Refuting Unfalsifiable Claims with Superior,
Incompatible Explanations
by Allan
Glenn, a.k.a WinAce
Last update: October 9, 2003
This article focuses primarily on the nature of rational epistemology and its role in debunking empirically untestable claims. Section 1 will briefly address the issue of unfalsifiable claims in general, highlighting a few relevant facts and setting the stage for the rest of the essay. After a short digression on the burden of proof, the rest of the article is introduced.
Section 2 aims to demonstrate that, contrary to popular maxim, we routinely dissect and refute such claims. Indeed, all semblance of reason would be lost otherwise. Disbelief toward a proposition is not necessarily ill-advised without actual evidence to the contrary. If it were, we would be forced to accept, or at least remain neutral towards, almost any assertion ever made, no matter how unlikely or glaringly fictitious. As I show here, this quickly becomes an untenable position.
In section 3, I will briefly analyze the nature of statements about reality, demonstrating that a genuinely unfalsifiable claim is extremely difficult to postulate. I also introduce the concept of falsification by acceptance of a competing, mutually exclusive hypothesis as a better explanation. Many paranormal claims are not, strictly speaking, prone to disproof by evidence; however, they can be effectively "disproved" by establishing an alternate, incompatible hypothesis which better explains the data.
Section 4 outlines a set of six important criteria to invoke when deciding which of a set of competing explanations is superior. This is the main body of the essay, and includes such considerations as parsimony, explanatory power and internal consistency. Various counterarguments are also addressed.
Finally, section 5 concludes with a summary and closing remarks. An extensive list of footnotes and some acknowledgements follow.
Index:
1. The Problem of Unfalsifiable Claims
4.1: Consistency with Well-Established Knowledge
4.2: Parsimony
4.3: Explanatory and Predictive Power
4.4: Internal Issues within the Claim
4.5: The Origin of the Claim under Questionable Circumstances
4.6: The Claim's Likelihood of Existence if Incorrect
5. Conclusion
6. Footnotes
7. Acknowledgements
1. The Problem of Unfalsifiable Claims
Empirically unfalsifiable claims are extremely common in human society. From faith healing, to water dowsing, to alien abductions, individuals and even entire industries often capitalize on our collective propensity to accept emotionally satisfying, magical and, above all, refutation-proof beliefs about the world.
In many cases, these are merely benign superstitions, such as knocking on wood to alleviate bad luck. On other occasions, they can cause legitimate, grievous harm, such as encouraging the sick to opt for useless therapies in lieu of medicine that could actually help them.
As humans, we generally value correct beliefs about Life, the Universe and Everything. Hence, we can seek out accurate, up-to-date knowledge on a variety of subjects even when it has no direct bearing on the quality of our lives. We want to know if that particular belief about black cats is true, or if strange entities from another realm have been interacting with us over the years.
To this end, we can investigate esoteric statements about reality and test them against competing explanations. A prime example of this is our quest for supporting evidence in areas such as extraterrestrial visitation and ghost sightings.
However, when asked to support their claims with strong evidence, proponents of many paranormal beliefs can fall back on their unfalsifiability. "I can't prove it, but neither can you prove it false!" is a well-known retort. This can be accompanied by requests for more funding, continued interest, or openmindedness in light of presumed uncertainty.
The desired implication is that acceptance and denial of the proposition is on the same level, which is a remarkably effective, if ultimately fallacious, rhetorical tool. Many in the audience will fail to realize that this is an argument from ignorance, a textbook logical fallacy. A claim is not true, or even likely, merely because it hasn't been adequately debunked. [1]
One could, of course, respond perfectly well that the claimant is responsible for demonstrating their views, especially if they're unorthodox or require much revision of what we know about the universe. If a self-proclaimed cosmic ambassador asserts he has an extra brain in his stomach that lets him telepathically communicate with Andromedeans, you're under no obligation to even consider his statements without appropriate support. [2]
Depending on the nature of the claim and other considerations, disbelief can be the appropriate, default position. In at least some cases, such as the above, arguing otherwise would cause the burden of proof to roll over in its grave quickly enough to cause a detectable anomaly on the Richter scale. [3]
But when faced with something that appears so contrived and obviously fictional, just how satisfying and appropriate is it to go on the defensive? The answer is, of course, "not very satisfying at all." While a belief can be reasonably rejected simply because it lacks evidence, some of them, by virtue of irrationality, the threat they pose to society and other factors, will cry out for a conclusive refutation.
2. The Solution - Inference to Better Explanations
So, how do we deal with heretofore unknown, but suspicious and possibly false claims? The answer is that a fine line between naiveté and dogmatism must be walked by every critical thinker. On the one hand, sheer insanity awaits if we remain undecided on every bizarre yet conveniently unfalsifiable claim we come across:
On the other hand, some claims that ran counter to initial expectations and
common sense have, indeed, been borne out. Assuming something is logically
possible, meaning it isn't internally contradictory, where does one draw
the line to declare it a likely fiction and, more importantly, false until
shown otherwise?
