How We Know What Isn’t So

By Phil Plait | April 11, 2005 10:04 am

I’m reading a book right now called “How We Know What Isn’t So” by Thomas Gilovich. It’s about how we think, and where thought processes go wrong. When that happens, we either believe things that are not true (like astrology), or don’t believe in things that are (like the Moon landings).

Gilovich has lots of interesting stuff to say, and so much of it is relevant to science and critical thinking! I have long said that it’s okay to dismiss some notions because they are so wrong they don’t deserve to be entertained. A lot of people say I am dogmatic when I do that, but really, when someone tells me that an invisible giant planet will destroy the Earth next week, I think it’s okay to dismiss that. Gilovich has this to say about this idea:

“People are inclined to see what they expect to see, and conclude what they expect to conclude… At first blush, such uneven treatment of new information strikes most people as completely unjustified and potentially pernicious… it brings to mind… groups blindly adhering to outmoded dogma. .. [but] it is also inappropriate and misguided to go through life weighing all facts equally and reconsidering one’s beliefs anew each time an antagonistic fact is encountered.”

In other words, some things really shouldn’t bear equal treatment.

He uses the word “fact” there in the last sentence. I think he should have used another word, maybe “idea”, or “piece of evidence”, because if a fact comes along that does contradict your belief, you do indeed need to rethink your belief. Even if evidence comes along, you need to weigh it. But not all evidence has equal weight either. Evidence that is repeated and independent is pretty heavy, while an anecdote bears little or no weight.

Then he says something I find pretty funny:

“In evaluating more clear-cut information, our perceptions are rarely so distorted that information that completely contradicts our expectations is seen as supportive.”

In other words, he’s saying that given great evidence against us, we rarely use that evidence to support our beliefs. To that I say, “Ha!”

He’s never met Bart Sibrel, James McCanney, or Richard Hoagland. These guys constantly use evidence that directly contradicts what they are saying, yet claim it supports them. Whether it’s the Moon Hoax, a physically impossible model of the solar system, or the face on Mars, direct evidence blowing these ideas away is used by those men as triumphs of their theories. It boggles the mind to think they actually believe what they say. Of course, if they don’t believe in what they say, then maybe Gilovich is correct anyway. I don’t know if they believe in what they say or not, but it doesn’t matter. Either way, what they are saying is wrong.

A final word: Gilovich goes on to say:

“We humans seem to be extremely good at generating ideas, theories and explanations that have the ring of plausibility. We may be relatively deficient, however, in evaluating and testing our ideas once they are formed.”

To that he is exactly right. We can always come up with an explanation ( “It was astrology! Planet X! Sunspots! Telekinesis! Homeopathy!” ) for some event, but as humans we’re pretty poor, maybe even lazy, about explaining our explanations to ourselves.


Comments (12)

  1. Homeopathy! Astrology!
    I work at a health food store and a lot of my co-workers are really convinced by these two, uh, systems of belief. It’s really tough to fend them off when they claim things like: “I can feel the restorative power of this homeopathic medicine right through the bottle,” or, “How can you say that’s not a glass worm?” or “Oh! Your sign is undergoing a lot of flux from Vespa today!”
    Will it ever end? Will I ever be allowed to work with rational people again? Argh!
    To the glass worm believers and the astrology mumbo-jumbo I can usually contrive a pertinent response, but to the homeopathy stuff, all I can usually manage is an open-mouthed expression of disbelief. No wonder I’ve been ill more often at this job than any previous job: no one takes their recovery seriously–they just buy the snake oil, consume it, and continue spreading diseases until their bodies are able to fight it off. ūüėõ Pah!

  2. Zamboni Schwartz

    Just goes to show that when somebody is desperate to believe something, for whatever reason, they will cling to whatever lies they have to in order to support the lie in their own minds.

    Case in point: Today’s episode of the Montel Williams show featuring Sylvia Browne.

    Usually I just roll my eyes and laugh at her cons, but today she went too far. A woman came on the show with a home video of her 2 year old son playing with blocks. The video showed a “dancing red orb” (her words, not mine) around the child. The woman thought it was a supernatural being, and wondered if (since it was the color red) it was an evil spirit trying to hurt her baby. Sigh….

