‘Preapproved Narratives’ Corrupt Science | Hacker News Reader

‘Preapproved Narratives’ Corrupt Science

wsj.com

80 pointswjb332 comments

This is very common and is basically one of the "ugly truth" that people get to know once they start doing academic research. If you want to live, you have to get published, and to be published, you have boots to kiss. The grant agencies and the reviewers are not going to be "open-minded" about all topics and there are things you are basically pressured into not saying even if you know that is likely how it actually is.

Journals and conferences would also refuse to host your studies if they find the contents to not align with "settled science", regardless of how or what the studies is actually about. And to an academic, that is potentially the end of their career once they got branded as someone too controversial to be associated with. No funding will come and no student will join. Unless they are big enough to carve out their own niche. But those are rare.

It's only tradition; universities and the education sector in general have a long history as religious institutions of one sort or another - either in themselves, or funded by them. It's just a different sort of belief system now, but there has always been a strong "faith" element.

There are reasons science tends to work this way. First, there are cranks out there. The Society for Neuroscience used to continually receive abstracts from a guy who claimed black people's brains are more like monkeys' than white people's. Is that a topic that they need to be open-minded about, or can they just dismiss him as a racist crank?

More seriously, there are organized efforts by well-funded groups whose purpose is to sow doubt about science that could hurt their profits. The fossil fuels lobby is the best example, but the alcohol, sugar, and various polluter lobbies are out there too. They fund scientists who will do research designed from the outset to be favorable to the funders.

Faced with an organized bad-faith effort to corrupt the scientific endeavor, journal editors and grant reviewers naturally start applying more scrutiny to anything on the "wrong" side. Then the lobbyists complain about "preapproved narratives" in the Wall Street Journal.

There are highly placed researchers, people who've climbed the ladder over the years and have lots of authority in their field. They have their preferences about the direction the field should go in, and who should be doing the work, as well as protecting their own legacy.

Whether their reasons are benevolent or not, the existence of these people is the rule in science. If the knowledge moves too far forward, or away from their expertise they lose their relevance, and so they have incentive to protect it. This leads to biases in scientific research that are politically difficult to overcome.

Science advances one funeral at a time. -Max Planck

Cranks, almost by definition, are pretty much instantly recognizable to an expert: no one needs to apply extra scrutiny to catch them.

Suspect industry funding is a problem in some fields, but the powers that be couldn't care less about the charge of the electron, and yet it took twenty years to correct a fairly trivial error in Millikan's original measurement.

> Cranks, almost by definition, are pretty much instantly recognizable to an expert: no one needs to apply extra scrutiny to catch them.

Correct, but you are missing a wrinkle. What if I disingeniously call people cranks? What if a lot of actual experts in one camp call actual experts of the other camp cranks?

In the hard sciences we can deal with it. If one camp can build working satellites and the other cannot we know who the real deal is. But in softer sciences? It's an unsolved problem as far as I can tell.

In some scientific fields like human nutrition there are way more cranks than experts. Some of the cranks have even managed to earn PhDs and publish multiple papers. The fact that there is a lot of industry funding for this field doesn't help the situation.

> Is that a topic that they need to be open-minded about, or can they just dismiss him as a racist crank?

Well, this is kind of the point of peer review, isn't it? You review the work, you reject it if it doesn't pass muster.

There’s an old joke that you finally get to do the research you want to do when you’re an emeritus professor

This feels short on facts high on opinion (almost like the WSJ has some pre-aproved narative!). I can't speak to the other pieces mentioned but I think it's worth noting that Brown's claims about a 'prefered narative' in Nature are rather undermined by some of the peer reviewers comments which encouraged him to broaden the scope of the research in precisely the way he claimed that Nature tacitly discourages.

A more indepth look at the claims and counter claims... https://www.carbonbrief.org/factcheck-scientists-pour-cold-w...

A beter example might have been the lack of funding available to alzheimers researchers not looking into the dominant (and now increasingly shaky) amyloid hypothesis. which to me looks like a genuine case of a "prefered narative" which may have held back progess for a decode or more.

> This feels short on facts high on opinion

Well, it IS the Opinion Page, so don’t expect much. You don’t really need facts if you are just going to blame a vague “Narrative” boogeyman.

It’s a common excuse from people claiming to be censored: “I can’t say this or that because it ‘goes against The Narrative.’” OK, so tell us more about this shadowy narrative that’s hiding around every corner! Is The Narrative in the room with us right now?

There's an entire media ecosystem for this sort of stuff now.

Step 1. Write a crappy paper or make an otherwise ill supported claim in the scientific community.

Step 2. Scientific community says "hey that work is not good."

Step 3. Run to media outlets that are thirsty for "look at how this community doesn't permit alternative thought" narratives.

Step 4. Pivot into reactionary grifting, rake in patreon cash.

