What is science? What is pseudoscience?

What is science? What is Pseudoscience?

by Donald E. Simanek

A visitor to my web site asks "What is the definition of pseudoscience?" That's a fair, but challenging, question. Normally one would expect the practitioners of a discipline to define it, but in this case the practitioners of pseudoscience don't recognize the validity of the label.

The question translates to "How does one distinguish between science and pseudoscience." Perhaps we should first settle on a definition of science. Even that isn't an easy task, for it has so many nuances. Whole books have been written on the subject.

The scientist might answer "I know pseudoscience when I see it." But the boundary between science and pseudoscience is murky. Sometimes it's hard to tell cutting edge scientific speculation from pseudoscience.

Let's recognize two uses of the word 'science'. First, it is an activity carried out by scientists, with certain raw materials, purpose and methodology. Second, it is the result of this activity: a well-established and well-tested body of facts, laws and models that describe the natural world.

Scientists accept that the observations and the results of science must be "objective." That is they must be repeatable, testable and confirmable by other scientists, even (and especially) skeptical ones. The edifice of law and theory that science builds must be representative of a "shared" perception that can be observed and verified by anyone equipped with good observation skills and appropriate measuring tools. Much of modern science uses language and concepts that go far beyond the directly and immediately observable, but there must always be logical links and experimental operational links between these concepts and things we can observe.

As part of the process of crafting scientific models and theories, scientists must brainstorm, innovate and speculate. That's the creative component of the activity. But they must also maintain a disciplined rigor to ensure that their theories and models fit into a logical and consistent interrelated structure. The final edifice called science allows deduction of predictions about the world, predictions that may be tested against observations and against precise measurements made on nature. Nature is unforgiving of mistakes, and when experiments disagree with the predictions of scientific laws and models, then those laws and models must be modified or scrapped.

Scientists' personal styles, prejudices and even limitations are ever-present realities in the process. But rigorous and skeptical testing of the final result must be sufficiently thorough to weed out any mistakes.

It's fairly easy to distinguish science from pseudoscience on the basis of the final product, the laws and theories. If the results (1) cannot be tested in any way, (2) have been tested and always failed the test, or (3) predict results that are contradictory to well established and well tested science, then we can fairly safely say that we are dealing with pseudoscience.

At the level of speculation, it's not so easy. Consider these two examples.

Is the notion that hypothetical particles (tachyons) may travel faster than light a pseudoscientific idea? Well this speculation was proposed by scientists with perfectly respectable credentials, and other respectable experimenters took time to look for such particles. None have been found. We no longer expect to find any, but we do not consider the idea to have been "unscientific".
Is it scientific to hypothesize that one could build a perpetual motion machine that would run forever with power output, but no power input? Most scientists would answer "No."

What is the essential difference between these two examples? In the first case, the hypothetical tachyons would not violate any known principles of physics. In the second case, a perpetual motion machine would violate the very well-established laws of thermodynamics, and also violate even more basic laws as well, such as Newton's laws, and conservation of momentum and angular momentum.

But are the laws and theories of physics sacred? Of course not; they represent part of the logical structure called "established physics" that is the culmination of our accumulated scientific knowledge. We fully expect that future discoveries and insights will cause us to modify this structure in some ways. This won't invalidate the whole of physics, for the old laws and theories will continue to work as well as they always did, but the newer structure may have more precision, power, breadth or scope, and may have more appealing conceptual structure. Such continual evolution and modification of physics is gradual and generally changes only a small portion of the vast edifice of physics. Once in a while, a "revolution" of thought occurs causing us to rethink or reformulate a major chunk of physics, but even that doesn't make the old formulations wrong within their original scope of applicability.

Certain principles, laws and theories of the earliest history of physics have survived unscathed. Archimedes' laws of machines, and his laws of liquids work today as well as they ever did. Newton's laws still work fine, even though relativity has extended the scope of classical mechanics tremendously. Even Ptolemy's now-discarded geocentric model of the solar system (with its cycles, epicycles, eccentrics and deferents) did correctly account for the data on planetary positions in the sky, and if anyone today cared to do so, that Ptolemaic model could easily be extended and improved to work with our improved data on planetary positions to predict past and future planetary positions. But it would be useful only for that limited purpose, since it does not take gravitational forces into account and it did not correctly model the distances of planets from us and from each other. Newton's mechanics and his theory of gravity gave far greater scope to celestial mechanics, unified it with terrestrial mechanics, and made it a powerful tool for understanding the entire universe, not just our local solar system.

