This week I am defending the pillars of skeptical philosophy from attacks by the champions of nonsense. The paranormal crowd relies upon a host of logical fallacies and flawed methodology to maintain belief in that which science has not validated. The skeptical movement has done an excellent job of exposing these flaws, including many intellectual lights such as Martin Gardner and Carl Sagan. It is a battle for quality and integrity in science and scholarship, and those promoting a spiritual or paranormal world view are fighting hard. They are trying to redefine science and even logic so that their flaws are not flaws, but virtues.

Yesterday I defended the proper use of Occam’s razor in evaluating competing hypotheses. Today I will discuss the “invisible pink unicorn” analogy. After explaining the actual logic behind this tactic I will then address common misinterpretations, again using this article by Mr. Wu as an example.

Perhaps the best known example of this argument was put forward by Carl Sagan in A Demon Haunted World, who used the example of someone who claims to have a dragon in their garage but answers each challenge for evidence of the dragon with a special reason why such evidence will not be forthcoming. This illustrates two logical principles important to science and skepticism.

Falsifiability

Scientific hypotheses must be falsifiable. In other words, they must be formulated in such a way that they make predictions about nature – about what we will find when we look at some aspect of nature in particular or what the outcome will be of a specific experiment. If a hypothesis does not make such predictions, then there is no way to test it. Testing ideas is the cornerstone of scientific methodology. Without it you are just practicing philosophy, not science. Ideas that cannot be tested are not necessarily false, they are just “invisible” to science, and are therefore worthless as hypotheses.

The “invisible pink unicorn” or Sagan’s dragon are used as examples of unfalsifiable claims, to illustrate this point of logic. Science cannot prove that unfalsifiable claims are not true – by definition. But this is not a reason to believe in them, any more than it is a reason to believe in any of an infinite number of potential claims that cannot be falsified.

Special Pleading

Sagan also used his dragon analogy to illustrate the logical fallacy known as special pleading – inventing a unique and special reason to explain why each type of evidence that could potentially validate a claim is lacking. There is no reason to speculate ahead of time that the phenomenon in question should have such features, they are just invented ad hoc to explain away the lack of evidence. For example: question: “Can I see the dragon?” answer: “No, it’s invisible.” Question: “Can I feel it? Answer; “No, it is non-corporeal?” “Can I measure the heat of its fiery breath?” No, it breaths heatless fire.” Etc.

To take a paranormal example, ESP researchers often make the excuse that ESP does not always work, that it exhibits a “shyness effect” which causes it to stop working when a scientist is present, that sometimes it has the opposite effect (so-called psi-missing), that the effect is too small to measure reliably, etc.

Again, from a purely logical point of view it is possible that a paranormal phenomenon exists that just happens to have all the characteristics necessary to render it unfalsifiable, but that would be an extraordinary coincidence. And further, once you have used special pleading to render a claim unfalsifiable you have also catapulted it out of the arena of science.

In short, the invisible pink unicorn analogy is meant to keep paranormal proponents or other pseudoscientists from using special pleading to retreat from scientific scrutiny, often all the way to the ultimate hidey hole of unfalsifiability.

What does Mr. Wu Say?

Well, Wu does not seem to quite get the logical points of this analogy, or that of special pleading or the need for scientific ideas to be testable. He characterizes the analogy as “severely flawed and ludicrous.” He further argues that it is a straw man argument and a false comparison. Here are his reasons:

False analogy

Wu argues that there is a difference between ESP, spirituality, and other “metaphysical realities” and invisible pink unicorns that make the comparison a false analogy. The difference, he argues, is that people have actually experienced the former, while the latter is just being made up by skeptics as a rhetorical trick. But this argument is a non sequitur. The examples are analogous to the extent that they cannot be falsified and can be defended with special pleading. Whether or not people actually believe in the alleged phenomenon is irrelevant. In fact, the invisible pink unicorn is invoked specifically to use an example outside of culture and common belief, so that the logical principles can be illustrated without the fog of deeply held belief.

Actually, Wu makes this same point four times (even numbering them as separate points) but they are just variations of this same point – spirits are different than unicorns because there are and have been people who actually believe in spirits.

Implicit in Wu’s argument is also the assumption that anecdotal experience and historical belief are of scientific value. This is a separate fallacy for another time, but suffice to say anecdotal experience has been thoroughly established as completely unreliable and is rightfully not admissible as hard scientific evidence for that reason.

The only other separate point Wu makes is this: “…just because something is unprovable does not automatically put it in the same category as everything else that is unprovable.” Well, yes, actually it does, by definition. Anything that is unprovable is outside the realm of science, period.

Wu goes on to say, “For example, I can’t prove what I ate last night for dinner or what I thought about. Without witnesses, I can’t prove what I saw on TV or how high I scored in a video game either. But that doesn’t mean that these things are in the same category as every story in the fiction section of the library.”

The examples Wu uses are curious. What he ate for dinner last night is potentially demonstrable, if anything identifiable is left in his stomach. There could also be inferential evidence, such as left-overs, empty boxes, and garbage. The video game example is especially weak since it is very common for such games to record their scores. I think this demonstrates that he does not completely understand what is meant by unprovable, or more accurately unfalsifiable – something for which no evidence or observation can be brought to bear that would increase or decrease the probability of it being true.

What he thought about the previous night is a better example, but let’s get to the real point. He seems to be saying that we take for granted everyday occurrences that cannot be demonstrated scientifically. He does not say it directly, but he seems to be implying with his “fiction” comparison that the inability to be proven scientifically does not mean that they are not true – that he did not eat fish and chips the night before while contemplating the British parliamentary system and scoring well on Donkey Kong. But this is also a non-sequitur, for as I stated above the scientific/skeptical position is not that unfalsifiable notions are untrue, just that they are outside the realm of science.

Wu finishes by saying, “The bottom line is that while it is true that no one can disprove the existence of invisible pink unicorns, the evidence to support God, spirits and psychic phenomena, although mostly anecdotal, is vastly greater, more significant, more relevant, and more sincere than the evidence to support invisible pink unicorns and other fictitious examples deliberately made up by skeptics.”

This is both a false premise and another non-sequitur. First, you cannot have evidence either for or against a belief that is unfalsifiable. Paranormal proponents often want to have it both ways – they admit any evidence, however flawed, that seems to bolster their claims, but then dismiss contradicting evidence or the lack of evidence by saying that their claim is not subject to such evidence. But in science, claims must be defined clearly, and they are either amenable to evidence or they are not.

If a claim is formulated in a testable fashion, then evidence can be brought to bear and all relevant evidence should be judged scientifically. If it cannot be tested, then it is outside the realm of science and in the realm of faith alone. Most people put God in the realm of faith. Spirits can go either way – some believe on faith, others claim to have scientific evidence. Psychic phenomena should not even be in the same category – it is a claim about human ability, something that should be testable. If psychic claims are insulated against falsification, that is just an illegitimate attempt to have one’s cake and eat it to – to pretend to be a science without risking falsification.

The invisible pink unicorn stands as a legitimate analogy to keep paranormal proponents from straddling the fence between science and faith, and using special pleading to insulate their claims from falsification. My only remaining question is this: if the unicorn is invisible, how do you know it’s pink?