Over the last year, I’ve had numerous blogospheric encounters with the conservative climate “skeptic” Anthony Watts, the author of WattsUpWithThat. In the process, I’ve been particularly struck by how Watts handles inconvenient evidence.

Twice now, I’ve seen Watts make a mistake, and then seem to rationalize it, rather than simply correct it. I’ve also seen Watts shift the goalposts, refusing to accept inconvenient evidence even after saying he would do so.

What’s up with that?

Look: We all make mistakes. And we all adopt beliefs that later turn out to be incorrect.  There’s nothing wrong with that per se; it’s actually quite natural. What really matters is what we do after we’re proven wrong. So let’s see what Watts does:

Research on Astroturfing. A while back, I introduced the blogosphere to a social science study on online anti-global warming astroturfing. Watts then leapt in, accusing the researchers of having “setup fake websites to gather fake data.”

I have no idea how Watts got his idea about the researchers setting up fake public websites. But it was incorrect. The researchers were not creating fake sites that could deceive unknowing web surfers. They were showing sites to research subjects in a lab setting–and of course, debriefing them afterwards, in line with standard procedures.

But the sites were not actually online, live for the world to view.

I pointed this out, and noted with some amusement that Watts and his commenters had been slamming the study based on a basic misconception–barking up the wrong tree repeatedly, until somebody on the thread bothered to read the actual paper. 

Watts then responded by further defending himself—implying it was the study authors’ fault that he had misinterpreted them, because they didn’t use the actual word “Intranet”:

Note the word “website”, which appears 56 times in the full paper. The word “Internet” appears once, in the bibliography, and the word “Intranet” does not appear in the paper at all. Why wouldn’t they mention that the study was conducted on a private Intranet and not on the World Wide Web?

Answer: because it is obvious, to anyone reading the study or familiar with such research, that this is a controlled experiment in a laboratory setting, in which research subjects are shown something on a computer screen but not pointed to an actual live URL.

Watts then quickly found another reason to bash the study:

In other words, they didn’t study websites in the wild, but copied wild ones and manufactured “tame” ones of their own design that never left the lab.

There would be some serious control problems with such a “live” experiment…not to mention potentially lending some strength to Watts’ initial complaint about the risk of deceiving unwitting web surfers!

Note the underlying point here. Watts launched a baseless attack on the astroturfing study. When his error was pointed out, he tried to blame the study authors, and came up with new criticisms, including protesting that they should have conducted the study in a way that he himself had previously claimed would have been deceptive and misleading, or even unethical.

On to episode two:

The BEST Study. I just wrote about this one, and it is quite telling.

A while back, Anthony Watts wrote of the headline-grabbing Berkeley BEST study that “I’m prepared to accept whatever result they produce, even if it proves my premise wrong.”

But when the study came out and he didn’t like its findings, Watts instead engaged in  a phenomenon called goalpost shifting. Look at how he now talks about the BEST work: In a recent post, he referred to the “incomplete and rushed, non quality controlled, error riddled BEST science.” 

Are we noting a common theme here?

On to episode three:

The Republican Brain: My next book will not be out for about 6 months. Nevertheless, much like the Astroturf study, Watts attacked it without reading it. He justified doing so by claiming that someone else had reviewed the book, so he could rely on that review instead:

Chris Mooney has come up with new book to explain why people like you and I are “abby-normal” for not unthinkingly and uncritically accepting all aspects of global warmingclimate change climate disruption. I haven’t read it, though the cover itself speaks volumes. I won’t commit the same dumb mistake that Igor Peter Gleick committed when he wrote his bogus non-review of Donna LaFramboise’s IPCC book, so I’ll let somebody who has reviewed it speak about it. Dr. Roger Pielke Jr.

But this was incorrect—Roger Pielke, Jr., had not reviewed my book, nor could he have because my book is not out. So Watts basically did commit the “dumb mistake” he was claiming he wouldn’t commit.

I pointed this out. So Did Watts then say, “Whoops, sorry”? Or did he come up with more reasons to criticize me?

The proof is here. Read it for yourself.

Why is all of this significant?

If you can’t admit it when you’re wrong, you also can’t know when you’re right.  If you don’t hold your opinions and beliefs tentatively, subject them to scrutiny, and then try to parse out which of them truly hold weight, then you run the risk of rushing headlong into all manner of self-serving biases.

And please note: This has nothing to do with whether or not you’re smart. Smart people (like Watts) are in fact particularly vulnerable to this problem, because they’re extra good at rationalizing their views. Even as they’re super awesome at finding apparent problems with the arguments of those who disagree with them, and arguing back against their opponents, they’re often oblivious of their own biases.

But it doesn’t matter how many great arguments you can spin out to defend what you believe, if you can’t also perceive where your beliefs might be untrue. Without self criticism, all your self-supporting arguments amount to little more than spinning your wheels–while you remain stuck in the mud.