The quick and dirty version of the concept of falsifiability is if you're going to make a claim, or present a theory/hypothesis... you need to know how to tell if you're wrong. That sounds like it should be simple and straightforward but it just flies right over the heads of an astounding number of people on a regular basis. Religious people in particular, but it's not just isolated there. I know a disturbing number of people who think of themselves as rational scientifically minded individuals and are actually quite skilled and knowledgeable in the sciences, who simply do not properly appreciate why falsifiability is so important. They reach the point of knowing they have to be able to tell if they're wrong so they can slap a "right" or "wrong" label on an idea and stop there. It's a little more than that.
See, the primary way we figure out if we're wrong about something is we use it to make a prediction, then test it. When we hit a situation where something is unfalsifiable it's generally for one of two reasons.
- It can be used to predict nothing.
- It can be used to predict anything.
No... that's useless.
In this context, "predict anything" means no matter what happens, no matter what outcome a test has, you can claim it was predicted by your "theory". To demonstrate why this is so spectacularly pointless, an example of such a "theory":
Newton Vs. The Matter Directing Elves
Let's say we have two competing "theories" about how stuff moves around the universe... Newton's laws of motion, and the Matter Drecting Elves theory that says magic invisble elves are actually moving everything around.
Now, the former makes specific, testable predictions that everything is going to behave in a certain way if it is right. It tells us where the earth is going to be in an hour, where that asteroid over there is going to be in a year, where that bullet you just shot is going to go, etc...
The latter predicts that absolutely ANYTHING could happen. No possible observation is inconsistent with its claims. It can never, ever, ever, ever be proven wrong. It is totally unassaiable. Completely ironclad.
If tomorrow the moon started orbiting the earth in the opposite direction? Newton's laws are in trouble, but the Matter Directing Elves theory is going strong! They just decided to turn around because they were bored always going the same way or something. Elves theory wins!
If tomorrow everything in the universe just plain instantly STOPS? Newton's laws have had it. The Matter Directing Elves theory on the other hand is golden. They obviously got tired and/or went on strike over their work conditions. Elves theory wins!
So if the Matter Directing Elves theory is consistent with ALL possible observations and is NEVER proven wrong... why do we throw it out without giving it a second thought? Because it's worthless. It has exactly as much informational content as if you just said "Something will happen". That is a statement that is always right, and tells you absolutely nothing whatsoever. Pretending that just because something you said is never proven wrong makes it a valid explanation of how the universe operates is absurd.
The real bar is that it is never proven wrong when it could have been if it was.
If an idea is not falsifiable, it is useless. It tells us nothing. It is an illusion of understanding... sleight of hand, nothing more. You're taking something you have absolutely no clue about, slapping a title on it like "Matter Directing Elves"... then pretending you have explained it just because you assigned responsibility for it to a name you made up.
And that doesn't just apply to when we're formulating official scientific theories.
Someone tell me what observation would falsify the hypothesis "God exists".