    It got worse from there. Sylvia claimed that there was “Nothing to be worried about, because it is an angel protecting your child.”

    Bullpuckies, I say. You know why? Because the “red orb” only moved when the camera moved, only jiggled when the camera jiggled, and was obviously (to anyone even slightly familiar with optics or cameras) a reflection on the lenses from a light source THAT WAS VISIBLE IN THE CAMERA’S FIELD OF VIEW.

    The show was being used to show how “wonderful and powerful” Browne was, but inadvertantly presented absolute clinching proof that she is, indeed, full of crap.


  3. Cass

    In the “13 things that don’t make sense” issue of New Scientist, Madeleine Ennis found that “homeopathic solutions – so dilute that they probably didn’t contain a single histamine molecule – worked just like histamine.” This is when tested on white cells directly.

    Can’t wait to see results from blind trials because I find homeopathy really hard to believe.

  4. Jarno

    To Cath – you might be interested to know that homeopathy has indeed been tested in a very rigorous double blind study (overseen by none other than James Randi), and it was found that homeopathically prepared “medicine” worked no better than untreated distilled water.

    I saw a BBC documentary on the subject actually, and it described several experiments with varying results; the common theme to the tests that produced negative results was that they were much more rigorous and well documented than the tests with positive results.

    BBC wanted to set the record straight by accepting Randi’s million dollar challenge (Randi has promised to pay a million dollars to the first person/institution to show him a supernatural event in a scientifically controlled setting… his money has been safe so far, despite countless attempts to win it). As the results where tabulated, the outcome was clear: no difference between the homeopathic medicine and distilled water.

    The individuals who got positive results pretty much seemed to either work alone, or be and work with people believing in the power of homeopathy; both things of course compromise the credibility of the tests: working alone in a single lab there’s certainly higher possibility of unintentional error and contamination of the samples than when working with a group in a carefully planned experiment. Working with believers, there’s very likely an element of error introduced by the natural bias of the people interpreting the results.

    So if you find homeopathy really hard to believe… that’s quite healthy, in the light of scientifically rigorous tests! :)

  5. √Ę‚ā¨ŇďPeople are inclined to see what they expect to see, and conclude what they expect to conclude

    In Terry Pratchett’s novel “The Wee Free Men”, the heroine has a special ability: first sight, or the ability to look at what’s in front of her and recognize it for what it is, rather than what someone might expect it to be.

  6. Eric

    I work at a health food store, and I sell homeopathic medicines as well as use them myself. I know from personal experience that they work. However, I have checked the dilutions of the medicines that I use, and they are not very dilute, mostly 3X or 30X, with only a few of the ingredients very dilute such as 200C. (And I don’t claim to be able to feel their effects through the glass bottle!)

    This is only my opinion, as I am not a scientist or doctor, but I think that, in order for homeopathic medicines to work, you have to start with the less dilute concentrations. You can then work your way down to the extreme dilutions where there are no molecules of the active ingredients left. In that case, they work just like the bell in Pavlov’s experiment. In that experiment, the dog would salivate when he heard the bell because he had been conditioned to expect food. A person using plain water would react because his body was expecting the active ingredient, so his body would react in anticipation of the ingredient’s being there. Note that, if, at the beginning of Pavlov’s experiment, he had simply rung the bell and not presented the dog with any meet, the dog would not have salivated.

    It is my guess that, if the double-blind experiment were to be repeated, but this time a less dilute preparation were presented to some of the subjects, and plain water given to the control subjects, and then the dilution gradually decreased until all the subjects were given plain water, then those subjects who were initially given the less dilute preparation would continue to respond as did Pavlov’s dog.

    As I say, I am no scientist, but I think that the above explanation is plausible, and I refuse to stop using medications that are helpful to me because some experiment failed to verify that they work (although, as I’ve said, the homeopathic medications that I take are not highly diluted, unlike the ones used in the double-blind test mentioned above).

  7. Tomasz Tkaczuk

    Isn’t that the placebo effect?

