There's a story that I read at some point several years ago, but I can't remember when or where. My best guess is that it was either in an out of print book of older SF stories (probably 30+ years old) or it was an online post someplace (probably less than 10 years old).

The premise, as best as I can remember it, is that galactic civilizations have decided (at some point in the distant past) that humanity is too dangerous and have effectively quarantined our solar system. But a couple of aliens are wondering just what could possibly be so scary about us. So they decide to build a laboratory environment that would be absolutely impossible to escape from and abduct one single human being and stick him in it, in order to observe him.

There are multiple levels of security ensuring that there's no way to possibly escape. But it turns out that the thing that distinguishes humans from all other intelligent beings in the galaxy is that they just can't comprehend the concept of something being "impossible". So, despite the fact that there's no chance of success, the human insists on repeatedly trying time and again to get out.

Of course, he eventually succeeds. And steals the aliens' ship. And they realize that they've made a terrible, terrible mistake.

Does anyone else remember this story? One thing I do know is that it's not "With Friends Like These..." by Alan Dean Foster (I checked that one already).

  • 1
    Possibly "Danger—Human!" as described in this question and this one? – Daniel Roseman Oct 12 '18 at 15:24
  • 1
    That's it! I guess that would make my question a duplicate -- should I delete it? – Admiral Jota Oct 12 '18 at 15:35
  • 3
    I've marked it as a duplicate. I wouldn't delete the question because duplicates can help people who are searching for the story. The duplicates all have slightly different details about the story so together they increase the target for searches to aim at. – John Rennie Oct 12 '18 at 16:58
  • @JohnRennie - I've been given to understand that story-identification questions aren't generally marked as dupe here until/unless there's an accepted answer that points to the same story. Have I misunderstood? – Jeff Zeitlin Oct 12 '18 at 18:24
  • 1
    @DanielRoseman - It's not possible now, as the question has been marked dupe, but when the querent confirmed it, you might have done well to make the comment a formal answer. – Jeff Zeitlin Oct 12 '18 at 18:35