Article Text

Download PDFPDF

A libertarian case for mandatory vaccination
  1. Jason Brennan
  1. Correspondence to Professor Jason Brennan, Department of Strategy, Economics, Ethics, and Public Policy, Georgetown University, 3700 O St NW, Washington DC 20057-0004, USA; jason.brennan{at}


This paper argues that mandatory, government-enforced vaccination can be justified even within a libertarian political framework. If so, this implies that the case for mandatory vaccination is very strong indeed as it can be justified even within a framework that, at first glance, loads the philosophical dice against that conclusion. I argue that people who refuse vaccinations violate the ‘clean hands principle’, a (in this case, enforceable) moral principle that prohibits people from participating in the collective imposition of unjust harm or risk of harm. In a libertarian framework, individuals may be forced to accept certain vaccines not because they have an enforceable duty to serve the common, and not because cost–benefit analysis recommends it, but because anti-vaxxers are wrongfully imposing undue harm upon others.

  • Political Philosophy
  • Public Policy
  • Rights
  • Right to Refuse Treatment

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Many people in the USA process scientific information in highly irrational ways. As a result, many now refuse to vaccinate their children. As a result, there have been outbreaks of deadly, dangerous or debilitating diseases, such as pertussis and measles, diseases that were nearly eradicated in the USA.

This leads many to ask: Is it permissible for the government to force people to vaccinate themselves and their children? In this paper, I will argue that that even libertarians can and should endorse mandatory, that is, government-enforced, vaccinations.

This thesis, if correct, is of interest for two major reasons. For libertarians, the problem is that many sense there are strong grounds in favour of mandatory vaccination. Many, if not all, would want to avoid having to bite the bullet and say that mandatory vaccinations are unjust infringements of individual liberty.

For non-libertarians, the thesis is of even greater interest. If mandatory vaccination can be vindicated even within a libertarian framework, that shows that the case for mandatory vaccination is indeed very strong. Starting with a libertarian framework seems to load the dice against mandatory vaccination. Libertarians hold it is unjust and impermissible to force or coerce individuals to act to promote the common good. Some libertarians accept that citizens have stringent but unenforceable moral duties to promote the common good, but they reject the view that governments may coerce individuals to do so. Libertarians place high value on individual rights and autonomy, and see rights as stringent side constraints that prohibit certain ways of promoting the common good (pp. 35–6)1 (pp. 28–34).2 Libertarians will not accept mandatory vaccination simply because cost–benefit analysis recommends it, that is, because the goods outweigh the harms. But suppose we can show that even libertarians can accept or champion mandatory vaccination. That means that the policy can be justified even if we start with premises that appear to load the philosophical deck against it.

As an analogy, consider one reading of David Gauthier's Morals by Agreement.3 Gauthier wants to show that it is rationale to accept moral constraints. But by starting with, and trying to ground morality on, value subjectivism and the assumption of egoism, he thereby begins with a starting point that loads the deck against morality. Someone who thinks these are implausible starting points might think the book is of little interest. However, an alternative way of thinking of the book is that if, at the end, Gauthier shows that moral constraints can be justified even with such premises, then the case for morality must be strong indeed. This paper means to do something similar on behalf of mandatory, government-enforced vaccination. I am not arguing that libertarianism is true, but rather that one can defend mandatory vaccination even if one begins with strongly individualist and antigovernment premises.

For the purposes of this paper, let us assume the vaccines in question (A) are highly effective, (B) have a low incidence of side effects, (C) protect against serious illnesses, and (D) that the evidence for A–C is strong and widely available, such that the overwhelming majority of people who might dispute A–C would be unjustified and epistemically irrational for doing so. Following the vaccination schedule recommended by the American Pediatric Council, this list would include vaccines against hepatitis A and B, rotavirus, diphtheria, tetanus, pertussis, poliovirus and varicella, among others.4

For the sake of simplicity, I discuss universal mandatory vaccination. But, of course, realistically, for most diseases, all that matters is that a sufficiently high and sufficiently well-distributed subset of the population receives vaccines. If enough people voluntarily accept such vaccines, then of course mandatory vaccination is not justified. If only a subset is needed, then the state might select citizens by lottery, just as it selects jurors.

