Cyber banishment: an old sanction for virtual spaces

ABSTRACT In this paper, I present the punishment called cyberbanishment. This sanction refers to the old-fashioned punishment of forbidding a person to be present in specific places, but instead of physical locations this sanction concerns digital spheres. The proposition of introducing cyberbanishment is based on a few observations. First, the current practices of punishment are problematic, and there is a need for alternative punishments. Second, this form of sanction may be a proper response to crimes committed online, like hate crimes or punishable threats. Third, the online sphere becomes more important each year, and this was even amplified by the recent pandemic. The development of virtual realities and metaverses could contribute to this trend. Fourth, as an instrument imposed by courts, cyberbanishment may increase judicial control over the limitations of constitutionally protected freedoms in the online world, which is currently decided mainly by the platforms themselves.


Introduction
In 2021, Facebook, Instagram, Twitter, and YouTube suspended Donald Trump's accounts in response to an attack that took place at the United States Capitol (Floridi, 2021). It immediately raised questions about issues of freedom of speech and the role of tech giants and social media platforms in speech moderation (see Ganninger, 2021;Macedo, 2022). Another example of a try of removing someone from the Internet is InfoWars, the right-wing news, multi-platform company founded by Alex Jones, which was available on many platforms and social media sites (Ottoni et al., 2018, p. 324). On one of his programs in December 2015, Alex Jones interviewed Donald Trump, the future president of the US (Williamson & Steel, 2018). On his platform, Jones was spreading conspiracy theories (Sellnow, Parrish, & Semenas, 2019, p. 131). Before the presidential election in 2016, the so-called 'Pizzagate' conspiracy, fake news linking Hillary Clinton and her associates to pedophilia and even murder, was trending on many platforms, partly due to significant amplification by Jones (Britt, Rouet, Blaum, & Millis, 2019, p. 95). In 2018, the most prominent social media sites collectively acknowledged that Jones had infringed the rules on hate speech, and they banned him from publishing his content on their sites (Sellnow, Parrish, & Semenas, 2019, p. 133). Leading technology companies, such as Facebook, Twitter, YouTube, Reddit, and Apple, which had previously allowed his content to be hosted, banned Jones from their platforms (Nicas, 2018). Due a chain reaction from this ban, he lost the opportunity to effectively be present in the public sphere, which is both dominated by and dependent on tech companies. However, this outcome was not due to a court decision but the consequence of independent decisions by each of these companies. Since it was a decision by private companies and not the government, the case could not be treated as a violation of the First Amendment (which protects free speech) (Holliday, 2019, p. 55). There have been suggestions that the government should step in during such conflicts, but this is not an easy task: There is great tension around the governance of hate speech between those who push for greater government oversight and advocates of letting the Internet regulate itself. There is a delicate balance between privacy, protection, and the importance of free speech, particularly when the concern is that of public interest. (Holliday, 2019, p. 63) Those examples could be interpreted from many different legal perspectives. In this paper, the perspective of criminal law is presented. Here, I argue that this kind of reaction to alleged misbehavior may function as a form of criminal punishment. It is similar to the biblical-rooted punishment of banishment -not as an exile from the physical sphere but the digital one. I further suggest that it could supplement possible reactions to a crime and support the alternatives for imprisonment. What is more, the imposition of such measures by the court, instead of technological giants, could mitigate the tensions resulting from the interference of tech giants in a matter of freedom of speech. Cyberbanishment could be a useful criminal measure for an era of increasing significance of virtual reality (on virtual reality, see Chalmers, 2022) or metaverses, and it may be seen as a way of preparing the law for that (Garon, 2022). This paper is structured as follows. After introductory remarks, I present the concept of cyberbanishment as an alternative punishment. After that, I focus on the requirements that a new sanction needs to meet to become a part of the legal system. I put special emphasis on the issue of efficiency. Then I discuss how cyberbanishment could look like in practice. Finally, the paper ends with conclusions. 1

Concept of cyberbanishment
One of the motivations for considering new forms of punishing people is that the current practices of punishment are highly unsatisfactory (Glasgow, 2015). It is pointed out that despite thousands of years of practice, we still do not know how to punish appropriately (Apt, 2016;Focquaert, Shaw, & Waller, 2020). One indication of this is the overuse of the punishment of imprisonment (Chiao, 2017;Reiter, 2017). High levels of imprisonment are found even in countries where there is a decline in the crime rate (Krajewski, 2016). One of the possible reasons for this is the lack of 'attractive' alternatives for incarceration (see Morris & Tonry, 1991;Tonry, 1996). Technology has the possibility to transform many aspects of our lives, including criminal justice and the sphere of punishment. Technology could be adopted for creating new punishments as well as supplementing old ones, like electronic monitoring (Bagaric, Fischer, & Hunter, 2018;Mamak & Zając, 2015;Nellis, Beyens, & Kaminski, 2012;Wolf, Hunter, & Bagaric, 2018). This paper presents a concept of punishment named 'cyberbanishment,' which refers to the biblical punishment of exile from the Garden of Eden. The idea is to forbid a person who is punished from going to specific places on the Internet. It is not, however, intended to be as punitive as the biblical model. Ideally, the person who is (cyber)banished may eventually return to the online community.
