Hacker News new | comments | show | ask | jobs | submit login
The easiest way to undermine good science is to demand that it be made “sound” (fivethirtyeight.com)
55 points by rectang 11 months ago | hide | past | web | favorite | 53 comments

> Now Congress is considering another way to legislate how science is used. The Honest Act, a bill sponsored by Rep. Lamar Smith of Texas, is another example of what Levy calls a “Trojan horse” law that uses the language of transparency as a cover to achieve other political goals.

Master of darkness, we meet again.

> He sponsored the Stop Online Piracy Act (SOPA), and the Protecting Children From Internet Pornographers Act (PCIP).

> As the Head of the House Science Committee, Smith has been criticized for promoting climate change denial, baselessly attacking scientific outlets and researchers, and receiving funding from oil and gas companies.[2][3][4][5][6][7] He was formerly a contributor to Breitbart News.[8]

I can only hope that next year after he retires, his district manages to pick someone who isn't even worse.

Unfortunately this is a losing battle because no matter what is said a laymen will latch on to any uncertainty presented to them regarding scientific discussion(s), be that evolution or climate change.

They don't understand that scientist by nature prefer to never say something is 100% one way or another. This is because we all know over a long enough timeline just about every conceivable outcome will occur and so scientists, rightly, try to be accurate when they talk about this stuff.

Politics isn't about presenting honest clear facts but rather wordplay meant to serve their donors.

To add to your point, one way to view this is that it reflects the immense difficulty of deeply understanding and discerning the difference between 90%, 99%, 99.9%, 99.99%, and so on.

These inherent difficulties make it very easy to spread confusion and mislead the layman.

Esp. if you are actually trying to spread confusion.

It's a depression ass state of affairs no matter how you slice it.

I sometimes wonder if a truly impartial AI 'judge' can solve this; but even if we achieve some form of advanced AI that could act as that judge I can imagine some people would still fighting tooth and nail because, of course, the creators of the Ai have somehow influenced it's 'thinking' and therefor it is invalid to them.

I don't think it will ever be solved. It will be a continual effort.

Some of the best tools I can think of:

* cultivating a culture of critical thinking and strong opinions weakly held

* increased awareness of cognitive biases and how to reduce them

* educating people how to evaluate evidence and understand how theories develop [1]

* open and honest discussion

* shaming intellectual dishonesty

* constant vigilance

[1] I've been watching the Ring of Truth on YouTube. It's fantastic. We need more content like this. Link: https://m.youtube.com/playlist?list=PL6MvlS2uHDjhadiMNqqTtZN...

Or GMO's... or Colony Collapse... beliefs unsupported by science exist everywhere...

They should apply the same rigor to the economic forecasts for their tax cuts. That would be interesting.

> in Europe, many decisions are guided by the precautionary principle

Unfortunately, European politicians apply precautionary principle also to cases where an overwhelming majority of scientist think that, for example, GMO crops are safe.

And after Fukushima, Germany decided to abandon nuclear power, although tsunamis are not really a thing in Germany. This has made Germany resort to burning more coal for energy and caused Germany's carbon dioxide emissions to increase.

These are cases where sticking to the precautionary principle has caused way more harm than good.

Germany decided to gradually phase out nuclear energy in 2002 (after two years of deliberations and negotiations), 9 years before Fukushima.

At the end of 2010, the center-right coalition extended the duration of the phase out; a few months after Fukushima, they essentially backtracked on that extension. This resulted in the oldest 8 plants being shut down, most of them built in the 70s.

As of now, there are still 8 plants operating. In 2016, it provided about 13% of overall electrical energy. In the same year, regenerative sources generated around 29%. At its peak in 2004, nuclear power provided around 32%, and regenerative around 8%.

> an overwhelming majority of scientist think that, for example, GMO crops are safe.

But what does it mean for GMO to be "safe"? From what I've been able to tell from what I've read on this, it just means that there is nothing inherently unsafe about GMO.

That doesn't mean that GMO cannot be misused.

Look at what the food manufacturers did with sugar. Used in small quantities, sugar is perfectly safe for use in food. But many food manufacturers have used it in large quantity in a large variety of foods.

GMO gives those manufacturers a greater range of things they can do to food, letting them achieve things that the might not be able to achieve with non-GMO techniques, or letting them do things faster and cheaper than the could do with non-GMO techniques.

Whether or not it is safe to let them do so is NOT a science question.

