Post-mortem privacy and informational self-determination

Post-mortem privacy is becoming a vital topic of public and scholarly legal concern. Post-mortem privacy is understood as the right of a person to preserve and control what becomes of his reputation and dignity after death. The assumption that the deceased does not qualify for privacy rights, because his bodily presence has been terminated, no longer holds in our networked society. In the digital age, the phenomenon of the digital legacy that an Internet user leaves behind after his demise, has led to new challenges for the legal system. The deceased is no longer in a position to exercise human autonomy as an active agent. The article reconsiders the notion of human autonomy with regard to these digital representations. Taking the point of view that the control over personal information (also known as informational self-determination) is essential in protecting one’s privacy in the antemortem life, the article explores whether this principle may have validity in the postmortem context. Legal philosophical arguments are advanced in a discourse about the quandary if digital personae of deceased persons can be bestowed with a legal basis of personality rights and concomitantly privacy rights. Therefore much attention is given to the problem of the subject, which does not seem to be functioning in the case of the absence of a living subject. Briefly referring to novel personae, it is argued that fundamental human rights need not be limited to the rights of living human beings.


Introduction
The notion of post-mortem privacy increasingly appears to be becoming a vital topic of public and scholarly legal concern. Post-mortem privacy is understood here as the right of a person to preserve and control what becomes of his reputation, dignity, integrity, secrets or memory after death (Edwards and Harbinja 2013). In the digital age, the phenomenon of "digital remains" or the digital legacy 1 that an Internet user leaves behind after his demise, has led to novel problems. Not only the detachment of private communications such as e-mails from the physical plane (McCallig 2013) but also the intangible nature of digital assets such as bank accounts, e-books and iTunes have raised intriguing new challenges for laws of inter alia ownership (Maimes 2013). 2 Laws of ownership will have to come to terms with the bare fact that there is no longer a physical artefact or manuscript to possess or in which to claim ownership.
The locus for the digital presence is no longer circumscribed by physical attributes but more by the content of 1 "Digital remains" is the collective term used to describe the expressions, possessions and impressions that a decedent leaves behind in digital media. 2 Before digital doubles, the same questions were posed about photographs (Warren and Brandeis 1890). The observation of the photographer of Henry Kissinger, Richard Avedon, is pertinent here. He asserted that his photograph was a Doppelgänger, a likeness with his own life, an inexact twin whose afterlife may overcome and replace the original [underlining mine] (Floridi 2005).
Abstract Post-mortem privacy is becoming a vital topic of public and scholarly legal concern. Post-mortem privacy is understood as the right of a person to preserve and control what becomes of his reputation and dignity after death. The assumption that the deceased does not qualify for privacy rights, because his bodily presence has been terminated, no longer holds in our networked society. In the digital age, the phenomenon of the digital legacy that an Internet user leaves behind after his demise, has led to new challenges for the legal system. The deceased is no longer in a position to exercise human autonomy as an active agent. The article reconsiders the notion of human autonomy with regard to these digital representations. Taking the point of view that the control over personal information (also known as informational self-determination) is essential in protecting one's privacy in the antemortem life, the article explores whether this principle may have validity in the postmortem context. Legal philosophical arguments are advanced in a discourse about the quandary if digital personae of deceased persons can be bestowed with a legal basis of personality rights and concomitantly privacy rights. Therefore much attention is given to the problem of the subject, which does not seem to be functioning in the case of the absence of a living subject. Briefly referring to novel personae, it is argued that fundamental human rights need not be limited to the rights of living human beings.
1 3 the information the Internet user provides. Indeed, the digital persona 3 creates, shares and collects content at a staggering rate. As the shift to the digital continues, careful stewardship of digital content, which can, in some sense, be said to be a rich reflection of you, is more and more necessary (Carroll and Romano 2011). However, when the Internet user wishes to assume the role of a responsible steward, he finds Internet providers barring the way. It is curious to note these providers pretend to do so exactly for the sake of protecting the privacy of their user. 4 Some providers even stake a claim of ownership in their customers' e-mail accounts under the guise of this being necessary to protect the user's privacy (Atwater 2006;Wilkins 2011). 5 The social norms for expressions of bereavement on the Internet over the loss of loved ones are still evolving. A novel aspect in this development is the persistence of the digital persona after the physical demise of the real life person it represented. 6 Today's digital natives 7 have begun to project their own sensibilities onto rituals and discussions 3 Cf (Clarke 1994), who defines the digital persona as "a model of an individual's public personality based on data and maintained by transactions, and intended for use as a proxy for the individual''. In Clarke's original definition, the digital persona is then not capable of independent action. However, Clarke in 1994 foresaw active digital personae and perhaps even autonomous digital personae (which we might call software agents or avatars). See also (Roosendaal 2013) Bernal sees the online identity "as an 'extension' of the offline identity and personality of the individual concerned" (Bernal 2012). 4 One of the first cases that led to a judicial verdict balancing contractual obligations and privacy interests of decedents, was that of U.S. Lance Cpl. Justin M. Ellsworth, who was killed in Iraq on November 13, 2004. He had a Yahoo! e-mail account that, according to Yahoo!'s policy terminated upon death with all contents therein permanently deleted. After his death, Ellsworth's family requested to get access to his messages but Yahoo! denied this, in accordance with their policy (Herold 2010). Case drawn from In re Ellsworth, No. 2005-296, 651-DE (Mich. Prob. Ct. 2005. See also (Kopetski). Google states unequivocally that they are "keenly aware of the trust users place in us, and we take our responsibility to protect the privacy of people who use Google services very seriously". Accessing a Deceased Person's Mail, GMAIL, http://mail.google.com/support/ bin/answer.py?hl=en&answer=14300 (last visited May 4, 2012). According to McCallig (McCallig 2013) a recent California case (In RE request for order requiring Facebook Inc. to produce documents and things), California ND, Case No: C 12-80171 LHK (PSG) confirmed that service providers cannot be compelled to provide the contents of a decedent's online account. Otherwise they would violate the Federal Stored Communications Act 18 U.S.C. # 2701. 5 But see (Walters 2009). 6 Kasket 2013 proposes to call the digital remains that has the potential to persist, as facilitated by the digital age, a 'res digitalis' or digital being. It falls somewhere "in the middle of physical being and being of mind". 7 Digital natives are understood as being born after 1980, have access to digital technology and have skills to use digital technology in relatively advanced ways. surrounding death. They are starting blogs, YouTube series and Instagram feeds about grief and loss. The conversation about bereavement and the deceased is now no longer a private affair but it enters a very public forum. 8 Most notable in this respect is the memorialization feature of Facebook.
