E. Glen Weyl (GW):
We spend a huge amount of time talking to each other over Skype. But one of the first things that I learned from you was how foolish it is to be satisfied with Skype, to assume that it offers any kind of substitute for being there in person with someone. I wonder if you could start us off by talking about technology’s failures, and why it’s important for us to be aware of them.
Jaron Lanier (JL):
Accepting any particular technology as being a given, as an inevitability, as beyond criticism, as just a part of the natural environment, means that technology has failed. Such a technology has failed to foster human engagement. It has failed to be integrated into human society in a constructive way.
To criticize technology is to love it. Technology is people getting better at doing things in the world—there’s really no need for a more elaborate definition. And so to criticize the technology—whether it’s Skype, or the way our cities are laid out or the way the English language is constructed—is not to hate technology. Rather, it’s to love technology by engaging with it.
It sounds like what you’re suggesting is that a “better” technology—a better Skype, for instance—isn’t actually better if it doesn’t provoke us to think critically about it.
The most successful technologies in history are what you might call living technologies. They engage the people who use them in an ongoing conversation in which both the people and the technology change. Or you could use the word evolve if you like.
I mentioned cities and the English language as examples. And the reason I used them is because they have a systemic quality that is a precedent to what we have to deal with in digital systems these days. People have been arguing about proper usage in the English language ever since it’s existed. The language changes with us: it’s a living, growing thing. And the same is true of cities. We’re constantly reconsidering what they’re meant to be. Right now, for example, we’re reconsidering the role of bicycles in American cities. This is also a necessary process: without it, the city would become useless.
The issue with digital technologies is that they get harder to change. They get locked in. The financial incentives for the people who run them become so precisely tuned to particular features that they become stuck. And it starts to feel like some stupid feature in a social network is an eternal thing, like the physics of a photon. That’s what’s so dangerous about digital systems: the sense that things can only be as they are.
We’ve talked a lot about your interest in taking a humanistic approach to technology. What does that mean? Does it involve letting technology change with us? Does it involve using technology to critically reflect on our own perceptions of how we see the world?
The funny thing about the term humanism is that it’s had a bad name in various times and places. But it’s strangely difficult to come up with vocabulary for the most basic and obvious things that are right in front of your nose.
Humanism means we’re willing to mystify people a little bit. We’re willing to say that there’s something we don’t understand, something extraordinary, something perhaps even a little supernatural about the way people experience what they’re doing. That they’re not just going about their behaviors as if they were automata. That there’s an internal life to people that is genuinely not something that’s ever been well reconciled with the kind of world we can explore with science and logic. It’s this other channel, this experiential channel.
So what I mean when I use the term humanism is the belief that people are special, that people are central, that everything else becomes absurd if you don’t believe in the specialness of people—especially everything we do with technologies for the benefit of these special things we call people.
For instance, the Saudi government recently granted citizenship to a female robot. That robot has been able to do things in public that Saudi women, actual human ones, are forbidden from doing. And that’s a great example of what happens when you don’t believe in real people. You end up in this world of make-believe where you don’t even notice how you’re screwing over real people.
So in order to be ethical, in order to be moral, in order to be decent, in order to be kind, in order to have a society that’s functional, in order to even tell if your technology is working well or not, you have to grant a specialness to that thing we call a person. And that’s what I mean by humanism.
Virtual Reality as a Medium
I love that. Now, I want to turn towards how these ideas relate to your work on virtual reality. As you know, I’ve never experienced virtual reality myself. But you’ve told me how much you’ve learned about human sensory perception from working on virtual reality. And I wonder whether an important part of the goal for you is not just to build better technology, but rather to learn more about what it means to be human and how we can embrace it more fully.
Virtual reality is a medium. It’s fundamentally no different than print or photography or cinema. It’s a technology we use to create a media experience.
I used to have a catchphrase many years ago when virtual reality was fresh in the 1980s: “Information is alienated experience.” In physics, we have this idea of “potential energy.” By lifting up a weight and sticking it on a shelf, you’re storing up potential energy. Then when you knock the weight off the shelf and it falls, you’ve released that potential energy.
