Examples of how anti-science forces use falsifiability

In my recent article in Scientific American, I wrote about one issue that I deal with more extensively in my book The Great Paradox of Science and that is that we should get rid of the idea of falsifiability being both a defining element of what makes a theory scientific as well as it being the driver of scientific evolution. I said that my argument that falsifiability is a myth that does not describe how science operates is borne out by a close examination of actual scientific history.

I also argue that getting rid of falsifiability takes away one of the main arguments that anti-science groups use to discredit the scientific consensus in areas like climate change, evolution, and vaccinations. They point to this or that piece of data that happens to disagree with the scientific consensus in those areas to argue that the consensus has thus been falsified. I argue that what is necessary to overturn a scientific consensus is a preponderance of evidence in favor of an alternative theory, and a few isolated pieces of evidence do not come anywhere close to achieving that.

In response to my article, I have got some private responses and also looked online and they fall as I expected. Historians and philosophers of science tend to have no problems with it because they have been discussing such ideas for a long time, while some scientists are not happy that I am attacking something they believe in strongly.

But the people who seem to be objecting the most are evolution deniers and climate change deniers, arguing that I am denying falsifiability because I am trying to save science from their critiques. You can see them in those posts making the very points that I said they depend upon in their efforts to undermine science using falsification.

My primary goal in writing the book and the article was not to save science from these anti-science forces. It was to enable people to have a deeper and more accurate understanding of the nature of science. It was an incidental side-benefit that it undermined the arguments of the anti-science forces. I am glad that it seems to have had that effect and that they are reacting with some concern that their weapon is being taken away.

It is interesting that some people in both the science and the anti-science sides are attacking me for writing the article and Scientific American for publishing it. It appears that it is not only politics that makes for strange bedfellows!


  1. garnetstar says

    I agree that merely repeating the reductive mantra “A scientific theory must be falsifiable” leads to a lot of confusion and trouble, among scientists as well as science-deniers. While the larger truth is there--what Mano states, a preponderance of evidence that conclusively supports an alternative theory “falsifies” the previous theory--the mantra is way too reductive, and so is wielded more as a sword to claim that one’s pet theory is the scientific one.

    Creationists are especially bad at this, as they completely misunderstand that mantra, and take it so simplistically. They claim that evolution isn’t scientific because it isn’t falsifiable, because we can’t do the experiment of repeating the evolution of all life forms from a single common ancestor. And, that one incident that isn’t (yet) explainable by evolution thereby falsifies it all. Also, they confuse data (or whatever ideas they imagine are data) that may be consistent with a theory, but is not conclusive of that theory.

    They fail to understand that what evidence there must be, and what must be repeatable, in order to conclusively support another theory, and “falsify” evolution, is the data. Not the event (evolution itself), or a single unexplained outlier, but all the data. Only a preponderence of actual, repeatable *data* that *conclusively* demonstrates creation would “falsify” evolution. As Mano says, one rabbit out of place isn’t enough. That one data point might be taken to be consistent with creation, but is not conclusive of it: it could be consisten with many other explanations as well. And also, it’s just one data point.

    Such rabbits must be repeatedly found, and also a multitude of other species, and there must be an entire body of *positive* evidence that demonstrates that the only explanation of these rabbits, and everything else, is creation (conclusive evidence). Negative evidence, like “Evolution can’t (yet) explain this one point, so it is false!”, isn’t enough.

    I once had occastion to try to prove that a sample of carbon was sp3-carbon (don’t ask, it’s too long to explain.) We easily aquired data that proved that no sp2-carbon (the only other possible kind of carbon that it could have been) wasn’t present. But, that is only *negative* data for the presence of sp3-carbon! We had to acquire data that positively and conclusively showed that sp3-carbon was present.

    Moreover, there must be *postive* data that conclusively confirms creation for every phenomenon that evolution explains, like carbon dating (and I’d like to see what body of positive evidence explaning *that* would prove creation) and everything else, in every other field. There must be positive evidence, and a body of it, that is conclusive for creation, in ALL the areas that evolution explains.

