5X5 Episode 57

Skepticism 101 - Moving the Goalpost
S: This is the SGU 5x5 and our continuing series on Skepticism 101 and this time we are gonna cover the logical fallacy of moving the goalpost. This is more of a debating tactic or a tactic of denialism, really, than a logical fallacy, and it involves setting a criteria by which you will accept a proposition or a claim. And then, when those criteria are met, you change the criteria to make it even more difficult. You move back the goalpost, so that the criteria can never be met. So, you're rigging the game ahead of time to achieve the denial of a conclusion, an outcome that you want a priori.

R: We joke about moving the goalpost a lot when we talk about creationists because they often say: "Well, just show me a fossil that's, say, halfway between a bird and a lizard," or something. And every time a fossil comes up that shows them exactly what they're looking for, they demand to see something else. And it's the joke where we say: "One fossil comes up and it creates two more gaps in the fossil record."

S: And not just that. They also change the criteria, they move the goalpost by switching them. They say: "There is no fossil that's halfway between a bird and a reptile." Then we get Archeopteryx and they say: "Well that's, you can't prove that that's actually an ancestor or transition between birds and reptiles. It just happens to be half-reptile and half-bird but that doesn't prove that it actually was in the process of evolving into them." So they just completely reject the criteria they gave you ahead of time and now they change it to something else entirely.

B: Moving the goalpost is also referred to as "raising the bar". And in that vein the bar can also be lowered in some cases. If someone claims, for example, that a drug can prevent colds but you get one anyway, they can respond that your cold would have been worse if you hadn't taken that drug. So it's kind of a different take on moving the goalpost &mdash; raising the bar.

J: So, the response of moving the goalpost is actually an attempt to disqualify the previous criteria.

S: Yeah, it's saying: "I don't believe your claim because this evidence doesn't exist." Then, when you produce that evidence, they say: "OK, well, you have that evidence but you don't additionally have this other evidence over here." And every time you produce the evidence or meet the criteria they set, they just add more things. The point is they're never going to except your belief. Or either accept or reject the claim. They've already made up their mind and then they're backfilling in the criteria, rather than establishing criteria ahead of time without knowing what the results are and then living with the results, whatever they might be. That's the only way that honest science can proceed.

E: Can another example of moving the goalpost, Steve, be taken when, say, perhaps, you know, a person goes and has expectations of visiting, say, a psychic, or something to that effect? And they sit down, the psychic does the cold reading and effectively they are moving the goalpost by inferring, obviously by doing the cold reading, inferring something that will turn into a hit. And by kind of moving it from a negative and turning it into a positive. Is that another way of moving the goalpost?

S: In a broad sense, yes. In that moving the goalpost is a really just one form of having poorly defined criteria ahead of time. So when you're doing a scientific experiment, you have to decide ahead of time what the specific outcome will be that will determine success or failure. And you can't decide afterwards if you've met the criteria or not. You can't decide what that, where the goal should be or where the demarcation line should be. Because you can always decide where to set the cutoff, based upon what happened with the psychic or based upon the data that you get with the results. So you're just deciding: do I want this study to be positive or negative? Rather than saying, from some reasonable first principles: this is where the cutoff should be, and then collecting the data and seeing, you know, letting the chips fall where they may. So the broader concept here is you have to decide ahead of time, before you know what the data is, what the criteria should be for success or failure. If you decide afterwards, you're almost always gonna move the goalpost to meet your preconceptions. And then we're just validating our preexisting beliefs, not letting the evidence dictate those beliefs. I'm gonna finish this with one more recent and very typical example. This is pure moving the goalpost. Ten years ago the anti-vaccination movement said that thimerosal in vaccines was causing autism. By 2002 the thimerosal was removed from the childhood schedule in the United States. And they all said, and are on record of saying: by 2005-2006, you know, four years after the thimerosal was removed from vaccines, we will see autism rates decline. Well, they didn't decline by 2006. They said: well, it will by&mdash; they moved the goalpost to 2007, and then 2008, and now 2009 and beyond. That was almost a literal example of moving the criteria back farther, and farther, and farther, when the data did not fit their preconceptions. And if you do that, you can always confirm what you want to believe. And that's not how good science works.