In the same way, the only thing that’s ultimately real about a communications medium is the experience it can result in. That’s what I meant by the phrase “information is alienated experience.” You can create some bits, and if they are knocked off the shelf, some sort of experience will result. You don’t know exactly what kind of experience. You don’t know exactly where those bits will fall. But you know there’s some potential that’s been stored by putting them there.
So virtual reality is a trigger, rather than something that’s meaningful in and of itself.
Virtual reality is an engineered well of potential. It’s only meaningful to the degree that the person who experiences it has some bridge of intelligibility with whoever else might’ve been involved in creating it.
It’s important to understand that virtual reality is not a substitute for reality. And in fact I think virtual reality gets more and more beautiful and more and more worthwhile the less you respect it. That’s in general true for everything digital. The degree to which you disrespect digital stuff makes it more interesting and vivid, and increases its potential—the more you say, “This computer isn’t real; it’s just a pattern-generating heater that I only understand because I share culture with the people who set it in motion.”
And that’s very, very true for virtual reality. In the early days we used to sneak a flower or a mineral or something interesting in front of a person while they were wearing a headset. So when they took the headset off they’d have this little piece of the real world there. They would have their senses freshened. They would recognize the depth of reality, which you can take for granted and fail to notice on a day-to-day basis.
That seems like an interesting example of a way that technology can wake us up to certain things that are outside of the technology. And it reminds me of how people often describe virtual reality as a powerful empathy-inducing device.
The idea of virtual reality as an “empathy machine” is a bit counterintuitive, because it seems like empathy shouldn’t have anything to do with technology. If anything, empathy might seem anti-technology. But perhaps technology, by helping us reflect on how we see things, can also help us see things from the perspective of another person.
So far as I know and recall, I brought the term empathy into dialogues around virtual reality back in the 1980s. Back then I had been thinking about empathy a lot in relation to philosophy. I used to hang out with the psychedelic crowd, and the idea of empathy had a lot of currency in that world.
But you know, there’s this problem of nerds robbing us of our vocabulary for talking about ourselves. Take the term “consciousness,” for instance. It used to refer to something perhaps a little mysterious or very mysterious about why there’s an internal channel of experience. But an imperial conquest has taken place, and the term consciousness now just means one part of a program modeling another part of a program.
People talk about when AI programs will become “conscious” as a matter of fact—as if we actually know what that would mean, as if that’s an event that could be known. So consciousness has been lost to our vocabulary. I’ve started to use “experience” as an alternative to consciousness.
I think the question is whether empathy is a term that might suffer the same fate. Empathy has had a little bit of a glow to it over the years, and perhaps it’s losing that through too much usage in high-tech marketing. Empathy might become something that you can measure the degree of in the latest redesign of Facebook. Whereas it should be reserved as one of those terms for something we don’t understand, as something that is at the edge instead of at the center of our craft.
Another concept that’s similarly mysterious, at least to me, is “proprioception.”
It’s something that people are constantly experiencing yet never aware of. And it’s precisely the failure of technology to capture proprioception that makes us more aware of what we’re experiencing. What is proprioception, and how does it relate to virtual reality?
This is another area where the terminology fails us. Proprioception has a precise meaning, which is sensations that come to us through muscles and tendons and joints. For instance, it could refer to the self-awareness of how much a muscle is being stretched, which can then be interpreted by the brain to tell you about your pose if you’re a dancer. Proprioception can even tell you about your motion: your whole body can start to function like a giant accelerometer tree, a tree of little flaps that are moving as you’re moved.
Then there’s tactile perception, which involves sensations that reach the brain through sensors embedded in the skin. And that’s in turn extremely complex because there are a great variety of sensors in the skin and it’s quite possible we don’t even know of them all. We’re learning new things about them and how they work together all the time. They’re strange. For instance, some detect sharpness, while others only detect if a sharp thing is moving by, not if it’s stationary against you.
In virtual reality, we sometimes use the term “haptics” to cover both touch and feel—which is to say the senses from the skin, and the senses from muscles and tendons and joints. But the terms are loose. Different communities use them differently.
What’s difficult is that this is a world that science has only begun to intrude upon. There are people who will say you can learn a lot from a handshake. There’s a famous pop song from the early days of rock about how you can tell if somebody is true from their kiss. We know there are channels of communication that exist tactically. But you don’t even have to be touching somebody to be communicating. Whenever you’re talking to someone, there’s an elaborate communication channel between people that’s based on their head pose.
