sans comic sans

(With apologies to Back to the Future‘s Hill Valley.)

Briefly, the uncanny valley is a theory from robotics that predicts we will have positive responses to robots if they are either a) indistinguishable from humans or b) obviously not human. However, there is a “valley” of uncanny-ness as a robot takes on more human characteristics. Near-human robots tend to create feelings of anxiety and mistrust.

Applied to the world of animated movies, this theory is one way to explain why Bee Movie looks cute and Beowulf looks creepy.

Beowulf’s animation style (called “performance capture”) lives in the uncanny valley, if early reviews are any indication. From the Times review:

Neither wholly animation nor live action, it is a sophisticated visual technique, and true believers see it as the future of movies, though really the most interesting thing about it is that it’s not intrinsically interesting.

To be honest, I don’t yet see the point of performance capture, particularly given how ugly it renders realistic-looking human forms. Although the human faces and especially the eyes in “Beowulf” look somewhat less creepy than they did in “The Polar Express,” Mr. Zemeckis’s first experiment with performance capture, they still have neither the spark of true life nor that of an artist’s unfettered imagination. The face of Mr. Hopkins’s king resembles the actor’s in broad outline, in the shape and curve of his physiognomy. But it has none of the minute trembling and shuddering that define and enliven — actually animate — the discrete spaces separating the nose, eyes and mouth. You see the cladding but not the soul.

I find the theory fascinating. It’s a useful way to think about why we react the way we do to things that appear lifelike. It’s also a way to connect objects and ideas that, on the surface at least, appear to have little in common: AIBO the robotic dog, the sculpture of Ron Mueck, Frankenstein’s monster, and of course, a naked Angelina Jolie.