In the rest of this essay, I will discuss just that: a set of methods that, together,
is arguably the most potent and feared bane of paranormalists,
conspiracy theorists, pseudoscientists, religious leaders, and other proponents
of refutation-proof claims. That methodology being the subject of this article,
appeal to superior yet mutually exclusive explanations.
This tool is perhaps the most valuable in the skeptic's entire arsenal. Coupled
with the available data and a reasonably effective 'baloney detector', [4]
it allows one to mount an unexpectedly powerful, frontal attack on claims
ordinarily unfalsifiable with contradictory evidence alone.
3. Falsifying the Unfalsifiable
Although the focus of this essay should now be clear, it may seem counterintuitive: if a claim is unfalsifiable, how do you
debunk it, presumably using evidence? Below, I will outline the key reasons that
make this apparently contradictory action not only possible, but exceedingly
common.
First, very few claims are entirely unfalsifiable as formulated. Let's take the case of
the hypothetical cult leader with an additional brain in his gut, mentioned above. At first glance, it would appear
his argument does make empirical predictions, such as the presence of
something unusual in his stomach, putting it within danger of scientific
falsification.
Hence, many unorthodox beliefs can be shown false, within reason, with the data
alone. But, we know the modus operandi of their proponents from extensive
historical experience. As soon as investigation reveals one of their statements
false, it conveniently gets supplemented with an unverifiable, often contrived
ad hoc rationalization [5] that
conveniently explains away the damaging facts.
For example, if an ultrasound reveals nothing unusual in our friend's interior except a burrito, the original claim can acquire a supplementary rationalization such as "The aliens installed an apparatus that foils your simple human tests."
By now, you have a genuinely unfalsifiable claim being made. However, the
presence of numerous, unevidenced ad hoc rationalizations absolutely necessary to reconcile
a statement with reality is itself an important point to be considered when
evaluating the claim's likelihood of being correct. On further reflection, if
the above situation had taken place, few would dispute the balance of
evidence overwhelmingly supports fraud in lieu of legitimate contact with
extraterrestrials.
Additional considerations, mentioned below, quickly enter the picture and give
us the excellent ability to judge claims incorrect that we can't readily debunk
using the evidence alone.
At the end of the day, nothing about empirical reality at all can be
demonstrated with that absolute certainty. And when we infer the best
explanation in literally any area, we automatically declare false an
infinite number of other, unfalsifiable hypotheses that account for
that same observation.
"The hallway door is open because my wife left it that way" can
be inferred as the superior explanation on the basis of the
evidence, and, once established, automatically refutes other, technically 'unfalsifiable'
hypotheses like "spirits from another realm telekinetically swung it open."
If the best explanation for the claims of our friend, the cult leader is that he's lying to
profit from his followers, you simultaneously infer that he hasn't had
legitimate contact with the aliens, that they don't plan to invade
anytime soon, and finally that they don't even exist.[6]
To put it simply, the evidence can support the hypothesis that a claim springs
from human invention or other sources, effectively refuting it in the process, even if it doesn't falsify it to
scientific standards. If this is successfully argued, we can be confident that alleged
phenomena like alien abductions and encounters with living dinosaurs, to name
but a few examples, are in
error. This can be concluded even if the claim
itself is intentionally made impossible to falsify with mere
contradicting evidence. In that counterintuitive yet stunningly
successful ability lies
the true power of the inference to the superior, mutually exclusive
explanation.
4. Important Diagnostic Factors to Consider
Although perhaps not the most exhaustive or effective set of criteria to apply, I
will put forth six tests--rules of thumb, rather--that I've personally found
invaluable in critically analyzing certain classes of beliefs.
None are, by themselves, reliable indicators of falsehood. Neither are
they meant to be. One or two of these criteria might raise an alarm bell, but still be
easily dismissed in light of the preponderance of evidence.
A majority yielding warnings can still be overridden, albeit
with greater difficulty. When you're faced with a textbook example of an
erroneous claim, one that fails several of these tests, ambiguous data and unevidenced "it might
have been" scenarios likely aren't good enough to accept it at face value.
When and if you observe all six tests in the red, however, you can be quite sure
that something is amiss. In that case, you can reasonably infer that someone is pulling a fast one, which
can, by logical necessity, imply
that their claim itself is false.
And now, I will discuss the tests themselves.
4.1: Consistency with Well-Established Knowledge
To put it simply, "Extraordinary claims require extraordinary
evidence."[7] This is possibly
the single most important of the rules I delineate. Depending on the nature of the claim being made, it can actually render further
considerations superfluous. "I ate something recently" needs no
supporting data other than our knowledge of metabolism and gastrointestinal
processes. Hence, self-evident claims can be established by this one test alone.
Mundane claims like "I just spoke with the mailman" or "I
had pizza for breakfast" are completely unaffected by this criterion,
and their validity should be decided via other considerations.