Commercial parties also seek to develop services that accommodate the assumed needs of this category of Internet users. Services are springing up that address the idea of digital immortality through various means like headstone QR codes 9 that lead to Internet shrines or services that allow you to program an avatar 10 that is intended to communicate on your behalf after your death (Limer 2011). Technology in the form of websites, like MyLastWish and Heavenote, allows the deceased to connect about their final dreams with or to send a final message to their loved ones, after they are gone (Maimes 2013). These kinds of websites give you some control over what will happen to your reputation, as it emanates from your digital presence after your death.
This article will focus on the privacy-related dignitary aspects of the Internet user's digital legacy. It will especially discuss whether there is such a notion as post-mortem privacy that is worth protecting. The primary issue is whether it can be upheld that a legal subject should enjoy post-mortem privacy, particularly in relation to personal information. Taking the point of view that the control over personal information (also known as informational selfdetermination) 11 is essential in protecting one's privacy in the ante-mortem life (Buitelaar 2012), this article will explore whether this principle may have validity in the post-mortem context as well. The research question this article seeks to answer, is: Does an appeal to informational self-determination, considering its importance for human dignity, justify 8 Seligson (2014). 9 Research is also being done on gravestone technology that allows you to hold a tablet/phone in front of the headstone, which it recognizes and brings you to a memorial website (http://nottinghamtrent. academia.edu/Philip/Wane). 10 The website ETER9 is a social network (in BETA stage) that relies on Artificial Intelligence. One's Virtual Self creates a Counterpart that will interact with the world just like one would if one were present. The more one interacts in the network, the more the Counterpart will learn. By in this way eternizing, one keeps one's thoughts and posts for all time https://www.eter9.com. 11 The German Constitutional Court traced the right to privacy to the fundamental right to the free development of one's personality. In article 2.1 of the German Constitution it says: the value and dignity of the person based on free self-determination as a member of a free society is the focal point of the order established by the Basic Law. The general personality right as laid down in Arts 2 (1) i.c.w. I(1) GG serves to protect these values (…).

3 that individuals exercise control over their personal data after death?
Although this question might be approached through an argument based on claims of living persons to respect their post-mortem dignity, this could only provide a limited form of protection, as it would be restricted to honoring antemortem expectations. The aim of this article is to present a more fundamental, forward-looking theoretical argument that may be able to provide a more comprehensive form of protection for the various forms of post-mortem presence that are likely to subsist in the future. It appears that the often-held straightforward assumption that the decuius does not qualify for privacy rights, because his bodily presence has been terminated, requires reconsideration. The article therefore spends much time on the problem of the subject, which does not seem to be functioning in the case of the absence of a living subject. By approaching the question from a more philosophical point of view (Mill's concept of liberty and Dworkin's concept of life-transcending integrity) an attempt is made to show that the objective rights, such as enshrined in the European Convention of Human Rights (ECHR) and the Universal Declaration of Human Rights (UDHR), cannot be limited to the rights of living human beings but persist in vicarious digital personae. To further underpin the argument, various other lines of argument are developed such as respect for the dead by way of being true to promises, respect for the reputation of the persisting public persona and the composition of your identity in narrative form which also persists after your death. Brief reference is finally made to recent developments in the legal and philosophical sphere, where the legal system increasingly is taking note of personae that have no bodily presence or are, at least, non-human 12 . The article suggests it may be promising to build on these construals a conception of post-mortem privacy. If this hypothesis holds, the digital remains of individuals subsisting in the form of digital personae would then be granted a kind of privacy in the sense that they should be allowed to control 'their' personal data, constituted by data flows created in their ante-mortem and in a limited sense, their post-mortem phases.

Background
Social networks are increasingly places where millions of users reveal a lot about the most intimate details of their 12 In this article no attention will be paid to the rights of the nasciturus even though many similarities can be discerned. Main reason to disregard the nasciturus is that he has no previous existence (Rombach 1963). lives. These details cover all aspects of life, but due to the fact that the average social network user was at first mostly of a relatively young age, death was not a particular concern for the participants. Users relatively seldom died. The social network sites accordingly did not really worry about the ramifications that follow the death of a user. In recent years, however, the problem has become more urgent because more and more people over 65 are signing up (Wortham 2010) 13 and by definition this age group has a shorter life expectancy than the so-called digital natives. Facebook 14 apparently wishes to accommodate its users also in this phase of their existence as Facebook client by providing for pages where they permit their bereaved to share their mourning process with other bereaved, the socalled memorial pages. However, it is not always clear for the average Facebook user what steps they have to take to have Facebook handle their profile after their demise according to their personal preferences. 15 The very personal nature of the information that users entrust to the various Internet platforms such as social network sites (e.g. Facebook and LinkedIn), gaming (World of Warcraft) and communication (e-mail, Twitter) poses a 13 This happens at a rate of about 6 million per month (at least in 2009). In Great Britain, the increase was one million in a year (2009) (Wallop and Roberts 2010). In 2014, recent data from the Office of National Statistics showed that internet use more than tripled for those 65 and older between 2006 and 2013. 14 When someone dies and Facebook is informed, the profile is "memorialized". No one can access it or change it, although those on the friends list can continue to view and interact with the profile as before, posting material on the Timeline or sending messages to the deceased. In February 2015, Facebook extended its rather strict memorialization feature by introducing (only in the US) a so-called legacy contact. The memorialized account is, in fact, turned into a more active entity because, to a very limited extent, this account manager (a friend or family member, selected by the Facebook user) can manage the account after its owner's decease. The legacy contact will be able to write a post at the top of the memorialized Timeline, announcing the memorial service, respond to new friend requests who were not yet connected on Facebook, update the profile picture and cover photo, download an archive of the photos, posts and profile information (Callison-Burch et al. 2015).Other interactive websites have a variety of policies (Mazzone 2012). Twitter, for example, promises to close the account of a deceased user and provide family members with an archive of the user's public Tweets. LinkedIn simply closes a deceased member's profile after receiving a "Verification of Death" form. Google knows an Inactive Account Manager that waits a certain period before sending a mail if the user remains inactive. If no response ensues, Google notifies a 'trusted contact' (James and Magee 2013). 15 Facebook approaches the death of a profile user as a way of tidying up inactive and redundant accounts. To a large extent, Facebook depends on the report of other users that a user has passed away. Facebook also uses a scanning approach of posts using terms like Rest In Peace but this is open to pranks. There is no section to correct the consequences of the pranks which sometimes lead to embarrassing and painful consequences.
