(C) Daily Kos
This story was originally published by Daily Kos and is unaltered.
 . . . . . . . . . .



A Defense of Science - Part 1 - How to "Do Science" [1]

['This Content Is Not Subject To Review Daily Kos Staff Prior To Publication.']

Date: 2024-09-02  

This diary isn’t about conducting scientific experiments, it’s about what’s necessary to prepare to conduct scientific experiments, what’s necessary to be taken seriously, and why things are the way they are in science nowadays. It also contains ideas about non-experts using science and the level to which they need to understand it.

(It’s a longish essay so maybe you’ll want to jump straight to The Auto Mechanic Theory of Skepticism near the end. It was the most fun part for me to write. :-) )

A Defense of Science: An Essay of at Least Two Parts

Part 0 – Introduction and Glossary

Part 1 – The Practice of Science

Introduction

7 decades after his death, Albert Einstein remains the iconic figure of science. Ask anybody on the street to name a famous scientist and there’s a very good chance that his name will come up even if they couldn’t specify why he earned his fame. While “He was really, really smart” is not inaccurate, it fails to explain his fame as there have been plenty of highly intelligent people throughout history who have made significant contributions to society yet who don’t have a hundredth the name recognition of Einstein (no citation needed, I’m sure).

Einstein’s fame is due to the paradigm-shifting contributions he made to science—contributions he made as an expert practitioner of science. Popular lore correctly recalls where Einstein was in 1905, the legendary scientific annus mirabilis rivaled only by Isaac Newton in 1666: He was employed in the civil service as a patent clerk in Switzerland. What popular lore often fails to note is that Einstein wasn’t working in a vacuum and that he wasn’t shunned by the scientific community because his genius supposedly wasn’t recognized by the stick-in-the-mud authorities of the time. Einstein was educated by well-regarded professors such as Hermann Minkowski at ETH Zurich, where Einstein did the equivalent of a Bachelor’s degree. He went on to earn his PhD from the University of Zurich, officially receiving the degree in 1906 working under the then-chair of the Physics Department, Alfred Kleiner. Einstein had a solid understanding of the process of science and he worked within the system to produce his vast body of work including his famous theories of Special Relativity and General Relativity.

Yet saying that Einstein “worked within the system” goes against the conventional wisdom in which a lone genius is oppressed and suppressed and repressed (and maybe threatened to get compressed!) for challenging the monolithic Ivory Towers that supposedly run all of science. Galileo, Edison, and Tesla are cited as other examples of visionary individuals working outside of the fold, being shunned for having the temerity to introduce new ideas that upset the status quo. But popular beliefs about the lives of these people have been suffused with a level of romance that substantially distorts what really happened. It is absolutely true that there are scientific innovations we accept nowadays that were once considered on the fringe, promoted by a small minority—or even by individuals—who had to swim upstream against a strong current to get their ideas widely accepted. Examples include the Theory of Continental Drift, which flitted about the rafters of geological science for several centuries before morphing into the currently-accepted Theory of Plate Tectonics, and Ignaz Semmelweis’ then-controversial hypothesis that the rate of post-operative infections in surgical patients would be drastically reduced if surgeons would simply disinfect their hands. It’s also true that some scientists were repressed or brutally ridiculed for advocating seemingly outlandish ideas. But such stories don’t simply highlight flaws in the practice of science, they also demonstrate some of its strengths because these ideas were still allowed to air and, more importantly, because they show the scientific process allows for correction.

Prerequisites to Practicing Science

Beyond employing the Scientific Method, what does the practice of science entail? First and foremost, just like in any other career there needs to be a significant amount of training. There is virtually no chance of some ordinary joe or jane walking off the street and become a starting running back in the National Football League. The same chance applies to the scenario of someone with no documented education publishing results from truly ground-breaking scientific research. A would-be scientist needs to spend years studying the basics of the discipline, following the work done by thousands of scientists going back to the beginning of the Enlightenment Era in Western civilization. They must become very familiar with the latest results in their discipline of choice, another task usually requiring years. Parallel to that, they need to develop an understanding of mathematics and the profound analytical tools it provides. It’s also quite likely that scientists need to develop computer programming skills and it’s a guarantee that scientists must have practical communication skills if they wish to convey their ideas.

