Proponents of vaccine mandates typically claim that everyone who can be vaccinated has a moral or ethical obligation to do so for the sake of those who cannot be vaccinated, or in the interest of public health. I evaluate several previously undertheorised premises implicit to the ‘obligation to vaccinate’ type of arguments and show that the general conclusion is false: there is neither a moral obligation to vaccinate nor a sound ethical basis to mandate vaccination under any circumstances, even for hypothetical vaccines that are medically risk-free. Agent autonomy with respect to self-constitution has absolute normative priority over reduction or elimination of the associated risks to life. In practical terms, mandatory vaccination amounts to discrimination against healthy, innate biological characteristics, which goes against the established ethical norms and is also defeasible a priori.
- philosophical ethics
- public health ethics
- right to refuse treatment
Data availability statement
There are no data in this work.
This article is made freely available for use in accordance with BMJ’s website terms and conditions for the duration of the covid-19 pandemic or until otherwise determined by BMJ. You may use, download and print the article for any lawful, non-commercial purpose (including text and data mining) provided that all copyright notices and trade marks are retained.https://bmj.com/coronavirus/usage
Statistics from Altmetric.com
Moral logic of harm prevention
Arguments in favour of mandatory universal vaccination rely on the premise that everyone who can be vaccinated has a moral obligation to do so for the sake of those who cannot be vaccinated due to age or certain immune system disorders, or because the public health benefits of universal vaccination are so profound that to refuse vaccination would be unethical. This line of reasoning underpins the ‘obligation to vaccinate’ (OTV) range of arguments. Brennan1 formulates a broadly representative OTV-type argument in terms of an ‘enforceable moral principle that prohibits people from participating in the collective imposition of unjust harm or risk of harm’. Brennan begins by positing that (A) certain vaccines have a low incidence of side effect and are effective at preventing serious illness; (B) it would be a disaster if a large majority of individuals failed to receive various vaccines; (C) individual freedoms can be overridden to prevent a disaster; therefore, (D) ‘it is permissible to force individuals to receive certain vaccines against dangerous illnesses’. The argument seems to imply that non-vaccination is a sufficient condition of a disaster (would make the disaster imminent) that mandating mass vaccination is a sufficient condition of preventing the disaster, and that the overriding of individual freedoms could not result in a disaster of a different kind. None of these conditions can be assumed to be true. We could also infer from the above premises that anything conceived of as harm could be classified as a ‘disaster’ and this would automatically give someone a legitimate right to override the freedoms of others, but this is absurd. We must, therefore, conclude that C is false: individual freedoms cannot be overridden just to prevent a disaster. Restrictions on freedoms can be justified only if they are reasonably necessary to preserve what makes human life worth living, because freedom is a necessary condition of a life worth living and, therefore, worth preserving. This is a conceptually appealing formula, but since the criterion of reasonable necessity is as elastic as the notion of disaster, it does not tell us much about practical moral obligations.
Brennan sidesteps this problem by proceeding to hone an OTV-type argument just in virtue of preventing ‘the collective imposition of unjust harm or risk of harm’. The reference to unjust harm makes his moral premise intuitively true but also compels us to identify the underlying injustice. Given that the existing vaccination technology is not risk-free (even if serious adverse reactions are rare) the alleged moral OTV implies that we have an obligation to reduce the risk to the health of others by accepting an increased or unknown health risk to ourselves. If I must accept an increased risk to myself in order to reduce the risk to others, because everyone has a moral obligation to do so, then justice demands that others must also accept an increased risk to themselves in order to reduce the risk to me, therefore, contradiction. This impasse can be resolved only by taking into account what set of ‘risk-permitting rules would tend to benefit everyone as individuals’1; a crucial question to which I will return.
