Tuesday, March 17, 2009

Doggerel #177: "Certainty"


Welcome back to "Doggerel," where I ramble on about words and phrases that are misused, abused, or just plain meaningless.

Various woos like to ask us if we're certain about scientific knowledge, or about our doubts on "supernatural" topics. The problem with the question is that it deals with a fundamental misunderstanding about science and skepticism. Certainty isn't for us (unless you're a mathematician).

Scientifically minded people work with confidence levels. And we generally work by falsifying the null hypothesis, not by developing an absolute proof of a theory. It's also rare for us to declare anything impossible. As such, I tend to see certainty as the province of woos, not skeptics.

Scientific hypotheses must be falsifiable: That means that if they're wrong, they can be shown to be wrong with the right evidence, and modified or abandoned to account for that evidence. A skeptic has to know what would prove him wrong. Usually, when I ask woos what it'd take to get them to admit they're wrong, they refuse to consider the possibility. Instead of only going for certain things, we go with whatever appears to have the highest probability and plausibility.

We also recognize that we're mere mortals. We can't attain special exceptions to our limitations. When you think you have certainty, it can often pave the way to barbarism. Just take a good look at religious fanatics and cults of personality. Having a little doubt in everything is a good thing. That's how we've made such great progress in expanding the knowledge of our civilization.

6 comments:

Akusai said...

Along these lines, I hate it when someone in a perceived position of power or scientific authority (like the current Canadian Minister of Science, I believe his title is, who said something like this recently) who say things like "We can't think that science knows everything."

I mean, it's one thing to understand that on your own and mean "We need to make the public at large understand that science isn't infallible and perfect and doesn't claim to be." But far too often, the powerful people who drop that line don't themselves understand science and are saying it in sort of a populist "Us vs. Those Damn Overconfident Scientists" sort of way, and so are creating a strawman of science and thus perpetuating this ridiculous idea that any scientist anywhere thinks that science, as it currently is, knows everything there is to know.

Dunc said...

What really annoys me about that is when people make the false inference that the statement "science doesn't know everything" implies that "science doesn't know anything", or "science cannot be relied on for anything." (Usually, when you see people crowing about how "science doesn't know everything", they're actually shooting for one of those false inferences. Sane people rarely bother making statements that obvious.)

We may not be absolutely, 100% right about gravity. However, it's a very safe bet that if you let go of that hammer it's going to fall, so make sure it's not over your foot.

Akusai said...

I think Stephen Jay Gould said something like "Tomorrow, apples may very well start to fall up. That doesn't mean we should teach it in science classes."

Lifewish said...

We also recognize that we're mere mortals. We can't attain special exceptions to our limitations. When you think you have certainty, it can often pave the way to barbarism. Just take a good look at religious fanatics and cults of personality. Having a little doubt in everything is a good thing. That's how we've made such great progress in expanding the knowledge of our civilization.

I feel the urge to channel JS Mill. There's actually two different levels of certainty at stake this paragraph:

1) Certain enough to support a position in debate.
2) Certain enough to support a position in debate regardless of what the other side says.
3) Certain enough to refuse to allow a debate in the first place.

The important thing from civilisation's perspective is that we avoid type 3 certainty - the certainty of censorship. As long as we have freedom of speech, a certain amount of fundie-ish type 2 certainty is bearable.

Obviously things move faster when we try to stick to type 1 certainty, but it's not strictly necessary.

Valhar2000 said...

Lifewish, though on the face of it, I agree with your exhortation to stick to "type 1 certainty", this can (and often is) misinterpreted to support the kind of he said/she said relativist attitude that does so much to polute science standards in american classrooms.

One can also rightfully expect a debater to attempt to make sure that the arguments he brings have not been debunked previously 100 times; when one does not display such courtesy to the other side, one should not be surprised to be unceremoniously dismissed.

Creationists in particular would do well to take this into account.

Wikinite said...

lifewish,

I would tentatively agree that your "type 3 certainty" is not desirable, however I would put a caveat on it. Once a position has been thoroughly debated, say in the case of cdesign, then the proponents of a position need to produce new evidence, arguments, etc.

Not accepting "debates" from positions that have been thoroughly debunked and destroyed is not exclusionary, it's progressive.