A Historical Interlude I

Let’s consider physics. While the study of this subject goes back to ancient times, the origins of the modern understanding of physics arguably has its seeds in Nicolaus Copernicus and his work De revolutionibus orbium coelestium which proposed the heliocentric model of the solar system. (Copernicus was not the first to propose a Sun-centered solar system—Aristarchus in ancient Greece, the Indian astronomer-mathematician Aryabhata in the first millennium CE, and al-Sijzi during the Islamic Golden Age all proposed heliocentric models. But Copernicus had the benefit of publishing his work after the invention of the printing press a century earlier.) The idea of the heliocentric model stimulated the work of Tycho Brahe and Johannes Kepler and Galileo Galilei. These three intensely observed and analyzed the motion of the Moon and the planets to understand in detail the motion of celestial objects. The prolific Galileo also began the (literally) down-to-Earth study of kinematics, how things move. He famously concluded that heavy masses fall towards the ground at the same rate as light masses (ignoring air resistance), contradicting ideas established nearly 2000 years earlier by Aristotle.

Building on the work of these scientists and giving birth to calculus in the process, Isaac Newton developed a unified theory of gravitation in which he showed that the celestial motion of the planets and the terrestrial motion of objects near the surface of the Earth were all due to a single cause. The brilliance of uniting what were thought to be disparate phenomena helped fuel a wave of ground-breaking scientific discoveries in the 18th and 19th centuries. Experiments in Optics, Thermodynamics, Fluid Dynamics, and Electromagnetism were conducted and reconducted, theories were proposed and scrapped and replaced by new theories, and arguments for and against these theories were dissected to their smallest details. By the end of the 19th century, these topics had been so thoroughly scrutinized that it seemed physicists could explain 99% of Nature’s fundamental laws and that finally accommodating the remaining 1% was just a matter of time and plain old boring and painstaking attention to detail.

Back to Prerequisites

A would-be physicist will spend at least a year in a survey of the work done through the end of the 19th century. But that survey only scratches the surface. Over the course of several years, an undergraduate physics major will be exposed to more mathematical techniques that allow for new insights to ever more complicated circumstances. For example, in the first year a student might study the motion of a simple pendulum swaying back and forth over a small distance. In a later course, they would learn the numerical techniques necessary to study a simple pendulum swaying back and forth over a large angle. Furthermore, they would likely examine the motion that results from having a second pendulum attached to the mass at the end of the first pendulum (a double pendulum; friendly advice—don’t click on that link if you are the least bit math-phobic). While there may only be limited practical applications to these specific devices, a physics major needs to develop the skills necessary to investigate more sophisticated situations. It’s the equivalent of a baby learning to roll from back to front, before learning to scoot, before learning to crawl, before learning to walk while holding onto furniture, before learning to walk independently, before learning to run, before learning to sprint, before competing in races against others, before vying for an Olympic gold medal in the 100-m dash. It takes years.

At the end of these years of training, it’s extremely likely nowadays for a scientist to have an advanced degree, at least a Master of Science or, more typically, a Doctor of Philosophy, from an accredited educational institution. It is entirely possible for a person with no such degree to be a functioning research scientist. But it’s also entirely possible that a random person is capable of performing dentistry—personally, though, I wouldn’t blindly trust some guy walking around on the street outside to fill a cavity in one of my teeth.