What is libertarianism?

‘Libertarianism’ refers to a range of closely related political views. What libertarians have in common is the view that each individual is endowed with an extensive set of strong rights against interference in her personal and economic decisions, and, as a result, the permissible scope of government is quite limited (pp. 5–6).1 Some libertarians, especially those outside North America, eschew the term ‘libertarianism’ and instead call themselves ‘liberals’ or ‘classical liberals’.

Libertarianism is sometimes described as the view that all individuals are self-owners, combined with the view that what follows from rights of self-ownership is that individuals have nearly absolute rights against interference, including interference with their property rights in external goods (which, somehow, derive from self-ownership).5 But, as Peter Vallentyne and Bas van der Vossen note, this is at most one major branch of libertarian thought:In the most general sense, libertarianism is a political philosophy that affirms the rights of individuals to liberty, to acquire, keep, and exchange their holdings, and considers the protection of individual rights the primary role for the state. This entry is on libertarianism in the narrower sense of the moral view that agents initially fully own themselves and have certain moral powers to acquire property rights in external things.6

This kind of libertarianism—perhaps best illustrated by Robert Nozick—derives from Lockean foundations. But not all libertarians are Lockeans. Some are Aristotelians,7 some are Rossian pluralists,8 ,9 some are natural rights theorists, some are consequentialists,10 some are Rawlsians11 and still others have different background moral theories. As Matt Zwolinski summarises:There is no single theory that can be safely identified as the libertarian theory, and probably no single principle or set of principles on which all libertarians can agree. Nevertheless…libertarians are generally united by a rough agreement on a cluster of normative principles, empirical generalizations, and policy recommendations. Libertarians are committed to the belief that individuals, and not states or groups of any other kind, are both ontologically and normatively primary; that individuals have rights against certain kinds of forcible interference on the part of others; that liberty, understood as non-interference, is the only thing that can be legitimately demanded of others as a matter of legal or political right; that robust property rights and the economic liberty that follows from their consistent recognition are of central importance in respecting individual liberty; that social order is not at odds with but develops out of individual liberty; that the only proper use of coercion is defensive or to rectify an error; that governments are bound by essentially the same moral principles as individuals; and that most existing and historical governments have acted improperly insofar as they have utilized coercion for plunder, aggression, redistribution, and other purposes beyond the protection of individual liberty.12

Jason Brennan similarly says that libertarianism is ‘not one uniform political philosophy, but rather a family of related philosophies’; what these philosophies have in common is that they conclude (on the basis of various moral or social scientific arguments) that individuals have an expansive sphere of stringent economic and civil liberties, and that respecting such liberties requires a relatively small role for government (p. 5).1

Libertarians may have any range of views about what moral obligations individuals have to be beneficent, to act on behalf of others or to exercise civic virtue. Some libertarians are Randian ethical egoists, and thus reject the view that individuals owe positive duties to one another, but probably most libertarians are not Randian egoists (pp. 20–1).1 Indeed, many libertarians are part of the effective altruist movement, a movement that among other things holds that individuals ought to donate significant amounts of their income to charity. However, libertarians generally resist the view that individuals may be forced to work for the good of others. A libertarian might agree with, for example, Peter Singer13 that we have overwhelming moral duties to assist others, and so on, but will resist saying that one can be forced to do so.

Two false starts: trumping rights and paternalism

Libertarians believe that every individual has an extensive set of negative civil and economic rights. Some libertarians think rights are absolute, but most do not. Rather, they hold that such rights can be overridden to avoid disaster or ‘catastrophic moral horror’ (pp. 40–1, 111–12)1 (p. 30).2 So, for instance, suppose you can stop a madman from unleashing a nuclear bomb on our city, but only if you steal my car. Most libertarians would accept that you would be justified in taking the car, even over my protests to the contrary. You cannot steal my car just because you need it more than I do or because you need it to help you volunteer at the food bank, but at some point, your reasons for taking the car are so demanding that they outweigh my claim right against you taking it (pp. 137–141).14

One might think that so long as libertarians are not absolutists, then it will be relatively easy to justify mandatory vaccination within a libertarian framework. The argument might go as follows:

  1. While individuals have strong presumptive rights to refuse medications, these rights are not absolute. They can be overridden to prevent disaster.

