Proof Probability & Statistics & AI Don’t Discover Cause

Proof Probability & Statistics & AI Don’t Discover Cause

Everybody has read, or seen adaptations of, Bram Stocker’s documentation of Dracula, the undead count, unlate of Transylvania. From these pages and films, we all know how to kill a vampire. In Van Helsing’s own words, “I shall cut off her head and fill her mouth with garlic, and I shall drive a stake through her body.”

Nipping off a noggin in plain enough, as is stuffing a mouth full of the stinking rose. But there is still ambiguity: what kind of stake?

Again, Van Helsing: “a round wooden stake, some two and a half or three inches thick and about three feet long.”

Important information, yet Stoker still failed to tell us what kind of wood. Luckily, Charles Dickens, a mentor and confidant of Stoker, had Scrooge give us the answer, an answer that was plain to contemporaries of both men, but which plainness has faded somewhat for us. Scrooge said, “every idiot who goes about with ‘Merry Christmas’ on his lips, should be boiled with his own pudding, and buried with a stake of holly through his heart. He should!”

So it’s holly. The stouter branches of ilex aquifolium, to be exact, that plant being indigenous to Europe.

But what if we tried maple, oak, or cherry stakes? Pine? Apple? Stoker, and Dickens with him, didn’t absolutely say it had to be holly, or that it couldn’t be any other wood. Or that the stake couldn’t be 3.1 inches thick, or only 2.5 feet long.

Or what color. Or of green wood or dried.

Obviously, we must experiment! Are we not told that the only sure path to scientific knowledge—which is like regular knowledge, but better, because it is certified by Experts—is the randomized controlled experiment? Do they not insist that anecdotal evidence, the data gathered by our own eyes and experience, does not count?

Therefore we must plan a randomized controlled experiment, calculate a p-value, and then, and only then, can we decide if other woods, lengths, colors, smells, thicknesses, pliabilities, bark status, left- or right-handed thrusts, stakes with and without vampire heads lopped off, head lopped off first then stakes through heart, and vice versa, black versus white garlic, raw versed roasted, and so on and so forth.

We need to gather a large number of undead, the exact number being decided by complex sample size equations, thrusting some through with holly, others with oak, and so on, all randomized. If we don’t randomize, the results we gathered would not count.

Why? They would not have the blessing on randomization upon them, a blessing which is needed in order to apply probability to the results. Or, rather, to extract it. The results themselves are imbued with probability by the magic of randomization.

If this sounds absurd to you, then you have not had statistical training, and if you did, it didn’t take, for which you may consider yourself lucky.

The rules are that randomized controlled trials must be conducted and that other evidence, while it can be considered, with a sniff, the nose arched, will never be conclusive. P-values—those are conclusive.

When I wrote Uncertainty, ISIS was making their tour in the Mideast. They were then crucifying their enemies. If memory serves, their victims were crucified on the standard wooden posts and crosses. Old fashioned, but effective.

I wondered what would happen if we switched from wood to metal. Obviously, using null hypothesis significance theory, or even standard Bayesian theory, we’d have no choice but to conduct “gold standard” randomized controlled experiments. Our prejudice that metal would serve just as well as wood just would not do. Only “data” generated in the appropriate matter can count.

Now, even a strict frequentist would agree that the substance of the crucifixion—wood, metal, or whatever—would not matter. What does is the exposure and rough treatment. These kill, not the cross itself. That conclusion is deduced. Which is a violation of statistical theory.

That’s not rare. All frequentists, for instance, always and without fail interpret confidence intervals and p-values as Bayesians would, a strict error according to their theory. This never seems to bother them, even as they defend their theory.

It should be clear we are always after cause, even if we can’t reach it. It should also be clear we do not need to conduct new experiments to deduce conclusions about cause. I stress “new” because we already have gained substantial knowledge of steel and iron, for instance, through other means, which allows us to deduce metallic crucifixion would work swell. We only conduct experiments in an attempt to confirm our suspicions of cause.

