The Skittles Argument

Here is a bowl of Skittles. If you don't like Skittles, substitute chocolate M&M's.

Hundreds of them.

Help yourself.

Oh. Yes.

Ten are poisoned. Fatal poison.

But dig in. Please. What are the odds?

You won't, of course.

The downside is too great and the upside is fleeting.

It's the reverse of that awful mess that is Pascal's Wager. This is the one where we should believe in God because the upside is infinitely wonderful and the downsides are minor inconveniences. (One flaw in the argument is that the upsides only happen when we are dead, so you have to believe in the after-life as well. Another is that Pascal is assuming that the way the Catholic Church says you should live is the way to heaven, especially the donations-to-the-Church bit. But you're not supposed to point that sort of thing out, as it is considered bad manners.)

There's a whole class of arguments like this. Where one of the choices has penalties or payoffs way over anything we are prepared to risk. Like death or permanent injury or disability.

Some people argue against nuclear power on this basis. Sure the odds of the reactor blowing up are minuscule, but then again, Chernobyl.

The Skittles Argument was recently used by women explaining why they refused to have anything to do with men. Because one might be an abusive ***hole and you can never tell. Until it's too late.

Now it's used by men to explain why they are steering clear of dating and marriage. They don't know if the Skittle is poisoned, but what you do know is that forty per cent of men who pick a Skittle end up paying alimony and child support.

One flaw with using the Skittles Argument is its assumption that we cannot identify the poisoned Skittle. However, people aren't Skittles: very often, we can spot the Poisoned Skittle, and even if we get it wrong, we don't lose much by getting it wrong.

I'll modify that a bit. Very often, a psychologically healthy person can spot a Poisoned Skittle.

Which leaves a lot of people who can't. Because they have faulty calibration and don't know what a Poisoned Skittle looks like (ask me how I know about that). Or because they have faulty goals, which the Bad Boys / Bad Girls meet so well.

A variation of this flaw is assuming that everyone will react to the Poison the same way. Maybe it only affects people with existing conditions, and most regular people will be asymptomatic. Most people with existing conditions know they have them because the Doctor has already told them. There is no test for being asymptomatic, so everyone else has to assume they run a risk of feeling rough for a few days. Which is probably enough to say that even the fabulous taste of Skittles isn't worth the risk of finding out you're symptomatic.

So when what's at stake is everyday contact with everyday Skittles? That's as if we had a bowl of things that looked like Skittles, but most were just gum, and only a few were real Skittles, and who would grab a handful then?