undefined | Better HN

0 pointsblagie4y ago0 comments

The best science, in climate change, in COVID, and in many other areas, just isn't very good science. That doesn't mean we shouldn't prepare. We make decisions on limited evidence all the time -- in business, investments, and in war, there is a certain fog.

I think where scientists went wrong -- and what contributed to doubt and fueled our inaction and inability to prepare to date -- was exactly from the type of manipulation you described. In the nineties, scientists correctly believed we should do something about the climate, but to affect change, they politicized science.

In the case of climate change, if you read right-wing media, they'll point to failed prediction after prediction expressed with great confidence 10 or 20 years ago, and use that to discount the whole concept of climate change. If that same data had been presented with confidence intervals -- and /real/ confidence intervals, not just based on statistical measurement error but on the possibility of models being wrong -- I think we'd actually be a lot further along.

If the choices were "six million wildly different possibilities of one in a billion odds", the risk would be 0.6%, and the case for acting wouldn't exist. As it is, I'd give perhaps 75% odds for the need to act on climate change. Precision is important.

But no, a lot of possibilities aren't hard to model or a reason not to act. We assign probabilities, and give a distribution. We have major wars perhaps every 50-100 years in most parts of the world. Recent past has been stable, but we've had windows of stability. Pandemics are on the rise, due to growing population and more commute. We can come up with likely scenarios, and they all have a lot in common.

0 comments

2 comments · 1 top-level

dahart4y ago· 1 in thread

You didn’t answer the question. How can we prioritize funding and actions? What are you going to fund, exactly, and how will you get enough people to agree to it, vote for it, etc.? A lot of people already agree with you that we should prepare, and still we haven’t prepared. Why?

The reason everything went wrong is that scientists haven’t used confidence intervals? Hehehe, I don’t think that has anything to do with it. Please be more specific, which papers, which predictions are you talking about? Who made them, and when exactly? Right-wing media has a tendency to misquote people and frame truth as failures too, and has been systematically trying to erode public trust in science and truth. I don’t think right-wing media’s political claims have any bearing on what we should do. Adding confidence intervals wouldn’t help anything; they’re opposed to regulation and to change of entrenched big business, it doesn’t matter what the science says, they’re against it because it affects somebody’s profit.

One big mistake you’re making here is suggesting that all science is making predictions, which is false. Most science, and much of the best science, doesn’t make any predictions at all, and doesn’t need confidence intervals, because it’s identifying and explaining past history, not the future. We don’t need science to predict that the temperature will go up, because we can see the trend over the last 10 or 50 years. Scientists aren’t telling us the climate is going to change, they’re telling us it already has changed. Adding a confidence interval won’t make that any more convincing to climate change deniers.

Hard disagree that the best science isn’t good, and that’s another hell of an armchair claim to make. I’m sure, like everything, some of it isn’t very good. See Sturgeon’s Law. But some of it is great. Watching Fox News will not help you identify the good stuff.

blagieOP4y ago

> Please be more specific, which papers, which predictions are you talking about?

What gets cited most is Al Gore's summary of science in Inconvenient Truth. He quoted a bunch of predictions. He streamed those off. Within 15 years, there will be no more glaciers in Glacier National Park. Etc. Many of those didn't come to pass. Fox News is pretty honest when it quotes those. Selective, but honest.

He expressed complete confidence throughout: "There's not a single fact, or date, or number that's been used to make this up that's in any controversy"

Well, guess what? It's a chaotic system. We knew the predictions weren't great; we missed most of the effects of ocean acidification. Every number was in controversy. We missed a bunch of effects, underestimated some, and overestimated others. It's easy to point to place where people were untruthful. Fox News rarely lies, but it often quotes the places where things didn't come to pass. Liberal media quotes the places where things did, or where things were worse than predicted. Neither side is particularly truthful.

> You didn’t answer the question. How can we prioritize funding and actions?

Well, there are two hypotheses:

1) We're going to lie and convince everyone.

2) We'll be honest and count on people to be reasonable.

We tried #1. It didn't work. How about trying #2? It's painfully hard, calling people on lying when we agree with them, but it's critical, on both sides of the isle. Right now, we have a race to the bottom. Republicans lie. Democrats lie harder. Republicans lie harder. And so on.

> A lot of people already agree with you that we should prepare, and still we haven’t prepared. Why?

A lot of people disagree, and can point to good evidence that the side advocating for change was lying. The conclusion was correct, but when the logic along the way is dishonest, that's not how it's received.

> The reason everything went wrong is that scientists haven’t used confidence intervals?

I don't think that's quite what I said. I said the political left -- including and especially the scientific establishment -- expressed 100% certainty in predictions which (1) didn't come to pass (2) clearly had enough assumptions that a sane scientist could have predicted that (and many, in private conversations, did).

> Right-wing media...

Do you read right-wing media, or merely read about right-wing media? I haven't found media on either side of the isle to be more or less truthful. I read both.

> One big mistake you’re making here is suggesting that all science is making predictions, which is false.

No, I'm suggesting that this is the part of "science" which has the greatest problems with truthiness, and makes it hard to trust the best science.

> Most science, and much of the best science, doesn’t make any predictions at all, and doesn’t need confidence intervals, because it’s identifying and explaining past history, not the future. We don’t need science to predict that the temperature will go up, because we can see the trend over the last 10 or 50 years. Scientists aren’t telling us the climate is going to change, they’re telling us it already has changed.

And if this is all that was being presented, I think we'd have acted sooner.

> Adding a confidence interval won’t make that any more convincing to climate change deniers.

It's not about adding, it's about taking out.

> Watching Fox News will not help you identify the good stuff.

It actually will. Try it. Treat it (and all other media) as a primary source. You'd be surprised at how often an outsider perspective will do that. Mostly, it will help identify a blind spot around bad science you agree with. Chinese media too, actually. It's pretty easy to read critically past the propaganda. Does NY Times have better scientific coverage? Indubitably. But it comes with the same blindspot. In science, you're trying to make a good faith effort to disprove the hypothesis, and people who disagree with you do that much better than people who agree.

j / k navigate · click thread line to collapse