Banishment is one of the oldest public punishments, which temporarily makes some places off-limits to an offender (see Firth, 1978;Warylewski, 2007). Banishment has also functioned as the practice of sending convicts to penal colonies, among others to Australia (see Anderson, 2018). This form of punishment is used by tribal groups in countries such as the United States (Kunesh, 2007) or Canada (C. Miller, 2004, p. 253). But not only indigenous groups use this type of punishment. It is still employed in several states in the US (Carpino, 2010). As a punishment, banishment is usually an alternative to a short prison sentence, which can be imposed if one returns to the prohibited area (Armstrong, 1963, p. 760). Banishment functions in some form in modern societies under different names as well (Borrelli, 2002). For example, the practice of deportation of aliens has many similarities with banishment (Miller, 1956, p. 365), and so do laws that forbid certain categories of people (for example, the homeless) to remain in specified urban areas (such as public parks) (Beckett & Herbert, 2010, p. 6). Another kind of banishment can be seen in the legal need for some sex offenders to mark their places of living (see Levenson & Tewksbury, 2009;Levenson, 2008;Mustaine, Tewksbury, & Stengel, 2006;Richard & Lees, 2007;Tewksbury & Ehrhardt Mustaine, 2006;Tewksbury & Lees, 2006;Tewksbury, 2005). Social pressure on an offender (if unveiled) from members of a community could force them to change their place of residence. There are also designated areas where sexual offenders cannot reside; if they are living in such a restricted zone (e.g. near a school), they are legally forced to move (Saxer, 2008;White, 2008;Yung, 2007;Zgoba, 2011).
There also functioning kinds of banishment imposed by police in specific places in US cities: A form of banishment was introduced in the public housing area in the United States where . . . policies grant police the authority to formally prohibit individuals from entering public housing properties and arrest them for trespassing if they then violate the ban. (Torres, Apkarian, & Hawdon, 2016, p. 1) The broken window theory was at the root of the introduction of this kind of legal instrument (Kelling & Wilson, 1982). Roughly speaking, it argues that permission for minor violations of the law may lead to more serious and harmful consequences (crimes).
There was an assumption that banning problematic people from entering certain areas could decrease the amount of antisocial behaviors. Researchers have tried to empirically evaluate the effectiveness of banishment in reducing crime in public housing (Torres, Apkarian, & Hawdon, 2016). In the cited study, researchers found a reduction in the crime rate resulting from the ban, but they note that it was associated with additional costs that may exceed the benefits. The observed modest drop in the crime rate concerned property crimes; however, the drop was also associated with an increase in arrests for minor offenses in communities of color and had no significant impact on violent crimes.
