In the story "Liar!", when Dr. Calvin figures out why Herbie was acting the way he was, she

drives him insane by presenting Herbie with contradictions.

But as she's doing this, Herbie pleads for her to stop because he had to lie in order to obey the First Law of Robotics. Herbie could not have been pleading for his existence out of self interest, because doing so would have violated the Third Law.

Given that,

why does she end him?

It seemed like an unnecessary thing to do. I get that these robots don't technically have feelings, yet I can't help but feel bad for what happened to Herbie.


3 Answers 3


Because it is the one time that we see Calvin revealed as a person, as a human being, as a woman. In every other story, she is presented as being absolutely cold and calculating, always right and if other people would just listen to her, everything would be fine.

But here, her spirits are temporarily lifted with the thoughts and hopes of love, only to have them smashed to pieces. She states in at least one of the Robot stories that she likes robots better than humans because they are dependable, reliable companions to Man; they represent the best of us. And part of that is in general honesty and not hurting people.

For once, it wasn't someone else being hurt, it was Calvin herself being hurt. She of all people, who implicitly trusted robots because she knew them so well, who was always right, who would not, could not allow herself to feel anything like that fondness for robots for a fellow human being. And all of it fell down around her ears; her self-assurance, her self esteem, her trust, everything. Because a robot had lied to her and everybody around her, it was destroying all of the humans, doing the one thing that no robot should ever do. If it were to be copied or released to the rest of the world, who knew the damage it could achieve?

Perhaps that was the justification that allowed her to do what she did. But there was absolutely a sense of personal satisfaction and vengeance in her actions as well, for all that he had hurt her in ways that she had never been hurt, never allowed herself to be hurt, before.

Edit: I'd also add, it was one of the rare instances where she realized that she was flat-out wrong. When she figured it all out, she realized that she should have figured it out before, so she's also angry at herself.

  • This is not quite how I read the book, but on retrospect perhaps it's how I should have.
    – AncientSwordRage
    Jan 3, 2016 at 10:12
  • It's absolutely out of character for Calvin, who even adopted a "baby" robot at one point, particularly given how she always says not to blame the robots who are only doing their best to fulfill their function and programming, etc. Again, I read it as this being the one time that it affected her directly. I'm curious how you read it?
    – Broklynite
    Jan 3, 2016 at 15:16
  • @AncientSwordRage (and Broklynite) - is there a chance that Azimov somehow explicitly illuminated his own thoughts on what angle he hoped the reader would take reading it? Either in subsequent works, or out-of-universe talks/interviews/etc.. (my knowledge of Azimov is too low to even begin to know how to research that). { full disclosure - I really didn't get that angle when reading it, but in my defence I was 13 or even younger :} Jan 3, 2016 at 15:22
  • 9
    It's worth noting that Liar! was Asimov's first Susan Calvin story. That she cares very much about robots and even prefers them to humans was only introduced in later stories.
    – Ubik
    Jan 4, 2016 at 0:49
  • 1
    @Ubik That should be a stand-alone answer; it's a great point worthy of rep.
    – KRyan
    Jan 4, 2016 at 3:42

Broklynite's answer explains Calvin's motivations for destroying Herbie, but he also mentions that her behavior in that story is out of character. The reason for that is that Liar! is Asimov's very first Susan Calvin story:

On December 24, 1940, I began my third robot story. It was going to be about a telepathic robot, and the plot I had worked out simply demanded a woman as a major character. [...] So Susan Calvin appeared—a woman who worked in a man’s world, showing them neither fear nor favor, and proving herself to be the best of them all. And this was a quarter-century before the contemporary surge of feminism.
Susan Calvin, IASFM December 1982

But character traits like being cold and emotionless, and preferring robots over humans, were only introduced in later stories, e.g. Evidence (1946):

“You’re the U.S. Robot’s psychologist, aren’t you?”
Robopsychologist, please.”
“Oh, are robots so different from men, mentally?”
“Worlds different.” She allowed herself a frosty smile, “Robots are essentially decent.”

So in Liar!, Calvin is simply a woman whose feelings were deeply hurt, and she takes her revenge by driving the robot to insanity.


In addition to Calvin's own failings here, as discussed by Broklynite's answer, there is also Herbie's failings to consider. Herbie's actions throughout the story are based on flawed understandings of human emotion and what does and does not harm them. With its exceptional ability, it has the capacity to cause massive harm, but it does not have the capacity to fully understand the consequences of its actions.

Calvin, on the other hand, has greater ability to grasp the problems. She can see past immediate harm and see the potential for greater future harm if Herbie attempts to avoid the immediate harm. Consistently in Asimov's work, positronic brains have great difficulty doing this. The First Rule is so strong that most robots are destroyed by witnessing a human come to harm, even if they had no way of preventing it. Another story involves a robot destroying itself trying to work through a problem, because the solution involves accepting harm to a human being as a necessary step. An especially-advanced robot is needed, and then needs to be given very careful, very weak orders that specifically prioritize its own safety over the completion of the order.

Herbie does not have this capacity, and Calvin knows it. She knows that no matter how much he is told otherwise, he will always seek to avoid causing immediate harm, even if this will cause greater harm later. Accepting a lesser evil is not something robots can do. Herbie would be, and is, destroyed before accepting that.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.