  8. Eric

    No, it’s different from the placebo effect in two important ways:

    1. With a placebo, you don’t start with an active medicine and then switch to a placebo; you start right away with a preparation that has no active ingredients.

    2. A placebo only works if the subject doesn’t know that it’s a placebo. But conditioning works even if the subject knows that he’s being conditioned. For example, suppose that I ring a bell and present you with your favorite food. I do this many times, and you salivate each time. Then I tell you, “Okay, I’m going to ring the bell, but I won’t present you with any food this time.” When I ring the bell, you’ll still salivate, even though you know that no food is forthcoming, because you’ve been conditioned to do so.

    In my training on homeopathic medicines (and I emphasize that it was sales training, not scientific training), we were taught that homeopathic medicines do not heal the body, but rather they train the body to heal itself.

    Remember when you were in school learning to read. You first learned to read large print, and then the print was gradually reduced until you were able to read small print.

    When homeopathic doctors prescribe homeopathic medicines, they first start with lesser dilutions, so that the body can learn to recognize the active ingredients. Then they gradually increase the dillutions as the body learns to heal itself better. Then (and this is just my hypothesis, but I think that it is a reasonable one), when the dilution is such that there is no longer any active ingredient left, the body has been conditioned to heal itself in response to the stimulus of taking the medicine, so that the active ingredients are no longer needed.

    Let me add that the homeopathic medicines that I use and sell are all over-the-counter and come in only one strength, which is only slightly diluted, so they still have the active ingredients in them. I have never used any dilutions that were so strong that they contained no active ingredients. But I think that the above is a reasonable explanation as to why homeopathic medicines failed Randi’s tests even though many people, myself included, claim that they do indeed work.

  9. I don’t want this comment section to become a back-and-forth on homepathy, but having said that…

    Homeopathy has no scientific basis whatsoever. The “Law of Likes” on which it is based has no evidence to support it in the way homeopaths claim.

    Eric, in your case, there is no way to determine if you felt better because of what you took, or if because you would have gotten better anyway. However, if you study it using hundreds or thousands of such cases, then you are on to something. But as far as I know, no unbiased study of homeopathy has ever shown it to work. There is a lot of info out there on studies. is a fairly strident website, but it does have some decent information, as does

  10. Eric

    Okay, fair enough.

    I checked the ratbags website and clicked on the link for quackwatch. On quackwatch, I did a search for two different topics, “homeopathy” and “circumcision.” Under “homeopathy,” there were pages and pages of pasionately argued refutations of homeopathy. Under “circumcision,” there were a few bland paragraphs that didn’t take much of a position one way or another.

    Circumcision is a form of quackery that has harmed millions and millions of boys. Why don’t those debunkers of quack medicine who argue pasionately and at length against practices such as homeopathy expend the same passion and energy arguing against the practice of circumcision? Might it have something to do with the fact that circumcision means big money for conventional doctors, whereas alternative medical practices threaten their revenue?

    If the same reasoning that is advocated in “How We Know What Isn’t So” is applied to circumcision, it can be seen to be based on falacious reasoning, but I don’t hear those arguments coming from skeptics and debunkers of quackery.

  11. Nelson Martin

    Dear Mr. Plait,
    I purchased and read Mr. Gilovich’s “How We Know What Isn’t So.” I did so as I read somewhere recently that within its pages is a reference to a question posed: If you fold a piece of paper 100 times, how thick would the resulting piece of paper now be after the 100th fold.
    Most people I have asked, guess at a foot thick, a yard thick, etc. The answer is 8,000,000,000 times the distance between the earth and the sun.
    Whether that is the case, but I folded a an entire page of a newspaper, and could only fold it 7 times, the thickness was then in excess of 1″. I did the math on 50 folds and the result on only 50 folds was in excess of the distance to the sun, almost two times the distance, as I recall.
    Can you help me??!! Please – I need the source of this query as my friends are throwing mud pies for starters, and now find myself unable to prove my case or come up with the original source of my info. If nothing else, could you give me a hint as to how I could contact Mr. Gilovich?
    Appreciate your help and enjoyed reading you site.
    Nelson Martin


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


See More

Collapse bottom bar