Lilian Edwards and Edina Harbinja (2013)
This is the pre-print of a chapter forthcoming in “Digital Legacy and Interaction: Post-Mortem Issues” edited by Vinicius Carpe and Cristiano Maciel (Springer, forthcoming November 2013).
This chapter aims to explore some of the major legal issues pertaining to transmission of digital assets on death. “Digital assets” within this chapter are defined widely and not exclusively to include a huge range of intangible information goods associated with the online or digital world: including social network profiles e.g. on Facebook, Twitter, Google + or Linked In; emails, tweets, databases etc; in-game virtual assets (e.g., as bought, found or built in worlds such as Second Life, World of Warcraft, Lineage, etc); digitised text, image, music or sound, such as video, film and e-book files; passwords to various accounts associated with provisions of digital goods and services, either as buyer, user or trader (e.g. to eBay, Amazon, Facebook, YouTube etc); domain names; 2D or 3D personality-related images or icons such as user icons on LiveJournal or avatars in Second Life; and not excluding the myriad types of digital assets emergent as commodities capable of being assigned worth (e.g. “zero day exploits” or bugs in software which antagonists can exploit).
In emerging legal discourse, disputes over digital assets on death can be usefully divided as related to either their pure economic value, or what might be called their dignitary, personal or non-economic value. Domain names, for example are an obvious example of an economic asset which may be crucial to the branding and thus the profitability of a business. In a family business, not only who inherits the domain name itself, but also who gets the email notifying of the upcoming need to re-register, may be controversial issues. Similarly many outlet businesses nowadays operate exclusively from eBay and, again, who inherits that account (i.e. the password and login), the money attached to the account, and any connected ongoing auctions will be a serious matter. Virtual assets in game worlds often represent the fruit of thousands of hours of labour (if game playing can be called labour) and there is already a substantial ethical and legal literature around their value, sale and conditions for transferability. Photos, blogs and text (e.g. spontaneous poems) within social network profiles by celebrities – or those who die and later become famous – are likely soon to be of economic value, just as author’s letters and unpublished novels are today.
But such assets may also be of what one might call sentimental value. Millions of photos exist on Flickr, Picasa, etc which are of very little value to anyone but the accountholder’s immediate friends and family – but to them they may be priceless. Similarly, access to the emails of a deceased family member may be of desperate importance to the bereaved, and this issue more than any has sparked public and legislative attention to the issue of digital assets, intermediary platforms and death (see below, sec. B, p. 6). Another crucial modern phenomenon is the trend towards “memorialisation” of social network profiles on Facebook and the like; effectively turning them into shrines to the memory of the deceased, where friends leave last messages and other friends gather to read them. In such cases, although money is not (usually) the issue, emotions can run very high and conflicts develop (do the friends or the parents decide if the profile is memorialised? What if the profile tells people something about the deceased the parents would rather suppress e.g. homosexuality, atheism, suicide?) and the first litigation relating to access to, and possession of, social network profiles is also beginning to percolate through.
Finally it is worth noting that society in general as well as specific heirs, family and friends has an interest in the legacy of the dead; e.g. authors’ letters (or emails, or blogs) have value to historians, scholars and critics as well as a market value to collectors. The range of stakeholders involved in digital assets cases is thus disparate and by no means limited to those in a contractual relationship with service providers.
In all types of cases, the legal issues around access, control, ownership and transmission are complex, and worse still, such answers as there are may vary considerably from legal system to legal system. It is important to note the law does not start here from a blank canvas. To some extent, existing laws will already partially regulate the issues. Almost all legal systems have rules relating to the transmission of property on death, under the name of wills and testaments law, succession, probate etc. Succession law is usually divided into what the law says when a will has been made by the deceased (testacy) and what happens when no will exists (intestacy). The law regulating the procedure to wind up the deceased’s estate – the ingathering and distribution of assets to heirs or legatees – is a separate branch of law again, usually called executry or administration. Unhelpfully, succession laws tend to be very localised – e.g. the substantive laws relating to transmission of assets on death vary wildly even between, say, the various United States, England and Wales, and Scotland, which are at least all three common law jurisdictions (though Scotland has considerable civilian influence). Civilian legal systems (eg as found in the nations of Continental Europe, Latin America, Louisiana, Quebec and Japan) differ even further. These differences matter: in one system, the spouse of the deceased may inherit or have the right to occupy the family home no matter what a will says – in others, the will may reign triumphant. In some systems, step-children may inherit nothing on intestacy, while in others they may be equal to full-blood children. Some attempts have been made to harmonise a few aspects of succession law globally – e.g. as discussed below, the laws on recognition of grants of administration obtained abroad in relation to assets in the local jurisdiction – but very little of the basic substantive law has been globally harmonised. As we shall see later, this means difficult jurisdictional problems are likely to arise in the nature of digital assets , for example, if an English person dies in England but with a profile on Facebook, a company whose HQ is in California but whose servers may be distributed globally. We touch briefly on this issue at section E.
