Etikk i praksis. Nordic Journal of Applied Ethics (2021), 15(2), 17-28

http://dx.doi.org/10.5324/eip.v15i2.4102

Early View publication date: 16 December 2021


Against draconian penalties for COVID-19 quarantine infringements


Elias Moser


University of Graz, Institute of Philosophy, Section Moral and Political Philosophy and London School of Economics, Centre for Philosophy of Natural and Social Science


In 2020, after the first Covid-19 lockdown, several countries implemented a policy of contact tracing and self-isolating for individuals who crossed borders or came into contact with infected people. To enforce these restrictions, some states imposed very harsh monetary penalties for people who violated them. Behind these harsh fines lies an instrumental rationale. They allow the state to avoid implementing a system of labor-intensive and costly surveillance and enforcement. In this article I argue that such severe penalties are extremely unjust. In order not to expose citizens to the risk of being excessively fined, governmental institutions should instead intensify controls. I argue that they owe it to their citizens to increase the surveillance of compliance with self-isolation obligations.

 

Keywords: COVID-19, Quarantine, Self-Isolation, Theories of Punishment, Economics of Criminal Justice, Proportionality

    

Introduction

Following the first shutdown of stores, restaurants and public places and restricted freedom of assembly in public places, governments all around Europe re-opened their economies and lifted the restrictions. To combat COVID-19, states shifted to a strategy of contact tracing and individual self-quarantine. Individuals who came into contact with infected people or who immigrated from countries with high incidences of infection were obliged to stay at home for some days until the probability of contagiousness was reduced to a minimum.

These self-isolation obligations were enforced differently across countries. While some governments relied more on self-responsibility, others imposed harsh penalties. For example, on 28 September, the UK government imposed a duty to self-isolate enforced with a fine of up to ten thousand pounds (gov.uk 2020a).1 In Norway, the Attorney General’s directives hold that negligent behavior may even be subject to imprisonment.2 In this article, I argue that these high penalties for infringements of self-isolation duties are excessive and unjustified from a perspective of justice. The size of the fines goes beyond any proportion to the risk imposed on others. An alternative mode of enforcement of quarantine duties would consist of a regime of low fines but increased monitoring and surveillance. I plead for this alternative as a more just solution. My thesis is that the state owes it to individuals to implement more frequent controls instead of deterring them with draconian punishments.

In Section II, I sketch the rationale behind a regime of harsh fines. This rationale provides incentives for law-abiding behavior while being relatively cost efficient. Morally, these harsh penalties can be justified on utilitarian grounds. In Section III, I outline this argument and provide some reasons why I refrain from a purely utilitarian approach to the justification for COVID-19 policies. I then contrast the utilitarian account with a deontological conception of the harm principle that justifies restrictions but requires proportional sanctions for infringements in Section IV. The utilitarian may argue that the justification for harsh penalties does not follow from their account. I will outline these arguments in Section V. In the penultimate section, I discuss some possible defenses of harsh fines and show why they do not hold up. Lastly, I conclude by restating the claim that states should increase controls rather than imposing draconian penalties.



The rationale for draconian punishments

The instrumental reasons for imposing harsh sanctions can be outlined with a simple economic model introduced by the famous economist Gary Becker (1969). He assumes that individuals, in deciding whether to obey or disobey a legal rule, behave as rational utility maximizers. They calculate whether it is worth being disobedient in terms of utility and compare the expected utility of their liberty not to obey with the expected disutility of being caught. A person becomes a perpetrator if the expected value U (where U stands for utility) of the crime is larger than the expected disvalue U of being caught and penalized. The expected value is the product of the two factors, p being a chance between 0 and 100 percent and U decreasing with the severity of punishment.


U(disobedience) > |p*U(penalty)|

Let me translate this into the deliberation of a person in quarantine. He or she will disobey the self-isolation requirement, leave home, go to work or to school or meet other people if the gains from his or her liberty are high enough to outweigh the risk of being caught and fined.

If the lawmaker assumes that the noncompliant person is a rational utility maximizer, he or she will take the person’s deliberation into account and adjust enforcement of the restrictions accordingly.3 The lawmaker thus implements a system of penalties that renders the expected disutility of being caught larger than the utility of enjoying the freedoms from not following the self-isolation rules.


|p*U(penalty)| >= U(disobedience)

In order to achieve that goal, the lawmaker has two options: either (1) to increase the likelihood p of catching perpetrators or (2) to raise the magnitude of the penalty. The former can be increased by controls, random inspections or surveillance. The latter can be increased by raising the fine.

Now, as Becker rightly asserts, controls (1) always come at a cost. Governmental institutions need to perform labor-intensive sampling and supervision of potential perpetrators. In contrast, raising the fine (2) does not impose any social cost at all.4 Therefore, it is always more efficient to impose a harsh sanction instead of intensive controls. Harsh fines set the incentive for the rational utility maximizer not to violate the restrictions, and they do so nearly without increased expenses for the public.

This rationale supports the draconian penalties for infringements of quarantine restrictions. Especially in mid-autumn 2020, many European countries faced an accelerating increase in infections, such that controlling individuals would have necessitated a high amount of personnel that was not available. Nevertheless, I will argue that if legal quarantine restrictions are in force (and if we accept these restrictions as justified), it is the state’s obligation not to enforce the restrictions with draconian sanctions. Instead, it should increase controls to ensure compliance with the restriction.

 

 

Utilitarian justification

From a utilitarian perspective, the legal duty to self-isolate is necessary (morally mandatory), if the sum of individual disutilities – given no quarantine rules are in effect – is larger than the sum of disutilities for the people whose liberties are restricted due to the quarantine rules. The former value is a function of the risk of suffering severe health problems or death. The latter term is defined by the individual utilities deriving from the use of liberties, such as assembly, social interaction, working, visiting school, etc.5 Restrictions that save lives come at the cost of individual liberties. As a widely received article in the Economist (03-26-2020) quite adequately depicted, all COVID-19 policies involve a major trade-off between “lives” and “livelihood”. Utilitarians commonly assume that these two values are commensurable. So, we can compare the disvalue of someone being killed with the loss of wellbeing for someone whose freedoms are impaired.

An ethically desirable policy from a utilitarian point of view attempts to minimize the net harm. To restrict liberties and to impose self-isolation on potentially infectious people needs to be less hazardous than the potential loss of one’s life or health. It is generally accepted that the disvalue of being killed and suffering serious health impairments is very high and, if we can avoid it, many restrictions of individual liberties are thus justified. Since there is widespread consensus that the disutility from the risk of infection largely outweighs the undesirability of a temporal restriction of individual liberties, I concede that net harm is reduced by COVID-19 quarantine restrictions.

In asking the question of justification for the measures, the utilitarian takes the disutility from restrictions of liberties to be an ideal term that is based on the assumption that all people subject to quarantine rules do in fact obey them. Regarding the question of how many people obey the utilitarian reflection is different. It asks which mode of enforcement is necessary, given the COVID-19 rules lead to a minimization of net harm. Because the net utility from risk reduction is under the above-made assumption always positive, optimal enforcement is reached when the maximum number of people obey the restrictions. The utilitarian imperative, provided the quarantine rules are justified, is therefore to minimize infringements of self-isolation obligations in a way that imposes minimal costs on society. Within a Beckerian framework, a utilitarian would therefore opt for an increase of penalties up to the point where cost-intensive surveillance can be minimized while ensuring general compliance.

Therefore, both the imposition of self-isolation requirements and their enforcement by means of harsh penalties may be justified on utilitarian grounds. But does the claim that self-isolation duties lead to a net harm minimization justify these duties?

In public health, pure utilitarian accounts are usually not defended. Although the accepted necessary requirement for imposing restrictions on individual liberty is that the projected benefits of the policy outweigh its costs, this is not assumed to be sufficient for a justification (Childress 2001; Kass 2001). Other considerations have to be taken into account. Fundamental rights – such as liberty rights, freedom of movement and assembly and the freedom to work and do business – are usually conceived of as ‘side-constraints’ on the utilitarian calculus. For instance, the International Health Regulations of the WHO can be interpreted as adopting a side-constrained consequentialist approach. According to its constitution, the organization’s objective is ‘the attainment by all peoples of the highest possible level of health’ (WHO, 2020a: 1), whereas: ‘The implementation of these Regulations shall be with full respect for the dignity, human rights, and fundamental freedoms of persons’ (WHO, 2005: 3, §1). So, there are reasons to be skeptical about whether a utilitarian account of COVID-19 measures and its enforcement is sufficient.

Utilitarian grounds for assessing policies focus exclusively on the value of the outcome of different actions. An exclusively utilitarian account of the justification for lockdown policies cannot explain why far-reaching encroachments on individual freedoms are in general regarded only as a measure of last resort, and why, as soon as developments permit, the restrictions must be lifted. According to commonsense morality, justification for a policy does not necessarily follow from the fact that the benefits of a policy outweigh the costs. Such a view would allow us to sacrifice individuals’ liberties (without any potential consent) for the greater good, whenever net benefits are maximized, or net harms are reduced. The view would thus be highly revisionist and, I believe, it would not prevail in a democratic political process.

Common moral convictions on the justification for restrictions and their enforcement are better captured in an account that is able to do justice to the common belief that net-harm minimization is not a sufficient ground for the legitimacy of constraints on personal liberty. In what follows, I will therefore defend a deontological view on the justification for COVID-19 measures that is better able to describe positive morality. It assumes that the restrictions are legitimate since the individual act of a person that infringes a quarantine duty imposes a risk on other people. The infringement is therefore a morally unjustified action. An action that is worthy of blame.



Deontological view

The harm principle

The imposition and enforcement of quarantine duties can be justified on the basis of the famous harm principle (Mill 1999: 51–52; Ross 2002: 21–22). It states that an action may only be legitimately restricted if it would cause harm to third parties. This contrasts, for example, with paternalistic reasons for restrictions. The freedom of individuals must not be curtailed to protect them from the consequences of their own well-informed and rational choices.

The justification for restrictions of freedom on the basis of the harm principle is directed against a utilitarian view. Restrictions are only permitted if the person’s actions would cause harm, and it is not permissible to restrict actions merely because the consequences of the restriction are desirable. There must be a potential harm caused by a person’s actions and this is the case when a person imposes an increased risk of infection on other persons. Self-isolation obligations are therefore justified on the basis that a person who is quarantined potentially endangers other persons’ rights to life and health (Harris & Holm 1995).

To infect someone by disobeying the quarantine rules usually is not an act of ‘malicious’ intent. People who do not stay at home when obliged to do so are not willingly pursuing the goal to infect other people. Furthermore, in most cases in which someone infects other people while not complying, I would assume that there is no ‘ruthlessness.’ The person who disobeys does have legitimate needs or desires – such as going to work or having social interactions – the fulfillment of which, under usual circumstances, is protected by fundamental rights. We should therefore not assess the moral disvalue of the act of infringing self-quarantine duties by attributing vicious character traits to the perpetrator.

Nevertheless, the act of infringing quarantine duties includes exposing other people to risk of infection and becomes therefore a case of ‘negligence’ if the risk materializes. Exposing people to risk is not a direct cause of harm. It is harm that occurs with some probability. In order to assess the disvalue of exposing someone to risk, the harm needs to be factored with the likelihood of its occurrence.

The resulting disvalue is therefore expressed as a specific expected value. It is worth noting that in most cases where a person infringes quarantine duties the probability of causing harm is particularly low. Calculating an expected value requires multiplying different probabilities. The more factors we incorporate, the lower their expected value becomes. The following non-exhaustive list outlines some reflections that are appropriate to assess the disvalue of quarantine infringements.

- The probability of becoming infected due to contact with an infected person or border crossings.
- The probability of infecting others while not respecting self-isolation duty.
- The probability of an actual infection showing symptoms that manifestly impair the infected person’s bodily integrity.