A Historical Interlude II

Not every scientist directly engages in hands-on observation and experimentation. But every scientist worthy of that title must know how proper experimentation is done (and every scientist should deeply understand how theories are developed). This is true even if a scientist is strictly a theorist, never once conducting a single empirical measurement. This contrast between experimentalists and theorists goes back to the days of Brahe and Kepler. Tycho Brahe made extraordinarily painstaking naked-eye observations of the motion of the planets relative to Earth. Brahe bequeathed the data to Johannes Kepler, who had theorized that the planets orbited around the Sun along circular paths. To his distress, Brahe’s data didn’t fit a perfectly circular path and Kepler ultimately was compelled by the data to theorize an elliptical path for planetary motion. Kepler’s theory enabled Isaac Newton to generate a more robust theory for planetary motion and astronomers subsequently used Newton’s theory to aid in the discovery of the planet Neptune in 1846. In the healthy practice of science, experimentation feeds theories and theories feed experimentation.

Now We Have an Expert. So What?

After years of training, a freshly-minted PhD scientist is ready to make their name. Perhaps they will start their investigations on their own or maybe they have or can develop connections and can become part of a collaboration with other scientists. Whatever the situation, they need to establish themselves in an environment that allows for serious scientific investigation. Colleges and universities and publicly-funded national labs are places likely to have the necessary infrastructure for this but there are also many corporations that vigorously fund and encourage cutting edge research.

Over time, an active researcher will achieve results they feel are noteworthy enough to share with others. The various frameworks that allow for this are open to the public—the results aren’t intentionally hidden. Admittedly, many topics tend to be so esoteric that the ideas are readily available only to those who are experts in the same discipline as the author. But it is in the best interests of scientific practice that anybody can have access to these results. This is especially true in instances where support for the research comes at least partly from public funds, e.g., government grants.

Peer-Reviewed Journals

The most formal and effective way of presenting results is through publication in a peer-reviewed journal. In this method of dissemination, the author submits a manuscript they prepared to the editor of a journal the author feels provides an appropriate venue for their work. Assuming the manuscript is otherwise in order, the editor will farm it out to other experts in the field, often those who have already published in the same journal. These experts will examine the manuscript and look for issues that could bring it into question. They double-check that the references are accurately cited and should be convinced that the author has done a search of prior work and isn’t just duplicating prior results. The experts will also determine whether the contents of the manuscript are truly appropriate for the journal in question. The editor will ask these experts to render their judgment on whether the manuscript ought to be published and explain their reasons. There can be three options: 1) Yes; 2) No, the manuscript fails to meet the journal’s established standards or is not appropriate for the journal; 3) Perhaps, but the author needs to make some changes such as providing some additional information in the citations or about how the research was conducted. The editor compiles these reports and then sends that information on to the author (removing the names of the experts for the sake of anonymity). If the answer is no, the editor will usually explain why the manuscript was rejected.

This process is not foolproof. Perhaps the editor plays favorites or the reviewers are less than diligent in their work. Maybe the author has committed fraud that slipped past the reviewers. These situations have gained some notoriety but they are rare exceptions with well-proctored journals as it’s very much against the journals’ and experts’ own interests. If the operation of a journal is shown to be corrupt via negligence or cronyism or whatever, it loses credibility in the scientific community. A journal that loses credibility subsequently means that the authors who published in the journal have their work debased and they risk losing prestige which means less of a chance for promotion and receiving grants to continue their work. As a result of having so much on the line, the rate of corruption in the system of peer-reviewed journals, particularly the most prestigious, is low.

A Historical Interlude III

At the dawn of the Enlightenment Era in Europe, investigations of nature began to prove themselves worthwhile to the rulers of the time. Great prestige could come to those scientists who generated new insights into astronomy or medicine or who developed new methods or instruments for use in such vital endeavors as agriculture and navigation. Those who came up with something truly novel would be showered with praise and, more importantly, significant largesse from royal benefactors. Perhaps the most overt example of this largesse was the castle built for the research purposes of Tycho Brahe, named “Uraniborg”. The construction of Uraniborg was estimated to have cost approximately 1% of the entire budget of the Kingdom of Denmark, which also provided the necessary upkeep for 20 years.