The key question is how far digital assets fit well into these existing legal paradigms, complex as they are, and whether new law is needed. Two preliminary issues complicate the matter. First, in general, assets only fall into the estate of a deceased if they are “property”. Property law, like succession law, is a well-established and ancient branch of private law. It has also been annotated by more modern legal conceptions including intellectual property (IP) law, data protection (DP) law, privacy law and information law. Property is most often recognisable as to what transmits on death. It has restrictions. In most systems, not all things we may want, or think we own, are property, although where the line is drawn is one of the most difficult issues in private law. Some items cannot be owned as they belong to all of us: the air and the high seas are usually in this category. More relevantly to digital assets, some items are too evanescent to be property. The hope of a future gain – e.g. an option to buy future stock, held by an employee when they die – may or may not convert into an item in their probate inventory, depending on the legal system and the exact details. The same may apply to a future right to sue on behalf of the deceased if he or she was libelled or wronged or dismissed from employment without cause during their lifetime. In many legal systems, such claims are viewed as personal to the deceased and title to sue in such cases dies with the wronged person and will not transmit to their heirs.
Some licences – contractual rights to use, as are commonly “sold” in relation to digital music or intellectual property (IP) in general – are explicitly given only for limited periods of time – which may mean they expire on death (or earlier) and so do not form part of the estate of the deceased. This point is very relevant to inheritance of MP3s downloaded from e.g. iTunes, or e-books downloaded to a Kindle. Finally, some assets are so novel it is simply hard to tell if they can be categorised as property, even if regularly “sold” or valued, and if so, what kind of property. The “zero day exploits” mentioned above may fall into this character, as might be on-line reputations e.g. the “karma” on Slashdot of a commenter, or a trader’s cumulative rating on eBay. We discuss some of these problems below, using emails and MP3s bought from iTunes as illustrative examples.
The second key complication is that many important digital assets are controlled, both practically and legally, by intermediaries. This is true in the offline world as well sometimes – e.g. access to a bank account is controlled by the bank when the user dies – but it is much less common than in the online world. Access to Facebook profiles, for example, is entirely controlled by Facebook. A user cannot set up a profile there without entering a contract with Facebook. This contract is usually formed when a user clicks “I accept” or similar, and is therefore deemed to have read and accepted the terms and conditions set by Facebook. (Such agreements – whether known as “end user license agreements” or EULAs (as is common in games and virtual worlds); or terms and conditions (T&C); or Acceptable Use and Privacy Policies are all basically legal contracts.) When the user dies, control of the profile is still effectively with Facebook. They can close or delete or memorialise the profile, according to their own internal rules or norms, while the heirs may not even be able to read it if they are not FB users, or even if they are if they were not “Friends” with the deceased (and how many young people have their mother, say, as a Friend on Facebook?) Legally as far as Facebook is concerned, the relationship between them and the user, even after death, is primarily regulated by contract. The contract may simply not contain any rules on what happens on death – Facebook do give some insight into their internal rules but many service provider contracts are simply blank on the matter, which may mean disputes are left to the discretion of abuse teams, or similar. Even where rules do exist though, there is a patent potential for conflict on death between the rules of contract and the rules of succession/executry. This major problem was first explored in case law in the celebrated Ellsworth and Yahoo! case, discussed in full below at section B.
A connected and less often aired problem is that contract rules are a matter of private bargaining not social policy. While inheritance laws may have evolved to try to balance the interests of, say, parents and spouse of the deceased, or spouse and best friend, or even society (e.g. ultimus haeres rules) , contract is unlikely to think about the public good or what value society places on family ties. This is particularly true in the context of social network or most online service contracts which are usually (as they say in consumer law) standard form contracts dictated to the user with significant imbalance of power and lack of opportunity to negotiate. In plain English, this mean that users rarely read the contracts they “sign” with online service providers and have no power to alter them even if they did; thus they are likely not to reflect their real wishes on death but merely what is commercially best for the service provider. There is often a lack of transparency as to what the service provider’s rules are, and a lack of consistency as to what the rules are from one service provider to another. This is why it is quite probable that what terms if any social network providers state about transmission of user accounts may well clash with the general law of succession; and that users are likely to have no idea what happens to their profiles etc when they die.
