The Stone Soup Gambit

Recently, I watched Jim Henson's The Storyteller. In one of the episodes, the titular character manages to con his way into getting a free meal out of a cook by making "stone soup". He puts the stone in a pot of water and boils it. As he takes a sip to "test" the flavor, he asks for a little salt, and the cook obliges. He repeats, getting more and more extra ingredients. After putting together a tasty stew this way, the cook is delighted and dumbfounded that someone could make delicious soup from "just" a stone. I think I can use this little tale to start a meme: "The Stone Soup Gambit."

When we argue for experimental controls and present alternative causes, it often feels like we're trying to convince someone that the stone contributed nothing to the flavor. There's rarely a shortage of alternative explanations: other ingredients in the soup. Essentially, what we want is a (double-)blind taste test between a pot of boiled water and stone soup, hold ALL the extra ingredients, to see if it's possible to consistently tell the difference (at greater than chance levels).

Instead, we usually get complaints that amount to saying it's impossible for the alternatives/ingredients to affect the outcome, therefore we can afford to be sloppy in our methodology. These efforts fall rather flat with skeptics, and the repetition is a leading cause of our frustration. Often, we get a few token concessions, controlling for whatever the field considers the token objection (see-through Zener cards, weather balloons, etcetera) and they expect us to be satisfied with the removal of so few alternatives.

It's a very simple concept. I don't see why applying such a basic principle invokes so much rage on our opponents' side. I certainly can't blame my skeptical friends for getting frustrated dealing with robotic, repetitious replies.