News & Politics

The World Is in the Middle of a Reproducibility Crisis in the Sciences

Recently in a private forum, someone talked about how the Earth is passing through a dark matter “Hurricane.” And one of my friends pointed out that it’s kind of interesting that we’ve moved from science discovering things you could see, touch, understand, and use to produce results, to the kind of nebulous theory that can never be completely proven or disproven.

In fact, apparently, no one is even absolutely sure that dark matter exists. But we have think tanks that instead of, say, assuming our theory of gravity might not be precisely true, posit dark matter to explain the discrepancies.

It’s kind of like the idea of parallel universes, say, which has become at least somewhat accepted, but which can never be proven or disproven.

And if it were just that, what my younger son (who is taking that kind of class) calls “physics when it becomes religion” wouldn’t be such a big deal.

But it is not.

We have laboratories, social studies think tanks, and countless and endless “experiments” going on, most of them paid for by the bountiful purse of Uncle Sam. And most of them irreproducible.

In fact, in an age of “evidence-based medicine,” which means, in practice, medicine where someone can cite some study to support what they want to do/is cheaper/fits their (mostly silly) Marxist ideas, most studies even in the hard sciences are non-reproducible.

This is not casting shade on the real scientists, who, yes, might take government money, but work very hard at making their study rigorous and absolutely reproducible. Heck no. Those people are important. They’re also a dying breed. The system encourages “science-to-order” generating things like the hokey hockey stick graph and thousands upon thousands of studies that are the equivalent of “climate science” or worse, rather than studies that are actually useful and applicable.

This is, in general, because government gets what it pays for, and what it pays for is the conclusions it wants. This is, of course, worse in the softer sciences, but it’s very bad in all sciences. Which is why the world is in the middle of a reproducibility crisis (though these guys would rather we called it a problem),  i.e. studies that when examined are wholly made up, or severely fudged.

Needless to say, this kind of fudged research that can’t be reproduced is very bad in the hard sciences. I mean, which of you wants a surgeon to go in and remove your tonsils because that will make your immune system stronger because some study shows…? Yeah, exactly what they did to me and my generation, and the exact opposite of what works. So, yeah, none of us wants that. We also don’t want new super-duper rocket fuels that don’t work at all. Or Solyndras that are financed by government, cost us millions of dollars, and produce nothing. (Okay, that was straight-up fraud, but I’m sure there are studies saying that solar energy can replace fossil fuels because there are studies saying practically everything.)

But it’s actually worse in the social sciences, because social sciences, by being called “science,” tell us who we are. Take Calhoun’s “mouse utopia experiment,” which informed the minds of generations about the perils of overpopulation. Let’s say the results weren’t precisely as interpreted. (I actually could only find this mildly critical article, while page after page of searches echo the original interpretation. I have a memory – and I’m sorry, it’s been years and it was a paper magazine – of reading that besides the fact humans aren’t rodents (duh!) there are other problems with the framing of the experiment itself, which explains why the results don’t apply to places like, oh, Singapore or Hong Kong.) But they were swallowed whole and regurgitated in things like “Logan’s Run” to convince normal human beings that more of us would be a bad thing, and that there was no point in wishing to give birth to and raise future generations.

Or take Zimbardo’s Stanford prison experiment, which was used by the left to justify why we need strong government/overseers to keep normal human beings from becoming monsters. It’s one of the foundational building blocks of lefty (and some righty) thought, buried so deeply that no one actually questions it, because most people aren’t aware of what their second- and third-order experiments are based on. And apparently, it’s a hoax. Which anyone who studied Zimbardo himself strongly suspected from the beginning. But few people bothered, or knew where it came from.

Then there is Milgram’s obedience experiment, also part of the lefty view on society and particularly on the right, and yeah… you kind of guessed it, also a hoax. Or at least severely fudged.

So, what is the problem with all these “soft science” experiments and studies that prove this or the other or yet the other thing?
Well, other than the fact that 80 percent of them, including foundational ones, result in policies… nothing much. Or to put it another way, “Nobody knows anything.” But we presume we know, based on results obtained in order to please – mostly leftist – institutions, like big colleges and think tanks, who take results they like and run with them, pushing policy decisions.

This process has started to infiltrate hard sciences and push out good science in the name of good-sounding bad science. I’m not going to propose we get all government out of research. Sure. I’d like to do that. I’m a libertarian. No money for government is my watchword. And that would mean no money for government to spend or pass on. But we do have a need for research in the interest of defense (among other things).

What I’m going to propose is that the softer the science the more we should look at it askance. Let’s start to ask what overpopulated mice have to do with overpopulated humans. Or like a recent study that “proved” that humans should eat almost exclusively carbs by feeding mice almost exclusively carbs, ask the “scientists” if they can live on cardboard and old clothes. Let’s ask that before making the experiment a foundation of social thought and policy.

And most of all, above all, let’s demand that experiments be reproduced. If they can’t be reproduced, nothing is proven and the hypothesis might remain a hypothesis but nothing more. (Yes, I do actually mean to say that the kinder-caucus now entering the house should stop agitating for policies to stop the Earth from burning up because, you know, if it hasn’t burned up on the schedule they were told when they were three, it’s probably safe. I mean, how stupid are — oh, Occluded Cortex Occasional Cortex Ocasio-Cortez. Very stupid. Never mind.)

Sure I’m very amused by physics when it becomes religion and by the idea that we might or might not be encountering a hurricane of dark matter, something that might or might not exist. But it’s pretty and shiny, and younger son – who is of the engineer tribe – rolls his eyes. But how much trouble can it cause us? (Okay, the occasional dark matter lab. Bah. Cheap at the price.)

On the other hand, stuff that tells us how to eat, how to treat disease, and particularly things that affect our fundamental view of humans and the world? Accept nothing on faith. That’s not how science is supposed to work. “What are the facts? Again and again and again – what are the facts? Shun wishful thinking, ignore divine revelation, forget what “the stars foretell,” avoid opinion, care not what the neighbors think, never mind the unguessable “verdict of history.” What are the facts, and to how many decimal places? You pilot always into an unknown future; facts are your single clue. Get the facts” (Robert A. Heinlein).

Irreproducible science is not fact. It’s glorified opinion.