Skip to content
October 31, 2016 | 6 Comments

A Deep Philosophical Account Of Probability

docsay

Another review of Uncertainty: The Soul of Modeling, Probability & Statistics, this one taken from an Amazon customer.

It was 7 or 8 years ago when I was sitting in my office at the university. At that time I was a Ph.D. student in the Machine Intelligence group at the computer science department. One of the department professors knocked on my door – he was looking for my supervisor because he needed advice on how to perform some test to verify some properties of his recent experiments. Well, Finn is not here, but maybe I can help. What’s your problem? He wanted to know about different statistical tests and how and why they work. Why they work? This question puzzled me immensely because I had taken many advanced classes in statistics, read countless books on the subject, but I never recalled any of them answering or discussing “why they work?”.

This lead us to Briggs’ new book “Uncertainty – The Soul of Modeling, Probability and Statistics”. It is a deep philosophical treatment of probability written in a plain language and without the interference of unnecessary math. This makes the book accessible to most university students. The books “Probability Theory: The Logic of Science” by E.T. Jaynes and J. Pearl’s “Causality” are the ones that have influenced my thinking most profoundly. Until now. Briggs explains why subjective, Bayesian or frequentist interpretations of probability are somewhat unfortunate and argues that the fundamental view should be to see probability as the extension of logic to the domain of uncertainty. I have met this view in other places before, but this is the first comprehensive treatment I have read. He also argues that all probability must be conditional – again, it is not the first time I have seen this view, but the first time I have seen a deep analysis of why that must be so. Now, it may not sound like much, but it really is. It has already allowed me to get a fresh perspective of one of my AI research problems that has plagued me for years.

Have you ever speculated what randomness really is? This book will tell you. Is there a mathematical definition of falsifiability? Oh yes. Do you ever wonder what the relationship is between probability and causality? And what is the role of statistical significance testing in relation to causality? A few years ago I read “The Cult of Statistical Significance” by S. T. Ziliak and D. N. McCloskey which is basically arguing against certain statistical practices while emphasizing the focus on effect sizes. Briggs’ book stand out because his analysis is much deeper (mathematically, philosophically) and because he goes much further by proposing that both p-values and relative risk should be abandoned, although he dislikes p-values the most. To read Fisher’s old gibberish, that led to this sad situation, is simply astounding.

I also have a little critique. First of all, I had hoped there was a section on the notion of “unbiased” estimators, but maybe Briggs can add that to the second version. Secondly, there are brief discussions of machine learning algorithms for causality. The reader could get the impression that people in this field think they can prove causality. If so, that is certainly not the case. From the little I know, then they always assume some kind of faithfulness of the distribution, or they take the graphical model as inductive knowledge (e.g. in the case of Pearl). Of course, the problem is, as pointed out by Briggs, that once the techniques get in the hand of less rigorous scientists, then they tend to forget that and immediately think causality has been proven. Briggs is kind to remind us that there is a difference between conditional and necessary truths, and once you start to assemble all your assumptions, the conditional truths may quickly become very uncertain.

In general, then this book should be relevant for anybody working with probability models and anyone consuming the output of such models. That’s a lot of people, including almost each and every scientist and university student. If you are a journalist, then read it too. It will give you a much better basis for accessing the nature and validity of all the research fluctuating in the media.

This lead us back to the question I was asked by the professor 7 or 8 years ago: “why do they work?”. I said that I didn’t really have a book on that (and I have many books). I think you might have to get the original paper(s) and see what’s in them. Then we discussed his problem a little, and I suggested a chi-square test and sent him out the door with a bunch of books, among them one with the reassuring title “100 Statistical Tests”. Today I would simply have given him Briggs’ book and said: they don’t work and here’s why!

Best regards,
Thorsten Jørgen Ottosen, Ph.D.
Director of Research


Here is my reply to his two criticism.

Ottosen is quite right that I can do a better job describing efforts to determine causality. My criticism in the book is not just that people take “machine learning” models, and all probability and statistical models, and assume they have proved cause, which they cannot, but that any model by itself can demonstrate cause where it is not previously known. I have more about this in the article “The Hierarchy Of Models: From Causal (Best) To Statistical (Worst)“. Automated processes can never identify cause because knowledge of cause is different than knowledge of determinism, and machines might be able to discern determinism in simple cases, or in multiple-choice (oracle) set ups.

About “unbiased estimators”, there are no such things. I mean, there are no such things as parameters at all. Parameters are not ontic. All parameter-based methods, which comprise the vast bulk of existing practice, should be eschewed. I argue this at some length in the book, but that I didn’t get the point across proves I need to be clearer here.

In the place of parameters and hypothesis testing? Understanding cause and making probabilistic predictions of observables.

October 30, 2016 | 25 Comments

Media Lying With Polls? Is The Pope…Er…Ah, Never Mind

4918917310_b41d65d3c4_z

Today’s post is at The Stream: Be Skeptical of the Polls: Is the media lying to boost Hillary?

Note: I’m putting the Summary Against Modern Thought Sunday series on hiatus until after the election, on 13 November.