> GMO crops are safe

My objections to GMO crops are more around an aversion to industrial scale farming, monocultures, corporate-controlled seed stocks, and ultimately authenticity in the food I eat - not the health effects of the food itself.

So I'll vote to oppose GMO crops, and I think it's right for politicians to reflect that opinion. Little to do with science, much more to do with the kind of life I want to live and the world I want to live in.

"authenticity in food" is a massive myth. The foods you eat have been genetically modified for centuries via breeding techniques.

"monocultures" and "industrial scale farming" are separate from GMO, and will exist without GMO, as they existed before GMO.

DRMed seeds and health precaution are they only relevant aspects of GMO.

>The foods you eat have been genetically modified for centuries via breeding techniques

This is my go-to whenever people start talking about organic products. This article does a good job showing what crops used to look like: http://www.businessinsider.com/what-foods-looked-like-before...

That’s disingenuous and straw-man’y. Selective breeding consists of favoring individuals with desirable naturally occurring variations in characteristics. Outright genetic engineering consists of cobbling together genes of totally unrelated species - if not kingdoms - to make something invulnerable to otherwise toxic amounts of poisons to increase production yield.

I’ll keep my breeding any time thank you..

>...favoring individuals with desirable naturally occurring variations in characteristics... >...cobbling together genes of totally unrelated species...

Both of these statements describe genetic engineering. But you're right, there is a big difference between them: one is basically just throwing things at the wall to see what sticks and the other is a scientific process backed by hundreds of years of research.

The reason why people assign a lower risk to selective breeding is that the amount of physiological change present in one or a few generations is small. You don't worry that the rice you eat had a trascription error and may be genetically altered because it is invisible to you. But when you look at genetic engineering you see the big changes in a short amount of time and you feel it is more unsafe even though those alterations are being performed by some of the most intelligent minds of our generation.

You talk of scientists "cobbling together genes of totally unrelated species" as if they were nice modular pieces that you can rearrange in any order. Just the mechanism by which genes are edited is a choreographed sequence of tools that make the steady hands of the worlds best surgeon look bad.

I am not nessesarily 100% pro-GMO though, I just dislike the FUD that people use to demonize it as something radically different that what humans have been doing for millenia. The only thing different is that now we have a clue what we are doing. Will some people abuse it to benefit themselves? Yes, of course, but that is independent of the methods by which they screw people over.

Artificial selection has already made plenty sure that our surviving crops are invulnerable to what otherwise used to be toxic amounts of poisons.

Our own DNA has genes of ancient virus cobbled together, and those aren't even a kingdom.

"Naturally occurring variations" can also be phrased as "random genetic mutations we have no control over and aren't sure if they're safe".

> My objections to GMO crops are more around an aversion to industrial scale farming, monocultures, corporate-controlled seed stocks, and ultimately authenticity in the food I eat - not the health effects of the food itself.

Those are also (except maybe “authenticity”, since it's not clear what that concretely means; but I suspect it still applies with any coherent operationalization) at least as much of an issue with modern non-GMO crop development, including modern corporate industrial-scale organic agriculture

“GMO” is a distraction.

It's revealing to notice how many replies don't understand authenticity.

One way to put it: do you know the name of the farmer who grew or raised your food? Have you met them or their family personally? What's your judgement of their character?

> It's revealing to notice how many replies don't understand authenticity

Only revealing of the fact that you are using a word that doesn't have a single well-established definition applicable to the context in which you've used it. Nor have you actually articulated a definition, only examples (none of which correspond to any usual sense of “authenticity”.)

In any case, the examples you've given all seem to be applicable to industrial farming, to which GMOs are an orthogonal concern.

Industrial scale farming and monocultures don't need GMOs to happen already.

Not sure what food authenticity means... that it's not a cheap Chinese copy?

Coal ash spills have caused big problems in my state, and the nuclear plant has caused none that I'm aware of. I'd imagine part of that is because people are more reckless with something that they view as safe, and vice versa but yeah dumb move for germany.

Nuclear is also more heavily regulated, meaning that the nuclear plant is required to follow very specific rules when handling the dangerous material

North Carolina with the Duke Energy spills?

> Unfortunately, European politicians apply precautionary principle also to cases where an overwhelming majority of scientist think that, for example, GMO crops are safe.

What percentage of these scientists allow the general public to modify the DNA of their food however they want?

> Unfortunately, European politicians apply precautionary principle also to cases where an overwhelming majority of scientist think that, for example, GMO crops are safe.

