The anatomy of online grief

If you're taking a plane to Ireland from an English airport, you'll find that you start your descent at right about the moment you reach cruising altitude. Arriving at your destination, you will hardly experience a revelation to the senses – architecturally, topographically and meteorologically, Ireland feels a lot like many parts of England. Less visible to the naked eye are the cultural differences, the often-radical step-changes in outlook that are so easy to underestimate when there's a common language. Less than two hours' travelling time can catapult you into an entirely different culture of grief.

The first time I crossed the Irish Sea, I was going to speak about mourning on social media for some organisations in Dublin. It was 2014 and online mourning was now a reasonably well-known phenomenon. I was a bit taken aback, therefore, when it garnered a considerable amount of media attention, and I voiced my surprise to one of my hosts as she was driving me to do a talk at the Psychological Society of Ireland. I think this feels quite new here, she said. Maybe we're just a bit behind? Hmmm, I thought. I wasn't sure why that would be. One of my PowerPoint slides that weekend featured a graph plotting Facebook usage in various countries, and there certainly wasn't that much of a difference between Ireland and England.

But then again, wherever I go, especially when I'm speaking with digital-immigrant audiences, I still encounter some level of concern about going online to grieve. Radio and TV presenters express their reservations in exaggerated, emotive ways, turning up the provocation to keep the audience tuned in: isn't this all a bit creepy? I mean, it's pretty morbid, talking to dead people online, isn't it? Everyone asks questions that pull for binary, black-and-white answers: is it good or bad to mourn online? Is it healthy or unhealthy? Should we be worried or not?

When I arrived at the psychological society, I fully expected to hear some of the same kinds of questions. Psychologists and other practitioners well versed in grief and mourning tend to simply frame their scepticism in a more sophisticated way, such as whether interacting with digital remains in mourning carries a risk of 'complicated grief'. Sure enough, this group of psychologists raised the usual issues, but there was something additional repeatedly rearing its head. It struck me because I hadn't heard it expressed to the same extent anywhere else. I was in the midst of pointing out a particular benefit of grieving online, which is that the bereaved can have access to a community of mourners twenty-four hours a day, seven days a week. A young woman – surely a digital native herself – raised her hand suddenly and with the air of having a very definite point to make. 'I don't see how that can be helpful,' she said. 'That can't be good. They should be with their families.' Others nodded, chiming in with similar comments. Yes, surely people in grief should not be on their devices, in their rooms, interacting with digital remains and talking to dead friends and fellow mourners online. It was inappropriate and possibly downright wrong. They should shut their laptops, put down their phones and go and grieve with their families.

I thought that I understood what I was hearing here. I was familiar with the attitude, particularly amongst psychologists and other kinds of mental-health practitioners who privilege face-to-face interaction. I'd even done a study of English psychologists' attitudes to digital technologies, and one of the major findings was that, whatever their age, they saw online interaction as an inferior type of relating, somehow not 'real', a substitute for authentic contact. In possession of these recently gathered data, I was confident in my response and encouraged people to reflect upon the underlying, knee-jerk beliefs they had about the 'right' thing to do in the face of grief.

'You're making several assumptions here,' I said. 'You're assuming that family members will be able to speak about the death, when they may be struggling with grief themselves. You're also assuming that they'll be willing to speak about it – people often believe that children and young people should be protected from death. The online environment might be the only place where someone feels comfortable expressing their grief or talking about the dead person; there may be spoken or unspoken rules against that at home.'

In other words, while I was exhorting them to reflect upon their assumptions, I was blissfully unaware of how many I was expressing myself. Climbing back into my host's car after the talk and grappling with a sense of déjà vu, I told her that I might have misjudged something about my audience. 'Can you tell me a little bit about death and grieving in Ireland?' I asked her. It was a topic that I hadn't researched before my arrival, because I didn't realise that it would be particularly at odds with what happens in England or the USA. I don't remember her precise words, but what she described was indeed utterly different from anything I knew. I had been describing online mourning as an antidote to the silent, isolated suffering that can come with death, but my Irish audience had reason to be flummoxed. They were trying to understand why a cure was needed for an ill they didn't recognise.

In 2017, Kevin Toolis published a book called My Father's Wake: How the Irish Teach Us to Live, Love and Die. In an article for a London newspaper drawn from that book, he writes, 'In the Anglo-Saxon world, death is a whisper. Instinctively we feel we should dim the lights, lower our voices and draw the screens. We want to give the dead, dying and the grieving room. We say we do so because we don't want to intrude.' In contrast, Toolis's own father, Sonny, had a traditional Irish wake. The body was cared for and kept at home, with the children playing in the sitting room at the foot of the coffin. People came from far and wide to socialise and gossip, eat and drink, and commiserate with the family in the dead man's own front room alongside his body. Wakes like this used to be a more widespread phenomenon, but Toolis identifies the forces of urbanisation, medicalisation and industrialisation as reasons for their steady decline in much of the West. But for whatever reasons, on Irish soil, the Celtic traditions persist; an 'ancient form of death sharing', as Toolis describes it.

My Ireland visit reminded me how hard it is to wiggle free of the automatic assumptions about grief, which may be highly local, particular to our culture, our religion, our community, or our own family. We assume that we know what constitutes 'healthy' grieving. We have opinions about what an appropriate legacy looks like, what a proper send-off is, how it's acceptable to respond to a bereaved person, or how we should behave in the face of death. We have our own experiences of what has and hasn't been helpful for us. So, with our personal looms already strung with the monochrome warp of our beliefs about death, we're now holding in our hands the multicoloured, jumbled weft of a much newer material: digital technologies. Our assumptions about the latter are highly variable, connected to too many contextual factors to count. Life online is either destroying our communities or fostering our connection with them. Constant connectivity either promotes isolation or prevents it. Social media either inspire us to be prosocial or turn us into rampant narcissists. And digital remains are either good for us in grief or not. What is the anatomy of grief in digital culture, within digital communities, and what rules are we meant to follow?

In February 2013, there was a story in the Australian press about the digital legacy of teenager Allem Halkic. After being cyberbullied via SMS messaging and on MySpace, Allem threw himself off a bridge in 2009. The ensuing trial was the first cyberbullying case to come before the Australian courts. After he died, his father Ali used to go into his son's closet, inhaling the scent lingering on his clothes, but after a while the smell faded – it became 'dust', as Ali put it. But his parents then discovered something that wouldn't fade: Allem was still logged onto his Facebook account. Ironically, this digital artefact became one of the most precious things to his parents, helping them connect with a son driven beyond the brink by the cyberbullying he'd endured. 'There is no way in a million years we would disconnect his Facebook profile, just no way,' Ali said in the news story. 'In the first year or so after he died, he was getting a message a day from friends and then it became slower. I got disappointed when people didn't write because I didn't want him to be forgotten.'

The author of the article remarked that no one could dispute the value of persistent social media accounts to bereaved family and friends, but he was wrong about this. Just a couple of months after this news story, the BBC reported that a grieving mother in Brazil had taken Facebook to court over the memorial page for her daughter. The precise things that had brought Ali and Dina Halkic joy were too difficult for Juliana Campos's mother to endure. 'This "wailing wall" just makes me suffer too much,' Dolores Pereira Coutinho said. 'On Christmas Eve many of her 200 friends posted pictures they had taken with her and recalled their memories. She was very charismatic, very popular. I cried for days.' Facebook tried the middle way first, making the tribute page friends-only, but that wasn't enough for Mrs Coutinho. The thing most feared by Allem's parents and most craved by Juliana's mother was the same event – the permanent disappearance of their child's Facebook profile.

