When psychologist and neuroscientist Paul Verschure traveled in 2005 to visit his grandfather’s final resting place, on the grounds of the former concentration camp of Bergen-Belsen, an hour’s drive south of Hamburg, he left disappointed. “I found nothing,” he said. “It was a beautiful park.”

After the camp’s liberation, in April 1945, a small section of the complex was selected for memorial landscaping, and the remaining structural remnants — a few fences, a watchtower, a demolished crematorium — were uprooted and replaced with shrubbery. Almost no trace of the original, abominable architecture was left — a 1991 excavation revealed only the foundation of several buildings. Stone monuments, an obelisk, a documentation center, and a memorial “House of Silence” mark the annihilation that once took place there, but, as the Bergen-Belsen memorial foundation notes on its website, “around two thirds of the camp’s historical area now resembles a park-like heath which reveals nothing about the camp that once stood there.” One could still easily stroll among the site’s crisp birch trees and forget the mass graves that fertilize their roots.

The chilling ambiguity of Bergen-Belsen’s vast pastoral landscape has also been found in Berlin’s Memorial to the Murdered Jews of Europe, a field of 2,711 gray concrete stelae adjacent to the American embassy and the Brandenburg Gate. When the city is teeming with tourists in the summer months, visitors can be seen gleefully snapping photos from within the somber field or climbing atop the stelae to get a better view of the Tiergarten across the street. “Yolocaust,” a media project from Israeli author Shahak Shapira, recently lambasted this impulse in a deliberate act of public shaming: Shapira mined social media to collect insouciant selfies taken among or atop the field of stelae, superimposed the photographers’ faces on historical photos of Holocaust victims, and posted the doctored images online. “I am worried that younger people fail to understand the importance of these memorials,” Shapira told the BBC. “They’re not there for me — for Jews — or for the victims; they are there for the people of today, for their moral compass. So they know not to elect the guys with the Hitler haircuts, because we could end up right where we were 80 years ago.” As Shapira prepared to launch the site, the German populist politician Björn Höcke condemned the memorial’s prominent placement, saying Germans are the “only people in the world who planted a memorial of disgrace in the heart of their capital.”

A “gray sentiment” incensed Verschure when he visited Bergen-Belsen: The site seemed to him a void, bereft of information, bereft of remembrance

The impulse to take a selfie at the Berlin memorial may be unrestrained by its jarring lack of specificity. In a review of the memorial, critic David Denby condemned the site’s elision of who murdered the Jews, where, and why. “Of course, the information is familiar, and few visitors would be unaware of it, but the assumption of this familiarity — the failure to mention it at the country’s main memorial for the Jews killed in the Holocaust — separates the victims from their killers and leaches the moral element from the historical event, shunting it to the category of a natural catastrophe,” Denby writes. “The mollifying solemnity of pseudo-universal abstractions puts a great gray sentiment in the place of actual memory.”

A similar “gray sentiment” incensed Verschure when he visited Bergen-Belsen: The site seemed to him a void, bereft of information, bereft of remembrance — an affront to the memory of what took place there. Yet such bereavement is the status quo on a continent littered with mass graves: “There are 42,000 of these sites around Europe, and the vast majority of them are invisible,” Verschure told me. “It’s a very generic problem.”

And as with many generic problems of our time, it seemed possible to address it with a tech solution. At the time he first visited Bergen-Belsen, Verschure, who runs the Synthetic Perceptive, Emotive, and Cognitive Systems (SPECS) group at Universitat Pompeu Fabra, in Barcelona, had begun experimenting with interactive installations or “intelligent spaces.” These immersive rooms are “equipped with a wide range of sensors and effectors” designed to affect and interact with those who wander inside, exploring, as his team of researchers put it, “how humans can act, exist, and behave in both physical and virtual spaces; the construction of socially capable believable synthetic characters; and the development of a framework for interactive narratives.” In 2002, Verschure debutedAda: An artificial creature” at the Swiss Expo, an “intelligent space” that detects the sound, feel, and look of “her” visitors in order to interact with them through patterns of music and light. Ada aimed to harness the brain’s continuous construction of the outside world — the infrastructure of consciousness — to teach a machine to identify and interact with humans “in a non-anthropomorphic way,” Verschure says.