  Step 1. Write crappy paper
  Step 2. Seek peer review. 
  Step 3. Wait months for peer review
  Step 4. Apply to journals for publish
  Step 5. Wait months for journal publish
  Step 6. Add paper to CV

Most research papers are garbage, and almost completely ignored, unless it can be used as a cite in a similar crappy paper. The major journals aren't going to take your paper unless you already have a reputation, or you're connected to someone with a reputation. They aren't going to take a poorly thought out provocation piece from a nobody, when they are swamped with poorly written garbage from researchers trying to make tenure.

Not sure why you're being downvoted. This exact thing happens in evolution, global warming, and more recently covid.

Lest anyone think this is limited to "soft sciences" we should recall that this "settled science bias" happens in hard ones, too: https://hsm.stackexchange.com/questions/264/timeline-of-meas...

I believe when this was published the review was also published, or at least part of it, because I recall that one of the reviewers had pointed out that he had biased his model by leaving out the stuff he left out. Which means he needed to argue why he left it out, which seemed rather duplicitous to me at the time. Feel free to correct me if I'm wrong.

Yeah, thats correct. He also threw his co-authors (many early in their career and now potentially reputationally compromised by association) under the bus to get his moment in the spotlight before leaving academy. The whole thing was pretty unedifying.

The reviewer's criticism didn't stop the article from being published. Sure, they didn't recommend it initially but that happens.

I'd call it somewhat mild attention to other factors from the one reviewer. The pointing out of issues in data balance, why an nn model, are also good criticism. Here's the review page if anyone wants to read it.

https://static-content.springer.com/esm/art%3A10.1038%2Fs415...

So the question becomes, is his claim of leaving out other factors not greatly influencing and it's just a grift and/or is nature's review process a dog and pony show?

And the epidemiologist mentioned at the bottom of the article said COVID lockdowns would enable a U.S. President to use the military to establish themself as dictator.

But this is the WSJ editorial page, which isn't exactly well-regarded, even by the rest of the WSJ... "WSJ Journalists Ask Publisher for Clearer Distinction Between News and Opinion Content" (https://archive.ph/te5Bd)

Maybe more generally it's just "Politics Corrupt Science"?

I doubt "preapproved narratives" are plaguing fields such as niche north american spider research (but maybe so if biased gatekeepers have reputations at stake).

But things like covid, climate change, lgbtq issues, and soon probably ai/ai safety are politically controversial and thus are much more susceptible to corruption.

But not all "preapproved narratives" are bad, though. Things like "murder is bad" is something almost everyone would agree on.

To a deontologist, murder is bad. To a consequentialist, it depends.

I'm not sure they'd actually call it murder in that case. They might accept some probability of murdering someone to preserve a strong negotiating position and mitigate inaction related failures, but most of the time the actual intent would probably be a non-murder killing.

https://archive.ph/Yttxo

"This is very common" seems to be a refrain this year with Harvard, Stanford and so on. Yet we have trouble accepting this is a real problem when we approach 'climate change' from an emotional or political standpoint. I say this as an environmentalist who doesn't understand why we can't be more straightforward. Why we keep losing talented people when they can't fix the numbers to fit the narrative? This has been a 'thing' my entire life but I feel like it is reaching a crescendo, adjacent to the lack of motion in QED.

“But makes cost effective and publishable news reporting, that we often use” - continued an alternate reality more open and transparent WSJ that likes to properly contextualize its stories.

> "No doubt the editors at the New York Times and ProPublica would say the same of their own pages."

Of course, the WSJ would NEVER have a “a preferred narrative”.

And so we should really keep burning as much petro as possible, because otherwise shareholders are being stolen from!

And this is obviously another "prefered naritive":

https://www.washingtonpost.com/health/interactive/2023/ameri...

"Forty years ago, small towns and rural regions were healthier for adults in the prime of life. The reverse is now true. Urban death rates have declined sharply, while rates outside the country’s largest metro areas flattened and then rose. Just before the pandemic, adults 35 to 64 in the most rural areas were 45 percent more likely to die each year than people in the largest urban centers."

I thought this was a good analysis of Patrick Brown's claims: https://www.youtube.com/watch?v=dXZUXQPqY3k

One of the best indicators of what the results of a psychotherapy study/experiment will be is what the scientist doing the research already believes. I doubt this is limited to simply psychotherapy.

https://slatestarcodex.com/2014/04/28/the-control-group-is-o...

It's easier to get a desired result in fields like psychology and educational sciences, since it's so hard to control the experimental setting. To add insult to injury, this makes that even when done correctly, studies often don't generalize.

Very funny that it comes from WSJ, which preapproved all of these narratives in the first place.

https://en.wikipedia.org/wiki/Self-criticism_(Marxism–Lenini...

Reply on news.ycombinator.com