Some authors2 defend OTV by appealing directly to the principle of primum non nocere (first of all, do no harm). The said principle, in order to be consistent, must apply not only to my actions with respect to others but also to the actions of others with respect to me and, arguably, to my actions with respect to myself. It is often overlooked in this normative context that mandatory vaccination violates body autonomy and thus constitutes actual harm (not merely a risk of harm) to any person made to accept vaccination under duress. This type of harm is not negated by any positive health effect of the procedure but constitutes a distinct category; it affects the ontological dimension of personhood. The threshold of reasonable necessity for medical coercion would have to be proportional to this harm and supported by a clear causative link between non-vaccination and serious harm to others. More formally, in order to justify coercion to vaccinate one would have to show that non-vaccination of X is a necessary and sufficient condition of an increased risk of harm to Y that exceeds the risk of harm to X associated with coercive vaccination, and that correcting this asymmetry of risks is reasonably necessary to preserve what makes human life worth living. A prospective benefit to public health does not of itself entail a reasonable necessity to infringe on personal body autonomy, which is one of the necessary conditions of a life worth living. Moreover, those who cannot be vaccinated have ways of mitigating the risk to their health other than by vaccinating everyone else, which of itself undermines the premise that coercive vaccination is reasonably necessary. Dubov and Phung3 contend that special normative criteria apply to healthcare workers. The choice to work in a healthcare setting comes with a set of ethical obligations, which include placing patients’ interests above our own. “When one decides to become a healthcare provider, he or she automatically decides to make certain sacrifices and assume some personal risks that come with this profession." Nevertheless, vaccination is not an occupational risk that is necessary for the effective performance of healthcare duties and the range of acceptable risks associated with a profession is not set in stone. “More than half of medical professionals decline annual influenza vaccination if given the choice. It is demeaning to assume that they are all uninformed or irresponsible.”4
Most research papers on vaccine ethics assume that vaccines are safe (which implies that the risks are negligible) and effective, and are therefore a public good that it would be irrational to refuse. This assumption is at best unproven and sometimes contrary to the evidence.4 Critically, the safety of the vaccines currently used for mass immunisation was not established via saline placebo controlled randomised trials in previously unvaccinated individuals.i The difficulty of establishing a clear causative link between vaccines and any late-onset health conditions, and the fact that vaccine manufacturers are typically not liable for the adverse effects of their products, allows the industry to give absolute priority to profits over consumer safety: this moral hazard constitutes an indirect health risk.
Herd immunity and the free-rider dilemma
The public health approach to OTV relies on the premise that there would be a statistically significant health benefit (herd immunity) to those with deficient immune systems if everyone else were vaccinated against a particular illness. This premise is disputed by some,4 but for the analytical part of my argument I will assume the most favourable view of herd immunity: an unequivocal public good. Nevertheless, the use of coercion or discrimination to achieve herd immunity faces a formal ethical dilemma. The risks associated with vaccination are not distributed in the same way as the benefits of herd immunity, with the vaccinated taking on all the risk whereas the immunodeficient partake equally in the public health benefit. Such ‘free-riding’ is legitimate as long the associated risk-taking is consensual, but would be arguably unethical in the case of mandated vaccination: the fact that an immunodeficient person is more at risk than others does not oblige anyone else to take on more risk for that person’s benefit, even if everyone were to get the same benefit. This logic extends to the issue of unequal distribution of harm, burdening some people with serious medical problems or even death due to vaccination while others reap the benefits.5 Furthermore, if the adverse reactions are a result of genetic traits then the initial risk is also unequal; some people may be able to take any vaccine with no negative consequences to their health while others could be incapacitated for life. In effect, some people may be made to pay a price that greatly exceeds their share in the associated public good—an absurd outcome. The public health approach to coercive vaccination cannot be deemed ethical if it is conditional on unfair or absurd treatment, especially if it involves a mandatory ‘sacrifice’ from the unlucky few for the sake of public good.
The free-rider argument is typically directed against those who refuse vaccination for passively benefiting from herd immunity without contributing to it.6 The US Supreme Court decision of 1905 (Jacobson v. Massachusetts), upholding the mandate for smallpox vaccination, is sometimes used in support of the claim that members of a civilised society are not entitled to be free-riders by refusing vaccines,4 but the underlying reasoning is controversial. The alleged free-riders did not have the choice to opt out of the benefits of vaccine-derived herd immunity but were involved involuntarily, by the collective choice of others. The vaccinated were themselves not obliged to contribute to herd immunity but chose to do so by exercising their agential freedom to discriminate between more or less valuable actions; they have voluntarily set up the free-ride and put everyone else on it. It would, therefore, be hypocritical for those who were voluntarily vaccinated to discriminate against the alleged free-riders for exercising their own agential freedom to not accept the cost of the ride they were put on without their consent. Moreover, the free-rider premise, taken at face value, applies also to those who cannot be vaccinated; it is unclear why they should be treated preferentially, at the expense of everyone else. Another way, medical reasons to not vaccinate do not negate any non-medical reasons to not vaccinate, nor do they automatically create the right to infringe on the autonomy of others. If this is correct then the free-rider argument implicitly contradicts the moral premise of OTV: the obligation to protect those who cannot be vaccinated does not exist, because they are also not entitled to take a free-ride. The same can be said about the vaccinated individuals for whom vaccines are not effective: they do not contribute to herd immunity but are explicitly committed to benefiting from it. It follows that the only individuals who are entitled to benefit from herd immunity are those who are effectively immunised and therefore cannot benefit from herd immunity, therefore contradiction, in which case the entire edifice of ethical compulsion to vaccinate for the sake of herd immunity collapses (the argument from consistency).
Self-constitution and intrinsic risk
The most undertheorised premise underpinning OTV is that our personal freedom to accept or refuse a preventive treatment for ourselves (or for our children, with whom parents have a unique ontological bond) has lower moral status or social value than the benefits of vaccination, and yet the freedom to discriminate between more or less valuable actions is demonstrably the logical foundation of all contingent value-commitments.7 "I must see myself as having unconditional value—as being an end in myself and the condition of the value of my chosen ends—in virtue of my capacity to bestow worth on my ends by rationally choosing them.”8 This consideration calls for an argument to prove that the relevant restrictions on our freedom are reasonably necessary to preserve what makes human life worth living. Specifically, we must consider whether individual contribution to herd immunity fully offsets the harm of coercively depriving a person of body autonomy with respect to a potentially life changing or otherwise irreversible decision about self-constitution. Since body autonomy is a constitutive condition of our existence as conscious rational agents and is also a necessary condition of a life worth living, it is as valuable as life. We, therefore, ought to regard every permanent violation of body autonomy or self-constitution as a partial destruction of individual agency, on par with a partial destruction of life. Moreover, considering that harm that is immediate is more ontologically significant than the risk of harm associated with individual vaccination refusal, the latter course of action has normative priority over the former. Preserving the constitutive conditions of agency trumps the obligation to eliminate or minimise any associated risks to life.
I further suggest that it is not unethical to expose others to the risk of dying insofar as it falls within the scope of risks intrinsic to human agency, or is the kind of risk that makes human life worth living. The underlying premise is that rational agency entails conscious acceptance of risks in order to act in the face of the unknown. "Almost everything a person does impose some risk on others", therefore “for us to live together and benefit from social cooperation, we must be able to impose some degree of risk of harm on one another.”1 Were we to nevertheless choose to act on the principle that agent autonomy (or any of the constitutive conditions of agency) ought to be restricted insofar as this would help minimise the risks to life, the said principle would encompass not just vaccines but every aspect of conscious agency; every possible action entails the risk of someone dying. Since the ontological minimum of anything is nothing, a commitment to minimising some property entails a commitment to eliminating all possible sufficient conditions of that property. It follows that risk-eliminativism about dying entails a commitment to the elimination of all human action, and therefore, to the non-existence of human agency; a position that is self-defeating.
More formally, (A) all human actions and social norms presuppose a commitment to the value of human agency—to reject this premise would be self-defeating7; (B) body autonomy is one of the constitutive conditions of human agency; (C) it entails exclusive ownership of our innate biological characteristics (these are constitutive of the kind of being we inherently are) and (D) discrimination on the basis of innate biological characteristics negates the value of human agency and is therefore unethical. Let as consider a hypothetical scenario of a treatment that would safely and infallibly prevent homosexuality in adults if administered to newborns. This hypothetical treatment may have been invented in response to a pandemic of a novel and potentially lethal pathogen affecting almost exclusively elderly homosexuals and for which there is no reliable cure. Would it be ethical to make this treatment mandatory? I suggest that under the existing ethical norms, based on respect for the innate characteristics of healthy human beings, the answer must be a resounding NO. A further complication in the above example is the absence of the capacity to give informed consent, leaving the decision in the hands of parents or the public health authorities. Since we are dealing with such a fundamental, irreversible change to the innate human constitution, it would be prima facie unethical for the authorities to mandate this medical procedure. It could amount to a crime against humanity. It is not obvious whether, in this case, even parental consent would be sufficient to ethically justify the preventive treatment.ii Nevertheless, biological parents do possess an unmatched interest in relation to the innate characteristics of their children in virtue of the fact that these characteristics are constituted on the basis of, and in continuity with, the innate characteristics of the parents (although not in an absolutely deterministic fashion). This is the intended sense of my earlier reference to the ‘unique ontological bond’ between parents and children, although the parent–child relationship also has a crucial phenomenological dimension.7
In the case of vaccination, we are dealing with the innate characteristics of all humans, with the characteristically human, natural state, and this adds further weight to the argument against mandatory vaccination. On the basis of the principle derived above (points A–D)—that it is unethical to discriminate against humans on the basis of innate biological characteristics—we can logically link the argument from the constitutive conditions of agency to vaccine mandates: (E) mandatory vaccination involves a range of discriminatory measures intended to augment the natural state of our immune system; (F) the natural state of our immune system is an innate and healthy biological characteristic of every human; (G) mandatory vaccination discriminates against innate and healthy biological characteristics; therefore, (H) mandatory vaccination of humans is unethical.
Vaccine mandates are a priori defeasible not because they limit individual freedoms and rights but because they discriminate against healthy, innate characteristics of every human. For this reason, mandatory vaccination is not ethically analogous to mandatory seatbelts10 or to using physical force to remove a dangerous substance from a child.11 Wearing seatbelts when driving or removing a dangerous substance from a child does not alter their individual constitution, but vaccines do. Vaccination is an irreversible medical procedure, not just a behavioural preference. The case of vaccine mandates is also relevantly different from the involuntary treatment of psychiatric patients. Vaccines are intended to permanently augment healthy, innate human characteristics whereas psychiatric treatments deal with pathological states characterised by already impaired agential capacities and aim only to re-establish those capacities. The emerging view in psychiatric ethics is that involuntary medical treatment is unethical under any circumstances.12 Crucially, any involuntary medical procedure intended to augment the innate human constitution would be just as unethical in psychiatry as it is in the context of vaccine mandates.
The argument from the constitutive conditions of agency is not affected by the balance of risks versus benefits associated with any constitution-augmenting procedure, or the medical circumstances under which such a procedure could be mandated, because it derives its normative force directly from the intrinsic value of human agency. Any form of compulsion or discrimination is unethical if used to facilitate, incentivise or normalise unwanted change in the innate human constitution. This is not only consistent with the established ethical norms (including, but not limited to, the first principle of the International Health Regulations of the WHO13) but, as demonstrated above, can be substantiated a priori.iii
Proponents of the view that some or all people have a moral or ethical obligation to vaccinate are implicitly committed to a further moral obligation, to develop a comprehensive and consistent argument in favour of the obligation to vaccinate, grounded in objective facts or a priori reasoning. This has not been accomplished as a matter of principle (the argument from consistency). I have developed an argument to the contrary. Vaccine mandates involve a range of discriminatory measures intended to augment the natural state of our immune system in the interest of public health. This amounts to discrimination on the basis of innate biological characteristics. The strongest mandate of compulsory vaccination would essentially make our innate biological state unlawful. There are ethically analogous hypothetical situations that are intuitively repugnant, for example, mandatory physiological alteration of healthy infants in the interest of public health. This would imply that all humans are born in a defective, harmful state. If this ethically analogous situation is unethical, a premise I have defended a priori, then mandatory vaccination is also unethical. The principle holds as a matter of logical necessity, in virtue of the intrinsic value of human agency, and is therefore not defeated by circumstances such as emergencies or pandemics. Moreover, it permissively justifies vaccine refusal by healthcare workers, despite their unique professional obligations, even for hypothetical vaccines that are medically risk-free.
Nothing presented here is meant to imply that vaccination ought to be refused; I have argued only that there is neither a moral obligation to vaccinate nor a sound ethical basis to discriminate against the unvaccinated.
Data availability statement
There are no data in this work.
Patient consent for publication
Contributors There are no contributors apart from the main author.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
↵Saline placebo is used in a small fraction of vaccine trials but without eliminating the confounding factor of other vaccines with similar ingredients being possibly administered prior to or during the study period.
↵Brown et al 14 argue in the opposite direction. Building on the premise that it is unethical to restrict freedoms of people who pose no or minimal risk of spreading SARS-CoV-2, the authors conclude that those who are immunised ought to be treated preferentially (granted unrestricted freedom with immunity passports), 'therefore' those who are not immunised can be justifiably discriminated against (their freedom restricted). This argument rests on a false dichotomy. While I agree that it is unethical to restrict the freedoms of people who pose no or minimal risk of contagion, I have shown that it is also unethical to discriminate on the basis of healthy, innate biological characteristics, and this latter category includes those who are not immunised (therefore immunity passports are unethical). These two ethical constraints taken together reveal a third possibility, which Brown et al 14 do not consider: it is unethical to restrict the freedoms of non-contagious persons, irrespective of whether they are immunised. The relevant restrictions may be ethically justified only if a person is presently contagious, only for as long as she remains contagious, and only if the pathogen is extraordinarily virulent (because restrictions on basic freedoms are themselves extraordinary, normally regarded as a form of punishment).
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.