by David H. Berkbile

    The term "settled science" has been applied to many things in the last few years, global warning or climate change among them.  The term suggests that meaningful debate on a particular point has closed, that consensus has been reached. But what does the term actually mean?

     Consider the example of "phlogiston," from the history of physics. The theory was that substances that burned in air were rich in phlogiston.  Burning was loss of the phlogiston content. When something was burned in an enclosed space, if there was enough material, it would cause the air to become completely phlogisticated, so it would no longer support combustion of any material.  The air was so saturated with phlogiston that it would no longer accept phlogiston from any material (Conant).  Breathing was thought to take phlogiston out of the body. 

    The concept of phlogiston was first proposed in 1667 by Johann Joachim Becher in his book Physical Education ("Phlogiston Theory"; "History of Chemistry: Robert Boyle"). Becher originally gave his concept the name "terra pinguis" rather than phlogiston (Becher 256); however, in 1703 Georg Ernest Stahl, professor of medicine and chemistry at Halle, proposed a variant on the theory and he renamed the substance phlogiston (Williams, Book 4, Chapter 1).

    Thousands of experiments were done to determine the amount of phlogiston in all sorts of materials, and large volumes were written containing the results of these experiments.  The experiments were repeated many times to improve the accuracy of the data, so you could look up the amount of phlogiston in red oak, white oak, hickory, pine, silk, cotton or almost anything that would burn in air.  Some scientists devoted much of the experimental work of their careers to expanding this data set and improving the accuracy of the numbers.  Coal turned out to be pure phlogiston because when it was burned there was no ash left.

    Phlogiston was settled science.  However, in 1753 Robert Boyle and then Mikhail Lomonosov burned a metal, and the ash was discovered to be heavier than the original metal. Lomonosov concluded that the phlogiston theory was wrong ("Lomonsov, Mikhail").  As far as the rest of the world was concerned, though, phlogiston was settled science, and scientists who had devoted much of their careers to researching phlogiston were not about to abandon this theory easily.  Some even created a negative phlogiston theory to explain away the troubling experimental results and the phlogiston theory continued to be used despite the data suggesting it was wrong.

    Antoine-Laurent Lavoisier's work on oxidation eventually replaced phlogiston as the theory of combustion (American Chemical Society).  Lavoisier did not start his work on combustion until late 1772, and it was not until 1783 that he published his full-scale attack on the then current phlogiston theory of combustion.  Phlogiston was settled science for over 100 years.

    An example of settled science with which all technically trained people are familiar is Newtonian physics.  In 1687, Sir Isaac Newton published Mathematical Principles of Natural Philosophy (to use its English title), said to be the most influential book in physics and possibly in all of science.  It identifies three basic laws that quantitatively describe bodies in motion. 

1. A stationary body will stay stationary unless an external force is applied to it, and the corollary, a body in motion will stay in motion unless a force is applied to change its motion. 
   
2. Force is equal to mass times acceleration so a change in motion is proportional to the force applied. 
   
3. For every action there is an equal and opposite reaction.  (Newton)

Newtonian physics remained settled science for centuries, and is still useful in very many settings.  I taught physics in the 21st century, and I taught the laws of Newtonian physics.  In Newtonian physics we know what motion is: the distance traveled divided by the time taken to travel that distance, or, in equation form: ∆d/∆t.  In the early 20th century, though, a Swiss patent examiner came up with a new and different theory.  His name was Albert Einstein.

    By Einstein's time, scientists had developed a very ingenious method of measuring the speed of light ("Michelson and Morley"). Researchers were using this method to determine the speed of light from stars.  In one set of experiments, light coming from a star at one edge of a galaxy that was spinning away from earth was measured, and light from a second star on the other side of the same galaxy that was spinning towards earth was also measured.  The measured speeds were the same, so scientists were trying to figure out why the experiments were wrong.

    But why did scientists assume there must have been an error? Think about speed in Newtonian physics.  Assume you have a person who is capable of throwing a ball at 60 mph.  If he was standing on a flatbed railcar in a train was moving at 60 mph, and threw the ball forward, and you were standing beside the tracks and caught this ball, it would feel like the ball was moving at 120 mph.  On the other hand, if he attempted to throw the ball towards the back of the train, to you, the stationary observer beside the tracks, it would look like the ball only stopped its forward motion and fell straight down.  Accordingly, Newtonian physics told them that the experimental results in measuring the speed of light had to be wrong. 

    Einstein, who became known for his thought experiments, asked himself this question: "What if the experiment was not wrong?"  

