What is a Skeptic?

What Do We Mean By Skeptic?

The words 'skeptic' and 'skepticism' create a lot of confusion. To the ancient Greeks, Skepticism was a philosophy that held that we can never truly know anything. Such a position suggests that we shouldn't bother trying to learn anything, as knowledge is unobtainable. 'Skepticism' of this era predated the modern understanding of science and referred to the word 'knowledge' in a metaphysical sense. This idea is not practical. It is actually paradoxical. How can we know that we can't know anything...if we can't know anything? In popular culture and the media, 'skeptic' is synonymous with 'cynic', or someone that has a perpetual negative attitude and always assumes malicious intent in the actions of others. These are not the definitions that we are concerned with here.

A skeptic is not a cynic, nor is a skeptic paralyzed by the inability to know anything. We are concerned with the modern concept of Scientific Skepticism. For us, a skeptic is a critical thinker who uses the methods of science to distinguish the known from the unknown and employs that knowledge to improve our world and society. The 'knowledge' referred to here is scientific knowledge (facts, laws and theories) as opposed to 'Truth with a capital T', as sought by the ancient Greek philosophers. In a simplistic way, skepticism is the philosophy of science.

Modern scientists produce scientific knowledge. They are natural skeptics. Most scientists don't spend much time thinking about skepticism. The physicist and great thinker Richard Feynman famously stated that the "philosophy of science is about as useful to scientists as ornithology is to birds." However, philosopher Massimo Pigliucci pointed out that although birds are certainly not aware of ornithology, the survival of many bird species may depend on it. Also, if birds could understand ornithology, they would likely find it beneficial. Those of us who are not scientists but use scientific knowledge to inform our practice (such as doctors and engineers) would benefit from learning the tools of skepticism as well.

As discussed above, science is the best method devised to determine facts and derive theories. Science sorts out what "is". Philosophy helps us sort out what "ought to be". One could think of skepticism as the philosophical idea that, when considering claims, we ought to use science whenever possible to determine their validity before putting the claims into practice.

"Skepticism is a perspective that does not accept or reject claims at face value but withholds judgment until sufficient evidence is available to make a decision." -Eugenie Scott

"Skeptic does not mean him who doubts, but him who investigates, or researches as opposed to him who asserts and thinks that he has found." (Miguel de Unamuno, "Essays and Soliloquies," 1924)

"Skepticism is a provisional approach to claims. It is the application of reason to any and all ideas. In other words, skepticism is a method, not a position. Ideally, skeptics do not go into an investigation closed to the possibility that a phenomenon might be real or that a claim might be true. When we say we are “skeptical,” we mean that we must see compelling evidence before we believe."

(Michael Shermer, Ph.D.)

Skepticism and Science

The skeptic must understand the methods of science and how scientific information is developed. The scientific method is messy. It moves in fits and starts. The skeptic understands that scientific knowledge is really a model building process that endeavors to create models of nature that can allow for accurate predictions. As new information becomes available, older models are either supported, modified or rejected. This messiness, it seems, may fuel the general public's misunderstanding and mistrust of science. People like certainty. Science is provisional. The fundamental uncertainty of science is perceived as a weakness by many. Modern scientists do not speak the language of the general public. Proponents of pseudoscience have charismatic leaders that exploit this perceived weakness in public forums. Hence, we have a proliferation of unscientific yet popular practices, especially in the world of healthcare.

Doctors and other healthcare providers must put knowledge into practice for the purposes of improving the health of patients. We are the liaisons for our patients and scientific information. As skeptical doctors, we 'ought' to employ the best scientific information to guide our practice. Therefore, we need tools for distinguishing scientific knowledge from pseudoscientific notions.

Skeptical Tools

As we will learn in later sections, there are many tools in the skeptic's toolbox. When dealing with empiric claims, four conceptual axioms are worth pointing out here: 'Hume's Fork', 'Occam's Razor'; 'Extraordinary Claims Require Extraordinary Evidence'; and 'Correlation is Not Necessarily Causation'. Other tools, such as logical arguments, and basic understanding of cognitive biases, heuristics and logical fallacies are covered at length in their own sections.


We learned this axiom in the Philosophy and Science section. When faced with a claim, we must first determine if it is the kind of claim that we can, in fact, know. There are two criteria, and a claim must satisfy one of the two.