Where this test really shines, however, is marking as suspicious claims that are
unprecedented, earthshattering, or contradict formidable areas of well-established
knowledge.
Obviously, "I just spoke with the mailman... who's been dead for
years" and "I had unicorn pizza for breakfast,"
given what we've learned about the world, make the claim much more worthy of
scrutiny. Both scenarios are logically possible, and could conceivably be demonstrated
with good enough supporting evidence. Nevertheless, few would dispute that the odds of
mistake, hoax, and other everyday possibilities is higher in some
types of claims versus others.
This can, all other things being equal, add significant weight to an alternate, more plausible explanation, justifying disbelief unless other considerations (such as powerful evidence) override it. For example, a strong argument can be mounted that the claim about unicorn pizza, in the absence of evidence for or against, initially belongs in the 'most likely false' pile on the basis of what we know about zoology.
The data claimed as support for a hypothesis should generally not be explainable with fewer assumptions merely by invoking other, already known
phenomena. This is known as Occam's Razor. [8]
If hypothesis A explains phenomenon X with three fewer
assumptions than hypothesis B, but is otherwise consistent with all
the data and doesn't set off more red flags, you might want to do yourself a
favor and accept A.
As an example, if an investigator documents a single set of footprints and fingerprints matching the main suspect,
finds his dead wife had a history of marital infidelity, no reasonable alibi,
and so on, he would probably not postulate an additional assailant who stood around, effectively did
nothing and left no detectable physical trace. Doing so is simply unnecessary
and yields no explanatory advantage.
This, of course, applies much strongly for assertions that set off the 'Extraordinary
Alarm'. A claim that armies of celestial warriors fought it out in the
clouds, conveniently originating in a volume written in prescientific times,[9]
can be readily explained without recourse to the paranormal. We would not be
justified in accepting its testimony, anymore than seriously considering the remote
possibility that ghosts framed a defendant for his wife's murder (although it
would make a great insanity defense). When they suffice to account for the data, it's wise to invoke existing
entities before postulating new ones.
By all means, invoke additional entities if
you have supporting data that needs them as an explanation.
If several television news crews broadcast, live, the aforementioned stellar armies, complete with spectacular, roaring weaponry,
discernible battle cries, and wounded giants falling to earth and crushing buildings--and
you didn't even wake up later--invoke those
extra entities, because you'll need them to account for the data.
Alternatively, had aliens arrived, settled on the White House lawn, and stated, as their first
proclamation to humanity, "Sorry, we had to leave in a hurry, and hope your ancestors
weren't scared out of their wits by our little wargame," you
would also be justified in rejecting your earlier explanation of simple human
error and/or exaggeration, which now wouldn't explain all the
evidence.
Parsimony can be disregarded in certain cases, as well. If a more contrived hypothesis predicts unique observations you can go and test, or if its underlying mechanism is sufficiently different from the alternatives that no straightforward comparison of their parsimony value is possible, other factors should take precedence.
The latter can occur when the underlying assumptions of two different explanations are completely at odds, or if each one requires fairly equal numbers of unsubstantiated assumptions to work. This can make parsimony a problematic criterion to apply rigorously in scientific studies, but that is a whole other discussion, and not the ultimate focus of this essay. Like the other criteria, this is only one consideration to take into mind, does not apply in some cases, and can never be definitive by itself.
Generally, however, paranormal claims are remarkably amenable to comparison with simpler alternatives. In many cases, they already incorporate the alternate hypothesis, but tack on a superfluous entity or two as a causal agent. This is how, for example, a blur on a photo that superficially resembles a face can be attributed to a ghost instead of lens flare or other, much more mundane optical phenomena.
As mentioned above, what seems unparsimonious can, conceivably, become the
simplest explanation that will account for all of the data. If one could visibly heal amputated limbs under controlled conditions
by invoking the name of a specific god, one would win James Randi's
million dollars and be touted as the
prophet to end all prophets.[10]
However, faith healers generally invoke, as their sole evidence, unverifiable, likely embellished anecdotes,
ambiguously documented 'improvements' that could be explained by the Placebo
Effect, and testimonies about healed conditions that spontaneously go
into remission on occasion regardless of "treatment" with magic
invocations, magnetic bracelets or sunflower seeds. This is not particularly
convincing to those familiar with the relevant medical literature and human psychology. [11]
Such "evidence" would likely exist regardless if the hypothesis was correct or
not, so it cannot be reasonably used as support of one possibility over another at
all.
4.3: Explanatory and Predictive Power
Here we temporarily stop by on the turf most familiar to professional
scientists. One of the defining marks of a
robust theory is a high ratio of explained facts compared to initial
assumptions, as well as the ability to successfully predict unintuitive, heretofore unknown
observations.