  2. It would be a disaster if no one, or if a large majority of individuals, failed to receive various vaccines against dangerous illnesses.

  3. Therefore, it is permissible to force individuals to receive certain vaccines against dangerous illnesses.

Consider an extreme case as an illustrative example: suppose a zombie apocalypse virus, like in 28 Days Later or World War Z, has begun infecting people. By hypothesis, allowing this disease is a disaster, greater than any disaster humanity has experienced. Suppose governments can stop the disease, but only if they forcibly vaccinate everyone. It seems the need to prevent the zombie apocalypse overrides individuals' rights to refuse medication.

In principle, this argument might work, but as of now, it probably does not do the work we need it to do or it would do that work only if things got much worse. Consider the situation in the USA right now: most Americans receive vaccines. A sizeable (and growing) minority refuses to do so, and as a result, certain diseases are once again spreading. But while the spread of such diseases is bad, the number of cases, and especially the number of resulting deaths or long-term complications, is still statistically very low, and so the long-term consequences are still not quite bad enough to be called a disaster.

Arguably, we are not yet at the threshold of badness in which the need to avoid disaster trumps individual rights. If very few citizens were being vaccinated, or if the diseases were much worse, this ‘Avoiding Disaster Trumps Rights’ argument might work. But without here trying to defend a theory of just where the line between bad and disastrously bad is, it seems unlikely that most libertarians would or must accept that recent outbreaks qualify as a rights-trumping disasters.

Advocates of mandatory vaccination might instead attempt to defend vaccination on paternalist grounds. ‘Hard paternalism’ refers to government policies that coerce citizens into performing certain actions, or avoiding other actions, for their own good.

But of course that is a non-starter for a libertarian. Libertarians believe that we must respect adults' rights to make stupid, self-destructive choices, though these decisions are, by hypothesis, stupid and self-destructive. Libertarians might accept that governments or other agencies should advertise that vaccines are effective and safe, and they might think it fine for doctors to refuse to take on patients who refuse vaccinations. But they would stop short of requiring that adults be vaccinated.

However, most of the important vaccines are vaccines for children rather than adults, and that significantly changes the paternalist argument. While libertarians (and liberals in general) might be averse to forcing adults to take medicine for their own good, it is not as obvious that the state cannot force parents or guardians to vaccinate their children for the children's good. But whether this kind of argument can be made to work is a tricky question. We have to determine what sorts of negative rights children have as well as what positive rights they hold against their guardians, and at the same time determine to what degree the state and children's guardians have rights to a say over what happens to the children. Whether mandatory childhood vaccines can be defended on paternalist grounds is further complicated by the fact that vaccines are largely a collective action problem. It is not literally true that any particular child must be vaccinated to protect that child from disease; rather, what matters is that a sufficiently high enough percentage of the population be vaccinated.

I will remain agnostic here about whether paternalism for children can (1) be shown to be compatible with libertarianism and (2) be used to justify mandatory vaccination. Instead, let us turn to a more promising line of argument, which holds instead that mandatory vaccination is permissible to protect individuals from the imposition of undue risk of harm.

Stopping harm to others

While libertarians reject coercive paternalism or using coercion to force adults to help one another, they do accept using coercion to prevent people from wrongly harming each other. To be clear, libertarians (and liberals in general) believe we sometimes have the liberty to hurt each other in certain ways. So, for example, they hold that speakers have the right to engage in ‘hate speech’, even if it hurts victims,15 ,16 or that businesses have the right to impose pecuniary externalities on their competitors, even though this may put their competitors out of business and leave them destitute (p. 143).17 However, we may not harm each other by initiating violence, poisoning one another or committing fraud. So, libertarians have to ask themselves some hard questions: Do individuals have a right to expose others to dangerous diseases? Is it ever permissible to use coercion to stop people from exposing one another to dangerous disease?