Unsurprisingly, research indicates that people who regularly use social media find it to be more useful in grieving, while those who don't are more likely to harbour unease and doubts about whether digital remains should continue to exist: a 'coping paradox'. The same digital artefact, such as a posthumously persistent Facebook profile, could mean inestimable comfort for an always-on, or wracking emotional pain for a digital pragmatist. How you experience digital remains, and whether it will be helpful for you to access them in your grief, will be determined by your relationship with digital technologies, woven together with your experience of a particular bereavement. If the last chapter got you musing about the former, the next bit of this chapter is designed to help you think about the latter. Before we go there, though, here's a caveat that will, by now, come as no surprise to you: in matters of grief, little is predictable, and nothing is definitive. Ask whether interacting with digital remains is a 'good' or 'bad' thing, and don't expect a yes-or-no answer.

***  

Many of us worry that we're not feeling or behaving correctly in the face of a loss, or perhaps we get concerned about other people, for the same reason. Open any search engine and enter the word 'grief'. If you're using Google, notice what it says in the 'People also ask...' box, for this is a window into common preconceptions about mourning. In the popular searches, you can discern the things about grief that people take for granted, but about which they can't recall the precise details. On the day that I search, first on this list is, 'What are the seven stages of grief after a death?', closely followed by 'What are the stages of grief?' and 'What are the five stages of grief and loss?' Looking at the search results themselves, you won't have to scroll very far down, if at all, to encounter multiple references to stage models of mourning.

The idea that the grief process consists of a predictable sequence of stages, eventually ending in acceptance, has brought comfort and reassurance to untold numbers of bereaved persons since 1969, when Swiss psychiatrist Elisabeth Kübler-Ross proposed it in her seminal book On Death and Dying. Kübler-Ross's five stages – denial, anger, bargaining, depression, and acceptance – weren't actually even based on bereaved people. Instead, they referred to the stages she had observed dying people themselves go through. Nevertheless, they were rapidly applied to the grief of those left behind. The reading public seized upon her stages with a kind of hunger, as though they had been waiting for half a century for someone to explain how to do the 'grief work' that Freud had talked about in 'Mourning and Melancholia' in 1917.

If people have been confronting loss and grief since the dawn of time, why the desperate uncertainty, all this searching for a proper course of action? Perhaps getting grief right has seemed particularly important since the mid twentieth century, when psychiatrist Erich Lindemann took the vague explanations of grief work in 'Mourning and Melancholia' and translated them into a modern psychiatric diagnosis. Thanks to Lindemann and the burgeoning medicalisation of human experience that has only grown steadily since his own efforts, failing to cleanly sever ties to your dead within a seemly period of time could land you in the doctor's office. You might even leave carrying a prescription for antidepressants and be labelled as suffering a 'major depressive disorder'. Once upon a time, if someone had recently suffered a loss, this would have kept them from receiving a depressive-disorder diagnosis. However, in the fifth incarnation of the Diagnostic and Statistical Manual of Mental Disorders, published in 2013, major depression no longer features a 'bereavement exclusion'.

Quite apart from the concern about going crazy, or about being perceived as going crazy, you can understand why we might cling to the idea of an orderly succession of grief stages, unfolding one after another until you reach the point when you let go, or when grief lets you go, once you've served your term, put in the work, done your time. Human beings struggle mightily with those things we cannot predict or control – it's not coincidental, then, that unpredictability and uncontrollability are two of the main ingredients in the mix when we experience stress and trauma. Imagine a bereaved person in the throes of unimaginable emotional pain, on their smartphones at 2 a.m., wondering how long they will have to endure this, trying to find out what they can expect to happen next, straining to see if there is any light at the end of the tunnel. Maybe that person has been you, at some point. You wouldn't even have to tap out 'grief', because autocomplete provides that as suggestion number one, as soon as you enter 'stages of'.

The fantasy of predictability, and the hope and sense of control that it brings, helps explain why Kübler-Ross's stages continue to dominate the discourse, on the Internet and off, even though they've long been regarded as a comforting fiction, 'practically folklore'. Part of what's implausible about the stages is the idea that they occur chronologically; later in her career, Kübler-Ross tried to loosen people's expectations around that by claiming that she'd never intended the model to be interpreted as linear. But the popular imagination is drawn to simplicity, and perhaps the thought of sliding back into that pain again, having thought you had reached the end, is too disheartening for many to contemplate. In our darkest hour, we hope that what Freud reassured us about is true: grief is supposed to come to an end, and it will, if you're doing it right.

Despite the strength of these canonical narratives of how we 'do' grief in the West, more and more people came to realise that perhaps we really had lost our way with death. By the mid-1990s, there was a groundswell of voices arguing for that idea, collected together in an anthology about a concept called 'continuing bonds'. Continuing bonds theory was the most not-new new thing you can imagine, given that continuing bonds with the dead has existed throughout history and even pre-history, across most cultures. Only with Freud had the West started to properly veer off on the slip road. Continuing bonds was a shot across the bow to the 'curious notion' of grief work – a construct that argued persuasively that continuing relationships with the dead wasn't worrying, it was just part of being human. Westerners who thought it was critical to move on and obligatory to sever ties were being told something that people in many other cultures have maintained throughout: keeping close to your ancestors is normal, adaptive and usually positive.

Continuing bonds are a bit of a free-for-all, in contrast to the buttoned-up positivism of medical and stage models of grief, and they happily morph to fit the context of each culture, subculture, family and individual. We may experience continuing bonds purely internally, in the form of memories or feelings of connection to the dead that aren't witnessed by anybody else. They may play out in our behaviour, whether our actions occur privately or as part of highly public rituals. In China, continuing bonds with particular ancestors may be non-negotiable, for funeral and remembrance rites are needed to ensure that your dead are content, happy ancestors instead of hungry, angry ghosts. If you're from a more individualistic culture, continuing bonds may be more organic, personalised and spontaneous, and you might only do it with those to whom you feel most closely connected. That could mean family, but not necessarily – you might retain a strong sense of connection to people to whom you have no blood ties whatsoever, whether those are friends, or people you've never met, like celebrities you admire or other role models who inspired you. You might think of a dead person as a guardian angel who looks after you, or as a soul that still touches your own.

Obviously, cultural context matters a lot here. The sociologist Tony Walter explains that some us have a 'care culture' of mourning. In care cultures, people believe that the dead still exist in some kind of spirit realm, and that they continue to need us to look after them in some way. He contrasts this with 'memory cultures', which hold that we must accept that the dead are gone, but in which we're encouraged to remember and honour their legacy, and to think about them as living on through their descendants. Memory cultures don't generally see the dead as separate entities with their own ongoing agency. East Asia and the global South are care cultures, and Western European cultures tend to be memory cultures. 'It is hard to talk about the European dead,' says Walter, 'without using the language of memory.'

Whether you come from a care culture or a memory culture, however, it all sits under the umbrella of continuing bonds. If you've ever lost someone close to you, you've probably experienced an ongoing connection with them yourself, in some form. You hear their voice, catch sight of them, feel their physical presence. You feel their guidance or support in your thoughts, your dreams, or the events of your life. You carry them with you – not just their memories, but their values, their characteristics, even aspects of their personality. You might make conscious efforts to do things as they would have done, or go places they would have visited. Emotionally and psychologically, they remain part of who you are, and they still play a part in the systems around you. 'Living people play roles, often complex, within the family and psychic system', write the editors of the new edition of Continuing Bonds. 'After they die, roles change, but the dead can still be significant members of families and communities.'