You’re constantly in motion. The head has to be in motion in order to perceive. If you put the head in a vise so that it can’t move, your visual acuity drops precipitously. The metaphor I always like to use with students is that the head is not like a Mr. Potato Head with little USB cameras and microphones stuck in it. Instead, the head is a spy submarine that’s out probing the environment. So when you’re with someone else, your two heads are probing, and you enter into a kind of dance that is almost always not undertaken consciously. And you become aware of how each other is perceiving in some way that we don’t really have the vocabulary to describe.
I wonder if one example of that kind of unconscious communication might be the stutter. I have a friend who has a little bit of a stutter. And it reveals her emotional state in a way that I find very powerful as a form of communication. It makes it hard for her to fully hide her emotions.
I’m just purely speculating, so don’t take this too seriously. But it’s not unreasonable to wonder if some of these things might not be flaws at all but rather adaptive. They might have been cherished in a deep evolutionary context.
One of the odd little research projects I’ve gotten involved in over the years is studying the relationship between olfaction, which is smell, and language. Going back in deep evolutionary time, you see the olfactory bulb in the centers of the brain that eventually turned into language centers. And if you look at the olfactory channel, there’s actually two of them: there’s the olfactory bulb, and then there’s a special channel that’s suppressed in humans but is present in most animals, which is meant for sensing other animals of that species. It senses things like defecation and sexual odors. And this pheromonic channel is a completely separate pathway.
A lot of animals have something called a Jacobson’s organ in their mouth where air is taken in. You’ll see these animals do a particular kind of rapid inbreath to sample air for this channel. Humans apparently have a vestigial Jacobson’s organ. We don’t use it but its traces might manifest in some way—it might be why we hiccup, for instance.
Are there people who lack language in whom this Jacobson’s organ becomes less vestigial or more active?
Well, we don’t know.
Language is combinatorial. You put words together, you get different meanings. And of all the core sensory modalities, the only one that’s combinatorial is smell. Because you have combinations of odors that create a subjective smell. Whereas the other senses are more spectral. You can see a color that’s somewhere between one blue and another blue, but you can’t really sense a smell that’s somewhere between one odor and another odor because smell is combinatorial. So smell, like language, is combinatorial. And that means that smell may have served as a kind of precedent for language.
But to get back to the idea of the expressive stutter: it’s conceivable to me that the stutter, like laughter, might have been some sort of social signal that long preceded language. And that we might be in the process of repressing it, very much as we repressed the Jacobson’s organ and who knows what else.
The modern human species really does focus on a certain kind of vision that’s commensurate with hand-eye manipulation and with language. And we tend to repress other behaviors and sensory channels that might still be with us. Maybe stuttering is a vestigial behavior that is related to empathy—some sort of way that people communicated their emotional state.
It’s interesting when you think about dividing these more mysterious senses from the better understood ones. What is it that makes senses like vision or touch less mysterious? It’s that we can both look at the same thing and think to ourselves, “We had the same experience of that.” Or we can both run our hands over the same surface and feel more or less the same thing.
But then you start getting into other sensations. Let’s say we both kissed the same person—would it actually be the same? Or, going deeper, could I ever feel what the inside of your stomach feels like? Could I know what it is to experience the back pain that you feel? Could I as a man know what it is to undergo labor?
That kind of empathy is not accessible to us with current technologies. But I’ve always thought that modern dance—something that you and I share a passion for—captures a bit of that quality. What I love most is watching dances where I feel like I’m getting inside the subjective physical experience of the dancer. That’s really what sets dance apart from other art forms for me.
Yeah, that’s beautifully said. I think we really are speaking at the edge of mystery here. There’s a way that people can experience something of another person’s subjectivity. But obviously it would be absurd to think that such an experience would be precise. It’s not. And yet it’s also not meaningless. It’s neither everything nor nothing. It’s somewhere in between.
