A Rape in Cyberspace, revisited

Thirty years ago, an online sexual assault challenged our very assumptions about the real and the virtual. With the rapid rise of the Metaverse, we are no closer to answering the questions it raised.

Over the course of the late seventies and eighties, as early adopters in the home computing craze struggled with just what to do with these amazing new machines, there was one group that had already figured out what computers were good for. These were the people who worked across the various Venn diagrams incorporating academia, research, or the defence industry, and who had access to the network of mainframes and dumb terminals that at that time made up the internet. 

And what did they use the internet for? Chatting, mostly. Email was the original killer app. But in addition to just talking, they did the sorts of things humans have always done when they able to communicate in groups or privately with one another: they debated politics and culture and science and religion, they played games… and they had sex. 

Sexting is certainly as old as writing (see: Napoleon’s letters to Josephine), but people quickly realised that the speed, immediacy, and opacity of computer-based texting could generate the kind of heady intimacy and biological responses that challenged the very idea of “virtuality.” 

But unfortunately, with the possibility of sex comes the prospect of sexual harassment and even sexual assault. If you can have sex online, can you also be sexually violated online? And if so, what does that mean, for law, for community, for speech, and for the very idea of the real/virtual distinction?

All of this came together in an article published in the Village Voice in 1993, written by the freelance tech journalist Julian Dibbell. The piece, whose full title was “A Rape in Cyberspace, or How an Evil Clown, a Haitian Trickster Spirit, Two Wizards, and a Cast of Dozens Turned a Database into a Society", described the repercussions of a “cyber-rape” that had occurred in an online community called LambdaMOO. 

To understand what happened, you need to know a bit about how things used to be. Before what we call the “graphical” internet — basically since the launch of the Mosaic web browser — almost all online interaction was done through scrolling lines of text. A popular form of text-based interaction was through what were called MUDs, or multi-user dungeons. First developed in the mid 1970s, MUDs were virtual worlds in which multiple users, or players, could interact in real time according to the rules and architecture of the MUD. If you have ever played a text based game like Zork or Planetfall or the Hitchhikers Guide to the Galaxy, you know more or less how these things work. When you log in, you are given a description of the environment and any people or other entities, and through various commands (go left, open door, drink water, shoot bug) you can explore the environment and interact with the other figures in it. 

A subset of MUDs, called MOOs (for “MUD, object-oriented”) allow users to manipulate the virtual space in which they are interacting. This can include activities such as creating new rooms and objects, changing the appearance of entities within the environment, and even generating new rules or algorithms that govern how the MOO operates. 

The oldest of these spaces is called LambdaMOO. It was created by a Xerox PARC researcher named Pavel Curtis, and its basic architecture was based on Curtis’s own house. It has a coat closet, which opens into a living room with a fireplace and a couch, as well as other rooms and spaces including a garage, a yard, a room with a hot tub, and so on. In these areas users can gather, represented pseudonymously through virtual avatars that can have whatever characteristics their users choose to give them. All sorts of identity-based cosplay is common, with users adopting whatever gender-, sexuality-, or species-bending roles might tickle their fancy. 

And so it was that sometime in March 1993, a LambdaMOO user named Mr. Bungle (self-described as a “ fat, oleaginous, Bisquick-faced clown dressed in… harlequin garb”) took advantage of a so-called “voodoo doll” subprogram in the environment that allowed him to take actions that would be attributed to other users in the community at the time. Over the course of a few hours, Mr. Bungle forced other members to perform increasingly degrading and violent sexual acts on others and upon themselves, while he holed up in another room and laughed. It was stopped only through the intervention of a senior member of the community who had the administrative powers needed to put a halt to it. 

The whole episode stunned the LambdaMOO community. There was no question some of the most basic norms that governed the space had been violated. But what, exactly, had happened? As Julian Dibbell wrote:

 No bodies touched. Whatever physical interaction occurred consisted of a mingling of electronic signals sent from sites spread out between New York City and Melbourne, Australia. Those signals met in LambdaMOO, certainly, just as the hideous clown and the living room party did, but what was LambdaMOO after all? Not an enchanted mansion or anything of the sort — just a middlingly complex database, maintained for experimental purposes inside a Xerox Corporation research computer in Palo Alto and open to public access via the Internet.

Even the victims themselves had trouble figuring out how to interpret what had been done to them. In the aftermath of the episode, one of them publicly scolded Mr. Bungle for what she called a breach of “civility”. But as she later confessed to Dibbell,  “posttraumatic tears were streaming down her face” as she was writing. But while many in the community were comfortable describing Mr. Bungle’s actions as an actual rape, there was never any question of bringing in the real world authorities. The community seemed to agree that what had happened in LambdaMOO would stay in LambdaMOO, and any punishment to be meted out would be decided there as well.

But there were no protocols in place, no laws or explicit codes –there was no constitution for LambdaMOO –  and before the infamous events in question, Pavel Curtis had made an executive decision to leave the running and governing of LambdaMOO to its members. So before they could figure out what to do about Mr. Bungle, they first had to figure out what the decision procedure itself would be. Most of Dibbell’s piece is devoted to chronicling the efforts by the community to figure out what to do, to invent self-government from scratch,  as the various factions – the so-called parliamentarians, monarchists, libertarians and anarchists – put forward their suggestions. (It all ends up resolved through a bit of administrative deus ex machina, with Bungle summarily banned). It is all super interesting, and there is a reason why Dibbell’s article quickly became a standard reading in courses on the burgeoning field of internet ethics or “cybermorality.” 

But there is one issue hanging in the back of all of this, and that is how to characterize the attack itself. On the one hand, Dibbell is frank about the low-res irreality of it all: “When all is said and done the only thing you really see when you visit LambdaMOO is a kind of slow-crawling script, lines of dialogue and stage direction creeping steadily up your computer screen.” 

But he also finds that, the longer he spends in the world of LambdaMOO, the less sense the supposedly clean distinction between the “virtual” online world of bits and the “real” world of atoms seems to make. Where he had once found it hard to take the idea of an “online rape” seriously, he ends up wondering how he was ever not able to do so. 

The precise status of the online world is the crucial question raised by Dibbell’s piece. What is amazing is that, thirty years later, we are no closer to arriving at a settled answer. Just last week, the Washington Post reported that, as virtual reality environments become more popular, reports of attacks including sexual harassment and assault are growing more common. Virtual reality is still a legal vacuum, though, which has led Interpol to call for police forces around the world to develop protocols for dealing with VR crime, including sexual assault. 

There remains an enormous amount of resistance to the idea that virtual sexual assault is in any way “real”, and one woman who wrote about her experience being sexually harassed in the Metaverse found herself bombarded with emails calling her stupid and ridiculous. But it is hard to see how this is a stable position. As some see it, it is the immersive nature of VR that can make the attacks feel as real as anything, but the issue goes beyond just the increased resolution of the technology. As Dibbell realised almost right away, the online world, even the one based on scrolling lines of text, has an ontology (and consequently, a sexuality) all of its own that we haven’t come close to figuring out.

Of all the prescient aspects of the original “A Rape in Cyberspace” article, the most important is Dibbell’s observation that the early 1990s marked  the “final stages of our decades-long passage into the Information Age, a paradigm shift that the classic liberal firewall between word and deed (itself a product of an earlier paradigm shift commonly known as the Enlightenment) is not likely to survive intact.”

---

Sorry for the delay getting this week’s dispatch out, but welcome new subscribers, and as always, thanks for reading. If you like this, I’d appreciate you sharing it with anyone who might it enjoyable as well. — ap