I seriously would like to see that majority of scientists live patching binaries in a networked environment (say, microservices) with ossified protocols and see how well their _improved_ applications run.

Hint: they will crash within seconds or create unexpected problems to the network much later.

(That is the software analogue to GMO.)

I hate this recurring tendency of HN commenters to myopically reduce every issue to one involving technology before forming an opinion on that issue.

Hint: the applicability of an analogy does not extend infinitely in all conceptual directions.

When you have to judge papers/findings by the motives and qualifications of the authors rather than by the data, you no longer have science you have opinion.

When the theory of relativity was introduced by Einstein, it did not matter what you thought about Einstein's qualifications or motivations. Einstein himself gave several ways the theory could be tested independently.

Because of the prestige that science garnered - it gave us airplanes, radio, tv, internet, landed a person on the moon, etc, many knowledge disciplines branded themselves as "x science". Thus, has the word "science" become devalued.

Yeah. If someone claims data, reproduce their experiment. If someone makes a point, check their math. If it's beyond the realm of repoducing experiments and checking math it's beyond the realm of human knowledge. Sorry, but time and time again our kind have proven to be liars; and worse, self-decievers. Nothing else is left in ourselves to trust.

> When you have to judge papers/findings by the motives and qualifications of the authors rather than by the data, you no longer have science you have opinion.

To a significant degree, that is exactly how the scientific community operates. Amount of funding, prestige of journal (in which the research is published), and acceptance of results depend highly on reputation. I'm not a scientist, but I've heard that from many scientists and I believe there is research and journalism that support it.

As in any industry, the ideal is not the reality. Think of programming; how much code actually follows the norms, ideals, and theories of how it should be done? Very, very little; OpenSSL comes to mind as an obvious example; OpenBSD tries to do it 'right', at least to a degree, but they make significant sacrifices in functionality and productivity, and many people think they are over-the-top, self-indulgent obsessives. The reality is that getting things done requires a lot of hacks; few people review any code, even open source; and if Linus Torvalds and I submit competing code to something, his will be accepted regardless of the quality - probably,

Mods / @dang / @sctb: I can't seem to edit the parent comment. When I made an edit, it cut off the end (about 1 sentence + 1 short paragraph). Now, no matter what I append in the edit, when I click submit it returns to the state above.

I'm telling you so that you are aware of a bug. This one comment isn't a big deal.

This is true and it would be ideal if all lawmakers were well-versed in the process of science, as well as the subject matter relevant to their decisions.

Unfortunately, that's an unrealistic assumption in the messy real world. In fact, I doubt even the most genius of researchers could be reasonably expected to verify the veracity of all the scientific claims that are advanced in defense of various policies that lawmakers must form opinions on.

> Because of the prestige that science garnered - it gave us airplanes, radio, tv, internet, landed a person on the moon, etc, many knowledge disciplines branded themselves as "x science". Thus, has the word "science" become devalued.

This is perhaps a bit off topic, but I'll just note that this includes Computer Science [1]. Also, you can replace "science" with "engineering" in that quote ("Software Engineering").

[1] https://en.wikipedia.org/wiki/Computer_science#Etymology

Do you understand the math of climate modeling? Have you checked the equations, data, and made your own measurements? If not, why do you believe climate change is happening? And why do you believe the scientists that say it is happening versus the ones that say it is not happening?

Does it matter? There doesn't exist a single person in the world which understands all of current science. We hold faith, and I personally hold higher faith in sciences which are reproducible either through more open standards or more rigor in their culture.

I understand the math, they have a bunch of simulations that produce wildly different results, they average these and make predictions based on a number of models running a number of simulations. The predictions have so far yet to have come true. (Although temp is rising, it's not to the degree predicted by the average of the models, despite there being more CO2 than the models were given input for, obviously the average of models are inaccurate. )

I believe it's not happening to the degree the IPCC said it was in 1992 which is why in the 2000s they observed a warming 'hiatus'. This is why people talk about consensus rather than data, or falsifications of the null hypothesis.

What were the predictions in 1992 and what level of confidence was assigned to them (the IPCC reports are very careful to assign confidence levels)? Also, there have been many generations of reports since then; why focus on 1992, 25 years ago? As far as I know, their predictions have been pretty good - though not perfectly accurate, of course.

