The rise and fall of scientific authority, and how to bring it back • Hacker News Board

Re: The rise and fall of scientific authority, and how to bring it back

#291

Post by oldgradstudent » Tue, Mar 19, 2019, 2:05 AM UTC

SketchySeaBeast wrote: post #187

balabaster wrote: post #164

Earlier quoted context omitted.

It's like hearing that science says that fat is bad... and then eggs are bad for you... and alcohol is bad... and then another study says that a glass of wine a day is good for your heart... and salt is bad... then sugar is bad... then carbohydrates are bad... and now fat is good as long as you don't overdo carbohydrates... and eggs have good cholesterol and are really good for you... and salt is necessary to balance…

That's not science though - that's a single study or two being blown out of proportion. It's an element of science, a snapshot of knowledge, but the scientific method requires that we continually refine those messages and ideas, and that's not happening - we do an experiment, and then without any more corroboration or reproduction it's released to the public. We are seeing tiny pieces of science, but the problem is w…

It's not a single scientific study blown out of proportions. It's the same at every level, including official reports by expert committees and scientific bodies:

Take eggs for example.

Here are the 1995 dietary guidelines:

https://health.gov/dietaryguidelines/dga95/9DIETGUI.HTM

> Choose a diet low in cholesterol

> The body makes and requires cholesterol. In addition, cholesterol is obtained from food. Dietary cholesterol comes from animal sources such as egg yolks, meat (especially organ meats such as liver) poultry, fish, and higher fat dairy products. Many of these foods are also high in saturated fats. Choosing foods with less cholesterol and saturated fat will help lower your blood cholesterol levels. The Nutrition Facts Label lists the Daily Value for cholesterol as 300 mg. You can keep your cholesterol intake at this level or lower by emphasizing intakes of grains, vegetables, and fruits and by limiting intake of egg yolks, including those used in cooking.

Here is the scientific report of the 2015 guidelines:

https://health.gov/dietaryguidelines/2015-scientific-report/...

> Cholesterol: Previously, the Dietary Guidelines for Americans recommended that cholesterol intake be limited to no more than 300 milligrams per day. The 2015 DGAC will not bring forward this recommendation because available evidence shows no appreciable relationship between consumption of dietary cholesterol and serum cholesterol, consistent with the conclusions of the AHA/ACC report, Cholesterol is not a nutrient of concern for overconsumption.

The sad part, is that there was never any good evidence to limit consumption of cholesterol and eggs.

Re: The rise and fall of scientific authority, and how to bring it back

#292

Post by hyperion2010 » Tue, Mar 19, 2019, 2:06 AM UTC

I've been thinking about this piece all day and I can finally articulate what my problem is with it.

Science is the rejection of authority in the 'authorial' sense of the word. Pre-scientific thinking relied almost entirely on authorities to determine what was right or true. Science is thus the rejection of the notion that an authority of any kind is a substitute for evidence.

Scientific knowledge, when presented as an authority is thus just as abhorrent as text from any other authority. We scientists trust it because we often have no other choice, but also because over decades of training we have all done the labs to actually see these principles operate first hand. Thus we accept the principles based on our own experience of them.

I hate to say it, but if someone hasn't had at least some exposure to a laboratory setting of some kind, or experimentation of some kind, then the fact that scientific knowledge is "right" is completely lost on them, and it is indistinguishable from any other knowledge from authority. Belief in the scientific process might be a substitute, but at that point it is no different that belief in revealed truth -- right for the wrong reasons means failure is just as likely.

Re: The rise and fall of scientific authority, and how to bring it back

#293

Post by tomrod » Tue, Mar 19, 2019, 2:06 AM UTC

charliesharding wrote: post #5

The way in which public considers science today is much more analogous to the way the public considered religion during the time of Galileo. Think of the way people are castigated for not believing in scientific "truths". I see signs in peoples yard saying that "in this house, we believe science is real". As someone who has studied biochemistry for years, this kind of reverence for science is truly troubling. The one…

People like to be right, and don't like to invest time in thought. Hence why the bandwagon effect works.

Re: The rise and fall of scientific authority, and how to bring it back

#294

Post by asabjorn » Tue, Mar 19, 2019, 2:08 AM UTC

Maybe nature should first take responsibility for their own role in discrediting science before asking others to do so, and stop publishing articles on discredited takes on complex problems such as the gender pay gap [1].

There is no excuse for a scientific magazine to publish articles with methodology that has been shown to misrepresent the problem at hand so many times. It certainly doesn’t help people’s trust in science when they do so.

[1] https://www.nature.com/articles/d41586-018-04309-8

Re: The rise and fall of scientific authority, and how to bring it back

#295

Post by foobarian » Tue, Mar 19, 2019, 2:11 AM UTC

joe_the_user wrote: post #150

balabaster wrote: post #124

Earlier quoted context omitted.

I believe this is a public perception problem. The reason that a troubling number of people no longer believe in science isn't that scientific method is flawed or that people don't believe in scientific method. The bigger concern is the way that scientific studies are mounted, paid for by interested parties that have conflicts of interest to produce results favourable to a desired outcome and then they're presented a…

I would say that an even larger problem exists. Among the public, science has always had some portion of people who understood it's processes as those of a secular priesthood and those who understood it's processes as discovering the truth through skepticism and experimentation. Even among this working scientists, this division has existed (and given a society that never discarded religion, anti-science along with an…

> the other aspect is the majority of people aren't seeing the material gains that science offers the world

I'm afraid this last bit is a big problem. I suspect there is a certain lag from discovery to perception. And then starting in the 1800 we had wave after wave of discovery, from electricity, magnetism, cars, airplanes, wireless, nuclear bombs, computers... stopping around the 1980s. At this point we can keep talking up science all we want but I fear something like fusion reactors or true AI would be needed to boost the faith.