Suppose Andy has been infected with a dangerous disease, for example, a zombie-apocalypse virus. I expect most libertarians would accept that people have a right not to be exposed to this disease and would allow for Andy to be forcibly quarantined to stop the spread of the disease. They would not think Andy has the right to impose such a serious risk of harm upon other people.

Thus, one might think one promising way to justify mandatory vaccination, consistent with libertarianism, is to argue that vaccines are a way of preventing individuals from imposing unjust risk of harm on another. And, indeed, I will something like this is the way to go. But this line of argument faces some serious complications.

Consider the following cases from Jessica Flanigan (pp. 7–8):18

  1. During a fourth of July celebration, your neighbour shoots his gun into the air indiscriminately. A falling bullet lodges in your shoulder.

  2. Suppose a friend brings her children who are infected with whooping cough to a party. She has not vaccinated her children. As a result of being exposed to her sick kids, your children get sick. (Your children are vaccinated, but the vaccine is only about 85% effective.)

Flanigan thinks (and I agree) that it is clear that the shooter in case 1 does something wrong. It would be permissible to use coercion to stop the shooter from firing indiscriminately in the air. Further, she claims, cases 1 and 2 are sufficiently analogous, so if it is permissible to use coercion to stop the shooter from imposing undue risk of harm onto others, it should be permissible to stop the mother from imposing undue risk of harm. In that case, though, to stop her from imposing such risk, we must require her to vaccinate her children. As Flanigan summarises:As in the case of celebratory gunfire, the non-vaccinators harm and impose risks on their neighbors. In both cases, the shooters and the non-vaccinators may never see the harm they cause to others. Both shooters and non-vaccinators may feel justified in exposing people to small risks of getting shot or infected with a contagious illness for the sake of their own freedom to fire guns or to refuse vaccination. Yet, neither shooters' nor non-vaccinators' rights entitle them to harm others, despite the fact that the risk of harm is of low-probability, their victims are unlikely to identify them, and they do not intend to injure their victims (p. 8).18

Flanigan explores a number of purported disanalogies between cases 1 and 2, and finds these purported disanalogies are either not present or do not make a moral difference. I will not explore her arguments here. Rather, the hardest problem for Flanigan's argument, even though I agree it is on the right track, has to do with the fact that the herd immunity that results from vaccination is largely a collective action problem. Unvaccinated people are not like bioterrorists purposefully trying to infect others. Many contagious people are unaware that they or their children are carrying an infectious disease. Some are like the mother in case 2 above, who should have realised her children were infected with whooping cough and should have taken steps to protect others from her children. But many times, those who carry and pass on infectious diseases are simply unaware they are doing so.

Further, vaccines are not like quarantines. Vaccines do not stop infected people from infecting others. Rather, they prevent people from becoming infected in the first place. Quarantines restrict people's liberty, but they are justified (in certain cases) because these people present a clear and present danger to others, or at least are at a very high risk, as individuals, of presenting a clear and present danger to others. But mandatory vaccination forces individuals who are not a clear or present danger to others nor are at a high risk of being a clear and present danger to accept a vaccine against their will. When a large group of people refuses vaccines, the group may impose a risk, but we cannot easily attribute the risk to any individual within the group. While quarantines focus specifically on dangerous individuals (because they are infected or are likely to be infected), mandatory vaccines are targeted at everyone. But the problem is that individuals as individuals make little difference. If everyone in the world were vaccinated except for Andy and Betty, Andy and Betty would pose no real threat to each other. Instead, vaccination presents a collective action problem, in which individuals as individuals are unimportant.

The clean hands principle

Libertarians, and liberals in general, do not hold that all morally wrong actions may be coercively prohibited. Instead, they hold that we sometimes have the right to do wrong. For instance, it would be wrong for me to, for no reason at all, forbid my parents from ever again seeing their grandchildren. It would be wrong for Harry to, just out of cruelty, to jilt Sally at the alter. It would be wrong to write a book advocating genocide. Nevertheless, libertarians and most liberals would conclude that the state may not enforce morality in these cases; in these cases, it may not use coercion to make individuals do the right thing.

