Negative Evidence

People, scientists included, since most of them are also people, are good at finding evidence for desirable propositions.

If you want to discover reasons why the earth is really flat, and not spherical, you’ll uncover lots of juicy stuff about why our rulers want us to believe in Old Roundy. What you’ll especially find are a host of counter-arguments to the obvious arguments against a flat earth. Flat-earthers have an answer for every objection you can come up with—short of launching them into space so they can see for themselves, that is. But even then, the true believers would discover reasons why their flight was faked, somehow.

I pick on these people, and almost all of them I like very much, only to prove the point. Finding evidence for what you want to believe is easy. (And if you’re inclined to laugh at them, I have just one question for you, did you voluntarily wear a mask during the covid panic? Incidentally, we also see this behavior in those cheering for “their” side in the latest foreign war: they lap up good “news” for their side, and dismiss with contempt all that coming from the other.)

It’s even easier for scientists who, some of them anyway, are at least modestly intelligent. At least they have had training that shows them how to find evidence, which is their job. Most of this, sadly, involves formulaic reasoning and conformity following. Even so, the point remains: your intellect can only be described as medically stunted if you cannot conjure up arguments for what you want to believe. Everyone not broken can do it.

Real intelligence arrives, however, when you learn how to find evidence which disproves what you want to believe.

Most are terrible at this, scientists are marginally better, or used to be before DIE sucked the life force from them.

I asked on Twitter if they still taught debate by having people take positions they didn’t agree with, and were thus forced to steelman (as opposed to strawman) positions they disagreed with. I didn’t get many responses, so maybe you know. I find it hard to believe the technique survives since the arrival of Feelings, Diversity, and suchlike. One fellow said the only thing they teach now was to talk really fast so opponents don’t have time to rebut all points. That I believe.

St Thomas Aquinas wrote Summa Theologica using the steelman formula. He’d list the best arguments against his position, state his opposition, and then, with the skills of a world class roboticist, he’d dissect the steelman and leave him in quivering, malfunctioning pieces on the floor.

As I said, science used to be that way, too. Scientists took great delight at eviscerating the arguments of their enemies. Nothing pleased them more (after winning an award, that is) than exposing the flaws of others. Even when I was in graduate school in the 1990s, it was still a thing to have some seminar audience member spout at the speaker, “That equation is wrong.” This was the masculine adversarial procedure applied to science. It worked.

Then came Feelings and Consensus. Would anybody today dare point out, openly, a flaw in the job talk of a Diverse Non-male? Maybe it still happens. Let us know.

Which brings us to a talk recommened by Marc Andreessen, from Dan Kahan. He showed this set of data to a bunch of people (this is a poor screen grab, I know, but good enough to make out):

Regular readers will have figured out the correct answer. It won’t be any surprise to learn, as Kahan says, that the more numerate (on some crude scale he had) people were, the more likely they were to get this. Same if he reversed the column labels. Meaning whether or not people got it right, it had nothing to do with skin cream working or failing per se.

In a clever move, he used the same table with different legends. He replaced skin cream with gun bans. Instead of “Patients who did use the new skin cream” in the first row, it was “Cities that did ban carrying concealed handguns in public”, and the obvious for the second row. Instead of “Rash got better”, it was “Decrease in crime” for the first column; or instead of “Rash got worse” it was “Increase in crime” for the first column.

The twist was he also separated out self-identified Democrats and Republicans. That had no play with the skin cream, but it did with the gun ban question. Lefties were more likely to get the question wrong when the correct answer was crime increases with a gun ban, and righties were more likely to get the answer wrong when the correct answer was crime decreases with a gun ban. Even if they were highly numerate.

Kahan said this was because political motivations can cause bias, and he is surely right about that. But I wonder how right.

The skin cream question was no more than skin deep (feel free to laugh! only the NSA, Mossad, the FBI, Facebook, Apple and Ford Motor Company are listening). Unless you’re in the pharmaceutical business, the question doesn’t elicit a lot of emotion or open up a host of other questions. So you stay on target.

There are, however, many emotions and questions surrounding gun bans. It’s easy to see why people would let these spurious matters (in the context of the table) distract them from the simple math of the problem. This far I am an agreement with Kahan.

But then he asked participants in his study, “How much risk do you believe global warming pose to human health, safety, or prosperity?”

It will be no surprise that Ds and Rs split here, too. But here Kahan has no table with the pre-computed right answers. Yet he does in the talk suggest that the right answer is the side taken by The Consensus, as issued from agencies such as the National Academy of Sciences.

That isn’t crazy. If you knew nothing about the thermodynamics of fluid flow on a differentially heated rotating sphere, it is natural and rational to look to Experts. Until, alas, we learn those Experts are not up to the job asked of them, as you and I have, dear readers, over many years (blog/Substack). Here Kahan could be guilty of his own bias. He either hasn’t taken seriously criticisms of Experts, or he considers those criticisms unworthy of exploration.

He hasn’t asked a good question, either. It has three parts: health, safety, or prosperity. Which of these is anybody answering? Are all three of equal risk?

Obviously not. And here, unlike the gun question, the outside matters become the real question. Because as I have long said, when some of us hear “climate change” we do not hear physics, but hear the motivations behind rulers wanting to sell “solutions” to “climate change”. The two are entirely different. You can agree about the physics (which I do not), but that physics in no way implies anything about the consequences or mitigation possibilities of the physics. To conflate the two is scientism.

The problem of Consensus is that when it become institutionalized, it becomes increasingly difficult to question. The scientists in it forget how to look for evidence which disproves their theories. They are certainly not encouraged to. And, in fact, often punished when they do.

Subscribe or donate to support this site and its wholly independent host using credit card click here. Or use the paid subscription at Substack. Cash App: $WilliamMBriggs. For Zelle, use my email: matt@wmbriggs.com, and please include yours so I know who to thank.