1 3 challenge to, and raises questions about, traditional experiences and concepts of privacy and how we regulate it. Facebook, for example, allows users to grant access to their profile to a chosen few while others offer even the intimate details of their personal lives to the general public (Kasket 2013). 16 Ever since Goffman (Goffman 1959), the sociological theory of role-playing has shown that everyone consciously or unconsciously is adopting behaviors that are designed to control others' levels of access to ourselves and to information about us. On social network sites such as Facebook, one attempts to take care of one's privacy by a constant dialectic between openness and concealing, depending on the context. Nissenbaum outlined a privacy framework of expected information flows, called contextual integrity. When the flow of information runs along the lines of established norms within a specific context, no uneasy feeling of a privacy infringement occurs. When the flow of information does not meet the contextual expectations, however, a violation has occurred and contextual integrity has not been maintained (Nissenbaum 2010). This violation is also often construed as an inappropriate transfer of personal data across boundaries of what we think of as separate spheres of access (Ambrose 2013). Contrary to the general opinion that Facebook means the end of privacy, it may well be argued that Facebook, with its evolving privacy policies, permits the user to pay much more attention to regulatory processes to safeguard his privacy than was the case in the analogue age. If the user takes the time to exercise these regulatory possibilities, he in a sense asserts control over transfers of personal data between spheres. 17 Privacy theorists call this the principle of informational self-determination [see (Buitelaar 2012) with references and a discussion of this concept].
When the user of the Facebook profile has died, the question may well be posed if privacy considerations for our digital selves still play a role. Most importantly, the Internet user is no longer actively able to make decisions about access to the digital storehouses of personal 16 For the methods Facebook users employ to control the information available about them, Kasket (Kasket 2013) uses the term "privacy regulation" within a psychological framework. It refers then to behavior that is designed to control access to the self. 17 A caveat is in place because the staggering number of users of Facebook seems to make it impossible for the individual user to determine precisely in which context the information about him is used. In fact, it is better to speak of a presentation of self in relation to the generalized other. Just as in the ante-mortem situation, however, a whole range of regulatory instruments and legal institutions are potentially suitable to protect post-mortem privacy, e.g. laws of privacy, breach of confidence, intellectual property, personality, publicity, defamation, succession, executory and trusts and data protection. It may be argued that thus the user is given the opportunity to exercise control over his presentation. I will return to this later. information he has left behind. Still, Facebook profiles in most cases persist. Facebook's memorialization policy can be said to repurpose these profiles after death as a site of mourning, memorialization and continued communication with the deceased (Kasket 2013). Besides challenges to the management of the privacy of these self-representations, a whole new set of questions in the area of ethics, psychology, cybernetics and law arises. It is evident that, for one, the deceased no longer is in a position to exercise human autonomy as an active agent in the traditional, legal sense. In light of the overwhelming presence of deceased human beings 18 and digital doubles on the Internet, it seems worthwhile to reconsider the implementation of the notion of human autonomy with regard to these digital representations of the self. This is especially so because many experience the vicarious version of their digital double as being almost a better representation of their selves. 19 Some theorists hypothesize that in the Internet era, this entity, created and facilitated by the digital age,-"a being that falls somewhere in the middle of physical being and being of mind"- (Kasket 2013) 20 persists after the death of the Facebook user. It will be a challenge for existing legal frameworks to accommodate these entities in a way that will safeguard their privacy in a manner that is executable and sensible in the Internet era.

Problem of the subject
Considering that the crucial problem with recognizing a right to privacy in the case of a deceased person is the lack of an active agent, it is pertinent to briefly discuss here what has been called the 'problem of the subject' as it pertains to the post-mortem phase of a person.
By the problem of the subject, I mean the implausibility of interests persisting in the absence of an interested subject. 21 It may well be argued that, when a person's public persona in the sense of his reputation, is falsely blackened after his death, the ante-mortem person is also defamed. Assuming that a person's immaterial interests may survive his death, and that by death or other subsequent events 1 3 these interests are not fulfilled or defeated, one may very well regard the ante-mortem person as harmed (Feinberg 1987). In other words, the interests that survive the person in question are harmed by thwarting or nonfulfillment even if the person does not suffer subjective disappointment. If this is so, the subsisting subject may be conceived of as a moral actor. However, this still does not permit attributing rights to the deceased but subsisting subject in the sense that it would bestow him with legal agent status. Indeed, rights are a particularly agent-oriented mode of moral discourse. Rights, on the other hand, do not exhaust moral reasoning (Winter 2010).
Feinberg has argued that postmortem persons cannot be harmed, for a moldering corpse has no feeling. He, however, builds an argument that posthumous harms can retroactively harm the interests of ante-mortem persons. Take for example the unfortunate event that the life-insurance company, where the deceased had protected the interests of his loved ones, collapses 5 min after that person's death, then his ante-mortem interest in his loved ones' security is harmed. In other words, it can be argued that knowledge is not a necessary condition of harm before one's death. In the case of the collapsing life-insurance company, it can at the least be argued that this person himself in his ante-mortem stage is misfortunate because his interests will be thwarted.
Feinberg argues that what was true all along has now become apparent. 22 Moreover, this raises the question if the situation really changes as to the precondition of being aware of an interest being harmed before or after death. It may well be stated then that the objection that dead men are permanently unconscious and that therefore they cannot be aware of events as they occur, and that they therefore do not have a stake one way or the other in such events, does not hold.