Let's suppose you wake up in the morning and try to turn on the radio for the
morning news. Yet,
nothing happens. Several explanations (of varying quality) immediately spring to
mind:
A pattern is immediately revealed. The first three hypotheses each involve just one common, reasonable event as their supporting assumption, and make numerous, distinct, easily testable predictions:
These explanations all have very good explanatory and predictive power. But what about the fourth?
Ultimately, just what do airwave faeries explain? What, if anything, do they
predict? And what, if anything, could ever empirically falsify them?
The answer, to all three questions, is nothing. The postulated process is
so vague that it tells us nothing we didn't know already. What's
more, by invoking it, you replace a mystery that, in principle, can be solved
(like a
dead radio) with one you can't even investigate!
This is an "explanation" that predicts no unique answers and is compatible with
any state of the data. After all, if you later find there was a blackout,
proponents of the airwave fairy explanation can assert their magic was
responsible for a frayed power supply to the city.
Ask them to get more specific, such as outlining the unique patterns fairy magic would leave on wires. Then you can investigate further, perhaps by comparing the state of the damaged wires with the predictions made by "rats chewed thru them," "simple decay was responsible" and other explanations.
Otherwise,
postulating an intentionally vague claim, to the point where its truth or
falsity has no bearing on empirical reality at all, is another warning
sign.
On a related note, a hypothesis that explains and predicts 40 diverse
observations, but requires an additional assumption for the 41st alone, is
superior in explanatory value to one that explains only 15 and needs 26 individual and unrelated
assumptions to deal with the rest. This may seem like a restatement of (2)
Parsimony, but it runs much deeper into the philosophy of science.
One could, in all seriousness, defend even the flat earth view with an endless
stream of ad hoc rationalizations, but the resultant mess of randomly invented assertions, united by no remotely plausible common assumption
or mechanism, postulated only to prevent a cherished paradigm from being
falsified (and not on the basis of independent supporting evidence), is totally dwarfed in explanatory and predictive power by
the mainstream view. [12]
Similarly, if 95% of Loch Ness monster sightings can be explained with recourse
to human psychology alone, it's prudent to look for explanations accounting for
the other few under the excellent working hypothesis that it's only a legend.
On the other hand, it would be quite a bit more difficult to square the existence of Nessie with such observations as:
Any and ultimately all of these damaging facts could be explained away by invoking enough ad hoc explanations, but there is simply no reason to do so given the evidence of an alternative, unambiguously better explanation.
4.4: Internal Issues within the Claim
This is a very simple test for general coherency. Even if a claim is
unfalsifiable by empirical evidence, it often contains internal problems that
render it unlikely or expose it as the product of human invention.
For example, if you had a guaranteed way of making millions of dollars in real estate, would you be
selling it for a measly "three easy
payments of $19.99" on late night TV, possibly diluting your own market in
the process? [16] If you were
clairvoyant, would you give time-consuming phone readings to
random callers for pocket change, or go to work on the stock market, lottery,
oil exploration, and similar avenues to great wealth? If you had the abilities, intelligence and benevolence of a
god, would the system you came up with involve most of your intelligent creations,
an eternity and a really hot oven?
More importantly, this can also be applied to the ad hoc rationalizations invoked to explain the original inconsistency. For example, a psychic may claim they're only interested in helping others. However, behavior that includes taking large sums of money from the grieving for purported contact with their deceased relatives can hardly be considered benevolent. In addition, if his excuse actually were the case, he could use his powers in myriad more effective ways, such as playing the stock market and donating all the proceeds to underprivileged children.
This factor, when properly taken into account, is remarkably effective at dispelling silly things; any toddler can make a difficult-to-falsify claim to explain away a raided cookie jar, but it takes a genius to invent one that's difficult to falsify yet makes a lick of sense. Like an evolutionary vestige, certain aspects of unfalsifiable beliefs can make perfect sense as a human invention, but come close to incoherence under the view that the claim is correct.
4.5: The Origin of the Claim under Questionable Circumstances
This criterion is fairly easy to understand, although it can
take a number of forms. In all cases, it will come down to one important
question: are
there indications pointing to human error or deceit as something responsible for
the origin of a claim?
For example, let's return to our earlier example of your best friend claiming he
had otherworldly contact with a dead mailman. Now, suppose he invites you over
to his house for a séance, and you, purely by accident, find a book beneath the
coffee table titled 'Pranks to Play on Best Friends'. You observe a horrified
look on his face, and begin reading.
When you do, you find that the chapter on spiritual pranks is
bookmarked, with the joke about a dead delivery man coming back to haunt his
route (remarkably similar to the claim your friend made) highlighted in red
marker. Your curiosity piqued, you look at the calendar and note that it's
April 1, otherwise known as April Fool's day, during which, as Western custom
dictates, you should play elaborate jokes on your friends.
Depending on how you responded to his claim earlier, you can either blush at
your indecisiveness, feel sheepish embarrassment at his successful deception,
desire to put him in the hospital, or feel smug at your critical thinking skills
and go to work planning one of your own jokes in revenge.