And we don’t quite know how to talk about things that are somewhere in between. We’re used to metaphors that come from the digital world: everything or nothing, one or zero. Either I understand you and I can say that I’ve gotten this information from you—or I don’t. But there’s this other in-between thing—it’s almost as if we can live a little bit through each other, but just a little bit. Like there are faint traces of us in each other. It’s a wonderful quality of life that’s difficult to describe and I don’t think is well addressed by any of our metaphors from science or math or technology. It’s this other domain, and it makes me sad sometimes that people growing up in technical culture don’t appreciate it.
One story of yours that really amazed me was how in virtual reality, you can attach a tail to someone and they immediately know how to control it. I wonder if that relates to the inbetweenness you’re describing.
The fact that you can take on different body plans and still control your body is one of the most startling and surprising results of virtual-reality research. It’s probably the most significant scientific discovery to come out of virtual-reality research, actually.
It was known before that you could alter the body to a degree, because of phantom limb research and other related investigations. But the notion that you could radically reorganize the body and the brain could still control it was not something that could have been tested before. And it is remarkable. One thing that’s probably going on is that the brain remembers body forms that our ancestors inhabited, because as far as the brain is concerned evolution is a slow, gradual process. So the brain remembers what it was like to have a tail.
There have been a few different experiments with putting a tail on a person in virtual reality. But the best, most rigorous work came out of Mel Slater’s lab at University College London. That particular tail was a really good tail. It was a long tail. And the task was to get it to whip around in front of you to hit a target. So it was a pretty non-trivial bit of athleticism with your tail. And people can just do it. I mean, it’s natural. Everybody’s brain knows how to run a tail.
The more striking experiments involved changing people into completely different creatures with different numbers of limbs, or with limbs attached strangely. And there you start to see a kind of jigsaw puzzle, where there are some body designs that brains can control and some that it seems that they can’t. What we’re unveiling is the brain’s own cartography. We’re discovering what world the brain thinks it’s inhabiting, or what body it thinks it’s part of. And the body that your brain thinks it’s part of is not just your body at the moment—it’s a multi-million-year stream of changing bodies.
Love, Sex, and Guessing the Number of Jelly Beans in a Jar
That makes me think of how this all relates to love and sexuality.
How do we know that something is a part of us? Usually because we can control it—we can move it, we get feedback from it, our brain is programmed to assimilate it to us in some way. Now, when it comes to love, we obviously have huge effects on the people that we love and who love us. But this research makes me wonder whether love might let us actually experience another person as a part of ourselves.
I have thought about this on many levels. To speak in the simplest descriptive terms possible: sex between people tends to be a longer and more elaborate affair than what a lot of animals do. Sex between animals tends to be a bit quicker, with less variation.
This feature of sex ties into a remarkable feature of the human species, which I think is our most precious quality: neoteny. Neoteny means we have very long childhoods and we’re helpless for longer than the youth of other species. As a result, we have both the need and the opportunity to take in learning from our parents. And that creates this cross-generational body of knowledge and experience and stories that becomes our culture and becomes our wisdom.
Part of neoteny is that the parents have to be more committed to their children than is typical for parents of other species. They have to be able to work with each other. So I think part of what’s going on in sex between people is not just pair bonding, which you see in other species like penguins, but a kind of learning to be each other. We’re learning a kind of a body empathy that then helps us remain connected for the long and difficult process of raising kids over many years.
I remember when I was young, and children and reproduction and families seemed like the last thing you wanted to think about when you thought about sex. People want to think about sex as this separate force, as this cool drive that powers everything, as this gasoline that fuels the psyche. But actually the whole cycle through which we reproduce is so much more interesting. I’m saying that now from the perspective of a middle-aged parent, so maybe my younger self would think I’m full of shit. But raising children is about ten thousand times more involving and more costly and more rewarding than sex or romance. You don’t realize that at first.
It seems like there are a lot of social trends these days that help box that realization out.
I really want to avoid falling into the fallacy of “the kids these days.” But it almost seems like romance and sexuality have become reflections of digital designs rather than reflections of our cultural and biological heritage—which is something so much richer and so much more profound. Dating apps turn people into these commodities that you swipe left or right on. Everybody’s caught up in this stupid rush for optimization that makes no sense at all. You have this false rationality about dating where people are supposed to analyze who’s right for them, and are supposed to follow protocols for getting to know each other, which seems weirdly bureaucratic from what I see. It seems strangely like a job market.