That quote explains why you can expect every single bereavement to be necessarily, inevitably unique. If you have four siblings, you don't have a generic 'sibling' relationship with each of them. If you have 200 Facebook friends, you don't possess an equivalent bond with every one of them. The relationships you share with your fellow humans may have similar themes and characteristics, given that the common denominator is you, but every relationship varies. Because relationships are unique in life and just as distinct in death, every loss experience will be different too, even if it can be classified within a particular type, or trajectory. Grief scholar George Bonanno, for example, has charted three patterns of grief reactions in bereaved people. Those with chronic grief reactions are so overwhelmed by loss that they may struggle to function in life for years; people who gradually recover are hit hard but eventually manage to reassemble and carry on with their lives; and the resilient are 'shocked, even wounded, by a loss, but . . . manage to regain [their] equilibrium and move on'. This doesn't mean, however, that each of us is innately a particular 'style' of griever, for depending on the loss, you may experience resilience, recovery or chronic grief.

Within each of these trajectories, however, there's a common characteristic – the intensity of grief judders up and down. Grief has peaks and troughs, waves of sorrow alternating with periods of contentment and even joy. Not only is it normal for these oscillations to happen, but it's typical for them to be utterly unpredictable and uneven. Some theories of grief attempt to provide a framework for this experience; the dual process model, for example, describes how we swing back and forth between loss orientation, when we're preoccupied with the dead person, and restoration orientation, when we're focused on aspects of life that don't include them. But Bonanno thinks that even this is too rigid, not giving sufficient credit to the sheer amount of oscillation that occurs. '[W]hen we look more closely at the emotional experiences of bereaved people over time,' he says, 'the level of fluctuation is nothing short of spectacular.'

The recap of all of this is pretty simple. Even though you may personally resonate with Kübler-Ross's stages, there's no right or wrong way to grieve. Whatever your process is, there is nothing abnormal about an unpredictable panoply of moods, or an oscillation between loss and restoration, between being sunk in grief and carrying on as though it never happened. Continuing bonds with your dead, in all its myriad forms, is par for the course, just as it has been for millennia. Whether you are actively caring for and interacting with your dead, or simply remembering and honouring them, it's still continuing a bond. Everyone's needs and desires are different, and that's OK. The choice of how you continue bonds, or not, should be up to you, not dictated by anyone else. That brings us to the sticky bit – the digital age has made continuing bonds easier and harder all at the same time.

 *** 

Let's start with the easy part: the digital environment is perfectly designed to facilitate continuing bonds. But is it a medium that supports a culture of remembering the dead, or a culture of caring for the dead? Well, before blogs, smartphones and social networks, hardly anyone built a substantive digital legacy in life, and so the Internet was almost exclusively equipped to support the memory type of continuing bonds. Very few ordinary folks in the 1990s had a digital reflection or a digital shadow with any vitality or substance; you would be unlikely, logging on, to find a particularly vivid representation of anyone that mattered to you. If anything happened on the Internet in the 1990s to help continue bonds with the dead, it was through online memorials that were compiled post-mortem.

When Web 2.0 came along, and gigabytes and terabytes of data started being stored on personal devices and distant servers, the Internet began to house extensive archives of millions of ordinary lives. Technologically mediated 'remembering' became incredibly easy, because our modern technologies save data effortlessly, by default, silently uploading them to the cloud for safekeeping, and cataloguing them cleverly for easy search and retrieval – by us, or by someone else who knows the location and holds the keys. We went beyond using the digital sphere to remember the dead and started using it to care for them: to feel in contact, to connect with them. In essence, because the dead live on through tech, we're now witnessing a shift in how Westerners think about them. But let's go back to the beginning, when it was still all about memory.

A young engineer called Mike Kibbee came up with one of the first platforms to commemorate a large number of dead online, in one place. It was 1995, and Mike would soon be dead himself, from Hodgkin's lymphoma. He faced his impending demise boldly and pragmatically; first designing his own coffin, and then collaborating with a friend to create a technology that combined cemetery, eulogy and obituary, and translating them all into an online form. His own memorial on the site, which still exists today, replicates the obituary that appeared in print in the Globe and Mail in Toronto. The obituary describes Kibbee's idea for the World Wide Cemetery as 'a stroke of genius', enabling a far-distant son to 'visit' the grave of a parent through his computer. It further described how the online cemetery mirrored the mourning rituals of offline life, enabling visitors to online graves to leave digital flowers, a poem, or a condolence message. 'The wonderful interconnectivity of the Web', said the obituary, 'made it easy to link deaths (and the marvellous details of lives) of family members who may have died years apart and in different countries.'

World Wide Cemetery touts itself as the oldest online cemetery and memorial site in the world. Just like offline burial sites, this 'elegant, peaceful and serene' space is a separate, dedicated place for remembering the dead, and visitors can engage in all the familiar mourning rituals: sharing stories, memories and photographs, leaving virtual flowers, and writing messages to the deceased. And, just like in the offline world, some cemeteries flourish while some become disused. When the World Wide Cemetery was launched, being the first phenomenon of its kind, it garnered a considerable amount of media coverage, and there was good uptake. Most of the memorials on the WWC are for people who have little or no other online presence.

Marc Saner, who took over the running of the cemetery when its previous proprietor tired of the task, told me that he did so because he couldn't bear the thought of his godson having no accessible digital legacy. 'After his death it took only about one year to erase him from the face of the Internet,' he said to me. 'This bothered me. My WWC venture started with my desire to give my godson – and my father – a permanent online presence.' Permanent it may or may not be, for these days, the cemetery is an exceptionally quiet place. It has set aside a hundred-year fund to attempt to safeguard its existence over the longer term, taking it to 2095, but visiting it feels a little like going to a small churchyard where no one is buried any more. In 2017, as far as I can work out, one memorial appeared. At the time of writing, with the year a quarter gone, 2018 didn't look as if it had seen any at all. When I spoke directly with Marc, he was reluctant to put a number on new memorials. 'It's small,' he admitted.

Other digital cemeteries may be more successful because they make sense within a particular context, fulfil a current need, or even meet the agenda of the state. An online memorial site in Hong Kong, for example, is run by the government with the hopes that it will encourage cremation and other sustainable burial practices. Someone can only create a memorial for their 'beloved one' if that beloved was cremated and/or interred in a Hong Kong cemetery or columbarium. Legacy.com is an online memorial of a different kind, a flashy combination of search engine and death-specific news desk, a for-profit enterprise that allows people to access obituaries from more than 1,500 newspapers and 3,500 funeral homes across the United States, Canada, Australia, New Zealand, the United Kingdom, and Europe – usually for a limited time, in accordance with their terms of agreement with various newspapers. It features tributes to celebrities penned by professional obituary writers on the Legacy staff, rather than by individuals who were close to the deceased.

In the virtual world Second Life, there are multiple memorial gardens, some dedicated to commemorating certain kinds of lives, or certain manners of death. On a rocky outcrop overlooking the sea, for example, with weeping willows and lights twinkling in the trees, is the Transgender Hate Crime and Suicide Memorial. Elsewhere you can find the Peace Valley Pet Cemetery, where there is no requirement for lost pets to have ever been a carbon-based life form: 'Walk through the graveyard and see the fun and touching ways Residents choose to memorialize their little lost loves, both real and virtual. In addition to the cemetary [sic], the build also comes complete with a chapel for delivering a fuzzy friend's last rites.' And if distance, schedule or infirmity prevents you from attending a funeral and paying your respects, you can often go to the funeral home's website to sign a virtual guest book.

All of these online facilities have much in common with their offline counterparts, while additionally doing what the Internet does best – making it possible for people scattered around the globe to share and access information and memories. Mourners should be able to engage in these rituals without let or hindrance, so it would be rare to be charged admission to a physical cemetery where the recent or contemporary dead are interred. Analogously, some dedicated memorial sites online are free to anyone with access to a browser. The World Wide Cemetery, for example, promises that they will never charge a fee or require accounts or passwords in order to visit the memorials. There are, however, many exceptions.