The other day on 27 October—this was all before the DickiLeaks business, an important point—another State media poll came out which had Hillary +9. Nine full points ahead of Trump. Nine.

State media? The media that colludes with the elites of both parties to maintain the status quo. The folks who cycle in and out of and cozy up to government. All the usual suspects.

Anyway, after the unbelievable nine-point-lead poll, there was a poll immediately after, which put Honest Hillary Clinton at +12. Twelve points up! Twelve! (I am writing this next number standing on my desk, shouting…) TWELVE!

These were my actual thoughts, in order, when I first heard that number:

  1. HA HA HA HA HA HA HA HA HA HA!
  2. Only graduates of Yale or Wesleyan are going to believe it.
  3. What is it going to do to the souls of those who know better but feel they have to defend it?
  4. State media has slit its own throat.

Now in these once United States, anything north of four or so is called a “landslide”. An idiotic name, but one which encapsulates the truth that it is difficult in the extreme to win presidential elections by large numbers. Consider that the Olympian himself, the Self-Anointed One, the Big O, beloved and adored far and wide, and genuinely popular in 2008, finished the race with 53% to McCain’s 46%, a huge 7-point lead (I’m rounding all numbers for ease of presentation).

A race that bettered this was Reagan trouncing Mondale 59% to 41%, an 18-point whopper of a margin. And then there was monumental year of 1972, which saw Nixon doing to McGovern what Germany did to Poland, crushing him 61% to 38%, a 23-point earthquake. A real landslide.

Since Nixon, victories average about 4 points. Yet in the +12 poll, we were asked to believe Hillary is so popular, so beloved of the nation, so exciting a candidate, that she could win by a rare and stunning twelve points. We were asked to believe Hillary was more popular than Obama was in his first run, when the State media was wetting itself over Barry’s pants crease. We were asked to believe the impossible.

When asked to believe the impossible, don’t believe it. Why? Because it’s impossible.

Whatever else happens in this race, it has been a pleasure watching the media destroy what little credibility they had left. Thank God for alternate sources, like the one you’re reading now.

Go there and read the details!

October 29, 2016 | 47 Comments

Why Is It Okay To Eat Pigs And Not Dogs? Science Does Not Have The Answer

4876969608_22634ee3b4_z

Opening sentence of the abstract in the peer-reviewed paper “When Meat Gets Personal, Animals’ Minds Matter Less: Motivated Use of Intelligence Information in Judgments of Moral Standing” in Social Psychological and Personality Science by Jared Piazza and Steve Loughnan: “Why are many Westerners outraged by dog meat, but comfortable with pork?”

The single-word answer which explains everything—bacon—never occurs to the authors.

Neither does the well known truth that dogs bite back when chased (pigs can be nasty, too). I’ve eaten both animals and I know. Pig is more versatile. Everything from chops to the loin is juicy and delicious—where would we be without pickled pigs feet?—whereas dog has fewer choice cuts and is generally stringy or greasy. And don’t even get me started on sausage, though some dogs do resemble walking sausages.

Dog as food is still found in parts of the world. If this article is right “China, Indonesia, Korea, Mexico, Philippines, Polynesia, Taiwan, Vietnam, the Arctic and Antarctic and two cantons in Switzerland” still serve up barking burgers. (I had a dog taco in Mexico, you racist.)

But maybe bacon isn’t the answer after all. Fetching a dog bone is (as many article say) taboo in the enlightened cities of the West (I say cities), and when the media reports on dog-as-chow there is usually what is termed a flap.

Before venturing into the paper, why would you say there is this excitement? Custom, say I. We’re long used to eating pigs and have for just as long and maybe longer valued dogs as companions. We consider it rude to eat companions. Whereas we’re fine with slicing and dicing an animal kept in a pen far from eyes and fattened for feasting. Plus you can’t entirely discount bacon. Call this the commonsense, or Custom answer. Now let’s see what our authors say.

It’s Piazza and Loughnan’s theory that our dinners’ intelligence should be the driving factor in drawing lines between what’s acceptable and taboo, figuring it’s fine to chew less intelligent animals and eschew more intelligent ones. Yet

people will actively disregard intelligence information when considering the moral standing of certain animals that pose a moral challenge to the consumer. That is, while evidence for an animal’s mind is generally persuasive, it is not compelling when a person is motivated to defend their use of the animal as food.

Piazza and Loughnan don’t appear aware that many people know that people are people and not animals, in the sense that people are rational creatures and animals not. This is why “Rise, Peter; kill, and eat” makes (made) sense to most of us. Considering a beast’s “intelligence” is a recent phenomenon, constrained to Westerners who have forgotten the distinction between men and animals.

Anyway, our duo took to the internet and asked 58 people to “imagine that in the distant future, scientists went on an expedition to another planet and discovered a new species called the ‘trablans.'” Half the people were told the trablans were intelligent, half not, and they were then asked questions like “Is it OK to start eating the trablans?”. Lo, a few more of the folks who were told the trablans were dumb said yes than did the people who were told the trablans were intelligent. A wee p-value “confirmed” the difference was in fact a difference.

Two other sets of people and questions were asked with similar results.