Re: The rise and fall of scientific authority, and how to bring it back

#296

Post by asabjorn » Tue, Mar 19, 2019, 2:15 AM UTC

charliesharding wrote: post #5

The way in which public considers science today is much more analogous to the way the public considered religion during the time of Galileo. Think of the way people are castigated for not believing in scientific "truths". I see signs in peoples yard saying that "in this house, we believe science is real". As someone who has studied biochemistry for years, this kind of reverence for science is truly troubling. The one…

The argument that science is a social construct is easily disproven by the device you are typing this on, which is build upon science. However, the claim that some science is more believable than other science is obviously correct.

Science is a network of knowledge [1] where central nodes can be assumed to be more likely an accurate representation of something predictive. The literature critical theory claim that it is arbitrary does not acknowledge those relationships, effectively missing the forest for the trees which also explains the shoddy science done in critical theory’s name.

Science is progressed by building new leaf nodes and supporting them through other nodes as evidence, or by showing that core nodes are not as predictive as you thought they were.

[1] https://youtu.be/bxdBRKmPhe4

Re: The rise and fall of scientific authority, and how to bring it back

#297

Post by tannerbrockwell » Tue, Mar 19, 2019, 2:32 AM UTC

asabjorn wrote: post #296

charliesharding wrote: post #5

The way in which public considers science today is much more analogous to the way the public considered religion during the time of Galileo. Think of the way people are castigated for not believing in scientific "truths". I see signs in peoples yard saying that "in this house, we believe science is real". As someone who has studied biochemistry for years, this kind of reverence for science is truly troubling. The one…

The argument that science is a social construct is easily disproven by the device you are typing this on, which is build upon science. However, the claim that some science is more believable than other science is obviously correct. Science is a network of knowledge [1] where central nodes can be assumed to be more likely an accurate representation of something predictive. The literature critical theory claim that it…

When you point to point to your computer, you are confusing a 'Technology' with Science. They are different. As for first principles, starting with Francis Bacon's Baconian method is the investigative method that would benefit here: [1]

"Bacon suggests that you draw up a list of all things in which the phenomenon you are trying to explain occurs, as well as a list of things in which it does not occur. Then you rank your lists according to the degree in which the phenomenon occurs in each one. Then you should be able to deduce what factors match the occurrence of the phenomenon in one list and don't occur in the other list, and also what factors change in accordance with the way the data had been ranked."

Thus we see that predictive is less important than reproducibility, which is currently in this era approaching a crisis point. [2]

[1] https://en.wikipedia.org/wiki/Baconian_method [2] https://en.wikipedia.org/wiki/Replication_crisis

Re: The rise and fall of scientific authority, and how to bring it back

#298

Post by asabjorn » Tue, Mar 19, 2019, 2:53 AM UTC

tannerbrockwell wrote: post #297

asabjorn wrote: post #296

Earlier quoted context omitted.

The argument that science is a social construct is easily disproven by the device you are typing this on, which is build upon science. However, the claim that some science is more believable than other science is obviously correct. Science is a network of knowledge [1] where central nodes can be assumed to be more likely an accurate representation of something predictive. The literature critical theory claim that it…

When you point to point to your computer, you are confusing a 'Technology' with Science. They are different. As for first principles, starting with Francis Bacon's Baconian method is the investigative method that would benefit here: [1] "Bacon suggests that you draw up a list of all things in which the phenomenon you are trying to explain occurs, as well as a list of things in which it does not occur. Then you rank y…

I agree reproducibility is an issue, especially in psychology that is slowly becoming a science right now. There it’s actially a good sign, since now psychology has a methodology we can use as a standard bearer to spot issues. In other humanities that like to call themselves science we unfortunately even lack such a methodology.

From first principles technology is definitely applied science through engineering, but technology is progressed through pushing the boundaries of science and science is the map of where to look. Eg quantum mechanics and optics helped create a map of where to look for new nano scale technology innovations. Technolofy shows the reproducibility of the theories it’s build upon.

The computer is build by innovations created using the predictive powers of material science, quantum mechanics and chemistry.

How do you think the optics used to create the processors were created? The advanced etching technologies to get to decreasing nm processor scale? Materials that respond reliably to store energy in batteries or power puxels in your screen?

Re: The rise and fall of scientific authority, and how to bring it back

#299

Post by lutorm » Tue, Mar 19, 2019, 2:56 AM UTC

charliesharding wrote: post #5

The way in which public considers science today is much more analogous to the way the public considered religion during the time of Galileo. Think of the way people are castigated for not believing in scientific "truths". I see signs in peoples yard saying that "in this house, we believe science is real". As someone who has studied biochemistry for years, this kind of reverence for science is truly troubling. The one…

"The one thing you learn in high level science is that nobody really knows what they're talking about, paradigms shift, and further study is required."

That's true, but it's the best we've got.

Not believing in science is different from disagreeing with conclusions of studies based on data. But just saying "science is flawed so I'm not going to believe it" makes no sense at all, because all other alternatives are worse.

Re: The rise and fall of scientific authority, and how to bring it back

#300

Post by evrydayhustling » Tue, Mar 19, 2019, 2:59 AM UTC

> Contemporary science deniers have not one (religious) motive, but many — greed, fear, bias, convenience, profits, politics — to which they cling with various degrees of sincerity and cynicism.

While doubtless describing many, this is a straw man when it comes to broad mistrust of scientific expertise. Issues like psychology's replication crisis and the continuously flip flopping advocate-science in areas like nutrition make people justly suspicious of the scientific establishment, regardless of how they would feel about an ideal scientific process.