An erstwhile work colleague found out about the death of her distant friend's husband on Facebook. Overwhelmed with grief, his widow had simply posted a link to the obituary, without further comment. As we sat together in a coffee shop talking about this event, my colleague picked up her phone and scrolled back to her friend's original post, expecting it to be an expired link. 'Obviously now it's going to be different, because it was four years ago,' she said. She clicked on it anyway. To her surprise, it did link to the funeral home's obituary and Guest Book page, and there was his name – we'll call him Ralph Buxton. 'You can keep the memorial website for Ralph Buxton online by contributing to its extension', the page said. Alerting us to the fact that other 'commemorative products' were also available, it enumerated a schedule of the fees payable for a three-month, one-, two-, five- or ten-year stay of execution for the online memories of Ralph. The ten-year option would set you back $499, Canadian. We sat in stunned silence for a moment. 'This Guest Book has been archived and is no longer available online,' we read. 'Restoring the Guest Book is a wonderful way to allow acquaintances and loved ones to express their sympathy and share fond memories.' The site went on to urge even persons outside the family to consider restoring the Guest Book as a 'special gift' to Ralph Buxton's nearest and dearest. In this case, in a rather depressing instance of the commercialisation of death, keeping online memories 'alive' came at a price.

Online memorials of these types have a critical feature in common: the image of the person that they present is biographical rather than autobiographical, subject to the editorial control of the authors. The deceased person, a non-participant in this process, isn't around to challenge what anyone says about them. Mourners are at liberty to do what mourners always do, including smoothing out the rough edges, sanctifying or vilifying at will, and deciding what's most important to remember. Ultimately, the people left behind negotiate and establish a durable biography, a reasonably true-to-life picture of the dead person that feels comfortable enough for individuals or communities to carry forward, and the deceased themselves won't have too much to say about it. When a person has built a substantial, visible digital legacy during their own lifetime, however, it's an entirely different kettle of fish.

***

Being at peace with the age that I am, I feel no shame in disclosing that I sit firmly in the 'digital immigrant' category, no matter how well I may have adapted to changing times. When I owned my first computers in the early 1980s (initially a Texas Instruments TI-99/4A, then a Commodore 64, followed by an Amiga), the most social interaction I derived from them was having my friends round to play Parsec. The World Wide Web hadn't been invented yet when I went away to university, but we had an intranet where I was studying, a space-age technological marvel that enabled students to chat one another up in real time via text messages, but that wasn't capable of revealing what the other person looked like, resulting in innumerable eventual disappointments.

After I graduated I went backpacking around Europe, during which time I effectively lost contact with all my friends and family for two months – it was before I had my first email address with America Online, and cafes at that time only featured coffee, not computers.

When Friendster and MySpace hit the scene in 2002 and 2003 respectively, I was far too busy with graduate school to notice or care. My first experience of an online social networking platform, therefore, was Facebook. I liked it from the start, enjoying the ease with which I could post photos as well as updates, and keep track of my widespread friends and family. It was a boon to an expatriate who'd lived in a lot of cities and still missed some people who'd remained in those places.

And the first time I encountered death on a social networking site? About half an hour after I signed up, while entering the names of my high school and university friends to see if they were on this new platform too. Some of them I was actively interested in reconnecting meaningfully with, but for some I was just curious about what they looked like these days, what they were up to now. I typed in the name of one old acquaintance – we'll call her Jessica Smith – and a handful of Jessica Smiths appeared, most of whom looked considerably younger than the thirty-seven-year-old I was searching for. One entry, though, stood apart from the others. It wasn't because I recognised the young blonde woman in the photograph; it was because it said 'In Memory of Jessica Smith' next to the image. Curious, I clicked on it.

That was 2007, and it was only the year before that Facebook had leapt the confines of educational institutions to become available to anyone over the age of thirteen with an email address. Prior to its public launch, it is difficult to imagine that Mark Zuckerberg and his colleagues ever sat around a table saying, right, guys, eleven years from now we're likely to surpass two billion regular users worldwide. Millions of those users could die every decade. What are we going to do about that? How many people do we need on board to respond to any special situations that arise?

I wasn't there, but I'm willing to bet that the above discussion didn't occur. As you would expect from a site that was set up to connect living users rather than to act as a mixed-use social networking platform and massive digital cemetery, at first there was no facility to memorialise profiles on Facebook, and it would have been impossible to differentiate dead users' and live users' accounts on the basis of design features alone. You could only have been tipped off by the content on the Wall (as it was then known) and the sudden cessation of posts and other activity from the account holder.

Clearly, this was a situation that wouldn't work over the longer term, and it didn't take Facebook long to realise that they needed to act. In 2007, a year after going public, Facebook started putting the profiles of dead users into a 'Memorial State', initially employing a one-size-fits-all approach. Once someone had contacted Facebook about a user's death and that death had been confirmed, no one could log in or change anything about the profile again. All of the posts would revert to 'friends only', even the posts that had been public before. Notifications from the account, such as birthday reminders, would stop. People could still post on the Wall to memorialise and to communicate with the deceased but, other than this, profiles were frozen in time and looked the same.

Facebook didn't seem terribly sure about in-life profiles repurposed as memorial sites, though. 'In the Memorial State, certain profile sections and features are hidden from view to protect the privacy of the departed,' said a Facebook representative in 2007. 'We encourage users to utilize groups and group discussions to mourn and remember the deceased.' So the page I had stumbled across wasn't the Facebook profile that Jessica had built and used in life. Instead, as the name suggests, it was a group started by her friends after she died, in some ways similar to something that could have been created on the World Wide Cemetery. I read the messages that people wrote and studied the photographs they posted, experiencing a sensation that I recognised, a kind of detached curiosity with a dash of poignancy. It was the sort of feeling you may have experienced when passing a place where a fatality has occurred, pausing to read the faded notes pinned to teddy bears, or tied to shrivelled bouquets of flowers and attached to a lamp post or railing – a slight voyeurism, mixed with a moment of contemplation on the fragility of life.

After a few moments clicking and scrolling, I started to wonder what had happened to Jessica's own profile, if indeed she had one. Facebook was pretty new; maybe she didn't have her own presence at all. With a couple of clicks, though, first on the name of the administrator of the group, then on the administrator's own list of Facebook friends, I found her. Not only did she have a Facebook page, but it was extremely well developed, and no wonder: twenty-one years old in 2006, born in 1985 in a technologically developed country, Jessica was a child of the computer age, a digital native. She and her friends had clearly been early adopters of Facebook, and their digital footprints had ranged wide and deep across the fledgling social networking site within only a few months. Her life was an open book: however risqué the outfit, however profane the language, however raucous the party, it was all there. She clearly lived life to the full, and she shared that life with a large audience, an audience that now included me.

Jessica's in-life profile was different from the memorial page. On the latter, people talked about her, but on the former they talked to her, carrying on conversations that had started in life. One photo showed her with her friends, lined up on lounge chairs in some tropical location. It had been posted before she died, and there was her own comment underneath: 'Oh my god, we look all skinny and tan!! I want to go back there!!!' After her death, her friends simply carried on the thread of conversation, as though it was the most natural thing in the world. The comment was casual, conversational, everyday. It was almost as though she wasn't dead at all, just away for a spell, not available for the annual jaunt to Florida this year, but maybe next time. I had worked with bereavement and grief in my clinical practice and studied it in my previous research, and there were lots of things related to mourning that sparked my curiosity about her profile page, but these are what captured my attention the most: the pedestrian, frequent conversations with her community of friends, persisting six months after her death, phrased in the second person and directed at Jessica herself.