Similarly, if you remember talking with a science fiction writer who made statements to the effect of
"I'd
like to start a religion. That's where the money is!", you might smirk if you find that he's
now raking in millions writing books explaining how to
exorcise ancient extraterrestrial spirits from your brain.[17]
4.6: The Claim's Likelihood of Existence if Incorrect
This is my final criterion to use when deciding if a better explanation is warranted. Could the claim readily originate regardless if it was correct or not?
We know, for instance, that we're naturally predisposed to
recognizing faces and patterns, even where none actually exist. Hence, could we
reasonably expect to see people making themselves look silly by staring, with
religious fervor, at puddles and tree trunks even if the Virgin Mary and Buddha weren't
engaging in such public exhibitions?
Likewise, if disembodied spirits with the power to affect the physical world and
unusual interests in human affairs, commonly known as gods, didn't exist, could
someone be expected to invent them anyway? And is there precedent for thinking
such an erroneous belief could come about?[18]
Could humans, the only known species capable of understanding and dwelling on
their mortality, have a powerful incentive to develop the belief in an
afterlife?
If you answered "Yes" to all three questions, you just might be
a critical thinker. What's more, you've now finished reading all six criteria
I've outlined! Learn to use them well, as they can be of great benefit in
evaluating the esoteric claims we encounter in our day-to-day lives.
A child tells you about their invisible, vase-breaking, cookie-pilfering friend.
A patient at a mental institution explains his encounter with the gang from a
fantasy movie. An upstart cult leader claims extraterrestrial invaders are conspiring
to destroy the earth, and his group holds the only key to salvation.
Q. What do these three things (along with a host
of others) have in common?
A. They are all, strictly speaking,
unfalsifiable with the simple experiments we would use on specific empirical
claims like "I have a 1932 Quarter in my pocket." Nevertheless, they demonstrably fail most or all of the tests above;
tests which, when unanimously negative, are fairly reasonable indicators of
falsehood, and which several other unintuitive claims that were later vindicated (such as
powered flight) could pass with flying colors even in their heyday.[19]
This will automatically lead a reasonable person to conclude, with a degree of
certainty reserved for few things, that the better explanation is obvious--said claims are false, products of their times, fallible
humans and/or historical inevitability. This is the adoption of an alternate,
mutually exclusive, superior hypothesis.
Inferring the best explanation is a practice we all perform from time to time,
even if unconsciously. It allows us to stay mentally afloat in a sea of human
error while gathering the few bits of value that swim by from time to time.
So the next time you get asked for evidence that vampires or ghosts don't exist,
you can just say "That's irrelevant when dealing with such
intentionally unfalsifiable claims--all things considered, the best
explanation is that both were invented by superstitious people when they
were working with limited observations and just didn't know any better."
Alternatively, say you received divine revelation that they're lying, and ask
them to prove you didn't.
1. Argument from Ignorance is an error in
reasoning that occurs if one implies that a proposition is true
because it hasn't been proven false. Robert T. Carroll of the the Skeptic's
Dictionary has written a concise summary of the fallacy.
^ Back to text
2. This is known as the Burden
of Proof, and ultimately derives from ancient principles of common law such
as 'innocent until proven guilty'. ^ Back to text
3. The Richter Magnitude Scale is a
mathematical model used to calculate the strength of earthquakes by interpreting
seismographical data. The U.S.
Geological Survey hosts an excellent explanation on how it works. ^ Back to text
4. 'Baloney Detection' is a phrase borrowed
from the famous astronomer and skeptic Carl Sagan, who, in his
monumental work The
Demon-Haunted World: Science As a Candle in the Dark, outlined a similar (but
by no means identical) set of criteria for detecting erroneous claims. ^ Back to text
5. An ad hoc hypothesis is tacked on to another
to explain damaging observations when the predicted experimental consequences of
the latter are contradicted. For example, if a dowser is unable to find water
more often than chance would predict in a controlled experiment, an appeal to a temporary
inability due to psychological pressure can be made.
It's important to note that not all uses of the ad hoc hypothesis are
fallacious: if the proposed explanation makes unique predictions and is
falsifiable independently of the hypothesis it's designed to protect, one
can scientifically test it. Examples of valid, testable and verified ad hoc
hypotheses include antibiotic resistant bacteria accounting for the failure of
penicillin in certain patients.
On the other hand, unfalsifiable ad hoc rationalizations, or ones that
are made without taking into account empirical data that contradicts them,
are one of the great hallmarks of pseudoscience. ^ Back to text
6. The last conclusion is entirely reasonable to accept
given the miniscule odds of a cult leader randomly inventing a set of claims
that, by a stunning coincidence, correctly describe the location, attributes, plans,
motivations, and technology of a hypothetical intelligent species a galaxy away.