You and I are big believers that markets are important for honoring people’s work and treating them like adults. And that a problem with the digital age is that it has removed money from much of the system, which has the effect of devaluing and infantilizing people.
But there’s a limit to that line of thinking. Because romance actually doesn’t seem to be well modeled by a market, despite the fact that everyone seems to be modeling romance by markets at the moment. If you approach every date as a five-minute transaction to evaluate someone, you’ve sort of vitiated the whole point, which is that through the process of getting into someone else’s head you determine whether it’s a head you want to be in.
So how do you reconcile those things? How do you believe in the importance of markets as a way of respecting people’s work and yet at the same time remain skeptical of the market model of romance?
I like to use as a starting point a demonstration that’s often done on the first day of business school. They have everyone in the class guess how many jelly beans there are in a jar. And the answer is often pretty good. The reason the demonstration is done is because it’s supposed to indicate a kind of a wisdom in the overall community that no single person might have. The idea is that a market might get at that collective wisdom.
But the interesting thing about the jelly beans in the jar is that it only works if each person starts as a real individual who is distinct. If you have people who can’t actually look at the jelly beans in the jar directly, but can only look at Facebook postings about the jelly beans in the jar—and if a lot of those postings have been written by bots—then it doesn’t work. If everybody is only perceiving the jar through some sort of intermediation that regiments them, then the effect won’t make sense anymore.
The same thing is true for voting in a democracy. This idea of getting people to work together in some kind of a collective abstraction only makes sense if each person is an individual in the first place. Otherwise the whole idea becomes ridiculous.
And there are limits. All you can do with these collective abstractions, like voting or marketplaces, is to work on one simple parameter at a time. So you can set a price for something, or you can choose a candidate, but you can’t, say, direct a movie. Whenever somebody tries to do these collectively created movies, they don’t turn out well.
Some people argue with me. They’ll say, “Look at Wikipedia.” But I feel that that makes my case. The example I like to use with Wikipedia is math. Before Wikipedia started, there was a whole movement of people making pages about math online. They became some of the most popular pages on the whole internet. When I was at Internet2, we used to have a contest that gave kids prizes for coming up with great websites. Each site had a distinct point of view and distinct authorship. It presented math or any other topic with a particular passion and a particular flavor.
Now when people want information about math they go to Wikipedia, and they get this dry, academic voice that completely excludes everybody. It has the precise function of turning math into a walled garden that only a few professional people can understand— just like Newton wanted it to be back in the day when mathematical notation was made terrible!
When you do things by committee, even if you think you’ll get something more inclusive, you inevitably end up with an institutional quality that excludes anybody who doesn’t fit the rigid model that brought the committee together. So these abstractions that bring people together, like markets, voting, and perhaps the internet, they only work for single-parameter investigations. And they only work if each person is a genuine individual with their own perspectives, their own earned knowledge of the world, their own legitimate separation from each other.
How do romance and sexuality fit into that?
Romance and sexuality are much closer to writing a movie together. They’re creative, not single-parameter. Now if we were just mealybugs, then sexuality would be more like voting for a candidate or setting a price, where we would just share our genes and that would be that. But since we have this high degree of neoteny, we must create together in order to procreate. And so the collective abstraction model fails and it turns into this horrible, bureaucratic, lifeless, scentless form of connection.
This brings us back to the specialness of people and to humanism. I mentioned at the start of our conversation that sometimes the term humanism is not held in high regard. There are circles, especially in Europe, where there’s the feeling that humanism represents a degraded form of individuality. And then there are religious people who have the same problem with humanism. They feel that in order to be a person, you should have a belief in something that transcends reason and what you can sense. That you should believe in some kind of higher spirit and calling and tradition. And so the humanist is their bogeyman as well.
But those two ways of disliking humanism are actually pretty similar to each other. They’re both saying that treating society as a channel to being yourself and being free and being an individual isn’t adequate. And I think it can be and it should be. I think society should be something sacred and beautiful. But just as guessing the number of jelly beans in a jar only works if everybody is a real individual, an authentic society only works if the people in it are fully authentic individuals. It only works if people have invented themselves.