The pivotal question with the 'problem of the subject', both in moral and in juridical terms, remains whether the individual dead can present practical or moral reasons or interests on which claims can be grounded. Besides interests that are squelched the moment the subject dies, the dead person's surviving interests must be the interests of someone or other who has a stake in them. How could a moldering corpse have a stake in these interests, it is often objected. Even so, there are many interests of living persons that can be violated without their ever becoming aware of it. Arguably, it is a plausible analogy that, since knowledge is not a necessary condition for harm in the antemortem stage, it is not a necessary condition either in the post-mortem stage. Feinberg posits convincingly that the interest one has in a good reputation is the best example of 22 J. Feinberg, Harm to Others, (Vol. 1), (Oxford University Press, 1984) as cited in (Winter 2010). the self-regarding category also when one is no longer in a position to be injured, wronged or harmed by it. Obviously, self-centered interests provide the highest degree of fulfillment, satisfaction and pleasure when one is able to achieve them in his life-time and that they are brought about by us, or if not by us, then for us. In case these interests are thwarted or even made senseless after one's death by the occurrence of harmful posthumous events, it can be said that this event is responsible for harming the ante-mortem person. However, this should be interpreted as entailing no more than that the ante-mortem person was harmed in being the subject of interests that were going to be defeated whether he knew it or not. My friends may then very well feel sorry for my descendants but also for me, even though I am dead. In this line of reasoning, it may be upheld that the harm principle also encompasses the retroactive harm that not only an ante-mortem person suffers by events that occur after his demise but also by vicarious or posthumous events that harm his persisting digital double even though he has no knowledge of it.
Applying the harm principle to the digital persona surviving the physical person's cessation, it is helpful to expand the argument by understanding the definition of personality to include not only our abilities to believe, learn, feel and so on but also the characteristics of a public persona, who subsists in speech, memory of living persons as well as in information held in impersonal media (Winter 2010). In a sense, the theory of the public persona that survives the cessation of the physical person may be said to lend support to the theory that a phenomenon such as posthumous claims are not inconceivable because from the above it may be deduced that there is a complex posthumous persona available to bear claim-ascription (Winter 2010). On the other hand, Winter also recognizes that a public persona that persists in media and memory differs little from a fictional character. It is hard to believe that a fictional character can present us with a moral claim.
Yet again, a moral-legal argument may be made for the plausibility of posthumous claims by referring to the thesis that the blocking of a person's interests contravenes the obligatory respect for a person's end-setting interests. A person's interests have a goal that lies in the future and as argued, quite often, extend beyond his physical presence. Obviously, interests will only exist if there is an interested subject, who is an entity that is able to make reflexive value judgment. There must be a close connection between "being interested" and "having interest". One can only have the latter if one has the capacity for the former (Winter 2010). Moreover, moral claimants such as human agents are special in contrast to non-interested things or even 'agents' like computers, because the entity must be the sort 1 3 of thing to which things can matter, i.e. that something could be of interest to the entity in question. 23 Indeed, things that matter to humans may lead to moral claims. In turn, these are necessary to interests upon which claims are grounded. These claims must, however, be of significance to the claimant. Because, during my life, the various aspects of my personhood form a more or less cohesive whole, the status of my reputation (my public persona) matters to me regardless of what I know concerning possible changes to it (the so-called no-effect injury) (Winter 2010).
In contrast to Feinberg, Winter does not believe that noeffect injuries provide a sufficient ground for the analogy to base posthumous claims on (Winter 2010). Posthumous defamation is still wrong, not only on legal grounds but more so on moral grounds. Swennen argues that general notions of the protection of human dignity in this context might be said to transcend the interests of a random legal subject (Swennen 2013). Winter's discourse leads then to the conclusion that it can be doubted whether an ante-mortem person's post-mortem fortune can ground posthumous claims. Winter quite convincingly states that the original problem is merely reposed by attempting to infer from the non-necessary character of experience (the no-effect injury) the non-necessary character of an interested subject (Winter 2010).

Personality rights-is there a ground for their post-mortem persistence?
Personality rights "Person" can refer to a sort of entity that is subject of moral predicates such as "right" and "duty", but it can also refer to the unity imposed on a diversity of mental elements, such as memories, loyalties, preferences and opinions. A person's physical, mental and moral identity is protected by the rights he holds as a juridical person and legal subject (Swennen 2013). This can be done in the form of a negative protective right or of a positive right, both in its horizontal and vertical dimension. The juridical person holds sovereignty over a spatial zone which is determined by his body. The discretionary competence over this spatial zone can be construed from a broader perspective in the sense that it is essential for a person's personal autonomy to determine by his own choice what enters his experience. From a rights theory point of view, the right to privacy can then be seen 23 Winter refers to C. Taylor, "The concept of a person" in Human Agency and Language; Philosophical Papers Volume 1 (Cambridge, Cambridge Press, 1985) p. 98. as a subjective right as well as an objective right. 24 Rules of objective right line out which attributes and goods constitute a person's unique identity. These can be both material as well as immaterial. Privacy as a subjective right shows that only the individual concerned holds the active right to decide who can interfere with his inviolable personality. Privacy as an objective right denotes an individual's inalienable interest in worthwhile elements of his personality, the inviolability of his own space in which to develop and flourish. It is asserted here that the right to privacy is, in fact, a mix of these two types of rights.

Post-mortem persistence of personality rights
In the English common law system, there is a long recognized principle of actio personalis moritur cum persona: personal causes of actions die with the person, such as defamation claims (Harbinja 2013). This pertains to especially moral rights of authors. Countries with a civil-law tradition appear to be more inclined to recognize the persistence of similar rights after death (Harbinja 2013). This is prominently so in Germany where the Grundgesetz (GG) is most outspoken in the defense of the inviolability of human dignity. In Germany, the protection of human dignity derives from its highest constitutional source, the German Constitutional Law. This has been applied to the protection of the dignity of the deceased. In the Mephisto case, the Federal Constitutional Court ruled that "[i]t would be inconsistent with the constitutional mandate of the inviolability of human dignity, which underlies all basic rights, if a person could be belittled and denigrated after his death. Accordingly an individual's death does not put an end to the state's duty under Art 1.1 GG to protect him from assaults on his human dignity" (BVerfG 30, 173).
Laws of testation, on the other hand, relate only to what happens to the assets with economic value someone leaves behind after his death. These assets are then attached to the inheritor who, in this sense, continues the personality rights of the decuius. 25 In so far as personality rights such as the 24 European Continental writers distinguish between subjective and objective rights (Campbell 2013). Subjective rights are primarily focused on the rights of individuals, conceived of as political and economic claims but also as liberties, faculties or powers. Natural law developed over the centuries into objective rights, a conception of the attributes that make the individual person unique, "a force of the soul" associated with human rationality. (B. Tierney, The Idea of Natural Rights: Studies on Natural Rights, Natural Law and Church Law, 11-50-1625. Atlanta, Georgia: Scholars Press, 1997) as cited in (Pennington 1998). 25 This is also called the principle of saisine: le mort saisit le vif which means that the deceased is continued in the person of the inheritor (Hartkamp 2013).
