June 26, 2009
One of my favorite psychological/theoretical/technological concepts out there is the Uncanny Valley.
The idea of the Uncanny Valley was developed specifically to talk about human / gaming and human/robot interfaces. Remember the Final Fantasy movie from about 10 years ago? At the time, it had the most photo-realistic CG characters ever. But they weren’t quite real. In fact, the movie was a bit of a flop and some people think one of the reasons was that the characters were just plain creepy — they were so obviously human, and so obviously not. Same with something like the motion-capture CGI The Polar Express. I think this is one of the reasons Pixar makes its people so cartoonish — because we actually like the characters better if they don’t look real at all than if they looked almost, but not quite, human.
The one-season wonder SF TV series Earth 2 had a great example of this. (One of many great ideas the show had that were kind of ruined in the shallow execution. The show had a lot of potential and if you want a rundown of why I think it failed (admittedly I decided this after only watching about 3 episodes) I can do that later.) The expedition leader’s kid suffered from a variety of ailments and weaknesses due to being the latest of multiple generations to be born in space (leading people to believe that humans need an earth, leading them to Earth 2 and a season’s worth of adventures). To help him walk, he has a mechanical exoskeleton. I absolutely hated that kid. First off, I generally dislike kid characters, especially when their primary purpose seems to be saccharine sentimentality, which this show had in spades. (It even had a cute puppy-eyed alien that I just wanted to take a hammer to.) But that exoskeleton: Uly didn’t look human. He didn’t move like a human. His strides were jerky, the servos whined. I mean, he was human. He looked human. But not quite. Not really. There was something subtly, horribly wrong with him. Which of course was the point, that humans raised off-world weren’t quite human. Unfortunately I didn’t think the show quite realized what it was doing with that thematically, or that they were illustrating a specific concept so well.
I think the concept can carry beyond the nebulous line between human and inhuman in artificial interface. I think it can also explain the uneasiness we feel when we confront anything that treads the line between natural and unnatural. Take this video for example: The Big Dog.
This machine achieves agility because its limbs are based on the movement of actual, natural limbs. But it’s clearly a machine. And it’s so damn creepy. I absolutely love this video, and I hate it. There’s something hideous about this device. This thing kind of looks alive. It kind of looks like an animal. But I get this simultaneous thought process: 1) No animal really looks like that, and 2) No machine should look that alive. It’s getting both thoughts at once that lands this machine in the Uncanny Valley.
The Uncanny Valley deals with liminality, a thing that is neither one nor another, or is both at the same time, that exists between two different states. I’ve often thought of urban fantasy type monsters as liminal as well — a werewolf is both human and monstrous. I think it’s part of why so many filmmakers developed the humanoid werewolf — the rubber-suit-looking werewolf that is monstrous but identifiably human. A vampire is human, but not in some very important ways. It’s not the Uncanny Valley as the concept is generally conceived and discussed. But it’s a useful way to talk about liminality and how these creatures can be so intriguing and horrifying at the same time.