The other thing I noticed was my own inner response. Reading Jessica's own words, following her conversations with her friends, viewing her images, and seeing her in the context of her relationships was a far different experience from looking at the memorial page. Trawling through her digital legacy, the one that she herself had contributed to, I started to feel a kind of familiarity, almost as though I knew her. In fact, I can still remember her face. I can still recall specific photographs on her page, even ten years later. That's not because I still look at the profile, though, for that's denied to me now. Later that year, when I was starting to formally research the subject, I looked for her profile but couldn't see it in search results. I didn't understand why, at the time, but now I realise that it had probably been memorialised and, in line with the policy at the time, rendered findable and viewable to friends only.

Losing track of that profile didn't bother me very much. I was just a researcher interested in a phenomenon, and I could always find other 'data' to use for my study; if not Jessica, then someone else. What I realise now, though, is that what was a neutral emotional experience for me, a minor irritation, would have been a completely different experience for someone else, someone who might have been close to Jessica but not on her Facebook friends list for whatever reason. What if her mother or her grandparents had been comforted by looking at her profile, and then woken up one day to find they couldn't see it any more?

I now have a massive digital footprint of my own. Facebook contains a personal archive – thousands of photos, videos, updates and commentaries – that shows who I am. My Instagram account shows what I find beautiful, my Pinterest posts indicate how I clothe my body and decorate my home, and Apple Music and Spotify records reveal the music that speaks to me most; but only a select few see all these. To the wider public, my blog discloses my thinking and opinions, my Twitter feed projects my professional identity and reveals my political opinions, and my website touts my therapeutic and writerly services to residents of greater London. In hundreds of thousands of words and thousands of images, I have written and continue to author my autobiography, telling the world who I am, what I do, and what and who I care about. This vivid, rich, multi-sensory record is almost completely analogous to my self-presentation in everyday life. Many years ago an academic in Korea described how digital being sits somewhere between the being of mind, res cogitans, and the being of body, res extensa, sharing qualities with both but qualitatively different. Of all the elements of my digital footprint, my data on Facebook most fully capture my personality, values, humour, image, and the last decade of my history. What value might that have to those left behind, if I were to die tomorrow? If my profile disappeared after I died, my res digitalis would disappear along with it, and what might that event mean to someone? Potentially, quite a lot. Ava, a digital native who participated in one of my research studies, lost her close friend to an automobile accident. She succinctly summed up the vitality that her friend's Facebook profile held for her: '[If the profile were deleted] it would feel like I wouldn't be able to talk to her properly,' she said. 'It would be deleting the last bit of her that's still almost real.'

Talk to her properly. I was hearing this sort of thing a lot, and it seemed that something more was going on than merely memorialising the dead. In the early 2000s, a study of mourners' communications on virtual memorials such as online cemeteries showed that only about 30 per cent of people talked to the dead. Ten years later, I tabulated that of nearly 1,000 posts on the five Facebook memorial pages I studied, 77 per cent were addressing the person who'd died. It was pretty obvious when someone had more of a memory culture, when they held the view that the dead couldn't hear them. For them, writing a message was about showing sympathy and support to the family, not getting in touch with the dead. 'I know he can't read anything, what we are writing here,' said one such visitor, and it's interesting that they felt the need to emphasise this. 'But I just want to share my feelings with his friends and family.'

On the other side, there were the remaining three quarters of posts. 'Even though it seems silly to talk through Facebook, I know u can see and understand every word I type,' someone says. 'I know u can read this, it just sux that u can't talk back . . . thanx for letting me talk to u again,' says someone else. Facebook is apparently so exclusively effective at facilitating connection that an inability to log on could entirely prevent one from making contact. 'Happy late birthday! I did not have computer access yesterday . . . but I did remember your birthday and thought about you all day!' says one person. Another apologises as well, even though it wasn't about missing a birthday: 'I'm sorry that I haven't written to you for a while now, I know the castle is luxury but not so much that it has the Internet.'

Is writing messages in this way just about obeying force of habit or following conversational convention, rather than being an accurate portrayal of what we believe about what happens and where people go after death? Or can we take these messages as a true reflection of writers' beliefs? Do people think that the dead are not only sentient, but actually reading their messages from some kind of Internet café in the afterworld? And might individuals' religious beliefs make a difference as to whether they believe that they're actually able to make contact? After all, we live in increasingly secular societies. In the ten-year gap between the 2001 and 2011 census in England and Wales, the percentage of people identifying as Christian fell from 71.7 per cent to 59.3 per cent, and the percentage of people reporting that they had no religion grew from 14.8 per cent to 25.1 per cent. Since 1990, it's been reported that the percentage of Americans reporting no religious affiliation has swelled from 8 per cent to 22 per cent, nearly tripling in just under three decades. In what may be a conservative prediction, it's thought that by 2020 there will be more Americans with no religion than there are Catholics, who made up 20 per cent of the US population in 2017, and by 2035 Americans in the 'no religion' category may exceed Protestants, currently the largest denomination in the United States, with 45 per cent of residents identifying as some type of Protestant. With such a decline in religiosity, does this mean that the concept of afterlives and the idea of the sentient dead are diminishing too?

You might think so, but actually this isn't necessarily the case. Interestingly (and contrary to what I've always assumed), belief in God and belief in an afterlife don't always go hand in hand, and even atheists or agnostics may believe in some kind of life after death. Although formal religious education, beliefs and affiliation have definitely all dipped significantly in the past decades, levels of belief in God and some kind of afterlife have remained high, and people express their beliefs so readily in the online world that scholars are finding all this far easier to investigate. Sociologist Tony Walter, who has conducted many studies using online material to better understand our current views about heaven, angels, souls and the afterlife, refers to the many 'surprising mentions of heaven in otherwise secular posts'. In Sweden, according to one researcher, a spiritual take on death and afterlife is widespread and shares more in common with New Age thinking than traditional religion, and the life-after-death concept 'glorifies individuals and extols the bliss of an afterlife existence free from any punishment'.

Walter has also observed a recent boom in references to the dead becoming angels, a notion that seems to be embraced even by many thoroughly secular people. An angel that used to be a human being, alive on earth, is a new sort of angel, completely unrelated to any sort of traditional church teaching; Walter thinks that belief in this sort of heavenly entity is a way of articulating relationships with the deceased, relationships that are facilitated by the online persistence of the now-angelic dead's data. 'The most common place where the angelic dead are encountered,' he writes, 'is online.' There is a remarkable similarity between cyberspace and angels, he argues, in that people seem to see both as ways of travelling between earth and heaven. '[U]nlike souls locked up in heaven,' he writes, referring to how we used to think about the dead, 'angels can read social media posts. Thus technological development affords a new space for, if not creedal faith, then spiritual discourse.' Whether the online environment is in itself a kind of new heaven, or just the medium for people to get in touch with the angels in some other sphere, evidence seems to suggest that both secular and religious people believe in the idea of the dead receiving our messages and looking out for our interests on earth.

This was borne out by my own research participants when I was studying the phenomenon of talking to the dead on Facebook, a medium that Walter describes as 'a particularly amenable place for angels'. I asked Ava, a secular young woman, if she felt as though Facebook was somehow different from, say, writing a letter and leaving it at her friend's gravesite. She didn't hesitate. 'I feel she will see it if it's on her [Facebook] wall,' she said. 'When I can't see what I wrote to her, I feel like she won't be able to see it too.' What about sending thoughts and prayers? 'You can think thoughts in your head, and think, "Oh, I'm hoping he can hear me",' said Ruby, who lost her cousin. 'But when you write something in Facebook, it's a more tangible way to communicate.' Not just remember: communicate.