So is it reasonable to expect that Newton's laws and the laws of thermodynamics will suddenly be made invalid by some backyard inventor's experiments to achieve perpetual motion? No. Such folks earn the label 'pseudoscientist' or even 'crank'.

The seekers after perpetual motion are a textbook example of the scientific impulse gone astray. They exhibit most of the qualities of pseudoscientists of all stripes. We list below a few qualities of, or symptoms of, pseudoscience. This is also a catalog of the many things that can cause mistakes and error in science. The history of science itself provides examples of some of these, but we hope that we have learned from the mistakes of our past history. Few pseudosciences exhibit all of these characteristics.

Pseudoscientists have deficient or superficial knowledge and understanding of well-established science.
Their proposals are therefore based on faulty understanding of very basic and well established principles of physics and engineering.
The inventors may not be at all aware of these flaws in their reasoning.
They feel that physics is unnecessarily complicated because physicists are 'blind' to simpler explanations.
Some complain that physics is "too mathematical" while others dazzle the innocent with mathematical gymnastics, mistakenly thinking that mathematics is physics, not understanding that it is only a modeling tool.
They obsessively focus on a narrow problem without grasping the powerful interconnectedness of physical theory. Therefore they may not be aware of the broader implications and consequences of their ideas.
They have inordinate confidence in themselves, plus an almost religious faith that their feelings, intuitions and hunches provide a reliable guide to scientific truth.
Anyone who fails to see their genius is labeled 'blind'. They love to compare themselves to innovators of the past whose ideas were initially rejected. "They laughed at Galileo, didn't they?"
Pseudoscientists are angry that their ideas are ignored by the scientific community. They behave as if scientists should drop whatever else it is they are doing to investigate speculative proposals, even though these proposals are not motivated by established scientific knowledge, and may be scientifically implausible.
Pseudoscientists have over-reliance on personal testimony of individuals, and other anecdotal evidence.
Pseudoscientists have an obsession with anomalous observations that seem not to fit established science theory.
Pseudoscientists often display an attitude of "If it feels right to me, it must be right."
Pseudoscientists feel that "Nothing is a coincidence."
Pseudoscientists have an obsession with finding "patterns" in data. Scientists must be pattern-seekers too, but it's a mistake to seek significance in patterns of things that have no possible connection or relation, such as patterns of stars in the sky (constellations), tea leaves, or ink blots.
Pseudoscientists often commit various abuses and misuses of statistics.
Pseudoscientists are motivated by considerations that lie outside the scope of science, or have already been thoroughly discredited. Example, the acupuncturists' acceptance of the reality of specific "energy pathways" in the human body. Another example: the creationists' view that science must be in harmony with their particular interpretation of the King James translation of the Bible.

Responses to Correspondents

Philosohical motivations.

A correspondent reminds me that I haven't said much about the motivations of pseudoscientists who believe in certain "paranormal" phenomena. Here's some that are commonly seen.

A feeling that the world described by science is too ordered and constraining. They long for the possibility of magic and miracles.
A conviction that there are in nature hidden powers that can be mastered by the human mind, if one is sufficiently dedicated.
Nothing conceivable is impossible.
A science that doesn't appeal to my common sense can't be correct.
True science should be understandable by anyone.

The first three of these attitudes have much in common with the Hermetic philosophy of the alchemists. They believed that if one was pure of spirit one might attain power over nature, a power equal to that of a god.

The last two are a reaction against the complexity of science, which does indeed require a good grounding in mathematics to understand. I devised a "joke" slogan to illustrate this: "If the gods had meant us to understand physics, they'd have made it simpler."

More motivations.

Pseudoscientist's questioning of science is frequently motivated by some one thing the questioner doesn't like, or doesn't accept, or something that conflicts with a passionately held emotional conviction. They are usually totally unaware that "correcting" the thing they don't like in a way that suits them better has far broader implications.