There is so much nonsense posted about climate science; it would be great if someone could back up claims, perhaps by citing the IPCC reports or some serious analysis of them (i.e., in a prominent journal).

1992 because it's the first report.

Here's a decent summary: https://en.wikipedia.org/wiki/Global_warming_hiatus

'climate is classically averaged over 30-year periods.[5]'

'Publicity has surrounded claims of a global warming hiatus during the period 1998–2013. The exceptionally warm El Niño year of 1998 was an outlier from the continuing temperature trend, and so subsequent annual temperatures gave the appearance of a hiatus: by January 2006, it appeared to some that global warming had stopped or paused.[2] A 2009 study showed that decades without warming were not exceptional,[6] and in 2011 a study showed that if allowances were made for known variability, the rising temperature trend continued unabated.'

Essentially in 5 years we'll have enough data to confirm / deny the 1992 predictions, currently it's an unproven theory because of a lack of forward data. We're essentially using consensus as a proxy for a falsification of the null hypo

If you ban opinions, it's impossible to get anything done that isn't very basic physics and chemistry.

Openness in science is a good thing that scientist should be striving toward. It's hard not to sympathize with "(o)nly studies whose raw data and computer codes were publicly available would be allowed for consideration" from the EPA. It's easier in some fields than others, but I believe it's a truly noble goal.

Just yesterday I finished Merchants of Doubt (the book; I didn't know there was also a documentary). It's a really good book, and very informative. Side effects include uncontrollable anger and hate towards certain industries.

> For instance, when the EPA was preparing to set new limits on particulate pollution in the 1990s, industry groups pushed back against the research and demanded access to primary data (including records that researchers had promised participants would remain confidential) and a reanalysis of the evidence. Their calls succeeded and a new analysis was performed. The reanalysis essentially confirmed the original conclusions, but the process of conducting it delayed the implementation of regulations and cost researchers time and money.

If the data had been open from the start, this wouldn't have been a problem.

Also, why don't we just sow an equal measure of uncertainty in the studies showing that these drugs/chemicals/externalities are safe? I bet if you demanded that the pro-industry studies should be "sound," you would discover a lot.

>> including records that researchers had promised participants would remain confidential

> If the data had been open from the start, this wouldn't have been a problem.

I mean, it sounds like there were serious legal obstacles going beyond "shit what did I do with that .csv file". And maybe the science wouldn't have existed at all if you demanded open access a priori.

I've always wondered why data for some huge meta-analysis studies isn't availabe. If you're not a researcher you can practically never verify the result that there's no link between vaccines and autism, or any other similar studies.

Most of these are just number crunching (increased risk of diet influencing disease, intervention experiments in psychology, medicine etc.), yet the number crunching is concealed and the numbers are the only thing presented, and I should trust the researchers that they didn't play with the numbers.

Not only is spurious doubt manufactured for commercial gain, as in the tobacco example, it's also dragging us down a path of environmental ruin, which is now widely acknowledged to be killing off the majority of other species on the planet, and yet the climate skepticism industry keeps churning out sunny articles offering various reasons not to worry about it.

Take for example this gem, which admits we're ruining the environment but says it's cool because it's natural: https://www.washingtonpost.com/outlook/we-dont-need-to-save-...

There's a soberly-worded climbdown on his own page at the university where he teaches, but guess which of these articles is going to be more useful for climate sleptics to cite: https://biology.columbian.gwu.edu/r-alexander-pyron

This is the problem with the much-vaunted 'free market of ideas' - as scientific endeavor moves outside the boundaries of scientific community into a subject of general interest, peer review and professional integrity are easily corrupted by marketing, to the severe detriment of the general public.

Some good ideas here, but we have to aim much higher than this.

I like some of the ideas Bret suggested in his call for climate change: http://worrydream.com/ClimateChange/#media

The easiest way to dismiss most nuanced discussions is to relabel one or both positions with broad emotionally appealing words.

Some good examples would be the "death" tax, the "Patriot" act, "Common sense" gun control, etc.

> These are the arguments underlying an “open science” reform movement that was created, in part, as a response to a “reproducibility crisis” that has struck some fields of science.

Because social scientists continue to make giant claims that fail to be supported by the evidence, and those claims are used to enact sweeping policies based on quackery.


Stereotype Threat. The Implicit Association Test. The Bystander Effect. Power Posing. These are examples of "science" that is based more on rhetoric and emotion than reality. But it is having real effects on policy.

If your claims are made up and based on shoddy principles, they deserve scrutiny.