In the case of COVID-19, all these factors are, compared to other diseases, very high. COVID-19 is assumed to be about ten times more lethal than any seasonal flu (Johns Hopkins Medicine 2020) and measures are of course appropriate to reduce the spread of the virus. However, if we are not assessing the legitimacy of measures as a whole but rather the appropriate penalty in the case of an infringement of the measure, we have to focus on the individual expected disvalue of violating the self-isolation duties. Presuming that all these factors must be taken into account, the negative expected value of someone’s infringement of quarantine duties is very low, and so the assessment of individual misconduct has to reflect the low probability of the harm caused to third parties.

In general, one might object to this calculus of expected disvalue. While the probabilities can be well expressed in numbers, the individual harm cannot. The latter includes the major philosophical problem of commensurability of different values (Griffin 1986; Chang 1997; cf. Gostin et al. 2020). Which harms are bad, and how bad are they compared to other harms? A utilitarian is expected to give clear answers to these questions, but for the purpose of assessing the legitimacy of harsh penalties for infringements of self-isolation duties within a harm-principle framework, we do not need to give a substantive answer. It suffices to compare the penalty with our actual moral and legal convictions on justified amounts of penalties in other cases of negligence. Thus, we may see whether it is proportional – that is, whether we can coherently claim that the severity of the penalty is justified.

In order to show that the severity is not justified, the enforcement of quarantine duties can for example be contrasted with the enforcement of traffic regulations. Let me assume for the moment that the amount of speeding fines is commonly accepted as proportional to the disvalue of the conduct. In the UK, for example, speeding penalties amount to one hundred pounds. The suspension of the driver’s license is imposed only in case of repeat offenses (gov.uk 2020b). These fines and penalties reflect the risk to life and health of other people through careless driving. Because the risk is low, the penalty is affordable. It is set within what may be called ‘retributive boundaries.’ Let me elaborate on that a little bit more.


Retributive boundaries of proportionality

If we assume that the prohibition of leaving one’s house is justified on a deontological basis – that is, if increasing the risk of infecting someone with the coronavirus is negligent – a sanction seems to be justified on the basis of the harm principle and thus a mala in se action. In contrast to the utilitarian account, this changes the way we conceive of the justification for the legal sanction. The sanction should no longer be regarded as a sort of fine but must be considered a ‘punishment.’ As such it stands in correlation with the disvalue of the misconduct. It is backward-looking in that it expresses an appropriate reaction to the performed act. Hence, the severity of punishment reflects the disvalue of the wrong done to someone else.

This view is clearly ‘retributivist’ as opposed to consequentialist theories of punishment (Murphy 1973; Morris 1976; Moore 1997). Retributivists regard the justification for punishment as contingent upon a perpetrator’s guilt and the severity of justified punishment as relational to the blameworthiness of the deed (von Hirsch 1990: 284). On the other end, utilitarians assume that the justification for a penalty depends on its merit for society. Legal sanctions may have incentivizing effects – rehabilitation of the offender, prevention of recidivism, or general deterrence. Overly harsh punishments can be justified as having desirable effects and the amount of punishment does not need to be proportional to the disvalue of the misconduct.

I do not attempt to settle the dispute between adherents of retributivist or utilitarian theories of punishment here. But it can be argued that proportionality of the punishment is usually conceived of as a requirement of justice. A just punishment needs to stand in relation to the deed. The draconian penalties for infringing self-isolation duties, however, lack any sense of proportionality. Being fined up to ten thousand pounds is beyond a reasonable proportion to the imposed risk of infection. The penalty puts citizens in financial straits and may seriously impede their pursuit of individual life plans. Therefore, the size of the penalty is at odds with commonly held convictions of just desert.

On the one hand, the potential risk of infringing on the rights of others in some cases is minimal. On the other hand, even if an infection occurs, the risk of suffering a severe impairment of the right to bodily integrity or life is still low. The assessment of the disvalue of the misconduct needs to account for these two variables – the risk of rights infection and the harm of infection – and the sanction system needs to be adjusted accordingly. Therefore, I conclude that, if the criminalization of self-isolation is justified with reference to the harm principle – that is, if there is a moral duty not to risk infecting others – the draconian penalty for violating the duty is disproportionate. The excessive punishment is unjust.