     How could the speed of light (∆d/∆t) be a constant (i.e. the same for light from both stars)?  What if distance and time are two components of the same thing?  Einstein developed the Special Theory of Relativity in 1905 and later the General Theory of Relativity.  In this theory both space (length, width and height) and time are dimensions of the universe within which we live.  We now call it space-time.

    When I took college physics, over 50 years after Einstein proposed relativity, we spent a lot of time studying Newtonian physics, but the Theory of Relativity was still not included in the syllabus. When I taught physics, I taught the Theory of Relativity late in the school year, and I would correct some of the equations of Newtonian physics. Today we accept that we live in space-time, and we have tested the theory of relativity in experiments on the space station. We now know the centuries old "settled science" of Newtonian physics was not settled at all.

     Nutrition is a more recent example. Many can remember when the "Food Pyramid" and dieting with a "calories-in/calories-out" logic were considered known (or settled) science.   How many calories you consume minus how many calories you use up was the key to weight control; to get these calories, one was supposed to eat so as to achieve a dietary balance in accord with the food pyramid.

    When Robert Atkins suggested that the system was more complex and created a new diet, he was ridiculed (Atkins).  Today, however, many diets follow some of Atkins' diet concepts.

    The calories-in/calories-out concept should not have made sense to chemically trained people.  Calories in food are determined by burning the substance in a calorimeter.  The idea that the amount of energy the body obtains from a food is accurately measured by burning that food in a calorimeter looks foolish when you consider starch and cellulose.  Starch, a major component of your potatoes, bread, cereals, rice, etc., accounts for a lot of the calories in many diets around the world.  Cellulose, as in wood, is not even digestible by humans or most animals.  Eating cellulose will give you no calories.  Cellulose is a part of plants we eat and it provides for part of the dietary fiber we need, but it passes through our systems basically unchanged.

     But look at the chemical structures.  Starch and cellulose are basically stereoisomers of each other. 

        (Fleser and Fleser, 962-63)

When burned in a calorimeter, they will give off equivalent amounts of energy. Their effects on the human diet, however, are quite different. The old calories-in/calories-out theory was flawed thinking.  Our bodies don't work like calorimeters.  Our bodies will digest some stereoisomers and not digest others (Study.com; Seattlepi.com; Breaking the Vicious Cycle website).

    In an opinion page column on this subject, Charles Krauthammer wrote:

When the federal government's 1980 "Dietary Guidelines for Americans" warned about the baleful effects of saturated fats, public interest activists joined the fight and managed to persuade major food companies to switch to the shiny new alternative: trans fats. Thirty-five years later, the Food and Drug Administration finally determined that trans fats are not just useless but unsafe, and ordered them removed from all foods. Oops.  So much for settled science. (Krauthammer)

Now the government has new recommendations to replace the old Food Pyramid and more changes are in the works.  A May 2015 article reported that the US Department of Agriculture Dietary Guidelines for Americans, which previously recommended that cholesterol intake be limited to no more than 300 mg/day, is being changed because evidence shows no appreciable relationship between consumption of dietary cholesterol and blood cholesterol. The article included this quotation: "US cardiologist Dr. Steven Nissen said, '—We got the dietary guidelines wrong.  They've been wrong for decades'" (Barr; USDA).

    When I was in college, my research advisor asked us to read Martin Gardner's book Fads and Fallacies in the Name of Science. Gardner made fun of things that he did not think were valid science—but some of the things he made fun of in the original edition (like acupuncture) are now more accepted practice.

    In his recent book Thinking Fast and Slow, Daniel Kahneman describes several theories in psychology and economics that have been experimentally proven invalid but which continue in use, much as phlogiston survived as a concept for 30 years after Boyle's experiments proved it incorrect and Newtonian physics continued in use over 100 years after Einstein proposed the Theory of Relativity. Kahneman refers to this as "theory-induced blindness."  Scientists become so emotionally attached to their theories that they cannot abandon them even when the data no longer support them.

    The problem starts with a general lack of understanding of the scientific method.  The scientific method of studying things is to first develop a hypothesis.  An essential rule for valid scientific hypotheses is that you have to have a test capable of proving your hypothesis wrong.