^ Back to text
7. This quote, while reflecting a reasonable
epistemological axiom found throughout the ages, is attributed to Carl Sagan. This
interview with the PBS series Nova, made while he was still alive,
expands on his reasoning. ^ Back to text
8. William of Ockham, while not the originator
of the concept, was a medieval philosopher who formally invoked the principle
that "plurality should not be assumed without necessity" on
numerous occasions. A short
biography of his can be found at the website of the University of Kent at
Canterbury, England. ^ Back to text
9. Flavius Josephus, an ancient Jewish
historian living only a few decades after the time of Jesus Christ, wrote down
these events, which supposedly occurred prior to the Roman sacking of Jerusalem,
in War
of the Jews, Book VI :
"... a few days after that feast, on the one and twentieth day of the
month Artemisius, [Jyar,] a certain prodigious and incredible phenomenon
appeared: I suppose the account of it would seem to be a
fable, were it not related by those that saw it, and were not the
events that followed it of so considerable a nature as to deserve such signals; for,
before sun-setting, chariots and troops of soldiers in their armor were seen
running about among the clouds, and surrounding of cities."
Such testimony about miraculous celestial phenomena can be found even in modern
times. Similarly, identifying what look like specific images in the vague stimulus that
clouds provide is a popular pastime to this day. This is known as pareidolia
in the psychology textbooks, and is the foundation of tests such as the
Rorschach ink blot.
It would come as no surprise to learn that the ancients, in a time of excessive
political turmoil,
could misinterpret naturally occurring phenomena as an omen, possibly
embellishing the events as the accounts were retold. Thus, we can explain their
reports entirely within range of known, well-understood phenomena, and need not postulate
unprecedented new ones. ^ Back to text
10. James Randi is a renowned skeptic and
infamous debunker of all things paranormal. A former magician and
mentalist, he's intimately familiar with the subtle psychological and mechanical
principles underlying many unfalsifiable beliefs.
The James Randi Educational Foundation, a group he started, has a standing
offer of $1,000,000 to anyone who can demonstrate a paranormal phenomenon
like dowsing, mindreading or telekinesis under conditions designed to
eliminate human error and fraud. Curiously, no one has ever passed even
the preliminary testing. ^ Back to text
11. Unorthodox medical treatments, in
particular, have a much higher antecedent likelihood of being ineffective than
those tested and approved by the rules of scientific investigation, which
include double-blind clinical studies.
Many conditions will go away on their own after a set period; even
serious ones like cancer have been known, rarely, to vanish (for poorly
understood reasons) without any treatment at all. If this occurs, by
coincidence, concurrently with administration of an ineffective
treatment, which is not at all unexpected considering their popularity
and the size of the population, the inference that one caused the other will be
an example of the Post
Hoc, Ergo Proper Hoc ("after this, therefore because of this") fallacy.
In a large enough number of trials, the statistically guaranteed occurence of
such individually improbable events is known as the Law of Truly Large
Numbers, and is explained by the Skeptic's Dictionary at this
URL.
In addition, there is little, if any way to diagnose the failures of these methods.
Even if 95% of the patients who receive a treatment report no improvement at
all, those cases can be attributed to lack of faith, improper ritual, or
other untestable ad hoc rationalizations. More importantly, such a sobering fact
will rarely be mentioned outside of the proverbial fine print. As a result, only the
'hits' are counted, but not the 'misses', as ultimately is the case with
virtually all testimonials. This is a very common psychological quirk
known as Confirmation
Bias, and can cause misconceptions about the treatment's actual
effectiveness.
Finally, the Placebo Effect can yield, in many cases, beneficial psychological
and even physiological results regardless of the effectiveness of a
treatment. If patients believe they'll improve as a result of the treatment, they
often will--at least temporarily. An upbeat attitude alone can alleviate
symptoms such as pain and hopelessness, and has been correlated with improved
recovery in a number of studies.
These facts, as well as additional ones, are noted in this article on spontaneous remission and the placebo effect by Stephen Barrett, M.D. As always, experimental controls, which are, regrettably, unavailable in anecdotal reports, are necessary to establish that a treatment's actual effectiveness is above that predicted by chance alone. Therefore, it would be downright naive to suppose that prayer, homeopathic medicine and even sunflower seeds (as treatments for, say, terminal cancer) would have a shortage of gleaming testimonials if they were, in fact, only as effective as a sugar pill. ^ Back to text
12. For an entertaining look at how proponents of a flat earth could reconcile their view with such things as astronomical and geological observations using a mess of disjointed ad hoc rationalizations, see this discussion at BlizzForums. ^ Back to text
13. A mass extinction that killed the
dinosaurs, as well as the majority of life on earth then in existence, appears
to have occurred towards the end of the Cretaceous Period 65 million years ago. A
gigantic meteorite impact, as determined from a crater around 125 miles wide
embedded in the Yucatan Peninsula of Mexico and other evidence, is currently
considered at least a partial cause of this event. The Hooper Virtual
Paleontological Museum has a detailed
webpage on the subject, while this
Space article expounds on the meteorite itself. ^ Back to text
14. In July 2003, the British
Broadcasting Corporation, in the hope of detecting the creature reported in
various sightings over the years at Loch Ness in Scotland, mounted a scientific
experiment that included 600 individual beams of sonar.