Ada laid the groundwork for Verschure’s lab to create the Rehabilitation Gaming System, a virtual-reality program that uses immersive technology to help restore brain function in stroke patients. Several studies have found that physical activity improves memory recall; the Rehabilitation Gaming System combines physical activity with interactive media to maximize neuro-rehabilitation through what Verschure calls “embodied goal oriented training” and has shown positive results in trials with more than 500 patients.

If interactive virtual spaces can have that sort of effect on individual memory, what could they do for historical remembrance? “This translates directly into how we think about commemoration,” Verschure told me. Ushering survivors and witnesses back into the scene of the crime, either in person or through virtual reality, could help them remember new details even a half-century later.

So in 2010, Verschure launched the Future Memory Foundation and set about creating a virtual and augmented reality version of Bergen-Belsen. His team began with primary resources, including aerial photos of the operational camp, historical photos and audio recordings, and interviews with survivors. By 2012, SPECS released a “box simulation” — a static, immersive presentation that panned over a gray-and-white landscape representing the camp as it existed in 1944: rows of barracks separated by watchtowers, a central road, and fences. “But it’s so peaceful, so beautiful, the heath, the trees, the birds,” explains Michael Gelber, a survivor, in a commentary that accompanies the simulation. “That’s not what it was back then. It was the exact opposite.” As he speaks, archival photos of the camp’s operational years are superimposed on Verschure’s simulated environment.

Over the past four years, the project has evolved from the box installation to an immersive virtual environment and augmented-reality tablet app that allows visitors to Bergen-Belsen to visualize where the camp’s machinery of death once stood. Used properly, Verschure’s technology prevents any visitor from mistaking their visit for a peaceful stroll. When Queen Elizabeth II visited the site, in 2015, Verschure was there to show her how to use his app: “Isn’t that a great thing,” she said. The pilot was a success.

“After this important validation of the Future Memory approach, our goal is to digitally reconstruct, enhance and link together at least 100 sites across Europe, to show the system-level organization of the murder machine created by the Nazis,” Verschure has written of the project’s mission. To illustrate the urgency of that laudable aim, he cites a 2014 study of British high school students’ knowledge of the Holocaust as motivation: Researchers found that the vast majority of students widely underestimated the death toll, didn’t understand why Jews were targeted, and did not understand the meaning of the word anti-Semitism. For Verschure, these findings prove that the way we remember, memorialize, and teach the past is not working. He is hardly alone in that conclusion.

But while Verschure hopes his project will help viewers understand Holocaust sites, he does not intend his simulations to evoke the actual, horrific experiences of life in the death camps. The Bergen-Belsen simulation is deliberately devoid of detail and color, relying on the historical record for texture, sound, and life. His augmented constructions allow visitors to see where the architecture of genocide once stood but deploys embedded audio testimony, historical photographs, and written survivor accounts to portray how it operated. The result is that the camp is not rendered at the height of its crimes nor at its current state of memorialized erasure — it is out of time, a historical document in 4-D. “We confront you with historical information, but meaning is not something you can dictate,” he says. “VR is not a silver bullet; it’s just a technology … The avatar does not replace the witness.”

It is not a great leap to wonder if future memorials will allow visitors to imagine themselves at the scene of the crime, to hop between perspectives of victims, perpetrators, and bystanders

Other projects are not so deliberately situated and contextualized. The Bavarian Landeskriminalamt, the state police, have created a similar VR model, of Auschwitz, for use in ongoing prosecutions of living Nazi officials. Not intended for broad public consumption outside the courtroom, the model is meant to place users in the midst of operating death camps, from the point of view of the perpetrator. The simulation claims to be accurate down to the last tree, allowing users to walk through the gates of the camp, to survey the barracks and the gas chambers just as an SS officer might have. It has already been put to use in the case of SS officer Reinhold Hanning, who was sentenced to five years in prison last spring at the age of 94. “The advantage the model offers is that I get a better overview of the camp and can re-create the perspective of a suspect — for example in a watchtower,” Ralf Breker, who created the model, told Agence France-Presse. “In two or three years, you’ll be able to enter the scene of every serious crime virtually.”