Imagine the faux pas of you leading a team of scientists to a distant planet where you end up serving what turns out to be the native rational trablans for supper. No such embarrassment need happen if trablans are mere animals.

So there is a test: do the trablans possess a rational nature like humans, or are they mere beasts? If the latter, fire up the coals and ice the beer. If the former, they are really like us whatever they look like, and cannibalism, though at times and places accepted, is against natural law.

That “times and places” is key, incidentally. If you argue against natural law, how many people and in how many places have to act a certain way to make that way “right” or “moral”? One? A thousand? Must they all live contiguously, or is geographic separation allowable? Without delving into it, it’s easy to see that the only consistent solution is natural law.

Let’s let the authors have their final word: “Smart animals deserve our moral concern, unless, of course,
we want to eat them.”

October 28, 2016 | 14 Comments

One World Without Borders Would Lead To Tyranny

16590286855_2845c714ec_k

Hillary, in her own words, said “My dream is a hemispheric common market, with open trade and open borders”.

She said it, but it’s not clear if she believes it. It’s unclear because she offered these views, many times, to rich powerful banks anxious to pay Hillary preposterous amounts of money for Hillary to tell them that open borders are good. Could it be—stretch your mind here—that these banks were not paying for her wisdom (have you thought Hillary wise?) but her future cooperation should she become president?

That’s a political question we can leave aside. More interesting is the call itself for the elimination of all borders. The call, once feint, grows lower.

Before that, consider the false but alluring ideology of Equality. Equality is the pernicious doctrine that (1) everybody is equal in theory but unequal in circumstance, and that (2) everybody could be made equal if circumstances were adjusted. The stronger the grip this ideology has, the more any inequality, no matter how trivial, outrages.

A person under the delusion of Equality will say it is unfair that one man was born here, and another there. The disparity will rankle and he will march. But it is, of course, impossible for both men to be born in the same place and time and under identical circumstances. This is why the equalitarian calls for the leveling of circumstance. This is why the equalitarian would punish, fine, or restrict one man if the equalitarian considers the man benefits from superior circumstance (it may not be superior in reality), and it is why the equalitarian would reward, embellish, and loose a second man if the equalitarian considers the second man suffers from inferior circumstance (again, it may not be inferior in practice). The equalitarian would insist on these actions against the will of both men.

Equality necessarily leads to tyranny. Why? Because the moment equalitarian controls are removed, inequalities arise, and since inequalities are unacceptable, the chains must needs reappear. Tyranny. Equality also always, and just as necessarily, leads to a massive inequality: there must under Equality be masters, the equalitarians, and slaves, the “equal.” This is obvious to equalitarians (as it is not for those not under the sway of the ideology), and it is the seductions of the great and vast powers they will wield that drives the equalitarians’ desire for Equality.

And so now we turn to The Atlantic, which published an essay by Alex Tabarrok, which gives the game away in its title (editors usually write titles), “The Case for Getting Rid of Borders—Completely: No defensible moral framework regards foreigners as less deserving of rights than people born in the right place at the right time.

There is the heartfelt cry against the inequality of circumstance. The cry is taken up immediately:

To paraphrase Rousseau, man is born free, yet everywhere he is caged. Barbed-wire, concrete walls, and gun-toting guards confine people to the nation-state of their birth. But why? The argument for open borders is both economic and moral. All people should be free to move about the earth, uncaged by the arbitrary lines known as borders.

Man is not born free, and man is never free in the sense meant by Rousseau and his followers (they mean Do What Thou Wilt, whereas true freedom is being unrestricted to do what is good). At the very least, infants are chained to their mothers else they die. And the same by analogy this is true of the rest of us. Equalitarians confuse the ties that bind us and make us stronger with chains that restrict desire. By a cruder analogy, you cannot take the innards of a functioning machine and toss them into the wind with the cry “Be free!” and hope the machine will still work.

Tabarrok’s only argument is that inequalities exist. “Nature’s bounty is divided unevenly,” he says. “Closed borders compound these injustices, cementing inequality into place and sentencing their victims to a life of penury,” he weeps. Tabarrok, a staunch materialist, can only put inequalities in terms of money. Spiritual and cultural worth is not quantifiable, money is measurable. And because it is measurable, it takes on meaning which far exceeds its importance.

No standard moral framework, be it utilitarian, libertarian, egalitarian, Rawlsian, Christian, or any other well-developed perspective, regards people from foreign lands as less entitled to exercise their rights—or as inherently possessing less moral worth—than people lucky to have been born in the right place at the right time.

This is confused because Tabarrok bases his morality on rights and not duties. Start from a false premise and you could end up anywhere. Worse, starting with rights assumes what it sets out to prove, that all should have the same rights (and equivalent circumstance): the argument is circular. What’s odd is that Tabarrok misidentifies moral worth as being in a state of equal circumstance, while simultaneously denying it. If moral worth is to be the key metric, as it should be, in judging a human life, then circumstance is always circumstantial.

It’s always most difficult to see what is missing, and what isn’t there in Tabarrok’s argument is the identification of who is to be in charge once all borders are erased. Ask him if he sees himself in the role of equality enforcer. He also forgets that when there is only One World, there is nowhere left to run to.