After the entire lake was probed and nothing was revealed, the team could only
speculate as to the root causes of the myth's persistence. As an additional
test, they hid a simple fence post underneath the water's surface and
raised it in front of tourists; most were not fooled, but several later sketched
monster-shaped heads instead of the square block of wood that was used. This
highlights the inherent fallibility of human eyewitnesses, whose preconceptions
can unintentionally mold mundane observations into meticulously detailed (and
incorrect) testimony. ^ Back to text
15. As Robert Carroll writes in his entry on
the Loch Ness
Monster, "... a minimum population of ten
creatures would be needed to sustain the population. The same report claims that
Loch Ness is incapable of sustaining a predator weighing more than about
300 kg (about 660 pounds) [The Naturalist, winter 1993/94, reported in The Daily
Telegraph]." ^ Back to text
16. This is in reference, of course, to
assorted get-rich-quick schemes whose proponents prey on our materialistic hopes
and dreams. The Federal Trade Commission has compiled
a list of advice and publications geared to helping us discern legitimate
work-at-home opportunities, which are few and far between, from the ever-present
scams.
Often enough, promises of fabulous rewards if you accept a proposition,
such as material wealth, or horrendous consequences if you refuse, like poverty
and gnawing doubt about "what might have been" for the rest of
your life, are enough to immediately shut down all critical faculties.
As 9 out of 10 professional con artists can attest to, wishful
thinking is one of our main psychological vulnerabilities. ^ Back to text
17. L. Ron Hubbard, former science fiction
writer turned founder of Scientology, reportedly said something similar to the
above, on the monetary incentive of starting a religion, in front of several different
witnesses.
This
well-researched 'non-Scientologist FAQ' by Don Lindsay outlines the evidence
and concludes that, while the most famous version of the rumor is unfounded,
Hubbard most likely did express such sentiments on more than one occasion, and
his church has used dishonest tactics to hide his blunder.
Later, Hubbard went on to start a religion that, apparently, claims an ancient
galactic overlord named 'Xenu' got rid of an overpopulation problem by
imprisoning the spirits of billions of subjects inside terrestrial volcanoes. These,
once escaped in modern times, purportedly possess us as 'body thetans', wreak
havoc by causing all sorts of psychological problems, and (conveniently) can be
removed by using the rituals and methods outlined by Hubbard, which also
supposedly open the way for our innate, suppressed psychic abilities to emerge. More information on
their beliefs can
supposedly be found here.
[Editorial: in this author's humble opinion, a science fiction writer who, in
all probability, said religion was a cash cow, starting a religion that sounds
like a 1950s B-movie... if there were ever a case where inference to a
better explanation disproved an unfalsifiable claim with more certainty than it
does Scientology, I haven't
heard of it.] ^ Back to text
18. The answer, of course, is a resounding
"Yes, belief in gods would exist if they didn't!" Assorted logical fallacies and quirks of our psychology,
many of which were examined above, can easily make us
err, interpreting the world around us as possessing driving intelligence
and decision making ability that can be affected with the proper rituals.
Additionally, historical precedent unambiguously supports the idea that we invent gods by
the thousands to explain unknown phenomena. One need only consider how common
belief in sun and lightning gods was throughout the ancient world, beliefs that
are (ironically) now considered debunked by the vast majority because of an
inference to a better explanation. ^ Back to text
19. A reasoned analysis of the claims made by
the Wright Brothers, using the criteria listed above from the point of view of a
resident of 1903 America, would likely benefit readers in understanding how
powerful they can be. A biography of Wilbur and Orville, complete
with timetable, is found on Time Magazine's 100
Most Influential People of the 20th Century webpage.
1. Consistency with Well-Established Knowledge:
Powered flight was observable by humans since the beginning of history in the
form of our soaring feathered friends, the birds, so there was nothing
inherently implausible about it, other than, perhaps, obtaining a power source adequate to support
itself and our own weight. By the time of the Wright Bros., even this had
been independently achieved in the form of engines.
While true powered flight had not been previously achieved by humans, other
forms had been. In the 18th century, Swiss scientist Daniel Bernoulli postulated
the Bernoulli
Principle, which could be used to calculate the necessary dimensions and
moving speed of a wing to lift a specific amount of weight.
Other forms of flight had also been available for centuries. In fact, the
airplane was largely an invention that utilized dozens of older developments. As
this
timeline illustrates, there were many achievements in this area prior to the
Wright Brothers:
In short, by 1903 there would be nothing particularly unusual about asserting
that powered flight had been achieved; at most, a citizen's surprise would rival
that of one
in 2003 learning of a cheap, affordable electric
automobile. Therefore, this test does not raise any red flags and we move on to
the rest.
