Statistics

DDP: Quack Cures For What Doesn’t Ail Us, Part II

I’m at the Doctors for Disaster Preparedness meeting in No-History City, Louisiana. Here’s Part II of (an earlier draft of) the speech I gave Saturday. Consider donating $1 for every typo you discover. Read Part I.

Theoretically speaking

I often say the root of all scientific evil is the love of theory. And it is so.

What is a scientific theory? Nothing except a set of premises—propositions, observations and the like—from which we deduce statements about the observable universe, where by “universe” I mean all the material + energy there is, including that in any so-called multi-verse or many-verse.

Here is a simple theory. Days following hot days are usually hot. This is as fully a scientific theory as that produced by the largest grants at the top indoctrination centers. From our humble theory, and given the observation “Today is hot”, we deduce that tomorrow is likely to be hot. There are no quantifications of “hot”, “usually” or “likely”, and there is nothing in the world wrong with that. The theory is understandable without quantification. Quantification is not what makes a theory scientific; indeed, over-quantification and quantifying the unquantifiable can ruin an otherwise workable theory.

Our theory does two things which all good theories should. One, it fits past data well. Two, it makes testable predictions.

Anybody can try our theory, and if decisions made relying on it work out for some individuals, they might want to adopt the theory as their own. Not everybody makes the same decisions. Some might want more precision in temperature; our theory will be of little to no use to them.

Should we only look for true theories? No.

What makes a theory true, and not just good or utilitarian, is when each and every premise in the theory is itself true (and the combination not self-refuting in some way). In our theory, it is true, given our observation, that today is hot. But it is not universally true that days following hot days are usually hot. That premise (that part of the theory) is only likely given other observations and premises. So we do not have a true theory; merely a good one.

I know of no scientific theory that is true in this absolute sense. To be a true theory, I reiterate, every single premise comprising the theory must itself be true, provably true given a chain of argument that indubitably leads back to unshakeable axioms and sense impressions. Candidates for true theories, then, are simple ones; indeed, the simpler the better. This is why particle physicists are much closer to true theories than fluid physicists.

True is a harsh, brutal, exacting word. Mostly true is not true; it is a little bit false. Mathematicians and meta-physicians speak of truth, and well they may. But scientists used to be, and now ought to be again, more cautious. They must re-learn to speak in uncertainties.

Richard Feynman? said, “When we know that we actually live in uncertainty, then we ought to admit it; it is of great value to realize that we do not know answers.”

Theories which are not true are thus always in need of fixing in the direction of truth. And since there are not absolutely true scientific theories, this is why the scientist who says, “The debate is over” is a bad scientist. He has confused a likelihood with truth, a telling mistake. He has conflated probability with decision, an all-too common error.

One thing immediately follows from learning most scientific theories are not true. It is this: most scientific theories are not false.

To be false, we have to prove at least one of the premises of the theory is itself false—and false is just as concrete and immovable a word as true—and this is unlikely because blatantly false premises are rarely found in scientific theories. Or we have to observe something the theory said was impossible. And when I say impossible, I mean just that: a probability of zero, something that no matter what cannot happen. Since most theories speak of predictions in probabilistic terms, they rarely or never say anything is impossible. Thus we cannot falsify most scientific theories.

Even the weatherman saying “Tomorrow’s high will be 88F” is not falsified by observing a high of 89F, because everybody, including the weatherman, knew there was a little unquantified plus-or-minus in his theory’s predictions. That “fuzz” is present in every scientific theory I know, including our favorites, like global warming. True falsification is as rare as a bureaucracy closing because they say they have fulfilled their mission.

Since truth and falsity will not be wholly found in scientific theories, though they are de rigueur for mathematical and metaphysical theories, what else can we use in their judgement? Usefulness.

Walks like a duck

Scientific theories make predictions about observables. Astrology make predictions about observables. Therefore, astrology is a science. It isn’t a very good one, but it is a science. David Berlinski makes this point in The Secrets of the Vaulted Sky: Astrology and the Art of Prediction.

Bigfootology, or xenobiology, is likewise a science since it makes predictions about observables. So is parapsychology. So is climatology. So is inorganic chemistry. So is radio engineering. So is homeopathy. So is any theory science that makes predictions about observables.

Yet why do we say astrology is bad and some others good? An astrological theory makes statements like this: “Given that the moon is in the house of Mars and Venus is rising, and that you were born son this certain date, this week you’ll experience greater fortitude.” The theory is the set of starry rules and the observation of the birth date. The deduced prediction is greater fortitude.

There is no mechanism, no sense of the cause of the greater fortitude, except by pointing to the rules, and the rules have only vague things, or nothing, to say about cause. Mars is the god of War, and in war you need fortitude, say.

Scientific theories do not have to say anything about cause. Every citizen before Newton knew to duck when throwing a rock into the air, and none of them (or almost none of them) had any theory beside the empirical, i.e. non-causal, theory What Goes Up Must Come Down.

Newton, and Einstein after him, and whoever comes next, did not obviate or destroy the citizens’ empirical theory. It remains as predictively accurate for most of mankind as it did for Adam, and will remain that way until the last stone is thrown. Rocks did not suddenly descend at different rates because a new theory was proposed. Our theories do not make the universe.