While the Auschwitz model has a specific juridical and nonmemorial purpose, the conditions of its use may one day change — the German prosecutor’s office charged with pursuing the last Auschwitz perpetrators will become an archive within the next 10 years, and one can imagine that the model will be included in that trove. It is not a great leap, then, to wonder if what the Bavarian Landeskriminalamt has created may be a preliminary manifestation of what future memorials could try to achieve: to allow visitors to imagine themselves at the scene — the time and place — of the crime, and to hop between perspectives of victims, perpetrators, and bystanders, or to explore a scene as a disembodied observer.

Augmented and virtual reality hope to change not only how we live in the future, but also how we view the past — and from whose point of view we inhabit history. Technology may have expanded the points of entry from which we can approach past wrongs, but that does not mean we should use all of them.

“What kind of circle is it, that aims to represent all sides of a horrible act?” Maggie Nelson asks in The Art of Cruelty.

Does drawing such a circle provide the most ethically thorough and fearless approach to a heinous deed, or is “true” ethical clarity achieved only when one privileges the experience of the victim? Does focusing on the POV of a perpetrator re-perform a cruelty, under the guise of a far-reaching empathy? How to cultivate the difference between an all-inclusive compassion, with freely given forgiveness at its base, and idiot compassion, which fails to assign or take responsibility or to protect us adequately from those who have done or would do us harm?

Virtual reality is that circle, and those are its stakes. It may very well be the ultimate empathy machine,” as Chris Milk calls it, but it can manufacture cruelty too.

“On the one hand you have this new body of VR films, that try to prompt empathy about victims of injustice, and on the other hand you have the gaming industry, which is all about the figure of the perpetrator,” said Matthew Boswell, a researcher of Holocaust memory at the University of Leeds. The tension between the two is at the heart of the emerging body of virtual memory projects. The Enemy, a project by photojournalist Karim Ben Khalifa, is a virtual-reality simulation that creates “a face-to-face encounter between combatants of opposing sides.” As the viewer walks between the enemy combatants, Ben Khalifa explains, “we will measure how they physiologically respond to the installation, and by using neuroscience research, we hope to shed light on what kind of empathy has been created.”

Amnesty International’s and Forensic Architecture’s online simulation of the Saydnaya Military Prison, a government-controlled detention site near Damascus where regime opponents are routinely tortured and murdered, invites viewers to “explore” the sights and sounds of the complex, including solitary-confinement cells and torture chambers, conveying the abuses perpetrated therein with disturbing accuracy. These are well-intentioned and ambitious projects, designed to cultivate concern, mourning, and recognition — one must first recognize others as human in order to grieve them, Judith Butler reminds us. But as gaming merges with mourning, similar projects risk adopting the aesthetics of, say, the murderous video game S.T.A.L.K.E.R.: Shadow of Chernobyl rather than the solemn Hall of Names at Yad Vashem.

Of course, simulations do have myriad advantages over static memorials and museum displays: They are protected from prejudicial defacement and wear from hordes of visitors, and the scenes they depict can, in theory, be made available to anyone, anywhere. CEOs at the World Economic Forum in Davos can enter the Za’atari refugee camp from the plush comfort of Switzerland. Virtual reality claims to be able represent what is supposed to be unrepresentable, or at least, untransferable: crimes against humanity, genocide, torture, war, solitary confinement — the list of horrors goes on. But technology has a tendency to fail and to age, quickly. When it takes on such condemnable subjects, the failure of the medium may be an affront to the victims whose reality it has seized.