1 3 right to publicity can be expressed in economic terms, these can be inherited by the inheritors (Swennen 2013). On the other hand, it can be asserted that the subjective rights and duties attached to the person of the decuius are not transferred. This is only logical because a personal, subjective right is exercised in relation to one or more persons. When that natural person dies, an important part of the personal right is extinguished. In this strict juridical interpretation, laws of testation do not shed any light on the question of who inherits the persistence of the non-economic, digital elements of the deceased's physical, psychic and moral personality as represented in the vicarious digital double of his ante-mortem self. They appear to die with the demise of the living holder of the personality rights.
However, as noted, the overwhelming persistence of digital personae on the Internet, without there being a living counterpart who can actively control the subjective rights associated with this persona, calls for somehow accommodating post-mortem digital personae with an appropriate locus in the legal framework that governs the survival or extinction of the rights and duties of subjects. In the context of the research question, the following, concomitant question becomes relevant: how can the issue of protecting the informational privacy of the digital double that subsists post-mortem be given a normative basis beyond purely economic terms, which at present is the predominant concern of laws of testation?

Human dignity
A promising approach is to take account of the opinions contained in German doctrine on the natural right of human dignity. As noted, the German Grundgesetz considers it only logical that the absolute protection of the inviolability of human dignity that a human being enjoys during his lifetime, persists after his demise. Immanuel Kant has stated in this respect, that everyone is entitled to adopt the defense of the bona fama defuncti of a decuius and that this should be regarded as part of the Recht der Menschheit. 26 Kant underscores, in other words, that at least the reputation of the deceased belongs to the innumerable moral judgements that enjoy nearly universal agreement (Mulder 2015). Later, I will return to the view of the digital representation of the deceased as, in fact, being comparable to the public persona the living person presents on the Internet. Indeed, the protection of the roles the digital person plays on the Internet, is considered by many theorists (Marmor 2015) today to be the protection of the right to privacy. As such, privacy can be regarded as a social good and not merely as an individual right (Regan 2015 Furthermore, the question may be posed whether the concept of informational self-determination, as it provides a conceptual basis for protecting the privacy of natural persons in their ante-mortem phase, can be transposed to the post-mortem phase of these natural persons, in order to serve as a starting point for the quest for a legal framework that takes proper cognizance of the persisting digital presence of the decuius. 27 In the current legal framework, the European Court of Human Rights has refused to recognize the right of privacy for the deceased, unless their privacy is connected to the privacy of living individuals. 28 Nevertheless, the Council of Europe Data Protection Convention (1981) does not mention natural persons when it reflects on the protection of privacy but rather covers "personal data" as being any information relating to an identified or identifiable individual (data subject); indeed, deceased as data subjects are not by definition excluded (McCallig 2014). This conception would then be in line with the general and sometimes deeply felt innate need to grant the deceased some form of legal protection. This pertains not only to the deceased's bodily component (the corpse) but also to his digital remains (Swennen 2013). Approaching the issue in this manner, i.e. from the human rights perspective, presupposes a general interest. As such, the natural human right of dignity component of personality rights might provide a worthwhile starting point for providing the building blocks for the hypothesis that there could be such a thing as postmortem privacy in the sense of informational self-determination.
To show that this exercise is not a priori doomed to fail, it is illuminating to take a better look at some jurisprudence where surprisingly some unexpected opinions on the dignity of deceased persons are expressed. The seemingly undisputed argument that the deceased are by nature unable to hold and exercise personality right turns out to have been questioned by reputable jurists. The dissenting opinion of 27 In the European Union, twelve states recognise deceased as entitled to independent rights (Bulgaria, Czech Republic, Denmark, Estonia, France, Italy, Latvia, Lithuania, Portugal, Slovakia, Slovenia and Spain). Four states expressly exclude the deceased (Cyprus, Ireland, Sweden and the United Kingdom). Ten states require there to be a need of a connection with a natural person (Czech Republic, Denmark, France, Italy (both natural and legal person), Latvia, Lithuania, Portugal, Slovakia, Slovenia and Spain). One state puts a temporal limit on this recognition (Estonia, namely 30 years on consent) 1 3 the ECHR judge Fura-Sandstrom in the Akpinar-Altun v Turkey case is quite revealing. 29 The judge states unequivocally that the State's responsibility to respect an individual's dignity and to protect bodily integrity "cannot be deemed to end with the death of the individual in question". 30 Judge Fura-Sandstrom argues from the Kantian principles that, just as in the Mephisto case, human beings must always be treated as ends and not solely as means. She finds this a compelling ethical principle to apply to Article 3 of the Convention. Indeed, from a textual point of view she can support her case further because Article 3 does not put any restrictions, qualification or exceptions to the blank statement that no person shall be subjected to torture or to inhuman degrading treatment or punishment.

Autonomy
To elaborate on the philosophical basis for the subjective/ objective rights approach to the search for an answer whether informational self-determination as grounded in the human right to dignity may play a role in safeguarding the privacy of the deceased, a discussion of Mill's concept of liberty is appealing and apposite at this point. John Stuart Mill was, I believe, the first to draw the distinction between self-and other-regarding decisions, which provides a useful guide for mapping the boundaries of personal autonomy. Mill states unequivocally that one of the elements of personal autonomy is the right to determine by one's own choice what enters one's experience. 31 This is one of the various things that is meant by "my right to privacy" (Feinberg 1983). Regan (Regan 2015) underscores that the Millian view on privacy regards this institution as important for the development of a type of individual that forms the basis for the contours of society that we share in common.