However, libertarians (and liberals in general) hold that some moral duties are enforceable. The state cannot use coercion to make me let my parents visit their grandchildren, but it can coerce me into feeding my kids. It may not coerce Harry to marry Sally, but it may coerce Harry to stop him from attacking Sally. It may not coerce me to stop me from advocating genocide, but may coerce me to stop me from actually engaging in genocide. Without here offering a full theory of why some duties are coercively enforceable and others not, I expect libertarians would hold that the state can use coercion to stop someone from actively spreading a dangerous disease. Doing so stops the individual from causing harm; it is not an instance of forcing an individual to help others.

The trick here is to justify forcing individuals to accept mandatory vaccination, even though we usually lack sufficient information to blame specific individuals for spreading dangerous diseases. Instead, at most we usually have sufficient information to blame the group of unvaccinated individuals, rather than any individuals per se inside the group. We can rarely point to individual parents and say, “Your refusal to vaccinate your kid caused other children to become sick.” In general, individual decisions to vaccinate or not have negligible effects on others. What matters is what most people do, not what individuals do.

To get around this problem, I propose to borrow a line of reasoning from Jason Brennan's The Ethics of Voting,19 a line of reasoning meant to explain how individuals can be blameworthy for participating in collective actions in which the group causes harm. He asks us to consider the following case:firing squadA band of 10 sharpshooters is about to kill an innocent child. They have been trained to shoot in such a way that each shot will hit the child at the same time, and each shot would be fatal on its own. You can't stop them from killing the child. They ask you if you'd like to join in and take the 11th shot (p. 72).19

Firing Squad presents a classic case of overdetermination. Brennan calls this a case of a collectively harmful activity, that is, an activity in which a group causes harm to another, but in which individual inputs into the group activity are negligible, in the sense that the harm will occur regardless of whether any individual participates or not (p. 71).19

Most people have the strong intuition that its wrongful to join in with the sharpshooters and shoot the child, even though, by hypothesis, the child is doomed anyways. Further, in this case, most people would accept that it is permissible to use coercion to stop you from taking the 11th shot, even though this will not save the child. What explains this, Brennan argues, is what he calls the ‘clean hands principle’: there is (sometimes enforceable) moral obligation not to participate in collectively harmful activities (p. 74).19 A collectively harmful activity is ‘a harmful activity caused by a group or collective, where individual inputs into the harmful action are negligible’ (p. 71).19 For instance, in the Firing Squad Case, whether an individual sharpshooter decides to participate or not makes no difference; the child will die either way.

It seems plausible that there is an obligation not to participate in collectively harmful activities. This provides one plausible explanation for the intuition that participating as the 11th shooter in Firing Squad is wrong. Further, as Brennan explains, the ‘clean hands principle’ seems to be a corollary of many background moral theories (pp. 73–4).19 Act utilitarians would hold, trivially, that it is generally wrong to participate in collectively harmful activities because of course one could instead do something that produces positive utility. Rule utilitarians hold that the correct moral principles are those principles that, when internalised by the overwhelming majority of people, lead as a whole to the best consequences.20 It seems plausible that such a code would include at least a pro tanto norm against participating in collectively harmful activity. (After all, a code that allows people to join in and cause collective harms in cases of overdetermination will produce lower utility than a code that prohibits it.) Further, it seems unlikely that ‘I will feel free to participate in overdetermined harms’ would be universalisable in Kant's moral theory. And so on.

Brennan further extends this line of argument to cover not only cases of foreseeable and known harm, but also ‘unacceptable risk’ (p. 81).19 Here, he follows Sven Hansson21 in holding that it is acceptable to expose someone to risk ‘if and only if this exposure is part of an equitable social system of risk-taking that work to her advantage’ (p. 81).19 Hansson's idea is that for us to live together and benefit from social cooperation, we must be able to impose some degree of risk of harm upon one another, by, for example, driving, flying, going to the store despite possibly having a cold or emitting at least some pollution. The idea here is that we ask what set of risk-permitting rules would tend to benefit everyone as individuals. Hansson's principle explains why we are free to expose people to risk through ordinary driving, but not allowed to, for example, leave a bomb on the street that has a 1-in-10 000 chance of exploding. A rule permitting driving (in ordinary ways) is expected to be to everyone's advantage, though it carries risk, while a rule permitting bombing is not.