As it was shown in the brief overview above, banishment is already functioning in the legal systems in different forms, either as a punishment with that name or as other instruments that could be interpreted in that manner. Thus, cyberbanishment is not completely new but the implementation of an old punishment, which is present in the system under many forms. Jones (2000, p. 14) has argued that limiting access to places that were previously open to an offender could serve as a punishment, namely, as controlat-distance: 'to the extent that sanctioning for misuse of electronic access systems involves withdrawal of privileges previously enjoyed by user within the system itself, this potentially introduces a form of (exclusionary) punishment.' His work was published before the dramatic advances of the Internet to become the form that we know it now. He discusses limiting access to systems -mostly having in mind the possibility of entering physical places that are protected by plastic cards, passwords, personal codes, or biological scans -but he also mentions access to virtual space. Jones has claimed that the limit of access to such places has 'punitive moments' and is a form of censure (Jones, 2000, p. 14). He has titled this sanction 'withdrawal of system access authorization' (Jones, 2000, p. 18) The author has concluded that the lack of possibility to access such places is a hardship, which he has called 'punitive moments.' In this paper, I understand cyberbanishment as a tool that forbids the perpetrator from being present in certain places on the web. During the punishment, they are forced to leave the online communities indicated by the judge in the sentence.
To sum up, cyberbanishment is similar to the classic punishment of banishment, which functions in different forms and ways in the legal systems. Cyberbanishment, like its predecessor, seeks to limit the perpetrator's potential locations, except that place is understood in a non-geographical sense, covering the digital spheres where contemporary humans spend their time.

Requirements for introducing new punishments
Despite the roots of cyberbanishment in the long history of punishment, the form I am proposing here is new and, in some jurisdictions, it needs to be introduced to the system before employing it as a way of reacting to crime. Bagaric, Fischer, and Hunter (2018) offer a set of requirements that need to be met before introducing a new punishment. According to them, each new punishment has three mandatory criteria: (a) it must constitute a hardship; (b) it cannot violate important moral limitations; and (c) it must be efficient to impose.
I briefly discuss all of these below. The first criterion is connected with one of the basic characteristics of criminal punishment. According to standard approaches to punishment, punishment needs to be painful and cause suffering (Hart, 2008, 4-5). There is no sense of imposing punishment that the perpetrator would anticipate. Pain (i.e. harshness, negative consequences) comprises an essential element of punishment in both of the primary theories of punishment, retributivism (just deserts in modern form) and utilitarianism (consequentialism) (see Canton, 2020). From the retributive perspective, pain is what the perpetrator should experience for the wrong that they did. From the utilitarian point of view, the pain associated with punishment is acceptable if a greater good may be achieved by its imposition.
One objection that could be raised regarding cyberbanishment is that it would not be universally painful for everyone. There are people who do not use the internet. A response to that is that cyberbanishment as a punishment is proposed for crimes committed via the Internet, so it would target those who are Internet users. In the process of imposition of the proportional punishment, the judge should evaluate how the perpetrator will perceive the pain of cyberbanishment. The severity of this punishment depends to a large extent on the scope of prohibition indicated in the judgment and on the knowledge concerning the customs of the offender. It requires in-depth insight into what the offender is doing on the Internet.
The universality of perceiving pain is not a requirement for introducing punishment to the system. In other words, there may be punishments in the system that would not be considered a hardship for all members of society. There are already examples of reactions to a crime that would only work for specific individuals, like a ban on driving a car would be perceived as painful only for those who drive cars. If the cyberbanishment would be perceived as painful for a certain perpetrator, this criterion (imposing pain) would be satisfied. The authors of the three criteria proposed the punishment of 'deprivation of the Internet.' Cyberbanishment would be something less than that, because it is not meant to deprive perpetrators of their access to the whole Internet but only certain places. However, some of these authors' thoughts are valid also in the case of cyberbanishment: it is clear that the denial of the internet is a punishment. . . . most individuals spend considerable amounts of time on the internet. It is a source of entertainment, connectedness, and vocational efficiency. Therefore, an inability to access and use the internet will necessarily frustrate the preferences of most people, thereby diminishing their well-being. (Bagaric, Fischer, & Hunter, 2018, p. 291) Going back to the theories of justification punishment, it could be claimed that the lack of access to online sites could be considered as a harshness that satisfies the proponent