In addition, Mill argues that discretionary control of body, privacy and landed property together do not exhaust a plausible conception of personal autonomy. The kernel of the idea of autonomy is the right to make choices and decisions-what to put into my body, where and how to move my body through public space, how to use my chattels and physical property, what personal information to disclose to others, what information to conceal and more (Feinberg 29 It concerns a case of a presumed Turkish terrorist who is killed by Turkish soldiers in 2007. After his death, his corpse was mutilated by the soldiers by cutting off his ear. He was returned in this state to his family. The ECHR, however, has not accepted the case as yet (Akpinar and Altun v Turkey 2007). 30 Reference is made to the German Constitution and the Mephisto Case. 31 J.S. Mill, On Liberty (Bobbs-Merrill Co. 1956). This work was originally published in 1859. 1983). Feinberg tries to explain this concept of sovereignty or autonomy in terms of personal domains that are analogous to the territory of a politically sovereign state (Feinberg 1983). This allows him to transpose this analogy from the spatial dimension or boundary of a state to that of the individual's body's 'breathing space' around one's body, comparable to offshore fishing rights in the national model. This breathing space concept extends the territorial sovereignty concept to a sovereign self-rule sphere and submits it as a tool for moral judgments. It can be said that Mill takes the right of self-determination to the extreme by arguing that it is entirely underivative, as morally basic as the good of self-fulfillment itself. 32 In order to demarcate the personal domain in moral terms, it is now useful to refer to Mill's principle of selfregarding and other-regarding decisions. Self-regarding decisions are decisions that primarily and directly only affect the interests of the decision-maker. Outside the personal domain are those decisions that are other-regarding decisions that directly and in first instance affect the interests of other persons. In moral terms, these self-regarding interests deal with vital aspects such as decisions about procreation. More importantly, perhaps, the individual should be in a position to enforce his right to self-determination so as to achieve his good of self-fulfillment. The subsequent question is, whether this conception of personal sovereignty not only marks 'spatial' and topical boundaries of the personal domain but also provides for shifting 'temporal' boundaries as well.
To illustrate the possibility of personal sovereignty not being limited by temporal boundaries, the following example is helpful. This is the case of the reformed murderer on Death Row (Feinberg 1983). Feinberg mentions the case of a convicted murderer who, after 7 years, had acquired an education, achieved genuine repentance, and even reconstructed his personality and character. He might well be described as "not the same person" as the murderer. Still, even he requires some sense of continuity with the past of and self-identity of an earlier wrongdoer.
This example is to be interpreted in the light of the just discussed Millian concept of self-regarding decisions, which an individual ought to be able to take as an autonomous person. However, these decisions are inextricably mixed with other-regarding decisions. They put an obligation on other persons to carry out the wishes made freely at an earlier time even if the promisor retracts them later. Indeed, it can be argued that it is within the realm of the principle of personal autonomy that moral obligations or 1 3 promises made in the ante-mortem phase should remain in force after the death of the promisor. Notwithstanding the complexity of weighing the prevalence of self-regarding versus other-regarding decisions as well as different types of moral reasons such as obligations or fidelity, the principle of personal autonomy does provide a clear guiding principle. It would be evading responsibility for what one is doing to permit one to say that the later self is not the same self as the earlier self who made the original promise. Likewise, an autonomous person cannot evade his responsibility either by arguing that he should not be governed by the dead hand of an earlier self. This is an illustration of the returning 'problem of the subject'.

Informational self-determination
At this point, I tender the suggestion to consider the two concepts of dignity and autonomy, as described, as together constituting the foundation of the concept of privacy (Buitelaar 2012). The legal concept of informational selfdetermination was developed by the German courts in the 1990s. The German Constitutional Court referring to the German Grundgesetz (Art 1, 2) established a direct link between the data protection regime and two basic values enshrined in the Constitution (Rouvroy and Poullet 2009). These are the constitutional rights to the right to respect and protection of one's dignity and the right to self-development. The concept of privacy, made up of these fundamental elements, can thus be positioned meaningfully in the ethical and philosophical discourse. The two constitutive concepts of dignity and autonomy, underlying privacy, permit modern man to pursue ideals of life and character, both before and after his demise, which would be difficult to achieve were privacy not safeguarded. Autonomy and dignity permit a person to see himself as a person with infinite, indeterminate possibilities. In this line of thinking, I follow Benn's argument that, if the individual has the confidence that he can be himself, he can believe in himself as a person (Benn 1984). A person in this sense is a subject with a consciousness of himself as agent, as a chooser and as one attempting to steer his own course through the world. The concept of informational self-determination can in that light be understood as the capacity to determine without coercion which information about him is and will be available and accessible.

Life-transcending integrity and informational self-determination
The question may be posed how this combination of autonomy and dignity into the concept of informational selfdetermination may help understand how the deceased may be positioned in such a way that he can control the information that is contained in the persisting digital double. Dworkin (1993) distinguishes two types of wrongs, the transgression of either detached or derivative values. He argues that derivative values, like rights including claims, are distinctive in arising from our interests. In the case of harming a person by raping him, from a dignitary perspective, it may be argued that a person has an interest in bodily integrity and this forms the basis for his derivative right not to be raped. People, in addition however, also embody the detached value of 'sexual dignity'. Different people express and respect this value in different ways. Persons interpret this detached value as applicable in broad or more limited confines. According to Winter, although everyone has a right not to be raped, based on the detached value of 'sexual dignity', no one has a claim to obligate another with regards to their sexual expression. Nevertheless, starting from the dignitary approach of respect, based on the derivative as well as the detached values, every person may derive moral reasons but also derivative rights to have his integrity respected even if he is no longer able to arbitrate his own agency. This is, for example, the case with vegetative persons but also deceased persons. According to Dworkin, these claims are based on an obligatory respect for the interests expressed by a person's life before his demise by death or otherwise. This is what Dworkin terms life-transcending integrity.
Feinberg asserted that people's lives are more successful if the interests they formed while alive, flourish when they are dead (Feinberg 1987). As noted, Dworkin embeds his attribution of posthumous interests in his theory of the integrity of life. Dead people hold claims that are grounded in an obligatory respect for the interests expressed during a person's life before it ends (Dworkin 1993). Many people find their most important interests in "life-time transcending interests" and they put much value on becoming known through these projects. Once undertaken, these projects can become part of these persons and, in fact, they provide them with a unique identity.