    Scientists make this mistake too! Extrapolation from one data point to “Your theory has been falsified!” is a big temptation to lazy reasoning and confirmation bias, even in scientists.

  2. Pierce R. Butler says

    Your denial of falsification falsifies the scientific method, and that PROVES the Bible!!1!

    (Sorry, but someone had to say it…)

  3. friedfish2718 says

    Mr Singham has not given any definition of what is “evidence”. One area of human endeavor which puts great focus on what is meant by evidence: the LAW.
    In the legal arena, one can file criminal lawsuit or a civil lawsuit.
    In a civil lawsuit, if the judge is convinced that more than 50% of the evidence favors the plaintiff, then the plaintiff wins. This is the PREPONDERANCE OF EVIDENCE, the standard of evidence. There may very well be solid evidence against the plaintiff but it is small (as seen by the judge) and is outweighed by contrary evidence.
    In a criminal lawsuit, a defendant is declared guilty if more than 99% evidence against him is proven.
    For proper science, the standard of evidence must be stricter than criminal law evidence (evidence beyond reasonable doubt).
    Mr Singham proposes that scientific evidence should be degraded to the level of civil lawsuit (PREPONDERANCE OF EVIDENCE) and yet he wants such evidence (and the hypothesis/theory it is supporting) to be undisputed.
    Mr Singham is the fox in the La Fontaine fable “The Fox and the Grapes” (the origin of the expression “sour grapes”).
    As science progresses, the uncollected fruit are hanging higher and higher. I daresay, the uncollected fruit are hanging EXPONENTIALLY (not linearly) higher and higher. I can see why Mr Singham is exhausted. Call it a day and declare the incomplete evidence to be full evidence.
    In the area of the evolution hypothesis, there are experiments which induced bacteria (initially incapable to synthesize, say, amino acid X) to be capable to synthesize amino acid X, by a gradually changing of the growing medium from 1 generation to the next. It took about 40 generations from what I recalled. I may be incorrect but there has been no successful experiments inducing unicellular lifeforms to evolve into obligate multicellular (say, 2 to 10 cells) lifeforms; successful experiments can come in the future.
    A commentator in this forum used a bad choice with the sp3 carbon. First, it is sp3 electron orbital. Second, when one says “carbon”, one thinks graphite and diamond; graphite has sp2 e-orbitals and diamond has sp3 e-orbitals; x-ray crystallography resolves the question. So? A more relevant example is cold fusion. To the cold fusion fans advocating, say, 1 kW cold fusion generator, that the underlying physics would imply a cosmos most different than what one sees today (Earth would still be a big ball of molten rock).
    Mr Singham’s arguments of “consensus of scientists”, of “preponderance of the evidence” are political arguments, not scientific arguments. Einstein was not impressed by an editorial by 30-odd german scientists condemning his work. Einstein stated that 1 person is sufficient to show his work to be incorrect.
    For centuries, Fermat’s Last Theorem (1637) remained in the hypothesis/conjecture stage. During that time, people did mathematical work on the assumption that the conjecture is true. In 1994, Dr Wiles proved the conjecture to be true. Now Fermat’s Last Theorem is an actual theorem. Question: outside of collecting fossil records, has anyone used the evolution hypothesis to predict the range of future species and the scheduling of the appearance of such future species? All I am hearing is extinction. Now and then one hears about a discovery of a new species. Is this evolution in play? Is the specimen of the discovered species the very first specimen ever? Can one find the immediate evolutionary parent species?
    Question: what is the difference between a mathematician and a philosopher?
    Answer: Mathematician uses pen, paper, and wastepaper basket. A philosopher uses pen and paper.

    Despite being a theoretical physicist, Mr Singham has become a philosopher.

Leave a Reply

Your email address will not be published. Required fields are marked *