There is no real way to check the premises of the astrological theory. They are stated as true; some even believe they are true. They are accepted or rejected by prejudice. All we have to go on are the predictions themselves.

Ask somebody whether they’ve experienced greater fortitude this week, and they might say yes, just to be cooperative. They’ll find some small instance where they asserted their bravery—not screaming when seeing a cockroach, maybe—and claim that. If they want to believe in astrology, and they knew of the prediction, they’ll search very hard indeed for this small instance. And they will nearly always discover it.

Given their discovery, the astrological theory has therefore made an accurate prediction, and therefore there is not only no reason to doubt the theory, but a very good reason to believe it. This is rational, as far as it goes, and it is the reason astrology is still and ever with us. It is easy, in a certain sense, to verify as accurate astrological predictions.

Now we know there are better ways to test astrological theory. Make predictions specified with exactitude, so that the predictions can be verified unambiguously. Spell out, in advance, just what fortitude is and what it isn’t. And keep the predictions hidden from those to whom they are applied. There is some trickiness in this which we can ignore here, but you get the idea.

Have we answered why astrology is a bad science? Sort of, but we didn’t state the key fault, which is this: ambiguity. It’s the ambiguity of the predictions which maddens. A skeptic will hear the forecast of greater fortitude and find just as many instances of its lack as a believer will find instances of its presence. The forecast will not verify for the skeptic, but will for the believer. Our pair are conditioning on different information. Both will be exasperated the other can’t see what is plain to them.

Dissuading the astrology believer is thus a Herculean task. We have to endeavor on a complete, thorough disquisition on the nature of evidence. We’re going to have to demonstrate how resolving ambiguity casts doubt on the theory’s veracity. We’re going to have to wade through mountains of case studies. We’re going to have to think deeply about cause.

And even then, even after all that, we will be left with two inescapable facts to which the believer may cling. One, we will not have proven the theory false; the premises of the theory are not capable of disproof; they may always be believed. And two, even after removing all ambiguity, we will be confronted by those times in which the astrological forecasts were accurate. Even monkeys throwing darts can pick good stocks.

There is no use putting such extraneous hits down to “chance”. There is no such thing as chance. Statistics, and probability, cannot prove cause. (I wrote a book on this called Uncertainty: The Soul of Modeling, Probability & Statistics.) There will be times when astrological predictions are accurate, and there must be some real, actual reason why they were accurate. And that reason might be the astrological theory is right! You cannot prove—and prove is another of those adamantine words—it wasn’t.

The only real hope you have of converting the believer is to change his metaphysical perspective. He believes astrology because he believes it is possible for the stars, or the universe, or whatever, to cause changes in his behavior or demeanor. This is something he wants to believe. To dissuade him from this means replacing that metaphysical position with another, such as the rank (and ultimately unsatisfactory) skepticism of the materialist—or the living religion of a theist. Everybody knows how difficult a task this is. Why, it often takes a miracle.

We’re all gonna die!

What holds for astrology holds for theories of environmental doom.

On his blog Real Climate Science, Tony Heller documents failed prediction after failed prediction using the very words of the doomsayers. In 1992 (he showed in an 18 July 2017 post) the warning was of an ever-widening ozone hole, which was an “alarming threat” which caused “the degradation of the conditions necessary to sustain life on this planet.” One scientist rang the familiar cry, “It’s far worse than we thought.”

And came the hard data showing the size of the “hole” has been essentially unchanged since 1990 (sometimes bigger, sometimes smaller). It must have been disappointing to some that the sky didn’t fall after all.

Paul Ehrlich in 1970 said that “The death rate will increase until at least 100-200 million people per year will be starving to death during the next ten years”. This did not happen. Something like the opposite of it did, in the sense that agricultural production has been rising beautifully ever since.

Was Ehrlich’s theory falsified? No. Was the theory about the ever-shrinking ozone hole falsified? No. Global-warming-of-doom? No.

It could be—and Ehrlich makes this very claim—that we have not yet reached the doom of which he spoke. The “fuzz” surrounding his dates, which everybody knew was there, was just a little wider than we first thought. Mass starvation is a live possibility.

The ozone might flee forever. The globe could burn up. All of it might still happen. Just as we might experience greater fortitude when our horoscope predicts.

How do convince believers these are all bad theories?

We could and should and must lead these believers through lectures on the nature of evidence, on what ambiguity means and how to resolve it in relation to their forecasts, as we did with the astrologer. We should scientists who believe in environmental doom what skill means and why they don’t have it. We can show how other theories make superior predictions.

But like with the astrologer, we can scarcely prove their theories wrong. And there will always be times and places, localizations, where their theories scored small hits. It will always be possible for folks to retain a tight grip on their cherished theories.

Again, just like with the astrologer, changing their mind requires changing their most fundamental beliefs. They must come to a new metaphysical view of the world; yes, come to a new religion, one which it makes their heart sing, and not darken, to hear, “Go forth and multiply.”