Utilitarian reaction

There is a potential utilitarian case against harsh penalties for infringing self-isolation duties to be made within Becker’s model. If we assume that individuals are risk averse or, in other words, that there is an ‘increasing marginal disutility’ in the magnitude of a penalty – there is no need to be overly harsh. An increase in the severity of the penalty would, under this assumption, create over-proportionally more disutility. People would then tend to be deterred from noncompliance with obligations more easily by less severe penalties. Therefore, the utilitarian defendant of the Beckerian model would not necessarily be inclined to support the conclusion that harsh penalties are necessary.

However, a Beckerian advocate of harsh penalties would refer to the observation that, in a regime of mild punishments, some people would not abide by the restrictions. Within the model, this observation can only be explained by the fact that these people are not sufficiently deterred. The assumption about the increasing marginal disutility in the size of the penalty is therefore either wrong or the decreasing effect is too small, such that harsh penalties are still necessary. Of course, no comparison groups are available to test how much the motivation to abide by the restriction is dependent on the severity of punishments. But in order to criticize harsh penalties from a utilitarian perspective, I believe one really needs to criticize the assumptions of the model.

One way of doing so is to outline ‘basic-needs constraints’ for the potential minimization of expected losses. For a very large group of people around the globe it is simply not affordable to stay at home and stop working for more than one week (The Economist 03-26-2020). Many countries are in fact not capable of financially compensating quarantined individuals for their income loss (Oxfam 2020).6 The disutility of the lack of freedom to go to work always outweighs the expected disvalue of penalties for these individuals, and they would never be deterred from pursuing work by the threat of harsh fines. To uphold the incentivizing effect of harsh penalties, the state would have to raise them to enormous heights. The policy would not only fail to incentivize but also cause much misery for those who are caught in noncompliance. Imposing such an exorbitant penalty would be highly unfair and could also not possibly be justified in terms of efficiency or utility.

Of course, people in Europe, in general, are lucky to be able to satisfy their basic needs even if they are prohibited from working for more than one week. But other desperate situations may also force people to leave their house and violate self-isolation duties. For example, quarantining may have highly negative psychological effects on certain groups of people. The psychological impact is widely acknowledged by scholars in the field (Brooks 2020). But this has so far not played a major role in the political decision-making process. The value of saving lives was always considered to outweigh the psychological duress of a few. However, the psychological pressure for affected individuals might be so substantial that they choose to ignore the self-isolation duties. In this case, even a harsh penalty fails to incentivize quarantined person.

It is also important to note that, even in Europe, only some members of society can easily afford to stay home for more than one week, are able to work at home and send their children to daycare or can be excused from work by their employer. Others are not in this fortunate position. Perhaps unsurprisingly, employees’ ability to work from home correlates positively with their income (Mongey & Weinberg 2020). This means that the working poor have to rely on the benevolence of their employer to not penalize them for being unable to show up at work due to publicly enforced self-isolation duties. From a perspective of distributive justice, therefore, the harsh penalty regime has a highly unwelcome effect. Poorer people are more likely to break the rules because they cannot be excused from work as easily. And the harsh fines hit the poor more often.

Although irrational in terms of expected disutility minimization, people under high psychological or economic pressure might be ready to violate self-isolation duties given the very low risk of being caught. The reasons for noncompliance provided by the utilitarian critic of the Beckerian model suggest that, due to specific circumstances, humans might not behave like rational utility maximizers. If that is the case, even the utilitarian does not support draconian penalties. Certain forms of irrationality might be involved, especially with respect to infrequent controls. In a number of papers, behavioral economists Amos Tversky and Daniel Kahneman (1974) found that when making decisions under uncertainty, people tend to rely on immediate examples that come to mind. They call this phenomenon ‘availability heuristics:’ If a person can recall something, this plays a more important role in his or her deliberation process.