For example, the perpetual motion inventor says, "I don't like Newton's third law, and I suspect that law isn't right. If it is wrong, I could make a reactionless engine and easily achieve perpetual motion." Yes, indeed, changing Newton's third would demolish conservation of momentum, conservation of energy, the laws of thermodynamics, and nearly all of what we thought we knew about chemistry, atomic and nuclear physics, etc. etc. Such people just don't realize the vast interconnectedness of everything in science. Now we ask: "Is it reasonable to suppose that all of these are wrong? I mean, seriously wrong? Are you saying that all the scientific experiments, precise measurements and applications of these scientific laws are wrong in ways no one had previously noticed? Are you prepared to rework physics from the ground up, trashing and replacing some 11 centuries of historical development? What is to be gained by doing this?"

Of course the pseudoscientist isn't equipped to do that monumental task, and really doesn't care about the larger picture—only about that one thing or small number of things in conventional science that he found unacceptable.

Now if the pseudoscientist could point to even one solid, repeatable, well tested experiment that conclusively demonstrates a flaw or exception to Newton's third law, that might be good reason for us to question the law, do more independent testing, and if the law is really found to be deficient, look for a way to modify the law to bring it into agreement with experiment. But the pseudoscientist never can produce such evidence. He only "thinks" or "hopes" or has "intiution" or a "gut feeling" that science is wrong in a way that could allow him to achieve his cherished goal of perpetual motion. Then too, pseudoscientists almost always have deficient understanding of the science they despise (or at least mistrust), which leads them to make errors of analysis or judgment that would embarrass a freshman physics major.

Another example is seen in those folks who wish to revive the classical theory of the ether. We ask them "Why?" Usually it is just that they can't conceive of space with nothing in it, or they can't accept action at a distance, or they don't understand that fields are merely mathematical models, not something "in" space. It's an emotional hang-up.

I asked one perpetual motion believer why he thought perpetual motion could be possible. He candidly wrote to me: "I'm not sure why I am convinced it's possible, I guess it's just because I want it to be possible." It's rare that such a person can actually articulate this.

The argument from incredulity.

One such person insists that kinetic energy cannot be (1/2)mv² because "a squared velocity is absurd". Another gripes that a falling body cannot accelerate by the law (1/2)gt² because "a square second is inconceivable".

I call this the "argument from incredulity". Here's another example: "I find it incomprehensible that bodies can exert forces on each other with nothing material in between, therefore there must be something in between." Is that much different from the creationist who says "I find it incomprehensible that design could arise from disorder without a designer, therefore the universe must have had a designer?"

Scientists are fully justified in dismissing these arguments as equally empty and irrelevant.

How is this different from the scientist who is so certain of conservation laws and the laws of thermodynamics, that these are never questioned? Even if some experimenter claims to have found a flaw in one of these laws, that person is dismissed as mistaken. Suppose someone claims that he dropped a body from rest from a measured height, timed the fall, and found that the body accelerated at 2g, twice the "accepted" acceleration due to gravity. He claims to have checked carefully to eliminate any possibility of other influences on the body's acceleration. Should he be believed? Of course not. We suppose that he made a mistake or blunder in measurement or calculation, or is trying to deceive. We do not for one second think he may have found a special place where the acceleration due to gravity was twice what it is elsewhere on earth. A pseudoscientist might claim that he had used "mental powers" to speed up the fall. A scientist would, quite properly, not even entertain the possibility that at the particular time when the experiment was done, the acceleration due to gravity had a momentary increase that went unnoticed everywhere else on earth.

This sort of thing actually happened in one of my mechanics laboratories some years ago. Two students were measuring the acceleration due to gravity with a specialized apparatus called the Kater physical pendulum. The mathematical analysis required is a bit tricky to understand. The students reported a value of the acceleration due to gravity of something like 4.8 m/s². Of course I knew they had made a blunder in the math, but they insisted they had double-checked everything. I said, "You'd better find the error, for I'm getting uneasy walking around in a laboratory where the acceleration due to gravity is only half what it should be." [I was subtly suggesting that if their value were correct, we'd all have noticed it when we went about our daily activities. But they didn't see the humor.] A week later they found their error, and, I hope, learned something from the experience.

First version of this document, May 2002. Latest revision, December 2009.

Input and suggestions are welcome at the address shown to the right. When commenting on a specific document, please reference it by name or content.

Return to the skepticism menu.
Return to Donald Simanek's front page.