of the retributive position. Imposition of this measure has the potential to cause authentic pain (hardship) to the offender. The threat connected with the possibility of loss of access to the digital communities could also serve as a tool for consequentialists. For example, there is a chance that it could act as a deterrence against committing crimes, since some people cannot imagine existing without the Internet. It could be used as a tool for incapacitation of the offender, who is no longer active on the Internet or able to commit a crime in cyberspace. The second criterion mentioned by the authors is a prohibition of the punishment violating important moral limitations: 'criminal sanctions should satisfy that they should not infringe important moral principles' (Bagaric, Fischer, & Hunter, 2018, p. 293). They claim that this requirement is not clear because we do not always know the content of moral norms, but they provide some examples of situations in which moral limitations are violated. They refer to punishing innocent people, punishing the family of the offender, or corporal punishments. Regarding the latter, they write, 'The deliberate infliction of pain to the body is morally repugnant, and hence corporal punishment is no longer a tenable form of hardship' (Bagaric, Fischer, & Hunter, 2018, p. 293). There is no need to spend much time on that issue. Cyberbanishment, understood as the prohibition of access to certain virtual places on the Internet, is in a sense a part of the collateral consequences of imprisonment (for more on collateral consequences, see Hoskins, 2018Hoskins, , 2019, a punishment that is actually in use. Prisoners do not usually have full access to the Internet (see Jewkes, 2008;Monteiro, Barros, & Leite, 2015;Reisdorf & Jewkes, 2016;Reisdorf & Rikard, 2018). Thus, if that state of affairs is acceptable, it is hard to argue why fewer limitations of freedoms would be morally unacceptable.
In the case of cyberbanishment, the third criterion -the issue of the efficiency of its imposition -seems to be the most problematic from a practical point of view. An intuitive response to the proposition to impose cyberbanishment could be that since it is impossible to control the enforcement of sanction, the perpetrator could have access to the communities anyway, and thus there is no sense in introducing it at all. It is necessary to consider to what extent the execution of measures that limit the possibility of the offender using technology could be controlled. Addressing the problem of efficiency, the authors note, 'Formally, it is easy to prohibit offenders from accessing the internet. A court order proscribing any internet use for a defined period of time has this effect. However, the more complex aspect involves monitoring and enforcing this restriction' (Bagaric, Fischer, & Hunter, 2018, p. 293).
Despite such a caveat, the authors are generally optimistic about the possibility of enforcing this sanction. They point out that the current technology offers more technical tools for blocking content on the Internet (Bagaric, Fischer, & Hunter, 2018, 294-95). The first step toward executing a sanction of the Internet is blocking access to it on the personal devices of the offender. They are also try to deal with the problem of the ways of bypassing any given locked device. This does not solve the problem of the possibility of gaining access in other ways. They mention three problems that they consider to be evident and should be discussed in order to support introduction of the sanction: (1) multi-user places and households; (2) telecommunications; and (3) external access points. (Bagaric, Fischer, & Hunter, 2018, p. 295) One obvious problem is the possibility of gaining access to the Internet through devices shared by housemates. They argue that part of the sentence could be 'the requirement for others within the household to deny internet access to the offender, with regular checks by corrections officers to ensure that this prohibition is not being breached' (Bagaric, Fischer, & Hunter, 2018, p. 295). This entails involving family members or other people who share a house with the offender in order to successfully execute the sanction. By 'communication,' they understand the possibility of using the other functions of a cell phone during the time of enforcement. They argue that it is not a significant problem, because an Internet connection is not necessary for such functions as texting or making phone calls. These two spheres are independent and there is still the possibility to use some functions of the cell phone even without access to the Internet, which could be limited at the same time (Bagaric, Fischer, & Hunter, 2018, p. 295).
The most problematic is the last issue, which is the possibility of accessing the Internet from external locations. They admit that it is hard to guarantee 100% success, but 'it is important to note that appropriate inspection and monitoring by corrections officers will ensure an adequate level of compliance' (Bagaric, Fischer, & Hunter, 2018, p. 295). They also point out that while there is no way to ensure that the offender will not access the Internet at all, this does not have to mean that such a sanction cannot be introduced to the system. There are other sanctions in the system that are difficult to control. They present the example of 'the cancellation of a motor vehicle license' (Bagaric, Fischer, & Hunter, 2018, p. 296). It is almost impossible to ensure that an offender does not drive a car during the time that the ban is in force; however, it does not prevent judges from imposing such a limitation.