Libertarians, of course, need not accept Hansson's particular theory of ‘acceptable risk’. However, they will need to have some such theory. As Robert Nozick asks (somewhat awkwardly phrased), ‘Imposing how slight a probability of a harm that violates someone's rights also violates rights?’ (p. 74).2 Almost everything a person does impose some risk upon others. Just by walking through the hallways at my university office building, I impose some risk of injuring and killing an innocent bystander, or of spreading the common cold or flu. Libertarians need some theory to differentiate between permissible and impermissible risk imposition. Whatever that theory might be, presumably libertarians will want to allow ordinary, unimpaired driving, but forbid, say, leaving a probabilistic bomb in the middle of the street.

Brennan argues that the clean hands principle not only forbids participating in collectively harmful activities, but also forbids participating in the collective imposition of unacceptable risk. To illustrate, consider this modification of Firing Squad:probabilistic firing squadThe sharpshooters from firing squad have captured a second child. They will roll 10 10-sided dice. If the resulting sum is exactly 21, they will shoot the child (with 10 simultaneously striking, fatal shots). Otherwise, they will let the child go. They offer to have you shoot with them if the roll results in a 21.

Brennan would conclude that it is at least pro tanto wrongful to participate as the 11th (or millionth) shooter in this case. Intuitively, that seems right: it seems plausible that one should not participate in such an activity. Further, it seems plausible that the state could intervene coercively to stop the sharpshooters, even though by hypothesis none of them as individual actors make any difference.

Let us modify this case again. Suppose the shooters were planning—on a roll of 21—to fire uranium shells at the walls of CDC BSL-4 laboratory, knowing that in doing so they risk spreading the diseases contained therein to the general population. The clean hands principle forbids this action.

Now consider one final variation, one that more closely approximates what unvaccinated people are doing.reckless astronautsElon Musk has just invented instantaneous interplanetary teleportation, and the technology is widely available. Suppose a group of privately-funded astronauts plans to visit a newly discovered planet, a planet that, for all they know, contains a wide range of deadly bacteria and viruses. When they arrive, they drink the water, without sanitizing it. They also give the possibly contaminated water to their children. When they arrive back home a day later, they refuse quarantine. Some of them visit Disneyland, while others immediately place their (for all they know, infected) children in daycare centers or schools. They could have taken steps to sanitize the water samples and to prevent themselves from contracting any alien diseases, but they decided not to do so, because they get their health advice from Jenny McCarthy.

In this case, by refusing to quarantine themselves and their children, the astronauts are not simply failing to help others. Rather, they are

  1. actively exposing themselves to potentially dangerous diseases, and then

  2. actively doing things that have a high likelihood of spreading these diseases to others, and

  3. actively choosing not to take steps to reduce the risks they are imposing.

Further, these conditions also obtain:

  1. They are not epistemically justified in 3; they lack grounds for refusing to take precautions, though they mistakenly believe themselves to have such grounds.

  2. The risk they impose on others cannot be justified as part of an equitable social system of risk-taking as it provides no advantage to others.

In this case, libertarians could hold that it is justifiable to use coercion to protect others from these astronauts. The astronauts, as a collective, are not simply refusing to work for the common good or for others' benefit. Instead, they are imposing unacceptable and unjust risk on others.

Now, to complete the argument, one just needs to show that anti-vaxxers in the USA are relevantly similar to the reckless astronauts. On the face of it, they seem rather similar: conditions 1–5 obtain for both.

Of course, there are some disanalogies. The reckless astronauts are exposing themselves and others to a completely unknown risk—it is possible that the alien planet's bacteria and viruses will turn out to pose no threat at all—while anti-vaxxers, as a group, impose a known risk. This particular disanalogy, though, seems to make the case against anti-vaxxers stronger than the case against the astronauts.