Technology has a tendency to fail and to age, quickly. When it takes on such condemnable subjects, the failure of the medium may be an affront to the victims whose reality it has seized

Virtual reality is hardly the only technology making forays into this troubled ground. New Directions in Testimony, a pilot project from the USC Shoah Foundation, uses natural-language processing to create an “interactive educational tool to permit students far into the future to ‘talk’ with Holocaust survivors about their life experiences.” A hologram-like figure of a Holocaust survivor appears in a resting pose on a screen, animating when it is asked a question. Once prompted, the computer sifts through a series of pre-recorded answers, collected over a five-day period of intensive in-person interviews with survivors, to find an appropriate response. The intention is to allow audiences to have an “authentic” exchange with a Holocaust survivor, even if this encounter happens decades from now, when the last witnesses will have passed away.

“In the resting pose, they offer a powerful metaphor: one that says something about our responsibility toward history, and toward the dead,” said Boswell in a recent lecture on the USC project. “Justice in the metaphysical, rather than the legal sense, will now depend on future generations recognizing their responsibility toward these strange digital revenants, and the role that they have to play in drawing out the stories that are buried in the electronic archive.” But future generations approaching the subject for the first time may not know which questions to ask the hologram-like figures in the first place, he notes. One can hardly probe the emotional traumas of past wrongs without having a baseline of knowledge of their events.

Sometimes the technology behind the simulation falters, altering the image and voice of the survivor. “It’s a reminder that you’re not having a conversation with a Holocaust survivor — you’re having an interaction with a computerized device,” said Rutgers Holocaust historian Jeffrey Shandler.

For some, such glimpses into the uncanny valley are the technology’s saving grace. Glitches break the scene. But other historians hope the technology will become more immersive, even customizable. Historian Wulf Kansteiner suggests that the fact that “consumers have generally no power over the conceptual framing, narrative emplotment, and visual display of the violent pasts which they are urged to remember” is a problem waiting to be solved. In a 2014 paper, Kansteiner argues that “we have to embark on the indeed somewhat frightening experiment of developing fully interactive historical worlds of large-scale persecution, ethnic cleansing, and forced migration. We have to offer consumers of these digital worlds the opportunity of three-dimensional and four-dimensional geo-immersion according to their own narrative preferences in the roles of victims, perpetrators, and bystanders.” Kansteiner complains that “the websites, displays, and animations dedicated to the dark side of history do not offer its users a chance to shape content according to their own aesthetic preferences.”

But what if someone prefers the aesthetics of the torturer, or of erasure? Kansteiner’s suggestion would undoubtedly open the door to all sorts of denial, perversion, and defacement of the past. Yet the logic of consumer choice is already applied to every emerging technology, including VR. In that sense, his suggestions are hardly novel. What he advocates is, essentially, the gamification of atrocity, which would allow newcomers to the darkest chapters of history to customize their encounter with the past. The past would be up for grabs.

“Memory is always housed in the technology of a culture, [and] each new technology raises specific ethical questions,” says Rachel Baum, a senior lecturer in Jewish studies at the University of Wisconsin-Milwaukee. “That doesn’t mean we have to throw the technology away, but it means we have to put the ethical concerns first.” As the technology improves, Baum wonders how the increasing realism of these simulations will be received. “Is it going to try to replicate Auschwitz in 1943 for someone in their living room? Are people going to leave that space and think that they’ve experienced it? Or are they going to think they saw a movie?”

Similar questions of spectatorship also puzzled Susan Sontag, who wrote this in 2003: “It is felt that there is something morally wrong with the abstract of reality offered by photography; that one has no right to experience the suffering of others at a distance, denuded of its raw power; that we pay too high a human (or moral) price for those hitherto admired qualities of vision.” Experiencing the suffering of others from a simulated proximity carries a more severe moral burden. One fears the emerging genre of VR-as-human-rights-document may take Sontag’s concern too literally.