David Hume advised us to consider if something can be known through pure logic and deductive reason. These things can be known through 'relations of ideas'. We can know them without experience, observation or experimentation (a priori). Knowledge in this category can be said to be mathematical (5+7=12), trivial (true by definition, e.g., 'all bachelors are unmarried men') or deductive (e.g., 'If A then B; A; therefore B').

Next, Hume asks us to consider if something can be known through 'matters of fact'. By this, he meant observation or an inductive / experimental process (a posteriori), In other words, can we consider the knowledge to be scientific?

We learned in the What is Science section that scientific knowledge can be classified as facts, laws and theories. Popper helps us identify scientific knowledge with the notion as falsification. Popper's falsification criteria is built-in to the second part of the 'fork'.

As few can match Hume's literary style, his "fork" is worth stating again from the original text.

"...let us ask,

Does it contain any abstract reasoning concerning quantity or number? No.

Does it contain any experimental reasoning concerning matter of fact and existence? No.

Commit it then to the flames: for it can contain nothing but sophistry and illusion."

(from An Enquiry Concerning Human Understanding)

This concept is named after William of Occam, a 14th century logician. Also called the "Law of Parsimony", it basically states that when faced with two or more competing theories that seem to explain a phenomenon, it is best to adopt the theory that makes the fewest new assumptions. In other words, we should adopt explanations that deal with as little unknowns as possible.

It has often been interpreted as favoring the simplest explanations. Although this is mostly true, it is not quite that simple. For instance, at first glance, supernatural explanations may be thought of as simple. However, such an explanation contains a mammoth unknown assumption. Any explanation that depends on a complex interaction of known facts and laws will always be superior, no matter how complex the explanation may seem.

A version of this concept is known to many medical students as..."If you hear hoofbeats, think horses before zebras". An explanation that employs variables that are known is favorable over one that employs unknown or unlikely variables. Any assumption that is necessary to make a theory work introduces unknown or unlikely variables. These variables decrease the certainty of the theory. This is subtly different than just picking the simplest theory.

For instance, a patient may appear to have less pain after being ritualistically poked with needles by an acupuncturist. One explanation may be that the acupuncturist manipulated the patient's vital energy to restore balance. Another explanation may be that the patient felt subjectively better because of a complex interaction of beliefs, expectations, showmanship, relaxation and psychology. Both explanations may explain the apparent reduction in the patient's pain. However, the concept of vitalistic energy is not scientific. It is not falsifiable. The first explanation would require a new assumption that such a force exists, even though no such thing has ever been observed. On the other hand, much is known about belief, showmanship, relaxation techniques and psychology. No new assumptions are needed for the second explanation and Occam's Razor suggests that it is the wiser choice to go with, at least until new information is available.

Another way of thinking about Occam's Razor is from a purely statistical standpoint. Every claim has premises or assumptions. Each assumption carries a prior probability of being true. The probability of a claim being true is just the multiplication product of the probabilities of each assumption within the claim. For instance, Jill might claim that she heard that Steve has a new, red Corvette. Jane might offer the claim that Steve has a new, red car. The first claim (Jill's) has 4 assumptions (Steve has a car, it is new, it is red and it is a Corvette). The second claim (Jane's) has 3 assumptions (Steve has a car, it is new, and it is red).

We might assign a probability to each assumption. (note -- the phrase "probability of" is shortened to "P")

P(car) = .95

P(new) = .80

P(red) = .70

P(Corvette) = .50

Jill's claim is the product of each of these (.9 x .8 x .7 x .5 = .25)

Jane's claim is the product of the first 3 (.9 x .8 x .7 = .50)

Therefore, it is more likely that Jane's claim is correct because it contains less assumptions. Even if we were nearly certain that Steve has a Corvette and assigned P(Corvette) a value of .99, the likelihood of Jill's claim would still be slightly smaller.


Carl Sagan popularized this phrase in his T.V. series, Cosmos.

It has its origins much earlier with Pierre-Simon Laplace (1749-1827) - "The weight of evidence for an extraordinary claim must be proportioned to its strangeness."

The sentiment is echoed in David Hume's statement from Of Miracles, "No testimony is sufficient to establish a miracle unless that testimony be of such a kind that its falsehood would be more miraculous than the fact which it endeavors to establish."