2. Parsimony
While only three newspapers initially published the report on their flight, and
their accounts were in some aspects embellished by overzealous journalists, that
offers little reason to discount the claim being made itself. To postulate an
alternate hypothesis, one would need to account for three independent reports in
the press, which would involve lies, bribery, conspiracy theories, and other
unevidenced assumptions. Most likely, it could not be done with fewer
assumptions than the obvious explanation.
Moreover, the
flying machine itself could be demonstrated to anyone who asked, and alternate
explanations would falter as soon as it was presented. The
evidence, in short, was quite sufficient to make their flight the simplest explanation
that could account for the data.
3. Explanatory and Predictive Power
Powered flight is an empirical claim, and makes quite a few specific
predictions. There's nothing vague or unfalsifiable here. In addition, since the claim was by no means extraordinary, postulating
alternative explanations, such as a system of invisible wires or the like, would
result in more unevidenced ad hoc hypotheses and unnecessary assumptions than
would accepting it at face value.
Similarly, no unique observations could be predicted from the hypothesis the
Wright Bros. had only pretended to achieve powered flight, while the hypothesis
that they did was enjoying remarkable predictive power as more and more
successful demonstrations under increasingly difficult conditions were made and
interest (as well as scrutiny) in their invention waxed.
This is now effectively over. Additional considerations from the other factors
will not override the overwhelming preponderance of evidence, even if the
inventors of the airplane had been sociopathic maniacs with a flair for
pathological lying.
4. Internal Issues within the Claim
A pair of brothers with a bicycle repair shop, who regularly tinkered with
mechanical devices and had an interest in flight from a young age, were perhaps
the most likely people to discover flight in history. It's to be noted that this
raises the likelihood of an incentive for fraud, because they had a vested
interest in inventing a successful airplane; but that risk would be minimized
upon further consideration, due to observations such as the tentative and unsure
nature of the assertions they made, which con artists would have likely
exaggerated for the publicity if they intended to fool the public.
At any rate, their dishonesty could be shown false upon detailed examination of
the evidence, and it never was. Instead, the exact opposite occurred,
their assertions were vindicated, and fleets of airplanes copied from their own regularly shuttle
tourist and businessman alike to and fro the far reaches of the world.
5. The Origin of the Claim under Questionable Circumstances
The Wright Bros., as far as is known, had never made any statements or been
involved in activities that would cast doubt on their integrity or make it
likely they would invent a claim about powered flight without possessing
the actual goods. And given the technology available in 1903, as well as their
background, both mentioned earlier, there is every reason to suppose their claim
was accurate.
6. The Claim's Likelihood of Existence if Incorrect
One could claim to have discovered
powered flight, but as that assertion is by no means vague or unfalsifiable, it
would be trivially easy to disprove with empirical evidence. In addition, unlike a religious view or cherished authority figure, no one had a
vested emotional or psychological reason in defending powered flight once it was
shown false.
Furthermore, the event was not historically unique, easily
stuck in the recesses of vanquished past memories and contradictory reports, but
an ongoing one. Where one could could make false statements about unique events in the past,
safe in the knowledge that one could feign sincerity, play off public emotion
and never be conclusively shown a liar, an ongoing claim is only good
while it lasts.
For these and other reasons, it's extremely unlikely the Wright Bros.
could have pulled off a deception for any length of time if they didn't actually invent
powered flight.
And with that, we see that one of the usual anti-skeptic assertions so popular
among assorted proponents of unevidenced and unfalsifiable claims, that of "your dogmatic attitude would
have prevented powered flight from being achieved," is categorically
proven false. ^ Back to text
I would like to thank the following online acquaintances, whose feedback is listed in chronological order, for their contributions, suggestions and constructive criticism.
Dark Magneto, who originally inspired this essay via an AOL Instant Messenger conversation on the nature of falsifiability and epistemology.
wade-w, who suggested several relevant footnotes and references.
Hugo Holbling for his extensive analysis of an earlier version of this essay at the Internet Infidels board.
Lobstrosity, whose masterful knowledge of the intricacies of English grammar should be an inspiration to us all.
StillDreaming, who suggested addressing the role of ad hoc rationalizations (and the internal contradictions they produce) in more depth than the original essay went into.
Grünfeld, a person whose ability to point out fallacious logic and subtle errors in reasoning is rivaled by few.
Cronoss, who told me to lighten up on the technical jargon. While an article of this scope can't completely eliminate it, his point in general was a strong one, and certainly helped to improve its readability in the long run.
Derek Mathias, who offered many suggestions on improved sentence structure and humor/subtle jabs.
All others who, due to my possessing the memory of a mere mortal, I forgot to list above.
© 2003 Allan Glenn. You may freely print, reproduce and distribute this document in its original, unmodified form for private purposes, as well as link to it from any webpage and translate all or parts of it into other languages. If you want to modify it for a different reason than translation, please email me first - I'll almost certainly agree, but I like knowing where different copies of this essay might eventually show up. All commercial publishing rights reserved.