In my article, dealing with the same issue, I consider a similar example. In the Polish criminal system, it is possible to prohibit a person from leaving the country during criminal procedures (Mamak 2011). Sometimes this requires the deposit of the accused individual's passport at the time of the ban. However, in the Schengen Zone, there are generally no border controls, and the offender who should stay in the country could go from Poland to Portugal without being stopped. If the procedural body discovers that the accused has crossed the border in spite of the ban, they face the possibility of more severe measures, including pre-trial detention (Mamak 2016, p. 46).
As the authors of the cited paper concluded: The efficacy of the sanction stems from the assumption that many offenders will generally comply with a court order, or if they do not comply, then many of those who violate the ban on driving will be detected through law enforcement measures, including license checks and targeted detection. It is not known what portion of suspended drivers adhere to the ban, but at least there is a clear, tenable mechanism for detecting those who flout the ban, and more severe sanctions for those who defy the terms of the initial punishment. (Bagaric, Fischer, & Hunter, 2018, p. 293) They take the view that a lack of guarantee that the punishment is fully executed in the case of every offender should not exclude that sanction from being adopted by the system. In the process of imposing a penalty, the judge could take into consideration the chances of the punishment being successfully executed, and if they deem that there is no way for the offender to follow the imposed ban, he or she could impose another sanction instead.
To sum up, the problem of enforcing this sanction is real, but first, it could be overcome, and second, it seems to not be an irreconcilable argument. There are technical possibilities to forbid particular devices to gain access to particular parts of the Internetor the whole thing. Furthermore, the development of the Internet would indicate that, with time, the tools of control will likely to be more efficient. There is also the possibility to engage others in the process of executing orders -be they housemates, family members, and professional correctional officers. Even if it is impossible to guarantee that an offender cannot find a way to circumvent the ban, that does not mean that such regulation is entirely unacceptable. There are examples of restrictions imposed by the state that are based on its authority combined with the threat that more severe restrictions may be imposed. This is the case with the cancellation of a driving license (with no confiscation of the offender's car by the police), or a sexual offender not being allowed in proximity of schools (but not always being monitored 24 hours a day). The potential issue of enforcing new sanctions should not prejudice against the idea of the proposed sanction.
There is also an element of symbolism embedded in this punishment, which has a communicative component (see Duff, 2001;Feinberg & Sugden, 1965). It may be important to 'just' communicate by the judgment that certain behaviors are not allowed, and there are consequences that may include a form of banishment from the online community.
One issue which could be problematic is the time in which decisions are made. Sometimes there may be a need to immediately banish someone from specific online spaces. However, imposing cyberbanishment as a punishment requires a sentence imposed after a legal process that could take months or even years. There would still be a need to ban that person in relation to the platform, and constitutional issues remain. My response is to use that kind of legal institution as a precautionary measure. Just as the time of pretrial detention served is deducted from the imprisonment sentence, in the case of cyberbanishment the time that it was imposed as a precautionary measure may be deducted from future punishment. In such a situation, this decision -like with the pretrial detention -may be made by the court almost immediately.

Introducing cyberbanishment to the legal system
Focusing on the practical issues of imposition, we could consider how cyberbanishment as an instrument of criminal law might be organized. The characteristics of traditional banishment include the requirement to not enter specified places and the temporality of the ban. Cyberbanishment could have both those features. An offender could be obliged to refrain from visiting internet sites, as determined by the court. The ban should be temporary (for example, one year). It should not cover the Internet as a whole. The court or another legitimate official should determine the list of places where the offender will be prohibited from visiting. The violation of such a ban should cause more serious legal consequences. Such a legal instrument does not require the agreement of the tech companies to ensure that the prohibition is respected, just as prohibition against driving a car does not require engagement with car dealers or car rental companies. The offender knows that if they breach the ban, they will be subject to more severe consequences.