Subscribe or donate to support this site and its wholly independent host using credit card or PayPal click here

18 Comments

  1. Ye Olde Statistician

    Metallic crucifixi could be re-used more easily; provided, however, they are thoroughly washed and sanitized between uses to avoid spreading covid among the cruciarii.

  2. Joy

    Briggs,
    You can get ointment for prickly heart

    Obviously, we must experiment! Are we not told that the only sure path to scientific knowledge—which is like regular knowledge, but better, because it is certified by Experts—is the randomized controlled experiment? Do they not insist that anecdotal evidence, the data gathered by our own eyes and experience, does not count?
     
    There’s a place for both and many other methods of finding out truth and all paths leading in that general direction.

  3. Sheri

    Apparently the statistical training did not take. One would have thought the calculus-based version would have helped but it did not……

    We can’t experiment with any of this, because in the infamous words of Scully on XFiles, there are no vampires “because they are not real”. No vampires, no experiments.

  4. Hagfish Bagpipe

    “I shall cut off her head and fill her mouth with garlic, and I shall drive a stake through her body.”

    Possible Covid treatment?

  5. Hagfish Bagpipe

    ”Scrooge said, “every idiot who goes about with ‘Merry Christmas’ on his lips, should be boiled with his own pudding, and buried with a stake of holly through his heart. He should!”

    No post on statisticalizing probabilitization uncertainology is complete without a plum Dickens quote.

  6. DAV

    Holly randomization, Batman! The stakes are getting higher!

    there are no vampires

    So then all of the methods tried worked. You don’t see any vampires around do you?

  7. GP

    Sloppy, sloppy Briggs, how can you call any study on the efficacy of crucifixion conclusive unless it is double blind? Test subjects easily might be dying simply due to preconceived notions on the lethality of crucifixion.

  8. JohnK

    > We only conduct experiments in an attempt to confirm our suspicions of cause.

    I’m wondering if that sentence is a mite contradictory. To me, the above sounds neo-Platonic — Platonic ‘suspicions’ of cause (unwisely) quasi-married to the Aristotelian logic. (We recall that for Plato, the Forms were forever beyond our reach; we can occasionally create ‘likely accounts’ — we can have ‘suspicions’ of cause — but never do we possess true Knowledge of things; but this is one of the radical disagreements between Plato and his erstwhile student Aristotle).

  9. GP

    JohnK
    There is no contradiction, because even if nothing can be truly certain an individual can easily believe the knowledge confirmed (and they often do).

  10. Heresy!
    Concepts such as “cause and effect” and “logic” and “reasoning from facts” are obviously cis-hetero-normative, white, Christian male oppression!

  11. Carlos Julio Casanova Guerra

    In 1930, Einstein Bohr and Heissenberg got together in Brussels. The bunch of Copenhagen explained to Einstein the way quantum physics was going… Einstein wasn’t pleased: due to lack of knowledge, it had to use statistics. Then, Einstein utter his famous sentence: “God doesn’t play dice”. Bohr answered: “yeah, but we cqnnit dictate God how to rule the universe”. The use of statistics anr probabilities in science us a matter of the subject studied. “Hard” science diesn’t use statistics, that is ridiculous. For real, it is infuriating to see all those ignorants wielding “science” in order to enslave the entire population of planet earth. So evil, so stupid….. BUT CUNNING AND DEVOID OF SCRUPLES AND LIMITS!!!

  12. BDavi52

    Indeed, “we are always after cause”, but to be clear — we are actually only after unambiguously recognizable causes….especially the singular cause….even more especially, the ‘recognized’ & ‘singular’ cause that fits well with the accepted narrative.

    But obviously we begin that search for cause by memorizing, first, the appropriate narrative.

