> I'll hazard it has a lot more to do with the behaviour than you want to allow. To keep on-track here, how many people do you know who did *not* cry at the ending of Old Yeller? And (spoilers, barely) that's just a non-living representation of a real dog, too. Nobody cries because he thinks the actual dog was killed; we cry because the representation that we've come to accept as the dog - the *character* of the dog - dies. And in this case, it's not even a dog representation that anybody in the audience has ever had a personal interaction with, which seems to put it one step farther removed from reality than the robodog, and yet people still cry. Why? Because Yeller was a good dog, and he was loyal, and the characters in the film were attached to him, so therefore it was a sad situation. If they remade the movie with a pet robot instead of a gen-yoo-ine dog, I suggest (if we assume that the movie would still, you know, make any sense this way) it would have exactly the same effect on audiences.

This reminds me of a conversation I had recently with a co-worker. We were talking about movies, and the movie "Artificial Intelligence" came into the conversation.

"I hate that movie," she said. "It makes you feel like crap."

She didn't hate the movie because it was poorly made or had a bad story or a boring plotline. She hated it because she felt sorry for the main character and didn't like what happened to him. Even though (and this is no spoiler) he was "only a robot."