Before the ascent of social network profiles, the perceptions of the experiences an individual has of his self-identity are bound into a unity by the narrative approach. These perceptions can be seen as bits or streams of information that an individual agent endowed with informational skills may use to attribute a unified self-identity to himself, at least in his self-conception. It is, of course, a serious challenge to explain how all the bits of information that compose a self are held together if there is no narrator to keep these disjointed bits of information together. Only presupposing a narrator would solve the problem, but this cannot be because the narrative theory of the self describes the narrator as the narrative (Floridi 2011). Kant, on the other 1 3 hand, has argued that the unified coherence of the informational bits could only be guaranteed by the unity of the very agent's self. These normative, informational activities thus permit a self-governing individual to constitute himself as a person and articulate his lifestories continuously and implicitly in ongoing autobiographical narratives, thereby ensuring the posthumous and proper persistence of their life's work (Buitelaar 2014). 33 An example of this is the Companions project (Wilks 2015). In this project a conversational agent was developed designed to interact with a person for a long time, learning their tastes and habits. In the process, the agent builds up a narrative picture of a person's life for posterity. The program not only manages the autobiographical material of the creator to leave some form of self-presentation of that person's life but also permits communication with descendants by assuming the voice and screen image of the deceased owner. Descendants can interact with the program and receive information about the deceased person in the way he intended to be remembered.
For a self-governing individual as composed of a unified coherence of informational bits, to assert legal and moral claims that are ascribed to a legal person, there has to be a subject whose standpoint enables a reflexive value judgment regarding the putative object of 'an interest' (Winter 2010). The corpse is by definition unable to exercise such activity but the question adopts a different hue if this precondition is imposed upon the subsequent existence of the public persona and some moldering bones. Winter posits that for there to be a case for generating a moral claim, there has to be an entity for which this claim is of significance. But could this also be the remaining public persona of a post-mortem entity to whom promises had been made in that entity's ante-mortem phase?
As noted, people do feel a strong obligation with regard to those now dead. It may be argued that attributing claims to a posthumous public persona is formally identical to attributing moral claims to a vital person because posthumous personae are similar in their discursive and textual ontology to living claimants. After all, moral discourse to a large extent deals with public personae. Our existence as a vital person consists of, for example, our abilities to learn, believe and to feel but, to be sure, also the way we are known and present ourselves in public. These facets comprise a dynamic, interdependent and organic unity that is called a person while alive. Our public persona during our lifetime as well as after our death subsists in speech, memory and information held in public media comparable to autobiographies. Indeed, after our demise it is exactly this discursive and textual ontology we presented during our lifetime, that persists post-mortem. Some may argue that the public persona in its discursive character that survives death is similar to that of a living claimant. From this ontological interpretation of a living claimant as constituted by his information, it ensues that a right to informational privacy can be understood as a right to personal immunity from unknown, undesired or unintentional changes in one's identity. Why should this right not be declared of equal validity as to the digital double of the living claimant in his post-mortem phase? I suggest that life-transcending integrity is seriously harmed when the integrity of the informational identity of the post-mortem persona is transgressed. Floridi even calls violations of informational privacy a digital kidnapping, irrespective of whether it happens in a public or private context (Floridi 2005); for that matter, I would add that the post-mortem and the ante-mortem phase can also be equated when such a violation occurs.
Recalling Dworkin's distinction between detached and derivative value and keeping in mind that morality is not exhausted by the simple connection between claims and obligations (i.e. moral pluralism), then an action can be judged wrongful despite the absence of an active claimant. From a pluralistic morality standpoint, it can be upheld that there is no basis for attributing rights to the dead; however, the dead can still be wronged. The normativity of deadregarding actions should perhaps be found in a broader conception of morality and not merely in the failure to fulfil moral claims. As said, morality is not exhausted by the opposite arguments of claims and obligations. Promises and known desires can also provide moral reasons for action without those reasons being claims. Furthermore, we may have moral obligations to fulfill projects of the dead as they are justified by the warrant of vicarious interests (Winter 2010).
Recent developments have widened the scope of the conceptual background of the legal subject in an attempt to determine the legal position of new actors such as robots, avatars and software agents. In the case of robots, the dilemma is whether these non-human agents can be held responsible for their legal actions or are the physical persons behind, or chronologically before them, to be held responsible? An answer is sought in a reformulation of the criteria for legal agency, whether human or non-human. Teubner (Teubner 2007) uses the framework of social systems theory whereby he believes it can be argued that nonhuman entities act as legal personalities. Entities can be attributed personhood by their environment under the minimum requirement of double contingency or double anticipation. By this Teubner means that in both directions of social interaction, there is uncertainty, unpredictability. The 1 3 entity or person emanates a resistance to predictable behavior while at the same time anticipating how others will 'read' their behavior (Hildebrandt 2014). In this construal, it may also seem tenable that the law can attribute legal personality not only to humans but also to the source of mere flows of communications, such as exercised by (autonomous) software agents and also by digital personae that persist after the demise of their owner. It is interesting in this respect to note that novel adaptations in contract law have been carried out, whereby it is made possible for electronic agents to form a contract by the interaction of these electronic agents without the human principal being aware of it. This fits these electronic contracting partners with mental-juridical capacities like the intention to act, the intention to bind oneself legally etc. In Teubner's account, furthermore, these non-human entities must have a capacity to deal with proto-meaning by which he (probably) 34 means a capacity to 'understand' communication with the environment without actually comprehending it. Are not the examples of gravestone technology and interactive programs that persist after your lifetime, to which this article alluded earlier, examples of the appearance a digital persona may adopt as a vicarious entity that can deal with proto-meaning?
Taking this one step further, digital personae surviving in this form their ante-mortem appearance might be fitted by legal techniques with a capacity for legal action. However, in the near future it does not seem likely that this will lead to full personhood. It may be useful to introduce for now different gradations of legal personhood, depending on the level of interaction (and thus, the degree of double contingency and capacity to deal with proto-meaning) that the digital persona displays. To offer a few suggestions of possible gradations in the case of post-mortem personae: mere interests, second-order rights or partial and full-fledged rights, concepts like demi-personalité juridique or situations where the bereaved can exercise moral rights of the decuius à travers le disparu. 35 The capacity to exercise informational self-determination in order to support the privacy of the digital remains in the sense this article understands it, on the other hand, may be a step too far for now. Nevertheless, the arguments outlined above suggest that novel interpretations of the legal subject, taking the constitutive components of informational self-determination, viz. human dignity and the autonomy of the 34 In the interpretation of (Koops et al. 2010). In this definition, domesticated animals and adaptable software agents become candidates for personae. 35 Rombach suggests a continuation of the human personality after its demise with respect to its inheritance rights and in matters of defamation (Rombach 1963). individual, as a starting point, can be of value in recognizing privacy claims of the digital remains of someone in the form of his subsisting digital persona. This concept of the post-mortem vicarious existence will need to be closely bound to its ante-mortem public persona. It will need a follow-up article to further investigate how and to what extent independent digital personae, such as artificial life-forms, could be attributed a right to privacy in legal practice, including the question how such a right could actually be exercised and enforced.