    Robert Aaron Long killed 8 individuals at 3 massage parlors in the Atlanta area back in March. Immediately the media was filled with speculations about ‘motivation’. Even the various Police and City Spokesmen would continually reference that search for cause and its critical importance: “It’s too early to know a motive…there’s still more work to be done”.

    But Long had already told us what ’caused’ his rampage. “He believed he has a sex addiction and that he saw the spas as “a temptation that he wanted to eliminate.” But never mind all that. That causal dog don’t hunt.

    The Narrative determined that the ‘real’ motive for the shootings (despite what Long himself believed)…had to be racism. The fact that the killer gave no indication of any racial bias didn’t and doesn’t matter. #StopAsianHate mattered. Plus the killer was a Whitely Toxic Southern Male. That mattered too. Well all that and the fact that he used a Gun because the Narrative tells us quite clearly that Guns cause killing.

    So what was the ‘real’ cause, absent the Narrative?

    Who the heck knows?
    We know what the killer said, what the media said, what the talking heads say. If we took the time we’d know what Oprah and Ellen and WhoKnowsWho might say. Take them all together shake them up and toss them on the table: is that cause? Or is ’cause’ as a philosophic concept applied in the real world much deeper than all that?

    Why was Robert going to massage parlors?
    He was a 21 yr. old Southern Baptist who already struggled with what he called ‘sex addiction’ (and had sought treatment at HopeQuest there in Atlanta, a Christian Therapy Center..where did that come from? He considered suicide (to deal, evidently, with these ‘personal failings’, why? Why didn’t he? Why did killing ‘the temptation’ become a higher priority goal? Why that day? that hour? Why those particular people?

    And let’s go deeper still. Why did those victims work there? Why those particular locations? Why those hours? Why did Robert struggle so much with his sexual desires and the expression of those desires? How can a ‘deeply religious’ Southern Baptist reconcile homicide as a solution to temptation? What did his parents do or not do as they raised him? What did he experience or not experience that bent him in this particular direction? Why, in the final result, did he choose to pull that trigger?

    We are always after cause — especially in incidents like this one (and negative experiences overall) because we like to believe that KNOWING a PROBLEM CAUSE allows us to create a PROBLEM SOLUTION and thereby avoid similar experiences in the future.

    How does that happen? We extrapolate from young Robert’s life ….we look at statistical surveys which tell us correlation (How many Southern Baptists commit violent crimes? How many Atlanta males commit violent crimes? How many known ‘sex addicts’ commit violent crimes? Does sex addiction correlate with broken homes and father figures? Does violence correlate with domestic violence in the home as a child? Does it correlate with income? with wealth? with college degrees? with hunting? with the availability of guns? How many ‘massage parlors’ are operated by Asians? If Asians represent x% of the overall population what is the % of Asian participation in sex work? And on and on and on. Every question triggers 10 more.

    And since randomized controlled experiments for murder, addiction, prostitution, etc. cannot be conducted on human populations we look instead to Sociological Studies applied to ‘randomized sampling’ and known outcomes relating to crime and economics (which are both regularly counted). We do all that while simultaneously KNOWING that the entire field suffers from significant Study Replication error.

    And more — we do all that knowing that CAUSATION is indeed the onion that can never be completely peeled, save by God.

    But all that doesn’t matter because the narrative already guarantees us: White Racism, Toxic Masculinity, and Guns caused the Massage Parlor Deaths! So there.

  13. C-Marie

    “…. Jesus said to his disciples: ‘I have come to cast fire on the earth and how I wish it were already blazing! There is a baptism with which I must be baptized, and how great is my anguish until it is accomplished! Do you think that I have come to establish peace on the earth? No, I tell you, but rather division.

    From now on a household of five will be divided, three against two and two against three; a father will be divided against his son and a son against his father, a mother against her daughter and a daughter against her mother, a mother-in-law against her daughter-in-law and a daughter-in-law against her mother-in-law.’” —Gospel of St. Luke (12:49-53)”

    Divisions among family, friends, acquaintances, strangers, due to the acceptance of the Lordship of Jesus, will only increase as times get more difficult. Jesus will see those through, who will receive His help.