Now, rare enforcement of people’s compliance with quarantine restrictions leads to the situation that people almost never experience being controlled and never hear of anyone being controlled. Hence, they tend to underestimate the expected disvalue of disobedience and the probability of harsh fines. Therefore, the rationale for Beckerian penalties might rely on the wrong anthropological model. Draconian fines are less effective than they would be if people were estimating the expected value adequately. Therefore, there is a potential utilitarian case against draconian fines.

But also from a perspective of fairness, the irrational evaluations of expected values are problematic. The harsh penalty seems to hit less rational people harder. By comparison, the enforcement of self-isolation duties that includes more frequent controls and affordable fines distinguishes less between rational and irrational people, since the availability bias is less severe. So if we assume for the moment that the deterrent effect is equally high in both modes of enforcement, the question arises whether it is fair to choose a regime of harsh fines. Irrationality of the agent by itself seems not to justify imposing a disadvantage on him or her. Therefore, opting for a harsh-fine regime seems to unjustly discriminate between rational and irrational people. In other words, a just legal regime would impose frequent controlling (and an affordable penalty) in order to protect irrational individuals from severe legal consequences
.



Possible defense of draconian penalties

Lastly, I address several arguments that attempt to justify the legal treatment of self-isolation obligations by means of draconian penalties.

(1) One argument refers to the fact that a pandemic is an extraordinary situation for our society. It can be described as an extreme case, in which the legal handling justifies implementing a form of emergency law. This does not necessarily mean that the COVID-19 rules in all countries have been formally introduced as emergency law, or that governments necessarily declared a state of emergency. The extraordinary situation is rather a moral justifier for specific legal rules. If so, the measures taken to combat the pandemic arguably do not have to meet the high standards of justice. The state has permission to disregard certain requirements because of the exceptional challenge. Thus, one could argue that the draconian penalties are well justified under the particular circumstances. One could also explain why rules in other areas of law that are concerned with minimizing danger, such as traffic, are not enforced with harsh punishments – because they do not involve a state of emergency.

However, a partial exception to justice requirements does not mean that all principles of justice are automatically thrown overboard. It is clear that by means of the COVID-19 measures, individual rights were severely restricted or violated. However, these measures pursued a specific goal, which, it is believed, could not have been achieved by other measures. However, in the case of enforcement, Becker’s model shows that the same effect can be achieved in two different ways – either through harsh penalties or through increased controls. The only argument for the former mode of enforcement is the higher cost of the latter. I concede that an initial shortage of staff might have made it impossible to carry out intensive controls in the short term. But as the pandemic continues, this argument becomes less effective. The large-scale testing initiatives, the vaccination campaign, the development and review of security concepts and the increased border controls are all very personnel- and cost-intensive measures that are actually taken by governments. I therefore believe the resources for just enforcement of self-isolation rules would also be available. Thus, referring to a state of emergency is more an excuse for the state’s unwillingness to pay the costs of increased controls.

(2) Another argument in defense of the regime of draconian punishments consists of the claim that the legal consequences for the penalized person might in fact not be as bad as they seem. In some countries, the penalty in COVID-19 regulations is defined as an upper limit up to which a person can be punished in the case of violation of duties.7 It is quite possible that the legal institutions of different countries with draconian penalties for violations of quarantine obligations may well refrain from imposing harsh sanctions – probably because of the above-mentioned considerations of proportionality and justice. But if this is the case, then the legal interpretation undermines the enacted norm, and this is undesirable for two reasons.

On the one hand, utility-maximizing legal subjects as in Becker’s model include the information they receive about the actual enforcement in their utility calculations. If the authorities never impose the maximum penalty, then the norm becomes generally known to be an ‘empty threat.’ Consequently, the announced legal sanction no longer has the desired deterrent effect, and in the long run, controls would have to be increased anyway.

On the other hand, it is a lawmaker’s virtue to ensure that the norm and the actual enforcement do not contradict each other too much. Should the citizens of a state fundamentally lose faith that the COVID-19 regulations are actually being carried out as they were formulated, this could cause great damage to the system of rules in general. Individuals would become much less observant of the rules, which in turn would have a devastating effect on compliance and the infection rates.

(3) One way to justify imposing harsh sanctions (regardless of whether the upper limit of the penalty is levied) is to emphasize the positive communicative effect. When a harsh penalty is imposed, the argument goes, it communicates to the general public that compliance with the norm is morally urgent. The penalty conveys the message of an imperative that everyone comply with the COVID-19 ordinances.

The communicative effect of punishment has been variously explained by criminal justice theorists and has also been regarded as a reason to justify punishment in general (Feinberg 1970; Duff 2001). The argument, however, not only assumes that imposing sanctions has a positive communicative effect, but also that the effect becomes stronger with higher levels of punishment. To support the argument, one would thus have to show empirically that increased controls, monitoring, and surveillance of compliance with self-isolation obligations are no better than harsh penalties at communicating to individuals that the matter is urgent. On the one hand, the reflections on the ‘availability bias’ instead point toward the opposite, namely that the communicative effect is lower when people are confronted less with a legal response to transgressions. On the other hand, given the obvious injustice of draconian penalties, I believe the burden of proof that harsh fines have a positive communicative effect lies with the proponents. They need to show why such a disproportionately high sanction is justified and support their thesis with empirical evidence.


Conclusion

As I have shown in the outline of Becker’s model of efficient enforcement of legal rules, a lawmaker has two options – either to implement frequent controls and to exact fines more or less in proportion to the misconduct or to reduce the number of controls and impose harsh fines. Some countries, such as the UK and Norway, chose the latter option to enforce Covid-19 self-isolation duties.

This approach may be more efficient, and the penalties may well be justified on utilitarian grounds. But they are highly unjust. If the risk of infecting someone is presumed immoral because of the ‘harm principle,’ harsh penalties are not justified as they are simply not proportionate to the risk that is being imposed on others. Some individuals, who are not or cannot always be utility maximizers are excessively punished for their disobedience.

Instead of enforcing quarantine duties with harsh penalties, I therefore propose that they should be enforced with the help of frequent controls, inspections, and surveillance. Although more costly, the state should not punish people excessively to enforce its policies. The state owes more intensive controlling to its citizens.



Notes

1 The Health Protection (Coronavirus, Restrictions) (Self-Isolation) (England) Regulations 2020 28.09.2020, (SI 2020/1045), §12. https://www.legislation.gov.uk/uksi/2020/1045/pdfs/uksi_20201045_en.pdf?utm_source=hootsuite&utm_medium=social&utm_campaign=post

2 Koronavirus (COVID-19) – Riksadvokatens retningslinjer og direktiver, 20.03.2020 (RA-2020-320). https://lovdata.no/static/file/1825/ra-2020-320-2.pdf

3 Throughout I assume the lawmaker’s goal is to reduce perpetration to a minimum in order to reduce the social costs of infections and the risk of people being killed.

4 Being fined simply includes a change in distribution from the individual to the state.

5 There are also very strong negative impacts of self-isolation duties that often have been neglected – for instance, psychological problems (Mucci et al. 2020; Rubin and Wessely 2020) or domestic violence (Russell 17-4-2020).

6 For example, the UK government has supplemented the introduction of strict fines with the promise to reimburse some of the financial losses of people who are forced to stay at home with a lump-sum payment of GBP 500 (gov.uk 2020a).

7 For example, in Switzerland, where fines of up to CHF 5000 can be issued. Bundesgesetz über die Bekämpfung übertragbarer Krankheiten des Menschen (Epidemiengesetz, EpG), 28.09.2012 (818.101), §83, 2. https://www.fedlex.admin.ch/eli/cc/2015/297/de#id-11t



References

Brooks, S. K., Webster, R. K., Smith, L. E., Woodland, L., Wessely, S., Greenberg, N. & Rubin, G. J. (2020). The psychological impact of quarantine and how to reduce it: rapid review of the evidence. The Lancet 395:10227, 912–920. CrossRef

Becker, G. S. (1968). Crime and punishment: an economic approach. In N. G. Fielding, A. Clarke & R. Witt (Eds.), The Economic Dimensions of Crime. London: Palgrave Macmillan. CrossRef

Chang R. (1997). Introduction. In R. Chang (Ed.) Incommensurability, Incomparability, and Practical Reason. Cambridge, MA: Harvard University Press.