That may very well take a miracle. Without it, we have a long, slow endless battle ahead of us.

Categories: Statistics

13 replies »

  1. The following statement in the article needs correction: “We should scientists who believe in environmental doom what skill means and why they don’t have it. We can show how other theories make superior predictions.”

  2. Richard,

    See the leading note about donations. These typos are placed by my enemies and are a tremendous annoyance.

  3. Evensongs:

    The Day Thou Gavest, Lord, is Ended,
    The darkness falls at Thy behest;
    To Thee our morning hymns ascended,
    Thy praise shall sanctify our rest.
    We thank Thee that Thy church, unsleeping,
    While earth rolls onward into light,
    Through all the world her watch is keeping,
    And rests not now by day or night.
    As o’er each continent and island
    The dawn leads on another day,
    The voice of prayer is never silent,
    Nor dies the strain of praise away.
    The sun that bids us rest is waking
    Our brethren ’neath the western sky,
    And hour by hour fresh lips are making
    Thy wondrous doings heard on high.
    So be it, Lord; Thy throne shall never,
    Like earth’s proud empires, pass away:
    Thy kingdom stands, and grows forever,
    Till all Thy creatures own Thy sway.

    https://www.youtube.com/watch?v=Pigh8VHr-ZE

    (Scream at crispy screatures if they’re close by. They never listen. A colleague put a bumble bee under my CCTV and enlarged it for my professional opinion on my return. . The Screen was brown and furry!)

    His cynicism is a kind of truth that fears hope. Hope is considered dumb.
    https://www.youtube.com/watch?v=p67d9F9nW2Y

    Go on Singing…hope springs eternal.
    https://www.youtube.com/watch?v=gE9b3f-OFGE

  4. Briggs ==> While it is perfectly true that Erlich’s theory was not falsified, his prediction was shown to be incorrect. The fact that he placed his prediction in a time frame, “in the next ten years”, means that the rest of us can point out that the predicted event did not take place within that ten years — thus his prediction failed.

    That his predicted event also did not take place in the subsequent two decades runs out the “well, maybe my timing was off” excuse. In fact, as you point out, not only hasn’t the predicted event taken place in the three (now almost four) following decades, bu t the opposite of the predicted event has occurred — survival rates are up all over the world — we can say with some growing sense of confidence that there may be something seriously wrong with the theory on which the prediction was based.

    The same is true of many gloomy-doomy predictions, including the failed predictions of James Hansen and the Climate of Doom crowd. 29 years after Hansen’s Senate side-show announcement, the Global Average Surface Temperature of Earth is a chilly 59°F or 15°C — sweater weather where I come from. (P.S. – This is why they use “anomalies” instead of reporting the GAST in degrees — no one except people of the Arctic think that 15°C is HOT).

  5. If Ehrlichs theory is not falsified because he made not one, but multiple false predictions using that theory, I fail to see why any other theory should be falsified after making multiple false predictions. Why the special treatment?

    For instance, all the nutcases computing the day of Armageddon have perfectly fine scientific theories too. Who cares no prediction was correct?

  6. day of Armageddon…Who cares no prediction was correct?

    Everybody. And nobody. That these predictions of date are incorrect is of no interest to anybody outside the nutcase community. That the world will end is a given, both from logic and revelation. The second source (but not the first) assures us also that we will not be able to predict it.

    Ehrlich is a different case, in that he was predicting (and goes on doing it!) both what will happen (based on no revelation and little logic) and when it will happen.

    The nutcases in both sets ought to be ignored.

  7. Mr Briggs should be expecting a commendation and a fat cheque from the Association for the Propagation of Nonscience for this one. He might even get appreciative mention from The New Age Philosophy of Anti-reason Society and The Academy To Destroy Science By Confusing and Debasing Terminology and Concepts.

    In any kind of valid scientific method a theory has a very specific meaning. It is an hypothesis (possible explanation for some observed phenomena) that is demonstrated to be consistent with known premises and is in accordance with all relevant observations. It is, therefore, a likely explanation for the what and how of the phenomena. If it can be proved by logic and experiment that there are no, and can be no, exceptions to the theory then it’s a scientific law.

    Mr Briggs has treated us to a litany of fantastic conjectures, irrational speculations, wild guesses he calls “theories”; the effect of which is to degrade physical science and the science of logic (philosophy) into a mere rationalisation of ideological fancies.

    Woe betide you, and your ideological cronies, Willy.

  8. @Andrew Brew

    A predection about an event has to specify the time it will happen, otherwise it is not a prediction. The time that an event is happening is one of its most important attributes, after all.

  9. I found this post was one of the best I’ve read ever.
    The word hypothesis is a give away in it’s construction,
    hypo thesis:

  10. “Scientific theories make predictions about observables. Astrology make predictions about observables. Therefore, astrology is a science.”

    Men have ears. Dogs have ears. Therefore dogs are men.

    “He believes astrology because he believes it is possible for the stars, or the universe, or whatever, to cause changes in his behavior or demeanor.”

    Not neccessarily. He might view astrology as an example of synchronicity.

Leave a Reply

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