Talking to Machines

Yes, that is the inner workings of a Ferbie you’re seeing up there— creepy, right!? The picture goes along with Part 2 of Talking to Machines, an episode of my favorite podcast, Jad Abumrad and Robert Krulwich’s Radiolab.

I’m not going to do much interpreting/reporting/storytelling here because Radiolab doesn’t need anything from me to be entertaining, informative, and thought-provoking. (Also, I’m jet-lagged and quickly crashing…)

If you care to listen, here are just a few quick things that I found particularly salient to topics of Automatravel:

1. When creating Ferbie, Caleb Chung drew upon his experience as a street Pantomime to help codify Ferbie’s facial expressions. As a pantomime, performing in a deliberately stylized and— I would argue— somewhat inhuman (or cartoonish) manner, Chung learns a complex language of simplistic gestural storytelling, which he then transfers into his inanimate creation, making it not just animate, but possibly empathetic. So just to plot that all into a simplified trajectory, that’s a human striving towards the inhuman, informing him how to lead the inhuman towards the human.

2. Sherry Turkle uses two terms that feel self-explanatory but could use some deeper delving: Relationships of Projection and Relationships of Engagement. She claims we relate with a doll— an object that is humanesque but clearly inanimate— through projection, giving it a story that reflects our own experience. By contrast, we relate with a robot through active engagement because it appears to be creating its own story. While I understand the theoretical framework of this distinction, I’m not sure the two sides can be so neatly delineated in practice. In fact, it wouldn’t be too much to say that one of the most powerful forces driving my Watson is rooted in my suspicion that when audiences think they are engaging, they’re still doing an awful lot of projecting.

3. (though this might be more of a 2b.) Throughout the episode, did you notice the numerous examples of peoples’ intellectual understanding of their experience ultimately seeming at odds with their in-the-moment feeling of the experience— a sudden knowledge that they have engaged in a sort of “inanimate disavowal," an unconscious: “I know very well that this thing isn’t alive, but…"

Those are three (possibly two and a half) things I noticed enough to type out, but I’m hardly scratching the surface here. So. If anyone reading this winds up listening to the podcast, making it through the hour, and subsequently finding themselves as intrigued by it all as I am, I’d love to hear you. Post a comment below!