What about going into the person's room, inhaling the scent of their clothes and being amongst their things, like Allem Halkic's father had done? 'It's strange,' said Clare, 'but part of me just feels like he sees it somehow. When I'm communicating with him on Facebook, there isn't that immediate reminder that he's gone. But when I see his name on his headstone in a silent cemetery or I see his room frozen in time, it's more in-your-face.'

If angelic entities still have agency, then they still have social agency. In the majority of posthumous activity on memorialised Facebook profiles, and perhaps to a lesser extent on bespoke memorial pages, I witnessed plenty of evidence of this assumption about the dead. They're physically dead, yes. No one seemed in denial about that. But socially dead? Not just yet. Unlike the hungry ghosts of China, demanding that they get the proper attention at the proper times and in the proper forms or else, the dead on social networks are generally perceived as benign, and connections with them – as is usually the case with angels – are experienced as positive. Nevertheless, they are perceived to have a few expectations; or at the very least, you might still feel certain social responsibilities to them. They could be capable of feeling disappointed or neglected if you don't go online to wish them happy birthday, or you don't say hi for a while, so you should probably apologise if you've been neglectful. They're still interested in hearing everyday news about the football game, the birth of a baby, or that gig that you went to, so it's nice to go online to share it. They might feel hurt, abandoned or rejected if you de-friend them, so excising a deceased friend from your social network feels complicated and strange to many. It's important to express your gratitude by going online to acknowledge and thank them for the messages they've sent and the help they've given you, although that's been via other channels – natural phenomena, for example, or intercessions to keep you from harm. 'Thanks for the dream you gave me, you weirdo.' 'There's been a really bright star in the sky lately and I know that that's you.' 'The car almost skidded over the median. Thank you for keeping me from going across all the way.'

Might something about our digital environment actually encourage our belief in its ability to help us traverse earth and heaven, just like the angels, and cross the barrier between life and death? Whether we are secular or religious, whether we believe in an afterlife or not, why might we have an intuitive sense that communicating on a social network is an effective way of reaching across death's divide to our sentient, socially invested dead? Well, we're now accustomed to technologically mediated communication with remote others. We don't need auditory cues, don't require seeing someone's face, to feel confident that our message has landed. That confidence can be derived from two blue ticks in WhatsApp, your friend's icon sliding down to the last thing you said on Messenger, or an automatic 'read' receipt in your email inbox. To send a communication into the ether is to assume it has been nearly instantly received, no matter how far distant your correspondent lives.

Have you ever sent an email or an SMS to an unintended recipient? A message with sensitive content, perhaps? On those occasions, do you comfort yourself with the thought that maybe they just won't get it, just won't read it? Of course you don't. You know that they've received it, that they could be reading it right now, and you panic. If boundary-less, limitless, instantaneous communication is all you have ever known and that is your everyday experience of technologically mediated communication, is it really so odd that this sense would stop with your friend's death? Not being able to see or hear them has never stopped them getting your messages before.

There's an additional reason that the connection remains so strong on social networking sites, and you've already seen a clue – it's in something that Ava said. 'When I can't see what I wrote to her,' she said, 'I feel like she won't be able to see it, too.' It sounds as though Ava is describing one type of internal continuing bond, a kind of merger or incorporation into herself, so that her friend is able to perceive things through Ava's eyes. But it also reflects something else, something fundamental to the nature of social networking sites. Their raison d'être is connecting people, positioning us all within an increasingly complex web of relationships, the algorithms suggesting new points of contact all the time.

The English translators of the German philosopher Martin Heidegger liberally sprinkle hyphens to convey our constant interconnectedness with other people – we're never just Being, we're Being-in-the-world-with-others, and all of us Beings are never getting out from under that particular existential given. Social networking sites are simply our Digital-Being-in-the-world-with-others. Sure, you set up the profile on your own in the first place, but from there onwards it's co-constructed, co-authored. When Mark Zuckerberg launched the new Timeline layout in 2011, he invited us all to write our autobiographies, generously offering his site as publisher. '[This is] an important next step to help you tell the story of your life . . . to highlight and curate all your stories so you can tell who you really are,' Zuckerberg said at the launch. Autobiographers on social media, though, inevitably have a whole lot of collaborators. If you were writing a hardback book about your life, you could maintain total creative control and commit it to print: there it is, in black and white, from the horse's mouth. But on social media you write your autobiography together with your co-authors, your friends. If you interacted with a friend constantly on social media and you visit that profile after she's gone, it doesn't just feel like 'this was her'. It feels like 'this was us'. Or perhaps even 'This is us'.

Facebook has continued to evolve its approach to dead users, with the avowed purpose of making it easier for people to continue bonds. Much more will be said about this later in the book, but for now it's enough to know that one of their more recent innovations has been the Legacy Contact, a role akin to a platform-specific will executor, who is nominated by the user in life. A Legacy Contact cannot delete any friends that the deceased chose for themselves, but they can add someone who was not on the original friends list, like a parent or grandparent, maybe someone new to Facebook who's only joining to connect with the dead person's legacy. If that person feels funny about doing that (and there's research to suggest that people who are less familiar with social media tend to find interaction with digital remains unnerving rather than comforting), the Legacy Contact can also download an archive of data, provided the deceased gave permission for that when they set up their legacy provisions. A downloaded archive makes it possible for people who don't have Facebook accounts to access the material. It's also possible for users to stipulate that they want their profiles deleted after death, but for that to happen a Facebook user will have needed to have the awareness and motivation in life to delve into their settings and tick that box.

When you combine the fact that Facebook profiles are durable by default with the huge number of users (over two billion and growing), it's little wonder that the dead are proliferating on the world's most popular social networking platform. Just how rapidly the roster of the deceased is growing, or precisely when the dead will outnumber the living on the site, it's hard to say. The years 2065 and 2098 have both done the rounds in the popular press, but on closer investigation these seem to have been the products of fairly quick and dirty calculations. Certainly, accurate statistical projections of when the tipping point will occur are challenging to make, especially when we can't predict the peaks and troughs in Facebook's future fortunes, whether these will be exponentially amplified by savvy business decisions or derailed by major scandals. In the second quarter of 2018, for example, the company's growth stalled, perhaps due in part to controversies surrounding the 2016 US presidential elections. Even so, researchers at the Oxford Internet Institute have done their utmost to produce a more thoroughly considered projection of the global accumulation of dead people on Facebook, using the best source material available: the demographic data of 1.9 billion current Facebook users, and population statistics from the United Nations. Even if uptake plunged through the floor and the last user signed up in 2018, they've calculated that 1.3 billion users would be dead by the end of the twenty-first century, with a pretty large uptick shortly thereafter. If the site were to continue riding the wave of success, on the other hand, racking up new users at an annual growth rate of 13 per cent, and continuing to retain the profiles of the deceased, 3.68 billion profiles on the site would be memorials by the waning of the century.

Whichever way you slice it, that's a lot of deceased users – the rumours of millions of dead folks hanging out and being rather less than social on social media are entirely true. While the World Wide Cemetery never posed much of a threat to physical burial places, Facebook has become the largest 'cemetery' in the world. Offline cemeteries and monument companies, doubtless wondering whether their own coffers will steadily dwindle to nothing, see Facebook as the competition. Many have decided that if you can't beat 'em, you might as well join 'em. David Quiring's Seattle-based monumental masonry company, for example, is one of several companies producing QR codes that can be embedded in traditional granite headstones. Their website explicitly positions their QR-to-website service as an alternative or adjunct to a memorial on Facebook, suggesting that those left behind could compile obituaries, information about family heritage, photos and comments on a Living History™ archive website. John Troyer of the Centre for Death and Society envisions a 'Future Cemetery' that you can walk through wearing a virtual-reality headset, encountering your reanimated ancestors along the way, and I've seen demonstrations of 'augmented reality' technologies enabling photographs to pop up on the screen like Pokémon when you aim your iPad's camera at a perfectly ordinary-looking headstone.

