

This is a very problematic view, for several reasons.


What one really wants

The first thing to notice is that the Chief Medical Officer’s statement refers to some kind of freedom from addiction, but it conflates two different ways of obtaining that freedom:  quitting once one is addicted and never starting in the first place. These alternative options, however, refer to different freedoms, which makes it problematic to treat them as equivalent with regard to the way they allegedly promote freedom.

The ban would prevent someone from forming an addiction. Not having an addiction is a type of freedom – a psychological or a physiological freedom, if you want. But what if someone who is competent enough to make autonomous decisions for themselves autonomously chooses to start smoking and accepts the cost of becoming addicted? That is a different type of freedom, i.e. freedom from interference with one’s own lifestyle choices. This is the freedom that the ban removes and, more generally, it is the type of freedom whose infringement requires a justification when it comes to public health restrictions.

The justification, however, cannot be based on weighing freedom from State intervention against freedom from addiction, or from disease, to determine whether, overall, the policy promotes or undermine freedom. They are different, incommensurable types of freedom. I could freely, autonomously start smoking cigarettes right now, even if I know that I will soon be addicted, if no one prevented me from doing that. It might be a stupid decision, but it is a free one nonetheless. The freedom I enjoy when I start smoking is not reduced by the freedom I lose when I become addicted, and the freedom I lose when I become addicted is not increased by the freedom I have to buy cigarettes. These are different notions of freedom characterised by different ways of measuring them. Of course, one might be ethically or politically more important than the other, but that is a different matter.

However, let me say something that questions the idea4  that the type of freedom that addiction would remove is ethically and politically more important than freedom from State intervention. The problem is that it is difficult to see how that idea could be implemented without opening the door to undue State intrusion in personal decisions.

Perhaps young people are, generally speaking, not autonomous, responsible, competent enough when they choose to start smoking at 18 because they cannot take in due consideration the fact they will become slave of an addiction. But if so, the obvious solution would be to increase the legal age for buying cigarettes, not to ban smoking at any age.  Some people take up smoking later in life, after all. Surely 40-year-olds should be deemed autonomous and competent enough to make decision for themselves, including bad ones, if that is what they want. The same applies to all norms that set age limits. Maybe people are not mature enough at 18 to buy alcohol or to drive a car. Drinking or driving might put themselves (and others) at risks that, if they were more mature, they would not take. If so, the solution would be to increase the age limit, not to phase out alcohol or driving until they are banned for all age groups.

The obvious solution would be to increase age limits unless, of course, the assumption here is that starting smoking can never be an autonomous decision, at any age or in any circumstances, because the loss of freedom entailed by addiction always goes against individuals’ best interest. This assumption would lend more support to the smoking ban. However, it is a questionable line of argument, for two reasons.

First, it presupposes that future regrets or future limitations of freedom that follow from a certain decision (such as the limitation of freedom entailed by an addiction) indicate that the decision was not really free or autonomous in the first place. That is questionable. A man deciding to have a vasectomy removes his future freedom to have biological children and might regret the choice. That does not necessarily mean the choice to have a vasectomy was not genuinely free or autonomous. It only means it was a bad one. For this reason, it is not too relevant that the Chief Medical Officer refers to ‘most people’, and not to all people, when he says that those who smoke wish they had never started or that they could stop. Regret and incapacity to get out of addictions do not make the initial choice of those ‘most people’ unfree.

Even if they did, there is a separate question as to whether it is justifiable to remove a certain freedom from everyone (the freedom to start smoking) to protect a different freedom that only some, or even most, will care about (the freedom from addiction). Many people make an unhealthy use of their freedom to buy sugary products and take related risks, including the risk of future limitation of their freedom from disease (say, from diabetes). That does not seem to justify banning sugary products. In any case, this is a matter of distributive justice, as it raises a question about fairness in the way different types of freedom restrictions and limitations ought to be distributed across different individuals. Even if a ban was justified on these grounds, it would be on grounds of fairness, and not on grounds of protecting freedom (this is an important issue, which will be the theme of a Medical Humanities conference taking place in Oxford at the end of this month).

Second, the assumption that starting smoking can never be a genuinely free or autonomous decision equates genuine freedom with a certain objective standard of rational choice and of best interest. It is stipulated that only people who comply with that standard are really competent, autonomous, and ultimately free. For example, only people who prioritize health over anything else are genuinely free, autonomous, rational. In the name of such equation, the idea that a State is imposing certain choices on citizens (only) ‘for their own good’ – which is a problematic form of hard paternalism at odds with liberal values – would be artificially transformed in to the idea that the State is imposing certain choices on citizens for the sake of their own real freedom. The language of freedom gives arguments for restrictive health policies a liberal undertone. However, one might wonder if in these cases language conceals rather than reveal the actual nature of such policies.