As a result of the potential largesse, scientists of the time were extremely hesitant about sharing their ongoing results with others—it was in their best financial interests to announce their discoveries as faits accomplis. So secret were these research programs, the detailed star chart Tycho Brahe and his assistants developed wasn’t fully released to the public until 1627, more than a quarter century after Brahe’s death!

It became clear to many that this method of hiding results from others was extremely detrimental to the practice of science. As a response to this situation, in 1665 the first Secretary of the Royal Society, Henry Oldenburg, began the publication of the Philosophical Transactions of the Royal Society. Oldenburg promised that the full weight of the Royal Society would support the claims of precedence made by authors of articles published in the journal. This method of presenting results even during the intermediate stages of research gained traction and became a standard in the scientific community. Other journals were initiated and began following the same practices.

The editors of the various journals were primarily responsible for reviewing submitted articles and determining the relevance to the journal’s readers. This became onerous. More and more, the editors began to farm out the process to experts well-known to them though the editors were still dominant in deciding the contents. As such, the peer review process was somewhat haphazard and prone to the predilections of the editors and their inner coterie of expert reviewers. This was complicated by the pace of scientific discovery, which soon reached a point where you could only claim true expertise in a particular field rather than being a polymath knowledgeable in all fields. But it was only over the course of the 20th century that the modern approach to peer review, using anonymous expert reviewers, began to become widespread among the major journals with the less prestigious journals following suit. Even nowadays, the system of peer review is constantly scrutinized and suggestions for improvement are commonplace. For instance, people selected as experts to review papers will generally not have any meta-knowledge about effective methods of peer review. A set of more knowledgeable and experienced reviewers has been shown to improve the breadth of authors submitting quality articles to a journal’s published articles.

Conference Proceedings, Seminars, and Colloquia

Somewhat less formal ways of disseminating the results of a research program are presentations at professional conferences and invited talks at individual institutions. Peer review in these circumstances is less stringent though a poorly delivered talk will certainly engender criticism. Conferences commonly consist of many different sessions of topics within a given discipline or sub-discipline. Some of these sessions are comprised of invited speakers—the organizers of the conference have identified certain experts as having something particularly worthwhile and relevant and want them to give presentations. Other sessions are more open. Anybody or any group can submit an abstract describing the presentation they would give. There is an element of peer review as members of the organizing committee sift through these requests and identify those that come across as professional and adhere reasonably well to the topics at hand. But this peer review is typically not as strict as found in professional journals.

Poster presentations are another common venue at conferences. Presenters will be allowed a space to display posters that summarize their work. There is usually a lot of detail on a poster: an abstract, a discussion of the hypothesis and any experimentation, a brief analysis focusing on the results and often including graphs or other pictures, a conclusion, and some acknowledgements. A poster presentation is otherwise static though it’s common for someone associated with the poster to be nearby and engage in discussions with any interested viewers. Poster sessions aren’t nearly as subject to the same level of peer review as potential oral presentations but are valuable for providing a cafeteria of ideas and for giving novice researchers their first exposure to developing presentations for a professional audience (remember the reference to practical communication skills made earlier?).

Research and educational institutions will frequently invite guests to come deliver talks. These guests are often collaborators or otherwise an acquaintance of someone at the institution; in other cases, they could be talks given by prospective employees applying for a position. Such talks are offered based on the prior work the invitee has performed and published so, in a sense, a level of peer review has already occurred.

Bertrand Russell on the Practice of Science for a Layperson

Obviously, not everyone is a trained research scientist—society just might collapse in a heap if that were the case. But science plays a huge role in modern society whether or not we consciously recognize it. And the science that has creeped into our day-to-day lives has undergone rigorous testing and deconstruction and reconstruction by well-trained experts so as to understand it as thoroughly as possible. At a superficial level, you may think of your cell phone as a “black box”, seemingly impervious to understanding and effectively operating under mysterious principles. But if you put even a little bit of thought into it, you know there are trained experts who are fully versed in the design and are continually working on improvements.