Childress, J. F., Faden, R. R., Gaare, R. D., Gostin, L. O., Kahn, J., Bonnie, R. J., Kass, N. E. et al. (2001). Public health ethics: mapping the terrain. Journal of Law, Medicine & Ethics 30:2, 170–178. CrossRef

Duff, A. (2001). Punishment, Communication, and Community. New York: Oxford University Press.

Feinberg, J. (1970). The Expressive Function of Punishment. In J. Feinberg (Ed.), Doing and Deserving, Princeton, NJ: Princeton University Press.

Gostin, L. O., Friedman, E. A. & Wetter, S. A. (2020). Responding to COVID‐19: how to navigate a public health emergency legally and ethically. Hastings Center Report 50:2, 8–12. CrossRef

Griffin, J. (1986). Well-Being. Oxford: Clarendon Press.

gov.uk. (2020a). New legal duty to self-isolate comes into force. https://www.gov.uk/government/news/new-legal-duty-to-self-isolate-comes-into-force-today. Accessed 29 August 2021.

gov.uk. (2020b). Speeding penalties. https://www.gov.uk/speeding-penalties. Accessed 29 August 2021.

Harris J. & Holm, S. (1995). Is there a moral obligation not to infect others? BMJ 311:7014, 1215–1217. CrossRef

Johns Hopkins Medicine. (2020). Corona disease vs. the flu. https://www.hopkinsmedicine.org/health/conditions-and-diseases/coronavirus/coronavirus-disease-2019-vs-the-flu.
Accessed 1 August 2021.

Kass, N. E. (2001). An ethics framework for public health. American Journal of Public Health 91:11, 1776–1782. CrossRef

Mill, J. S. (1999) [1859]. On Liberty. Peterborough, Canada: Broadview Press.

Mongey, S. & Weinberg, A. (2020). Characteristics of workers in low work-from-home and high personal-proximity occupations. Becker Friedman Institute for Economics white paper. https://bfi.uchicago.edu/wp-content/uploads/BFI_White-Paper_Mongey_3.2020.pdf. Accessed 29 August 2021.

Moore, M. S. (1997). The moral worth of retribution. In M. S. Moore (Ed.), Placing Blame: A General Theory of Criminal Law. Oxford: Oxford University Press.

Morris, H.: (1976). Persons and Punishment. In H. Morris (Ed.) Guilt and Innocence. Berkeley: University of California Press.

Mucci, F., Mucci, N., & Diolaiuti, F. (2020). Lockdown and isolation: psychological aspects of COVID-19 pandemic in the general population. Clinical Neuropsychiatry 17(2), 63–64. CrossRef

Murphy, J. G. (1973). Marxism and Retribution. Philosophy and Public Affairs 2, 217–243.

Oxfam. (2020). Shelter from the storm: the global need for universal social protection in times of COVID-19. https://www.oxfam.org/en/research/shelter-storm-global-need-universal-social-protection-times-COVID-19. Accessed 29 August 2020.

Ross, W. D. (2002) [1930]. The Right and the Good. P. Stratton-Lake (Ed.). Oxford: Oxford University Press.

Rubin, G. J., & Wessely, S. (2020). The psychological effects of quarantining a city. BMJ 368:m313. CrossRef

Russell, A. (2020-04-17). The rise of coronavirus hate crimes. The New Yorker. https://www.newyorker.com/news/letter-from-the-uk/the-rise-of-coronavirus-hate-crimes. Accessed 29 August 2020.

The Economist. (3-26-2020). The coronavirus could devastate poor countries. https://www.economist.com/leaders/2020/03/26/the-coronavirus-could-devastate-poor-countries. Accessed 29 August 2021.

Tversky, A. & Kahneman, D. (1974). Judgment under uncertainty: heuristics and biases. Science 185:4157, 1124–1131. CrossRef

Von Hirsch, A. (1990). Proportionality in the philosophy of punishment: from ‘why punish?’ to ‘how much?’ Criminal Law Forum 1:2, 259–290. CrossRef