All these technologies could certainly make physical cemeteries more engaging and meaningful places but, as the years pass, will we continue to be motivated to visit actual places of interment, when instead we could easily continue bonds out of the rain, on our phones, in the comfort of our own homes? But cemeteries with no visitors aren't the only concern here, because just as digital remains and digital technologies afford the potential to continue bonds, they can disrupt them too. So let's look at the other side, through a couple of stories that sound as though they would never actually happen.

***

First, call to mind your nearest, dearest, most lifelong friend. Ideally this is a childhood companion, someone you met at school perhaps, whom you have known and cherished for as long as you can remember. It could be a friendship of more recent vintage, but key to this exercise is that this person is integral to your life. If they had not been there for you, and there with you, you would be a far different person today. No one knows you like they do, and in your time you've generated a lot of memories together. You couldn't possibly count how many conversations you have had over the years, how many messages you have sent one another, or how many photographs you've accumulated together. Imagine that you keep all these relics of your friendship in a large shoebox, with your friend's name written on the lid. It's a large box, because everything is in there – the origami-folded notes you passed under the teacher's nose, letters in their stamped envelopes from when you moved to separate cities, and stacks of photos that span the decades.

One day you get the worst news that you could ever receive: your friend has died suddenly, in an accident. You are numb with shock and disbelief. They have always been there – a part of you has died with them. Grief-stricken, and aching for a sense of connection with them, you rummage in the top of your closet, bring down the shoebox, and place it by your bed. In the days and weeks after the death, you pore over the box's contents many times a day. It's not the same as having your friend back, but everything in that box is redolent of not just who your friend was, but who you were together. Its contents make you laugh, cry, smile. Kind messages flow in from friends and family who know of your terrible loss, and you place these in the shoebox as well. After some months, you notice that you are spending less time looking through the box, but that's all right – you know where it is if you need it, if the impulse takes you, if you want to find that thread of connection again. The box is just there, within arm's reach.

One day you hear a knock at your front door. Opening it, you are surprised to see your friend's parents, whom you have not seen since the funeral, and who have travelled many hundreds of miles to arrive at your front door unannounced. Silently, and without even acknowledging your presence, they push past you and walk down the hallway to your bedroom. They unceremoniously bundle up the shoebox and, as abruptly and wordlessly as they arrived, slam their car door and drive away. You are overcome, frantic; you are desperate to retrieve your memories. When you are finally able to contact these people about their inexplicable behaviour and the whereabouts of your shoebox, they have terrible news: they have destroyed the box, and all its contents. Your blood thrums in your ears, then feels as though it's drained from your body. You can almost feel your friend's hand slipping from the grasp of your own. You've lost them all over again: it's like a second death.

This time, instead of being the friend, you are the parent – the parent, say, of a teenaged son or daughter. One day they go out as usual; several hours later, the police are knocking at your door. Reeling with grief, you begin to do what you must: you pull yourself together to inform everyone of what has happened and to make the necessary arrangements. As a parent, you naturally assume that you will have a central role in arranging the celebration of your child's life, an occasion where everyone will come together, remember, and support one another in mourning this profound loss. When you take the difficult step of picking up the phone to call your closest family and friends, however, you are floored to discover that they already know, that they heard about it virtually as soon as it happened. Some of your child's friends have far more details about the accident than you have yourself. The news isn't yours to tell.

Although you're rattled, somehow you manage to resume your planning. You search for photographs to illustrate your child's life, but you realise that you don't have any recent ones. Once again, you get in touch with your child's friends for help. To your astonishment, you learn that all the photographs they have are currently in use at a memorial service for your child, a memorial that is taking place right now. It's being attended by hundreds of people, and it's being held in a location where you've never been before. You rush to get there, but you find the gates locked. Beyond the closed doors, you can hear people laughing and crying together – they are showing photographs, sharing stories, supporting one another and talking about your child. You knock, but no one answers. You pound louder and kick at the door, pleading for access and threatening to burn the place down if you don't get it. That's your child. You should have been a part of this. You should have organised this. At the very least, you should be there. 

But no one lets you in.

*** 

If you're thinking that these scenarios are absurd, you're right. Anyone who enters someone's home without invitation to steal and destroy a box of important documents could be arrested and charged with trespass, theft and wanton destruction of property. And what sort of people would band together to plan and attend a memorial service for a deceased friend that deliberately excluded the friend's parents? But, in essence, this is what's happening. When digital remains cause a disruption in continuing bonds, it's almost always about one or both of these things: access, and control.

There are three broad categories of troubling experiences around access, all of which can interfere with a mourner's capacity to determine for themselves the kind of continuing bonds they want, or don't want. The first is the threat of the digital remains vanishing entirely, for ever. Mechanisms like automatic backups in the cloud may lull us into a sense of security about the persistence of all those data, but don't be fooled. Digital beings have an anxiety-provoking, paradoxical dual nature. On one hand, they can stick around for ever, and on the other they can vanish without a trace. Because there are no more shoeboxes in the attic, the fear of losing access to a loved one's digital remains is intense, perceived as a second death, a thief in the night that can steal the last hope of connection. 'I would be close to inconsolable,' said one of my own young research participants. 'Having something that may seem so small to some people is everything to me. [His profile] is the one last thread of him that I have. If we lost it, it would be like losing him all over again. There are just certain things that rip the wounds open.'

Your connection to someone's ongoing digital 'self' can be severed for ever through a host of mechanisms: a software glitch, an expired webpage subscription, terms and conditions of a particular platform that stipulate deletion on the account holder's death, hardware malfunction or obsolescence, or the next of kin removing a social networking profile. If you're a digital native whose communication history and photographic memories are stored completely on Facebook, your best friend dies, and his or her profile is deactivated: someone has entered your house, stolen your shoebox, and destroyed every last memory in it. And it's all perfectly legal.

The second problem with access occurs when you know the digital remains are out there, and you would love to get to them, but you can't. While in typical circumstances it is unlikely that anyone would ever lock you out of an offline memorial service, and no one could ever bar you from laying flowers at a headstone in a public cemetery, you could easily find yourself in a situation where your loved one's persistent digital reflection, perhaps the most comprehensive representation of them that survives in this world, lies frustratingly near but beyond your reach. Perhaps you weren't a Facebook user, but your child was, and they left no Legacy Contact that could add you. You would hope that someone could sit down with you, take you through it, and screenshot or download the bits you want; on the other hand, people may exclude parents or others from their social media for good reason, and loyal friends may be aware of that. In addition to exclusion from material, however, you'll also be excluded from the community of mourners that exists in that place. One mother, a participant in another researcher's work, describes the devastation of precisely this kind of exclusion. When her son died, she said, she was surprised to receive no cards, no letters of condolence from any of her son's friends. During the funeral, she was told by his best friend that there was a Facebook page memorialising her son, but without a Facebook account of her own, she couldn't get to it. All she had, she said, was an 'empty mantle', the place where she would have placed cards containing words of solace. She didn't know what she could do about it.