Given all that, how important is it for a non-expert to understand the scientific details of a particular phenomenon? The answer is that it is not at all important provided two conditions are satisfied:

Being able to understand the phenomenon is not important. Most peoples’ lives go on whether or not they understand how the Nernst Equation applies to batteries. As long as the people who design the batteries understand how to apply the Nernst Equation, all is well. Ignorance of science is not a sin if it’s not important to someone. Experts* within the scientific community have put in the effort to understand the phenomenon and can provide explanations as needed.

* - To be an expert requires extensive study and training as discussed above. This means, for example, that a physicist—even a Nobel Prize-winning physicist—should not automatically be considered an expert in biology (page no. 36 in link). Furthermore, an astrophysicist who wins the Nobel Prize in Physics for a discovery in stellar processes should not automatically be considered an expert in some other field of physics such as condensed matter. Such a physicist will almost certainly know more about condensed matter than the average person, but that has limitations when it comes to cutting-edge research.

What about when the science is important to someone who is not an expert? Someone can read up on the subject but to become a true expert requires a huge investment of time and practice. Since no one person can become an expert in all areas of science, it becomes necessary to listen to those who are experts in the field in question. What’s the best approach?

In his work On the Value of Scepticism, the philosopher-mathematician Bertrand Russell suggested the following:

“1. When the experts are agreed, the opposite opinion cannot be held to be certain.

2. When they are not agreed, no opinion can be regarded as certain by a non-expert.

3. When they hold that no sufficient grounds for a positive opinion exist, the ordinary man

would do well to suspend his judgment.”

The phrasing of the first maxim is interesting—it doesn’t demand that the layperson simply agree with the expert opinion, but it does strongly suggest that going against the experts is a silly idea. Anybody who holds to the opposite opinion of the experts and wants to be taken seriously had best have some compelling reasons that support their rationale, that credibly undermine the rationale of the experts, and that can effectively address the counterarguments the experts make against the opposite opinion. (If a person is truly capable of rising to these tasks, it would be hard to call them a “layperson”.)

A complication can arise because, depending upon the subject, there can be many people who have a valid claim of expertise. In a sufficiently large demographic, there is a high likelihood for outlying opinions. So, it’s not terribly unlikely that there are nominally qualified experts who go against, say, 95+% of the other experts on a particular matter. Does that put Russell’s first maxim in jeopardy? If there is such an imbalance in opinion among the experts, the interested layperson needs to apply some healthy rational skepticism. It’s ok to examine the contentions of the minority of experts—argumentum ad populum is a logical fallacy that one should avoid. But one shouldn’t stop there no matter how tantalizing or convincing that minority opinion is. It is necessary to check for the counterarguments made by the 95+% of experts; on important issues, it’s almost guaranteed these counterarguments will exist some place. After all this, the layperson has to come to a conclusion but not before some self-introspection and honesty while asking themselves how well they understood the pros and cons. It’s quite likely that an honest answer to that is, “Not as well as the experts.” If this is the layperson’s answer, then Russell’s first maxim survives. While it seems that there is disagreement among the experts and Russell’s second maxim should apply, such a pronounced imbalance tilts the situation towards the first. Admittedly, there’s no hard boundary between “balanced” and “imbalanced” but when it’s 95% compared to 5% it’s reasonable to side with the 95%. If, on the other hand, the layperson feels genuinely certain they understood the pros and cons of each side, then they should render a judgment that doesn’t take their biases into account.