Imbuing arguments for restrictive policies with the language of freedom risks opening the door to cheap justifications for liberty restrictions. After all, would you really want to drink that beer, if you knew how bad alcohol is for you? Would you really want to eat all that sugar, if you knew how high your risk of diabetes is?  Would you really want to refuse that vaccine, if you knew how bad a certain disease is? The list can be rather long. If we ask that kind of question about smoking, we need to ask it about many other things, whether or not they cause addiction. If we think that the answer is relevant to a justification of State-imposed restrictions, assuming one universal standard for what counts as autonomous and genuinely free choice might lead to conclusions at odds with liberal values.

In his criticism of the notion of ‘positive freedom’ as the freedom attributed to an idealized notion of the self, Isaiah Berlin warned against the danger of adopting what he called a ‘monstrous impersonation’ of the “real” self. That is the self as possessing a “higher nature” whose values and desires are different from the values and desire of actual individuals. As he wrote,

“the real self may be conceived as something wider than the individual (…), as a social whole of which the individual is an element or aspect (…). This entity is then identified as being the ‘true’ self which, by imposing its collective, or ‘organic’, single will upon its recalcitrant ‘members’, achieves its own, and therefore their ‘higher’ freedom’ (…). But what gives such plausibility as it has to this kind of language is that we recognise that it is possible, and at times justifiable, to coerce men in the name of some goal (let us say, justice or public health) which they would, if they were more enlightened, themselves pursue, but they do not, because they are blind or ignorant or corrupt”.5

Or perhaps, one might add, too young or too immature or too irrational to choose what they really want when it comes to lighting up that first cigarette.



 My concern here is with arguments supporting restrictive public health policies that appeal to the value individual freedom. We have seen them applied to recent pandemic restrictions and we are seeing them now applied to smoking bans. I have tried to suggest these arguments are very problematic.

Importantly, I have said nothing about other types of arguments for restrictive public health policies, such as those based on preventing harm to others or protecting public goods. Such arguments might be available to those who want to defend the smoking ban or any other public health restriction. That is up for discussion, but it is not a discussion I have got into here.

My point is simply that in offering justifications for restrictive public health policies, it is important to be clear about what values and principles are available in support of restrictions and which ones are not. This applies to bans on smoking as well as to any other ban or restrictions, for instance those on recreative drugs. I have suggested that, often if not always, individual freedom is not a principle in support of restrictions. At the risk of stating the obvious, if restrictions like smoking bans are ethically justified, it is in spite, and not because of individual freedom.



  1. Kieran Oberman , Freedom and Viruses, Ethics 2022, 132,4: 817-850
  2. g. Gerald Dworkin, The Theory and Practice of Autonomy (Cambridge:Cambridge University Press, 1988);Harry Frankfurt, Freedom of the Will and the Concept of a Person, in Frankfurt H., The Importance of What We Care About (Cambridge: Cambridge University Press, 1988), ch. 2
  3. Ronald Dworkin, Life’s DominionAn Argument about Abortion, Euthanasia and Individual Freedom (New York: Alfred Knopf 1993)
  4. Andreas T. Schmidt,  Freedom of choice and the tobacco endgameBioethics 2022, 36,1: 7784.
  5. Isaiah Berlin, Two Concepts of Liberty (1958), in I. Berlin, The Proper Study of Mankind, London: Vintage Books 2013, 191-242, pp. 204-5.
Share on

2 Comment on this post

  1. The paternalist paradigm is, for me, the key consideration. Specifically, the two forms in which it manifests. As mentioned, there is no sound argument to objectively determine what criteria a person needs to meet in order to be able to make a ‘voluntary’ free choice. And, as pointed out, the fact that the phrase; “most people” was used; importantly tells us that the choice has been made voluntarily by a subset of people (under their definition of voluntary choice). Assuming this, we can then ask; does regretting a decision make it involuntary? The answer almost certainly depends on factors pertaining to the individual at the time of making the decision. But we can also assume that regret itself does not constitute grounds for a decision being involuntary. Which implies there are instances where people make voluntary decisions which they later regret. Therefore, I would conclude, such a ban is an example of hard paternalism. I see this as an affront on liberal values.

  2. The public policies should be self-restrained. They should consider an individual to be free from public’s interventions.

    But we are on the slippery slope. We are afraid to be responsible for us.

    We rather want the state power to decide instead of us, to protect us, to supervise us. In some way it is more comfortable.

    It is all very essential and gradual menthal change in our behaviour. At the old times the individual was on top. Now it is over. On the top there is some general aim, like PUBLIC (not private) health, public order, public interest etc.

    The smoking ban seems to be minor thing. We would not talk about it if we had to solve more serious problems.
    So on the one hand this discussion about this kind of ban shows our welfare.

    But on the second we slowly move to the state guardianship. Because if the smoking is harmful why not eating hamburgers, drinking Coke or going to pubs?

    In Finland they want to ban smoking absolutely. These efforts are called “Game Over”. This title is quite right. At the end of our road to perdition there will be sterile, not-living, brave new world. Game over.

Comments are closed.