The 'Extraordinary Claims' axiom is related to Occam's Razor in that it refers to claims that require new assumptions, especially assumptions have not been established, cannot be established, or directly contradict established knowledge. In such a situation, the claimant has a large burden of proof, in that he/she must present proper evidence that will either create new lines of knowledge that are coherent with established science, or completely overturn established lines of knowledge that contradict said evidence. The axiom calls for a standard of what is considered proper evidence. We will consider proper evidence in the Types of Scientific Studies section. The axiom established places the burden of proof on the one making the claim. The skeptic who questions the claim is not burdened with disproving the claim (although many delight in doing so).

Extraordinary claims are often made by advertisers of health-related products. For instance, the sellers of bracelets and socks embedded with magnets claim that such products improve circulation by attracting the iron in blood cells. This claim is extraordinary because it contradicts known laws of physics. The natural state of iron in a blood cell is non-magnetic. For the claim to be true, proponents have the burden of proving that our understanding of magnetism is false. They also would have to prove, with proper evidence, that the health benefits of such products are observable under proper conditions. Holding such a belief without meeting these burdens of proof is not science. It is pseudoscience.

On this site, this concept will be referred to as the "Extraordinary Claims Axiom". At its core, it is really a narrative of Bayesian probability. A claim is extraordinary if it has a low prior probability, based on accepted knowledge. The likelihood of an 'extraordinary claim' being true is proportional to its 'extraordinary evidence'.


This axiom protects us from falling into the intellectual trap of the Post Hoc Ergo Propter Hoc logical fallacy (after this, therefore because of this). It is natural to think that if an event (Y) occurs either with or after another event (X), then X must have caused Y. It is used in conjunction with 'Occam's Razor' and the 'Extraordinary Claims' rule to sort truth from illusion. Just because Y always seems to follow X, does not mean that X causes Y. X might cause Y, but not necessarily. The 2 events may occur independently of one another. They each may have a common cause. Perhaps another event (Z) causes both X and Y. Perhaps Y would occur regardless of X. The best way to know is to observe for Y without X. Here we see the logic for controlled experiments.

Let's return to the example of acupuncture discussed above. It seems that pain reduction (Y) follows the administration of needles (X). Does this imply that placing needles along meridian lines reduces pain? To acupuncturists and to those patients who like it, the answer is "yes, absolutely". However, could other factors (Z) actually be doing the job? The only way to tell is to remove the placing of needles (X) from the equation without the patient's knowledge, so as not to alter the other factors (Z). In this case, Z may refer to complex interactions of beliefs, expectations, showmanship, relaxation and psychology. This has been done in blinded sham experiments, where some of the patients were randomized without their knowledge to undergo the entire ritual of acupuncture, but without the actual placing of needles. These control patients received "sham" acupuncture involving either fake needles that retract into the handles, or, in one study, simply tapping the skin with toothpicks. The results showed the same pain reduction (Y) without the placing of needles (X). From this, we conclude that it is more likely that other factors (Z) actually cause Y. Here we have examples of randomized, blinded, placebo-controlled trials.

Confusing correlation and causation has been at the heart of numerous tragedies. Of recent concern is the belief that vaccinations cause autism. Vaccines are given to infants (event X) at a time in the child's life before autism can be diagnosed (event Y). Using the Post Hoc Ergo Propter Hoc fallacy, a causal link has been made. Numerous studies have shown this to be false. Every time the issue has been investigated no causal link has been found. Autism occurs at the same rate in unvaccinated kids as in vaccinated kids. In this case, the two events are linked only in time. The actual causes of autism are unknown, but research continues. Unfortunately, the media, general public, and some outspoken doctors confuse correlation with causation. The results have been tragic. Many diseases, preventable by vaccine, have increased to epidemic proportions in some cases. Kids have needlessly died of meningitis and whooping cough and have been damaged by measles. Polio was nearly eradicated, yet has returned in some countries to problematic levels. For a detailed account of the vaccine controversy, please read further under Pseudoscience in Healthcare.

Limits and Ethics

Skeptics realize that science is an ongoing process. There are many gaps in our scientific knowledge base. We must be careful not to fill those gaps with unsubstantiated claims.

Comedian Dara O'Brian said, "They say well, science doesn't know everything. Well, science knows it doesn't know everything, otherwise it would stop!" Skeptics are very comfortable saying, "I don't know".