The Practice of Science for a Layperson and the Auto Mechanic Theory of Skepticism

To put this in everyday terms, let’s consider the Auto Mechanic Theory of Skepticism. Suppose you think you might have problems with your car and you’re not very knowledgeable about the inner workings of them. You take your car to a mechanic to get their opinion. But you’re a cautious person who wants to spend their money wisely. So, you get a second opinion and maybe a third. For the sake of argument, let’s extrapolate—you see 1000 different auto mechanics about your car and get a diagnosis from each of them. You tally up their diagnoses and get the following:

# of Mechanics Diagnosis Consequence 973 Transmission Problems “Uh-oh.” <Cha-Ching!> 20 There isn’t anything wrong “I can keep driving? YAY!” 4 Something vague about its horn “Ummmm, really?” 3 Older models had fuel line issues “What year do you live in?”

The mechanics who bring up the horn are clearly incompetent. The mechanics who talk about the older models might have been correct about those cars but you have a newer model so they’re not up to date on things. The vast majority of mechanics say there’s a transmission problem but that means a major hassle and you really, really, really don’t want to spend the gobs of money it would inevitably take to repair it. The claim by those 20 mechanics that there’s nothing fundamentally at issue is awfully tempting because there’s no disruption in your life and it costs a grand total of zero dollars. But if those 20 mechanics are right, why would 973 mechanics give you such erroneous information? Let’s consider the most plausible reasons for this:

The 973 mechanics are in the business to make money, so they’ll tell you exactly what will get them the most bucks. The 20 mechanics who say there’s nothing wrong are altruistic and are the only ones who are honest about things.

The 973 mechanics are kind of wishy-washy and subject to peer pressure—they think most other mechanics would say it’s the transmission and nobody wants to look stupid so they’re just going with the flow. The 20 mechanics who say there’s nothing wrong are the only ones who have the courage and self-respect to be honest about things.

The 973 mechanics are scared—if those mechanics don’t say it’s the transmission, then thousands of agents from Deep Transmission, who are so good at hiding that there’s no concrete evidence of their existence, will find ways to ruin their business. The 20 mechanics who say there’s nothing wrong are the only ones courageous enough to have seen through this vast conspiratorial scam and can be honest about things.

The 973 mechanics say it’s the transmission because they will get into the fabulously luxurious Transmission Mechanics Club which flies their jet-setting members on first-class flights to exotic locales like Bali for week-long “conferences” about transmissions. The 20 mechanics who say there’s nothing wrong are above such temptations.

The 973 mechanics are simply ignorant for reasons you don’t understand. You aren’t a car expert yourself so you aren’t qualified to judge this, but your gut instinct is that it’s only the 20 mechanics who say there’s nothing wrong who truly understand the subject and are honest about things.

The 973 mechanics are flip-floppers. With earlier models of cars, those same mechanics would have said there’s a fuel line problem but now they’ve flip-flopped to saying it’s the transmission. You obviously can’t trust them. It’s only the 20 mechanics who say there’s nothing wrong who have steadfastly held true to the same diagnosis and are honest about things.

After considering these possible reasons, you have to make a responsible decision as a car owner. If you choose to fix the transmission then you have to deal with the hassle and expense. On the other hand, if the 973 mechanics are right but you do nothing about it then there’s a real chance that you could suffer a catastrophic malfunction while on the highway. Even in the best-case scenario, you could find your car doesn’t start one morning and you’d have to scramble to find alternative transportation at the last minute in order to get to work (and you’d still have to pay for the transmission repair plus have the car towed). Maybe, just maybe, that 2% of all mechanics are right about the car being fine but how much are you willing to stake on that?

Possible Upcoming Essays

The Malpractice of Science

At the Intersection of Science and Society: Religion

At the Intersection of Science and Society: Politics

At the Intersection of Science and Society: The Future

(Maybe others depending upon the exertions of my fevered imagination)

[END]
---
[1] Url: https://www.dailykos.com/stories/2024/9/2/2267398/-A-Defense-of-Science-Part-1-How-to-Do-Science?pm_campaign=front_page&pm_source=more_community&pm_medium=web

Published and (C) by Daily Kos
Content appears here under this condition or license: Site content may be used for any purpose without permission unless otherwise specified.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/dailykos/