    For example, Einstein thought that a gravitational field would warp space-time and therefore cause light to bend.  He did not propose this hypothesis until he could dream up a test that was capable of proving his hypothesis wrong.  Einstein's test was to use the gravitational field of the sun (the largest gravitational field in the solar system) and look for the light from a star that should appear just behind the edge of the sun.  He hypothesized that light for such a star would be bend so that it would appear beside the sun.  This light should be visible during a total eclipse of the sun.  He calculated for an appropriately positioned star for an upcoming solar eclipse and published his hypothesis, with a prediction of how much out of place the star would appear unless his hypothesis was wrong.  The fact that the eclipse took place in the middle of Europe in the middle of a world war only meant that other scientists had to calculate for a different eclipse that happened in the southern hemisphere to test Einstein's hypothesis.

    I used to teach the scientific method by using a non-working flashlight.  I asked students to develop a hypothesis for why the flashlight was not working and to provide a valid test for their hypothesis. Students almost always propose that the flashlight bulb is burned out as the first hypothesis and that the flashlight batteries are worn out as the second hypothesis.  When I pick the flashlight bulb as the hypothesis and ask for an appropriate test to make it a valid hypothesis, the first proposed test is frequently to put a new bulb in the flashlight and see if the flashlight works.  When I push for a second proposed test, the most frequent answer is to test the light bulb in either a working flashlight or in a light tester.

    These two proposed tests are completely different.  The first is typical human behavior but is totally invalid science.  You cannot possibly prove the hypothesis is wrong unless you test the suspect bulb and show that it is not burned out.  Suppose the actual flashlight problem is corrosion on the contacts within the flashlight.  By removing a perfectly good bulb and replacing it with another perfectly good bulb, you might knock off enough corrosion to make the flashlight work.  This could deceive you into thinking you had found the problem and proven that your hypothesis was correct.

    Now consider the other test, placing the bulb in another flashlight to see if it works.  Obviously, if the bulb works in that flashlight you have successfully proven that hypothesis wrong.  However, if the bulb does not work in the other flashlight, you have not proven the hypothesis correct.  You could have not inserted the bulb correctly.  The bulb may not be compatible with the other flashlight.  The other flashlight may have a corrosion problem, etc.

    The key point: a proper test can prove a hypothesis wrong, but a test can never prove a hypothesis correct.  When a test fails to prove a hypothesis wrong, it is considered supporting evidence for the theory but not proof of the theory.  When several different tests fail to prove the hypothesis wrong, the theory is elevated to a working theory and begins to be used as a successful theory.  However, at any time a single test may prove the theory wrong, at which point the hypothesis should be immediately reexamined.

    Even many working scientists do not seem to fully understand the scientific method.  Science is a self-correcting discipline, but only when people, especially scientists, are willing to suspend their personal attachment to their old conclusions or opinions when considering new ideas.

    A recent example: in December of 2015, an article on NASA's website described recent research on "blue straggler" stars. Using the Hubble telescope, astronomers had studied the 21 blue stragglers in star cluster NGC 188.  Of the twenty-one, Hubble found that seven had white dwarf companion stars.  So, the article claimed, "This confirms the binary star theory for their origin" (NASA). While this experiment did add some evidence for one theory over competing theories, it certainly does not confirm or prove the one theory is correct.

    Even great scientists can be wrong.  Consider the statement, "There is not the slightest indication that nuclear energy will ever be obtainable.  It would mean that the atom would have to be shattered at will."  So said Albert Einstein in 1932. Going back even further, in 1883, Lord Kelvin, President of the Royal Society said, "X-rays will prove to be a hoax."

    Our current society almost worships science and believes if scientists say something is true, it must be true.  But it is in the every nature of the scientific enterprise that new information keeps dislodging old assumptions. Unfortunately, even many scientists do not seem to understand how science is supposed to work.  When someone tells you something is "settled science," know that they do not understand. 

    Scientific theories are useful if they can correctly predict future outcomes, and scientists continue to use even theories that have been proven invalid if the theory successfully predicts outcomes with a high probability.  But when theories are touted as proven science and they fail to accurately predict outcomes, they should be abandoned.

    I would like for you to be more skeptical of anyone who tells you they are absolutely sure of any scientific theory.  Understand that there is no such thing as settled science.

    A retired scientist, manager, and teacher, David H. Berkbile earned a PhD in organic chemistry at the University of Massachusetts.  He worked as a researcher, inventor, and technical specialist for DuPont for over twenty-four years.  During his career, he also served in several management positions for Teledyne, General Instrument/Commscope, and Alcatel. 

    Since his retirement from industry, he has taught chemistry, physics, and general science for seven years at West Nottingham Academy.

    His paper "Alcohol: Green Fuel or Farm Subsidy?" was published in The Torch magazine in 2010.

    The original version of "What Is 'Settled Science'?" was presented to the Torch Club of Delaware on January 20, 2016.