This issue can disproportionately affect family, rather than friends. Once upon a time, when someone died, it was the family, the legal next of kin, that had privileged access to the dead: to their bodies, to information about them, to their personal effects, to their writings and photographs. Law of succession was everything. Having privileged access also meant having control over who else had access, and family members were the keepers of that gate. In contrast, friends were at the opposite end of the access spectrum. However important their relationships might have been with the dead person, friends were in danger of being left out, of becoming disenfranchised mourners – a term that describes those whose grief is unseen, their right to mourn unrecognised or unacknowledged. The long-time secret mistress who wanted her letters back would likely not have got past the front porch. School friends might never hear of the death or might not know in time about the memorial ceremony that they would have wished to attend. In an extremely short time, the balance of power has flipped, for the era of social networking is the Age of the Friend. If a mistress is on Facebook and Instagram, she'll have far more access to the digital legacy than a grieving but technophobic wife. A father might certainly find himself with no physical photographs, appealing to friends for images to display at the funeral. Scenario number two doesn't seem so absurd any more, does it?

The third issue with access is when there's too much of it. Now that memorialisation of profiles is more widely utilised, it's somewhat less common to encounter digital presences that 'act alive' – unmemorialised profiles suggesting that you wish your dead friend Leslie a happy birthday or propose that you like Glamping Holidays Inc. because your dead friend Keith quite enjoys them too. Profiles that are unmemorialised also act alive through being accessed and managed by others, with consequences that may prove difficult for many mourners. Some people, through technical ignorance or psychological naivety, have no idea how much their activity on a dead person's site can discombobulate others, but sometimes they're aware and proceed anyway.

A heart-wrenching illustration of this latter scenario is the story of Vanessa Nicolson, who derived comfort from being on her late daughter's unmemorialised Facebook account, which was still logged in as though Rosa were still alive. One day, Vanessa saw a significant status update from Adam, Rosa's boyfriend at the time of her death. The post made clear that he had started a new romantic relationship. Writing in the Guardian, Vanessa explained what happened next. The feeling that her daughter had been 'replaced' gave Vanessa excruciating emotional pain. Almost without conscious thought, Vanessa posted a comment underneath Adam's status update. 'Don't you think it's a bit soon to fall in love again?' she said. As soon as she had done the deed, she snapped back to her senses, and realised that it would appear on the stream of comments underneath the post as a message from Rosa, Adam's dead girlfriend, complete with her profile photograph. Panic was added to the horrible cocktail of emotions Vanessa was feeling in that moment, and the impact on Adam and other friends is hard to imagine.

Indeed – unsurprisingly – when a spouse or parent is controlling the profile, research indicates that it has an almost universally negative effect on other mourners' ability to cope. Even if a profile is memorialised, however, with no danger of startling messages popping up from beyond the grave, it could still prove problematic. There's no separate cemetery on Facebook, one that you can choose to visit or not – although the design for memorialised profiles is always evolving, at the time of writing, even memorialised 'Remembering' profiles are integrated within the list of living friends, listed alphabetically with everyone else. This makes oscillation between the dual processing model's 'loss orientation' and 'restoration orientation' as simple as a click or a swipe – living friend, dead friend, and back again. It's like encountering tombstones dotted here and there on the busy streets of a bustling community. The possibility of connecting with a Remembering profile is always there, and it may not always be easy to decide: is this a good time for me to stop, remember and connect, or a good time to just keep walking?

*** 

The other potential disruption of continuing bonds derives from challenges to control. Even if you do have access to part or all of a person's digital legacy, there may be troubling things about it that take your agency away, and that disrupt your ability to continue a connection. Problems with lack of control come in myriad forms, and one is death notification. Social media transmit news more quickly than any police force on earth, so just like in the second scenario above, news of a death may appear on Facebook or Twitter before immediate family members know about the death themselves, before they have any opportunity to process the loss, or before they can make decisions about disclosing it. Those notifications can then form part of the enduring footprint, a perpetual reminder of the pain of that moment of realisation.

Encountering unexpected bits of a digital legacy can also upend what you thought you knew about someone, shredding the integrity of the bond to pieces. Even in the case of a digital legacy that is largely positive and a source of comfort to a griever, the presence of upsetting or traumatic elements can taint the whole picture. The profile photo might be all wrong. A family might not even be able to change a profile that is plastered throughout with depictions of the perpetrator(s) responsible for their loved one's suffering and death: a woman killed by her partner, for example, or someone who ended their life after being abused or bullied by a friend or family member. Less dramatic, but just as problematic for continuing bonds, the digital footprint might not constitute the kind of durable biography that an individual mourner can live with – and there may be nothing they can do about that. Any time that multiple parties with disparate needs and preferences are accessing a central set of digital remains, it's likely that not everyone will be happy with all that appears there. Who gets the last word on who the dead person 'really' was?

Sometime in the early 1970s, Aidan decided to leave the UK and start a new life. In so doing, he left behind a dependent family, including a six-year-old daughter, Cathy, and that family struggled on in poverty after his departure. 'He was an alcoholic wife beater who deserted his children and left my mum pretty much destitute,' says Cathy. She never forgave him for leaving, and she never saw him in person again. Some years later, however, something discomfiting happened – she received a Facebook friend request from him. Having never considered her errant biological dad a father at all, much less a 'friend', Cathy ignored the request. Aidan persisted. Eventually, Cathy relented. What if he were planning to return to the UK for a visit, or for good? She was from a relatively small town, and she wouldn't want to be surprised by running into him in the high street. What if he had changed? And even if he hadn't, what if he died? Even if he didn't play any role in her life, shouldn't she know?

Eventually, Aidan did die, but even though they'd never actually interacted on the site, his digital presence there proved difficult for Cathy. One of her siblings had reconciled with their father, much to Cathy's displeasure, and the people who had known him in more recent years used his memorialised profile to glorify him. 'There's this profile picture that I've always hated of him with my [siblings' kids],' Cathy explained, 'that's never going to go away. And it's just always there. There are all these people who are sainting this man who is so far from being . . . I'm not saying he didn't have good qualities. But history's just been completely rewritten in this page that just sits there, and there's nothing I can do about it. And I can't de-friend him, because then I won't ever be able to see what's going on . . . but at the same moment I can't bear to look at it, because it just kind of makes me vomit when I read all this stuff that isn't him.' Aware that she was in the minority opinion, she was loath to post her own version of events on her father's Facebook profile, and she is still grappling with her feelings months after his death.

*** 

Like Robert Louis Stevenson's famous character, Aidan was both Dr Jekyll and Mr Hyde. Both personae were equally true, at least to the people that knew each of them. For the American members of Aidan's family and social circle (according to Cathy, who confronted some of them via Messenger about what she perceived as their hurtfully inaccurate or at least incomplete portrayals), his persistent digital self was a good fit with the man they knew. His digital remains facilitated their own bond with the father, husband and friend they'd lost. For Cathy, Aidan's enduring biography was nothing short of a travesty, a constant niggle and source of pain, a thorn in the side in her grief process. 'It's not complicated grief,' she insists – she's familiar with the terminology of bereavement – 'but it's complicated.' Complicated, indeed. When a digital legacy is a salve for some and a wound for others, what should happen to it? Who decides? Well, more often than you might expect, it's corporations. 

- Elaine Kasket is a digital immigrant masquerading as a digital native, an American masquerading as a Brit, and a committed urbanite masquerading as a chicken farmer, but she is a genuine psychologist, writer, and public intellectual. A former principal lecturer in counselling psychology, she left formal academia behind to spend more time communicating to general audiences about the topics that most fascinate her. When she isn't writing or speaking about the murky and weirdly compelling junctures where life, death and the digital meet, she provides psychological therapy, is a producer for the Mortified live storytelling project, tells tales on stage, and makes unsuccessful attempts at learning the banjo. She has appeared in various print, broadcast and online media, to include BBC television, ITN Channel 4, Radio 4 and the Mortified podcast. She lives in East London with her husband, daughter and a revolving cast of troublesome bantams. Her book All The Ghosts in the Machine, published by Robinson, is available now. This extract is reproduced with kind permission.