Another disanalogy is that the reckless astronauts are an organised group, while anti-vaxxers are a disparate and disconnected mass of individuals. This might make some difference in attributing intentionality to the group's behaviour, but it is not obvious that it makes a difference in attributing responsibility. After all, individual anti-vaxxers can still determine, with a minimal amount of research and a basic degree of epistemic rationality, that their collective behaviour causes an undue imposition of risk of harm to others. They might not act as a tight group making a deliberate effort, but they can, with minimal epistemic effort, recognise that they are part of this dangerous group. Anti-vaxxers are not in general people who have failed to put in epistemic effort in determining that vaccines are safe. A person who simply does not pay much attention to medicine would normally just following the vaccination schedule her physician suggests. Rather, anti-vaxxers nearly always put in active epistemic effort to reason about vaccines in an irrational way. They tend to seek out misinformation and actively reinforce each other in reasoning about vaccines in irrational ways. So, while anti-vaxxers are not a fully organised group like the reckless astronauts, individual anti-vaxxers are in general people who actively chose to behave in certain reckless ways, rather than, say, people who simply failed to choose to behave in certain publicly beneficial ways.

Libertarianism and the clean hands principle

One important question is whether libertarians really can, or even must, accept the clean hands principle. Brennan happens to be a type of libertarian; that the cleans hands principle comes from a libertarian thinker is at least some evidence that it is compatible with libertarianism (p. 25).1

Of course, that is only weak evidence. Many people have inconsistent views, and self-described libertarians might end up endorsing principles incompatible with some of their background commitments. That holds for any philosopher in any tradition. A Marxist might end up endorsing ideas incompatible with Marxism or a left-liberal might end up endorsing illiberal ideas.

However, we can add a further argument here showing that libertarians not only can endorse the clean hands principle, but probably must. The clean hands principle is meant to explain why it is wrong to participate in collectively harmful activities, where collectively harmful activities are activities that cause wrongful harms, but the outcome is overdetermined, and so any individual agent could withdraw from participating without this, thereby stopping the harm from occurring.

Suppose a libertarian refused to endorse the clean hands principle. Suppose this libertarian, out of a commitment to methodological individualism, said that collectively caused harms cannot in any way attributed to the individuals who participate in the collective, that there is no moral duty to avoid participating in collective harms and, further, that it is always impermissible to use coercion to stop people from participating in collective harms. If the libertarian claimed all that, then her position would compatible with radically antilibertarian policies, provided these policies were enforced through overdetermined collective action.

An innovative fascist, imperialist, warmongering, socialist and paternalistic political party could then say, “We'll engage in war, censorship, mass murder, and impose high taxes, mass regulation of the economy, and ramp up the war on drugs, all things which libertarians hate. But, to make sure we don't run afoul of libertarianism, we'll always impose these action and rules through collective activities. For instance, when we shut down libertarian website, we'll train a team of censors to do so simultaneously, such that the actions are always overdetermined, as in the Firing Squad case!”In order to avoid this absurd result—that is, that enterprising fascists could avoid running afoul of libertarianism just by making sure they always have collectives rather than individual government agents engage in rights violations—it seems that the libertarian must endorse something like the clean hands principle, and, further, must accept in some cases it is permissible to use coercion against the individuals who constitute the collective performing the rights violation or causing the harm.

The government failure objection

This section responds to what might be the most powerful libertarian-minded objection to the argument above.

In the introduction, I asked readers to assume that the vaccines in question (A) are highly effective, (B) have a low incidence of side effects, (C) protect against serious illnesses and (D) that the evidence for A–C is strong and widely available, such that the overwhelming majority of people who might dispute A–C would be unjustified and epistemically irrational for doing so.

A libertarian, or really any political economist with a basic awareness of public choice economics,i might note that in arguing for mandatory vaccines, I implicitly rely upon another assumption. Call this E: when a government is imbued with the power to force people to accept vaccines, it will use that power in a just, competent and public-spirited way.

Libertarians will say that this is a dangerous and often silly assumption. On the contrary, they would claim, just as markets can fail, so can governments. Governments may be entrusted with power in hope that this power will be used to promote the common good. However, libertarians claim, government agents might use that power incompetently or in bad faith.