> objective knowledge is not enough to resolve environmental controversies.

This article is literally trying to build a case of "feels" over "reals". Transparent objective facts should guide the debate, not twisted political rhetoric masquerading as "science". If your position is incorrect and does not match reality, it must be dropped, even if you would prefer it to be true to push a point.

I can't believe this kind of garbage is coming out of 538

You fundamentally mischaracterize the article's central thesis; you're attacking a straw-man.

The article goes on, immediately following your quoted sentence, to draw an important distinction:

> But they’re also used as talking points by politicians who are working to make it more difficult for the EPA and other federal agencies to use science in their regulatory decision-making, under the guise of basing policy on “sound science.” Science’s virtues are being wielded against it.

> What distinguishes the two calls for transparency is intent: Whereas the “open science” movement aims to make science more reliable, reproducible and robust, proponents of “sound science” have historically worked to amplify uncertainty, create doubt and undermine scientific discoveries that threaten their interests.

And goes on to discuss how the term "sound science" was invented (or at least popularized) as FUD by Phillip Morris to discredit medical science on the health effects of second-hand smoke.

> undermine scientific discoveries that threaten their interests.

Scientific discoveries aren't scientific or "discoveries" if they don't reflect reality. If you can't defend the fact that your results don't replicate, or that your methodology is flawed, or your math is wrong, then you can't continue to promote your "discoveries" as if they are true.

Again, you completely miss the point of the article.

The article is not about Science! The article is about public perception of science. About how very valid criticism of bad science can be used to paint with an overly broad brush, and about how that rhetorical trick works well when convincing voters and/or policymakers to ignore good science.

BTW, the piece quotes Steven Goodman -- one of the "good guys" leading the charge against unproducible social science -- making exactly this point.

No one is making the case for letting bad science slide, least of all Steven Goodman.

The things you are saying are true. But they have absolutely nothing to do with the article that they're attached to.

There is a whole paragraph from someone who "who researches the sociology of technology and scientific knowledge at Colorado State University, wrote in a 2008 paper about why objective knowledge is not enough to resolve environmental controversies"

> These controversies are really about values, not scientific facts, and acknowledging that would allow us to have more truthful and productive debates. What would that look like in practice? Instead of cherry-picking evidence to support a particular view (and insisting that the science points to a desired action), the various sides could lay out the values they are using to assess the evidence.

This is pushing the idea that the facts don't matter. What matters is "values" and emotions. It's an elaborate setup to push a view from certain segments of the humanities and social sciences to dismiss the epistemology of science and an "objective reality" entirely.

> This is pushing the idea that the facts don't matter

Yes, but you seem to have confused the author's positive statement as a normative statement.

The author of that quote is not saying that's how the world should ideally be. He's saying that's how the real world really fucking works. In other words,

"The distinguished representative from Texas doesn't give a shit about science; he's made up his mind on climate change because there's a shitload of money at stake. If he can discredit good study X to weaken his opponent's point, or marshall shitty study Y to strengthen his own point, he'll do so. Because winning the political battle is more important than being 'correct'. So you're not going to convince him to change his mind by throwing science in his face."

The author is NOT defending the not-so-hypothetical distinguished representative from Texas. He's just observing the fact that this is how certain people process scientific claims. Not as scientists, but as motivated reasoners and political operatives with a priori agendas.

See also the discussion of the precautionary principle above for a less cut-and-dry example -- what's happening, in that case, is basically that two groups with different values are making different default assumptions about the level of confidence necessary to justify a decision. I.e., they have different values, and those values color the way in which they interpret the confidence intervals in studies on e.g. GMOs.

You're confusing a description of the status quo for a defense of the status quo. This is a ridiculous straw-man.

> What matters is "values" and emotions

Again, the claim is not "values/emotions are objectively more important than science".

The claim is "voters/lawmakers are more likely to be swayed by values/emotions than by science".

Those are two very different claims.

> It's an elaborate setup to push a view from certain segments of the humanities and social sciences to dismiss the epistemology of science and an "objective reality" entirely

That's quite the conspiracy theory, given that the overall tenor of the piece is decidedly pulling in the other direction.

There are now a hundred years of research regarding the social sciences and humanities.

Your scare quotes betray your knowledge on the matter. Perhaps you should research equal "science" claims such as the anchoring effect or such? Illusiory superiority?

The nonlinear social / contextual matters make the social sciences extremely complex.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact