Å·±¦ÓéÀÖ

Isaac Asimov is less a writer of books than a puzzle creator. Each of the linked short stories in I, Robot poses a problem, like an SAT logic problem, and works its way to a (usually) clever answer. Foundation, his most famous work, is the same thing. Asimov has less in common with the other Big Three mid-20th century science fiction writers than he does with Encyclopedia Brown.

But this is the book that invented the Three Laws of Robotics, which are so famous that basically no one has ever talked about robots again without dealing with them. They've impacted fiction - I'm revisiting this book as I watch AMC's robot drama Humans (it's okay), which refers to them frequently - and they've impacted reality: Google has had to try to in its self-driving cars. Here they are:

The Three Laws of Robotics

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey orders given it by human beings except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

What they don't deal with is what we call

A trolley headed down the tracks is about to hit and kill five people. If you pull a lever, it will switch onto another track, killing one person. Do you:

a) take direct action to kill that one person?
b) do nothing, allowing five to die?
c) my favorite,

The answer is technically obvious. The question is, can we possibly ever be comfortable getting into a car that's prepared to make that decision for us? Even if we know its judgment is accurate - and Google's cars are already much better at driving than we are - does it go against some basic factor of humanity to abdicate life or death?

I, Robot doesn't get into that as deeply as I'd like - it's presented and shrugged away in Evidence. Really, Asimov uses them mostly as a framework against which to throw a bunch of his puzzles. So this is an uneven collection: some stories get into the really interesting questions about what it will mean for robots to enter out lives, and some are just riddles. It's all pretty engaging, and some of it is great.

Robbie
Sentimental story about a little girl who loves her robot. Studies show that people build easily.

Runaround
Starting a trend that will shortly become boring, Asimov sets up a situation where his three rules cause an unforeseen conundrum - in this case, a robot running around in endless circles. The solution involves invoking Rule #1. This is not very exciting.

Reason
One of my favorites, about faith and evidence: a robot takes the available evidence and comes to the logical conclusion that the ship's engine is God and humans are deeply inferior. You're like okay, how will Asimov talk his way to out of this? How can they prove that they're really the robot's creator? Humans realize, after much fluster, that (view spoiler)[actually who cares. (hide spoiler)]

Catch that rabbit
A new kind of robot that controls several other robots goes wrong, why, who cares, this one is pretty dumb.

Liar!
A robot who can read minds may be lying. Turns out (view spoiler)[it's the only way for him to avoid harming people. (hide spoiler)] Here we learn that women think about love and men think about careers, as Asimov follows the First Rule of Science Fiction: never understand women.

Little lost robot
A batch of robots has been secretly made with altered first laws: while they still can't harm humans, they can now stand by and allow humans to be harmed. One of the altered robots is hiding. How can he be picked out of a crowd? Parallels to slavery are pronounced as humans call the robots "boy," which succeeds in making you uncomfortable; they'll continue doing this but Asimov never really digs into the idea. The story is one of his better ones, although the puzzle solution is just okay.

Escape!
Robots help us invent light speed travel, with unforeseen and unconvincing side effects that cause problems for the robots working on it: (view spoiler)[it somehow requires us to temporarily die. (hide spoiler)] Forgettable.

Evidence
A man running for office is suspected of being a robot. This is the first appearance of robots that look like humans, and also the story in which a version of the trolley problem is very briefly dealt with. I liked this one a lot.

The Evitable Conflict
Somehow, robot-directed industry is making mistakes. Why? (view spoiler)[Robots are framing anti-robot agents - creating mistakes that get blamed on them - to make them lose their jobs, because anti-robot agents are acting against the best interests of humans. (hide spoiler)] This story deals with the singularity, the moment when robot judgment becomes better than ours. Asimov seems unconcerned, as am I. This is another one of the better stories.