Conclusion
Taking into account the ethical/moral discussion so far, the question may now be reposed, which are the legal consequences of biological death for the possibility of transposing the concept of informational self-determination, grounded in human dignity in the ante-mortem existence, to the post-mortem phase of a digital persona?
Most European privacy laws deny the possibility of the deceased holding privacy rights. This dominant view finds its origin in the above-discussed "problem-of-thesubject" (Feinberg 1987). Death, being not only the terminating boundary of one's biological life, also means the end of the living subject with his long-range goals and hopes for his own achievement and personal enjoyment. This ante-mortem subject had an interest in protecting the information about these unique personal identity-determining character traits in order to flourish and realize his intrinsic human dignity. Without there being a subject to exercise this autonomous right, current European privacy laws hold that this right extinguishes with his death. It is the position of this article that this simple view no longer holds true in the light of the networked society developments.
The possibility is proposed that non-human entities in the shape of flows of communication can also adopt the appearance of subjects that can be attributed legal personality (cf. Teubner 2007). It is argued that the same might hold true for the vicarious subsistence of active digital personae after the demise of their ante-mortem 'owner'. Perhaps, the owners of social network sites might enable the owner of the digital double during his ante-mortem phase, as it appears in his social network profile, to design a selection of his digital remains into an independent life form. However, the present legal system would most likely not be able to accommodate the right to privacy of such a life form as the continuation of the privacy rights of the ante-mortem owner. However, it cannot be ruled out that it 1 3 may start to come about somewhere in the coming decades. 36 Still, it was also argued that it is not perhaps such an odd question to pose whether the assumption of personal sovereignty for the post-mortem presence of a human being, not only in the guise of flows of communication but also as the narrative of a person's life, in which his death is only an event, may hold equal validity for shifting temporal boundaries beyond the spatial boundaries of the personal domain. From the point of view of the narrative technique, it can be argued that a self-conscious individual imposes a linguistic unity on the events that he has experienced in his autobiography. In general, the individual attributes agential capacity to the personality that he deals with in his autobiography. These normative, informational activities thus permit a self-governing individual to constitute himself as a person and articulate his life stories in autobiographical narratives. Social network profiles, which in various forms can subsist after death, can very well serve the same goal. Indeed, autobiographies are in many ways similar to the public personae as they are composed in the form of social network profiles. Assuming that the hypothesis that posthumous personae are similar in their discursive and textual ontologies to living claimants (Winter 2010) is correct, it is a reasonable proposition to argue that attributing claims to a posthumous public persona is identical to attributing moral claims to a vital person. Or, as Kant argued, the bona fama defuncti deserves protection because it is a general human right. 37 As has been noted, it might seem that the absence of will in the post-mortem persona would preclude the possibility of granting it rights because it is incapable of exercising sovereignty (Wenar 2011). However, ordinarily, we would not doubt that the intrinsic value of human dignity, underpinning human rights, as well as unwaivable rights, such as the right not to be enslaved are indivisible parts of the legal personality of mankind. Indeed, it has been argued that the personality right of the deceased or at least his personhood continues to exist as long as the interest of the deceased makes this necessary (den Hartogh 2010). Today the abundance of social network profiles suggests privacy should be framed more as a common good while at the same time the 36 Brownsword expects that engaging in a speculative discussion on this topic at the turn of the next Millennium or the Millennium after that, will have a little more "practical purchase" (Brownsword 2008). Cf, however, the earlier mentioned social network ETER9 and the Companions project (Wilks 2015). The technical basis for this project was, as a matter of fact, 'machine learning'. 37 Feinberg posits convincingly that the interest one has in a good reputation is the best example of the self-regarding category also when one is no longer in a position to be injured, wronged or harmed by it (Feinberg 1987). normative underpinnings of the value of privacy are shared among members of society (Regan 2015). This conception of privacy mutatis mutandis might be said to apply to the post-mortem privacy of socially constructed selves in a networked society.
These strong human rights lay a claim on others to respect your sovereignty merely because you have these rights as human being and they are good for you. In sum, if the premise holds, that (1) the discursive and textual human ontology of a living claimant is characterized by being composed of strands of information and (2) that this tenet would warrant bestowing human beings with human rights [in contrast to non-human beings (Miller 2015)], then, if (3) the post-mortem's existence ontology is likewise characterized by strands of information, the conclusion may be drawn that the ontology of the post-mortem existence also warrants granting this existence human rights such as human rights grounded in human dignity. Thus this ontological approach helps in subsequently providing a basis for recognizing a right to privacy for post-mortem personae. This, in turn, ties in nicely with the observation that respect for the dead is generally considered to be an ethical obligation.
Furthermore, if in the future a positive effort by social network providers, 38 for example, is made to construct applications that may allow the ante-mortem person to maintain, shape and design his identity as informational agent (Floridi 2005), this can be construed as putting into practice the Kantian view of the cultivation of moral agency, of moral will yielding authentic freedom and responsibility. Thus, it can be argued that the post-mortem person could acquire a right to privacy in a sense and manner, comparable to the way by which the concept of informational self-determination endows the ante-mortem personality with a right to privacy. This article has given an indication of a few legal solutions that are directed towards overcoming the seemingly insurmountable obstacle to endowing deceased persons with a right to privacy due to the absence of legal subjectivity. While it is often assumed that the lack of this quality makes it impossible to ascribe the legal possibility of exercising control over its informational presence to the postmortem person's digital subsistence, the argument developed in this paper presents at least plausible arguments that evolving legal doctrine in conjunction with novel perspectives on personality may also accommodate the concept of informational self-determination as a basis for providing a continued warranty for human dignity, for the digital remains of a person after the demise of his bodily 1 3 presence, which increasingly may take the form of an interactive public persona. Thus, post-mortem privacy, albeit at first limited to the nature of for example a demi-personalité juridique, is a plausible perspective.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.