The instrument could be used in many circumstances (for example, in the case of members of a group of football hooligans, pedophiles, gambling addicts, racists, or neo-Nazis). Research has demonstrated that participation in online groups can lead to radicalization (see Koehler, 2014). There are places and communities on the Internet that have the potential to impact people negatively and strengthen extremist attitudes (Holt, Freilich, & Chermak, 2017, 5-6). It has also been suggested that '[c]ompared to other "socialization institutions", such as offline group activities, music and concerts, rallies and political trainings, the Internet appears as the most important element driving individual radicalization processes' (Koehler, 2014, p. 131). This instrument could potentially cut off specific people from such places.
This punishment, as suggested before, could be used for the crime that was committed via the Internet. Examples of such infractions are hate crimes, defamation, and punishable threats. There is also a discussion about the criminalization of spreading fake news (misinformation/disinformation), and in some countries it has already been partially introduced (Beech, 2018; Helm & Nasu, 2021;Statt, 2019). Elsewhere I propose the criminalization of spreading fake medical news (Mamak, 2021b(Mamak, , 2022. Cyberbanishment may be a useful punishment for such crimes. Here I want to show how an instrument could be worded. For that, I use a legal instrument from the Polish legal system that realizes the purposes of the abovementioned concept of cyberbanishment. 2 In the Polish criminal code, there is a way of reacting to crime that is called 'prohibition from associating with specific social groups or appearing in specific locations' (Wróbel, Zontek, & Wojtaszczyk, 2014, p. 39). This legal instrument may be used already as a described concept of cyberbanishment; however, it needs to be noted that its applicability to online spheres is my own suggested interpretation, which so far is used for such purposes. 3

Conclusions
In this paper, I presented the punishment called 'cyberbanishment.' This sanction refers to the old-fashioned punishment consisting of forbidding a person from being in specific places, but instead of physical locations this sanction applies to online spheres. The proposition is based on a couple of observations. First, the current practices of punishment are problematic, and there is a need for alternative punishments. Second, this sanction may be useful for punishing crimes committed online, like hate crimes or punishable threats. Third, the development of virtual realities and metaverses indicates that people will be spending more time online.
Moreover, the introduction of such a punishment may increase judicial control over the presence of citizens on online platforms. Bassini (2019, p. 182) has argued that there is a shift in the paradigm of fundamental rights protection on the Internet, which manifests itself with the delegation of enforcement of such rights to private Internet operators. The significant advantage of the imposition of such sanctions by the court is judicial control of that decision.
Cyberbanishment is a punishment that may contribute to making criminal law more adapted to the changing reality. To show that the law is also working in the online sphere and to make the online world a better place.

Notes
1. Before I go any further, I have one reservation. The proposed sanction is designed to fit with modern democracies where the rule of law is respected. It is intended as a sanction to assist prospective criminal reactions to crime, as an alternative to incarceration, and as a means of avoiding more serious crimes and thereby causing more harm. However, it is easy to imagine how the proposed punishment could be misused, either to silence inconvenient journalists or to fight political opponents. The idea presented in this paper was first articulated in a chapter of my book published in Polish (Mamak, 2021a). 2. It should be noted that there are fundamental differences concerning how new punishments are introduced. There are legal systems in which a judge can create completely new punishments and those that require the punishment to be described in the law before its usage, which means that it is also impossible to impose punishment that is unspecified by law (see Waltoś & Hofmański, 2020). The Polish legal system belongs to this second group. In Poland, the means of reacting to crimes are divided into two categories: punishments and penal measures (Wróbel & Zoll, 2014). The penal measures under the previous criminal code were called 'additional punishments,' and the change was, among others, motivated not to treat them as additions to punishments but also as an independent means of reaction to a crime (see Górowski & Szewczyk 2016). Thus, this instrument is a punishment in the broad sense, as a punitive reaction to crime. 3. The mentioned legal instrument is listed in Article 39 of the Polish Criminal Code containing the lists of penal measures. I have argued elsewhere that there are no obstacles to interpreting it in relation to prohibiting participation in online communities (Mamak, 2019, 173-77).

Disclosure statement
No potential conflict of interest was reported by the author.

Funding
The work was supported by the Academy of Finland [333873].