On this point, a libertarian might argue that if a government has the power to control health issues, pharmaceutical companies and others can and will compete to control that power for their own ends. If governments can force people to take certain medications, then pharmaceutical companies will have strong incentives to engage in socially destructive rent seeking. In the most neutral sense, ‘rent seeking’ refers to non-voting, non-criminal activities that individuals or firms engage in with the purpose of either changing the laws or regulations, or how the laws and regulations are administered, for the purpose of securing a benefit. A firm engages in rent seeking when it seeks to gain an economic privilege or advantage from governmental manipulation of the market environment. Rent seeking occurs when agents seek to gain not by producing and trading value for value with others, but by rigging the competitive environment in their favour.

In short, if we give politicians the power to force us to take certain medications, politicians will have at least some incentive to use this power to reward their allies and donors at the expense of the common good. Libertarians are quick to remind us that governments are staffed by real people, rather than justice fairies. It is one thing to justify giving public-spirited, competent bureaucrats or politicians a degree of power; it is quite another to justify giving real-world bureaucrats or politicians that power.

For a specific instance of this worry, consider the vaccine Gardasil, which immunises patients against certain strains of the human papillomavirus (a virus that can cause genital warts and cancer). In 2011, some Tea Party conservatives in the USA accused Texas Governor Rick Perry (who was attempting to secure the Republican nomination for president) of selling out to Merck, the manufacturer of Gardasil. Perry supported the vaccine, but had received a $5000 campaign contribution from Merck, which had also contributed hundreds of thousands of dollars to the Republican Governors Association.23 However, even in this case, the accusations seemed to be little more than that. As Rick Perry himself argued, he received millions of dollars in contributions from various sources, and so it was implausible he could be ‘bought’ for so little.23 Further, empirical work on rent seeking generally finds it results from postelection lobbying rather than campaign contributions per se.22 ,24

In the abstract, the government objection might be powerful. But it does not obviously justify rejecting all government power, even the major proponents of the government failure objection agree.25 The point of identifying the main sources of government failure is not to then say that governments fail severely all the time. Rather, the point is to caution political thinkers to weigh government success against government failure, as well as market success against market failure. Public choice theory recommends a cautious approach to imbuing governments with power. Programmes should generally start small and be scaled up only as it becomes clear that government failures will not be sufficiently bad.

While I take the government failure argument seriously, historically, it appears that mandatory vaccination regimens work extremely well, and that government failures in this particular endeavour are relatively minor. While there is a vast literature showing ample rent seeking in, for example, agricultural policies or even in other aspects of medicine, I could find no papers showing serious rent seeking in mandatory vaccination.ii Thus, this objection is not strong enough to prohibit mandatory vaccination.


One way to test whether a policy can be justified is to start with a background political theory that would seem to make it difficult to justify such a policy. If the policy can be justified despite that, then that is a strong count in favour of the policy.

In this paper, I have argued that even with libertarian assumptions one can justify mandatory, coercively enforced vaccinations, not on paternalistic grounds, but instead on the grounds that individuals may be stopped from participating in the collective imposition of unjust risk of harm. This argument ultimately has a libertarian source, and here I just apply it to on behalf of mandatory vaccination. If mandatory vaccination can be vindicated despite beginning with premises one would think would show it unjustified, then the case for mandatory vaccination must be strong indeed.



  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • i Public choice economics is the subfield of economics that uses the tools of microeconomics to explain political behaviour. (‘Public’ choice is contrasted with ‘private’ choice, ie, most choices made on a market in which the costs and benefits of decisions tend to be internalised.) The basic insight of public choice is that individuals working for governments or non-profits are not in general less selfish or corruptible than individual consumers or producers in a market. Accordingly, when asking what governments or non-profits are likely to do, researchers should not presume that agents will only or even predominantly aim to promote the common good. See Mueller22 for an overview.

  • ii For example, the journal Public Choice has a paper explaining coordination in eliminating smallpox, but no papers showing that rent seeking is a serious issue in government-mandated vaccinations.

Linked Articles

Other content recommended for you