    God bless, C-Marie

  14. Uncle Mike

    Isaiah Chapter 60

    13 The glory of Lebanon shall come unto thee, the cypress, the plane-tree and the larch together; to beautify the place of My sanctuary, and I will make the place of My feet glorious. (JPS 1917 Edition)

    13 The glory of Lebanon shall come unto thee, the fir tree, the pine tree, and the box together, to beautify the place of my sanctuary; and I will make the place of my feet glorious. (King James ver)

    13 The glory of Lebanon will come to you, the juniper, the fir and the cypress together, to adorn my sanctuary; and I will glorify the place for my feet. (New International ver)

    It is said that Isaiah prophesied the Crucifixion in this verse, and the “place of my feet” refers to the foot board on the True Cross. Various versions list many species.

    The “pine” is likely Aleppo pine [Pinus halepensis] but it’s unlikely that it was used for crosses or much else in the Holy Land. Not common, not strong, not large enough for timbers.

    The “cypress” is likely Italian cypress [Cupressus sempervirens] which grew in abundance in western Asia. An excellent timber tree frequently mentioned in the Bible, it was allegedly used by Solomon to build the Temple. Good candidate for the True Cross.

    The “plane-tree” may be Old World sycamore or Oriental plane [Platanus orientalis] a deciduous tree that ranges from the Balkans to Kashmir including western Judea. A lovely tree, much valued as an ornamental, makes poor timbers (but figured wood is sometimes used for cabinets and other furniture). Unlikely candidate.

    The “larch” and “fir” are likely not those genera at all, but probably Cedar of Lebanon [Cedrus libani]. Cedars grew extensively in the Holy Land, are mentioned in the Bible 70 or so times, and were used for timbers, ship building, and palace construction (including the Temple). Commercial use (and foraging by goats) led to near extirpation, but Cedars of Lebanon are making a comeback today. IMHO the most likely candidate for the True Cross.

    The “juniper” and “box” are likely Syrian juniper [Juniperus drupacea]. It is the tallest of all the junipers (20m though rare specimens may reach 40m). It may have been used for timbers in the past, but more so as a food (the berries) and an ornamental. Not a likely candidate.

  15. Robin

    Not sure what’s been said in the article holds for all possible techniques in statistics or AI, if you include something like Neural Networks or multi-variate analysis; used in a negative sense (to eliminate possible non-influential accurately measurable parameters).

    It would be possible, I suppose, to discover causation in that case even without any prior knowledge, purely in an exploratory/predictive sense, rather than a traditionally hypothetical one, as long as the predictive outcome contained sufficient repeatability.

    Not very efficient way to go about it though.

  16. Jim G

    Neural networks, and by extension MV analysis, are not magic but only models, and as Briggs says, models can be designed to prove anything. They are simply a state machine, fed selected data features which they partition into a multi-dimensional output space of decisions. Data can be easily biased to produce any output desired by the author. This leads to the condition of garbage in = garbage out in many cases.

    Modern deep networks simply allow a larger number of adjustable parameters to allow fitting more training data. They are still state machines and don’t think for themselves. Areas in that multi-space that don’t contain data can flip into any desired output condition. For example, your self driving Tesla might see the moon one night at an intersection and think it has a green light instead of a red light.

    The problem with humans and their actions is that they don’t always fit into neat little boxes.

  17. C-Marie

    Joy … Re the discussion about Churchill here on Briggs … Churchill by Roy Jenkins has arrived!! Riveting reading!! Thank you so much for the recommendation!
    God bless, C-Marie

  18. Joy

    C Marie, bless you,
    Thank my Dad…(actually)
    If you ever come to London, see the war rooms, then Chartwell, in Kent, there’s still a ginger tom cat and you can see his paintings in the workshop and his wall! His view of the countryside…

Leave a Reply

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