Given the complexities described above, this chapter will mainly draw comparatively on the laws of the US and England and Wales, with a few references where relevant to EU law. In general as a matter of private law, succession and property laws in the EU are a matter for national law. Some harmonised EU law, e.g. data protection law, is however relevant. Similarly in the US, succession law is largely a state not federal matter. As already noted, the lack of international (or often even national) harmonisation in this area is a particularly acute problem when talking about “globalised” or delocalised assets such as tweets or Facebook profiles. Another key problem is the simple novelty of the area, meaning there is a lack of legal precedent as well as good practice among solicitors/lawyers. Part of the plan for future Edwards/Harbinja research is to provide a one-stop shop guide for users and lawyers seeking guidance on devolution of particular digital assets on death.
 Edwards L Law and the Internet (3rd edn, Hart, 2009), ch 21, pp 687-690.
 Playing games as piecework to create commercially transferable assets (or levels of play) is often called gold-farming, and usually done by developing world players on low wages to sell to time-poor developed world players. See further Dibbell J “The Life of a Chinese Goldfarmer”, New York Magazine, June 17, 2007 available at http://www.nytimes.com/2007/06/17/magazine/17lootfarmers-t.html?_r=0.
 See e.g. Castronova E “Virtual Worlds: A First-Hand Account of Market and Society on the Cyberian Frontier” CESifo Working Paper Series No. 618 (2001) , available at SSRN: http://ssrn.com/abstract=294828 , Fairfield J “Virtual Property” B.U.L.Rev. (2005) 85, Blazer C “The Five Indicia of Virtual Property” Pierce Law Review, Vol. 5 (2006), Lastowka G and Hunter D “The Laws of the Virtual Worlds” Cal. L. Rev. 92 (2004) 1, Heeks R “Understanding “Gold Farming” and Real-Money Trading as the Intersection of Real and Virtual Economies” Virtual Economies, Virtual Goods and Service Delivery in Virtual Worlds, February 2010, Volume 2, Number 4, Westbrook T.J. “Owned: Finding a Place for Virtual World Property Rights” 2006 MICH. ST. L. REV. 779, Vacca R “Viewing Virtual Property Ownership Through the Lens of Innovation” 76 TENN. L. REV (2008) 33
 See e.g. “The inside story of Nabokov’s last work”, Guardian 17 November 2009 concerning the publication of a posthumous novel against Nabokov’s wishes at http://www.guardian.co.uk/books/2009/nov/17/inside-story-nabokov-last-work ; see MSN News, 13 December 2012 at http://news.msn.com/pop-culture/charlotte-bronte-letters-sell-for-dollar296000-2 reporting the sale of 6 letters by Jane Austen for nearly $300,000 and the sale of a previous unpublished novel by her for $1.1m in 2011.
 “Memorialisation” usually involves freezing all posts at time of death, preventing adding of any new Friends and rejecting further login attempts, but allowing existing Friends of the deceased to add comments. See e.g. Kasket E “Continuing bonds in the age of social networking” Bereavement Care, 31(2) (2012) 62-69 or Stokes P ‘Ghosts in the Machine: Do the Dead Live on in Facebook?’ Philos. Technol. (2012) 25:363–379. As of March 2012 it was estimated there are at least 30 million profiles on Facebook relating to dead people: Readwrite, March 6 2012 at http://readwrite.com/2012/03/06/i_wanna_live_forever_or_how_we_die_on_social_netwo .
 See e.g. Facebook discovery case, below n 77; Janna Moore Morin case discussed at http://www.deathanddigitallegacy.com/2012/02/20/nebraska-is-latest-state-to-address-digital-legacy/ , February 20 2012, (conflict between family and friends over whether FB page should be deleted or memorialised, see also BBC video interview with family, 31 January 2012, at http://www.bbc.co.uk/news/magazine-16801154 ); see generally Mazzone J “Facebook’s Afterlife” (2012) 90 N Carolina Law Rev 143.
 See further Desai D. “Property, Persona, and Preservation” (2008) 81 Temple Law Review 67.
 See n 38 and sec. C below.