Skeptics are concerned with ethics. Skeptical doctors are especially concerned with healthcare providers whose decisions are informed by pseudoscience. By definition, pseudoscientific ideas have the trappings of science, but are false. At best, they are unlikely to be true. Pseudoscientific practices are therefore deceptive. The deception may be deliberate, practiced by con artists. The deception may be practiced with good (but misinformed) intentions. It is likely that a large proportion of pseudoscience practitioners have no malice. They actually have good intentions but lack the skeptical tools to understand the deception. Hence, they deceive themselves as well as patients. Practices based on pseudoscience may cause direct unnecessary harm or prevent the patient from receiving treatments that may actually work.

It is very true that science-based medical decision-making carries risks. But it also carries quantifiable, objective benefits beyond subjective improvements. Medical decisions are hard, and doctors must always be aware of risks and benefits. Patients should be honestly told of the risks and benefits. The information that is used to assess risks and benefits should be the best information available. That is, the information should come from scientific knowledge. To make recommendations based on ideas that are false or implausible is deceptive at best and potentially deadly at worst.

Skeptical doctors recognize the deception. Most of us feel that honesty is a virtue and deception is wrong.

Conclusion

The modern skeptic is a scientific skeptic. We use proper science and basic ethics to inform our decisions. We withhold acceptance of claims until proper evidence is presented. We use axioms such as 'Occam's Razor', 'Extraordinary Claims Require Extraordinary Evidence ', and 'Correlation is Not Necessarily Causation' to evaluate claims. The skeptic recognizes that humans are prone to biases. We recognize that people defend their biases with logical fallacies. Skeptics understand that we are all prone to such biases and fallacies, even skeptics. To overcome these tendencies, we must learn about biases and fallacies in order to spot and correct them, especially in our own thinking.

Skeptical doctors are ethical. We feel that it is ethical to avoid deception when practicing medicine. We owe it to our patients to use skepticism to differentiate science from pseudoscience and nonsense.

John Byrne, M.D.

References and Links

"Skepticism - Wikipedia, the free encyclopedia." 2003.

<http://en.wikipedia.org/wiki/Skepticism>

"Scientific skepticism - Wikipedia, the free encyclopedia." 2003.

<http://en.wikipedia.org/wiki/Scientific_skepticism>

"Richard P. Feynman Quotes (Author of Surely You're ... - Goodreads." 2008.

<http://www.goodreads.com/author/quotes/1429989.Richard_P_Feynman>

Massimo Pigliucci. "Rationally Speaking: The future of philosophy of science." 2010.

<http://rationallyspeaking.blogspot.com/2010/04/future-of-philosophy-of-science.html>

"Essays and soliloquies - Miguel de Unamuno - Google Books." 2011.

<http://books.google.com/books/about/Essays_and_soliloquies.html?id=okEtAAAAIAAJ>

"Skeptic » About Us » A Brief Introduction." 2008.

<http://www.skeptic.com/about_us/>

"Hume's fork - Wikipedia, the free encyclopedia." 2004.

<http://en.wikipedia.org/wiki/Hume's_fork>

"Occam's razor - The Skeptic's Dictionary - Skepdic.com." 2006.

<http://www.skepdic.com/occam.html>

"Extraordinary claims require extraordinary evidence - WikiSynergy." 2009.

<http://wikisynergy.com/wiki/Extraordinary_claims_require_extraordinary_evidence>

"Pierre Simon Laplace - Quotations." 2006.

<http://www.todayinsci.com/L/Laplace_PierreSimon/LaPlacePierreSimon-Quotations.htm>

"Miracles - The Skeptic's Dictionary." 2002.

<http://skepdic.com/miracles.html>

"Correlation does not imply causation - Wikipedia, the free ..." 2005.

<http://en.wikipedia.org/wiki/Correlation_does_not_imply_causation>

Goldman, Rose H et al. "Acupuncture for treatment of persistent arm pain due to repetitive use: a randomized controlled clinical trial." The Clinical journal of pain 24.3 (2008): 211-218.

<http://journals.lww.com/clinicalpain/pages/articleviewer.aspx?year=2008&issue=03000&article=00005&type=abstract>

Cherkin, Daniel C et al. "A randomized trial comparing acupuncture, simulated acupuncture, and usual care for chronic low back pain." Archives of Internal Medicine 169.9 (2009): 858.

<http://archinte.jamanetwork.com/article.aspx?articleid=414934>