January 11, 2013

Is statistical reasoning too novel for contemporary humans?

In a post below, I quote a bit from David Brooks of what is considered sophisticated modern thought. As an example of "serious discrimination" and "racism," he reports:
Both blacks and whites subtly try to get a white partner when asked to team up to do an intellectually difficult task. ... Clearly, we should spend more effort rigging situations to reduce universal, unconscious racism.

It's possible that Brooks is being satirical here, although there's no evidence that his commenters notice the absurdity. 

In general, I suspect that our moral reasoning methods just haven't caught up to the vast advances in statistical reasoning over the last couple of centuries. 

49 comments:

Anonymous said...

"It's possible that Brooks is being satirical here, although there's no evidence that his commenters notice the absurdity."

Well, it's too 'subtle' for me too.

Anonymous said...

We must rig 'subtle' satire to be more accessible for universally dumb folks of which I am a part.

Tom S. said...

Seems the whole point of "moral reasoning" is its independence of statistics and everything else concrete.

Humans clearly make use of statistical reasoning all the time without even knowing it. We make a lot of decisions based on expectations about what is going to happen. We just don't use the word "statistics" for most of them.

Moral reasoning is one of those other processes that frequently opposes statistical reasoning. This is probably one of the reasons people love it so much. It makes them feel special to say, "See, look at me, I'm strong enough to resist what seems obvious!" They may also get some kind of internal ascetic pleasure from it.

Anonymous said...

It's possible that Brooks is being satirical here

That was my first thought...the use of the word "rigging" is very suspicious.

Anonymous said...

I don't know if he was being sarcastic. Brooks seems like he really believes in both "universal unconscious racism" and the ability of people like himself to get rid of it through social engineering.

Anonymous said...


Both blacks and whites subtly try to get a white partner when asked to team up to do an intellectually difficult task.


Is this really true? I bet Brooks is exaggerating or obfuscating. I bet that the truth is that both blacks and whites try to get a white male partner when asked to team up to do an intellectually difficult task.

Chicago said...

The idea of 'rigging' things so people think the way someone else wants them to is indicative of a dishonest outlook. If some sub-groups create certain impressions about themselves then in order to change that shouldn't they simply do things differently, thus creating a different perception? It seems to be all about manipulating perceptions rather than actually doing something about the reality of things. A lot of people have appointed themselves champions of another group without any of it's members having ever actually asked them to do so.

WMarkW said...

I was looking at this article today about "America's most diverse cities," along with a county map of diversity, as measured by the percent of the largest ethic group.

http://homes.yahoo.com/news/the-most-diverse-areas-in-america-222906910.html?bcmt_s=m#ugccmt-container

And here's an unrelated map of county homicide rates:

http://www.nationalatlas.gov/articles/people/IMAGES/crime_murder.gif

Seneca said...

"Tom S. said...
Seems the whole point of "moral reasoning" is its independence of statistics and everything else concrete."

There is a whole branch of moral reasoning called utilitarianism developed by John Stuart Mill that uses statistics. It advocates creating the greatest good for the greatest number. It was a very significant political philosophy in England in the 19th Century used to justify important government policies.

Interestingly, Steve's Citizenism is a modified form of utilitarianism in that it advocates the greatest good for the country's citizens (not the greatest number which would include every potential immigrant too).


Even contemporary philosopher Rawls' use of the idea of setting policy under a veil of ignorance (not knowing if you would be born rich or poor, smart or dumb, etc...) has some statistical moorings.

What Brooks is advocating is nothing more than Cultural Marxist hogwash about creating the "new man" free of White racism (Jewish racism is or course okay).

In short, the Scoth-Irish even Anglicized ones like Brooks feel safer in a multicultural society so that is why they have been strong proponents of diversity and immigration, no matter how much it harms people already here by driving down wages or increasing crime or social disorder.

Keeping a lid on things requires that people not notice things.... which is what Brooks and his ilk hope people continue to do.

It is also why so few read the Times anymore..

desert lady said...

I have come to the conclusion that over 95% of all internet political comments are posted either by idiots or by paid shills.

Anonymous said...

Isn't it more about unconscious self-interest here? If Brooks legitimizes even just a little discrimination, it could lead to people with different preferences wondering why they can't discriminate according to their preferences. And some of these people may have preferences that aren't held by Brooks, or may discriminate against Brooks.

ziel said...

That was a weird column of Brooks's. I've always assumed you were right that Brooks reads your blog and knows better, but this column suggests otherwise. But maybe, in fact, he is saying not so much "we need better ways to root out subtle racism" but "we need to understand the underlying reality that leads to these subtle, unstated assumptions."

ziel said...

That was a weird column of Brooks's. I've always assumed you were right that Brooks reads your blog and knows better, but this column suggests otherwise. But maybe, in fact, he is saying not so much "we need better ways to root out subtle racism" but "we need to understand the underlying reality that leads to these subtle, unstated assumptions."

Anonymous said...

Whenever I have to lift something really heavy, I subtly seek out a strong person to help me.

Such terrible unconscious prejudice on my part.

Assistant Village Idiot said...

Let us strive to be fair - not necessarily to Brooks, who makes his living by stating opinions and has some responsibility to think them through, but to the general reader. We grow very attached to our world-views. One chink in the armor is never enough for any of us to change course. We all juggle contradictory and even hypocritical ideas all the time. They are easy to see in others.

Definitive, slam-dunk, positive proofs of anything are rare. We are usually confronted with repeated instances where the evidence points against us, not unanswerable disproofs of our ideas. When the discomfort is too great in the aggregate, we begin to move our opinions to the area of greater comfort.

Belief in not-worse performance on a variety of measures by NAMs is protected first by the belief in environmental disadvantages and institutional racism to explain the observed phenomena that crime, IQ, illegitimacy, etc are in fact worse. That is a double hurdle, and most of us would rather just wave it off, thinking that maybe there's something to it, but it would be too socially expensive to say the truth, so let's pick a restaurant instead.

What we can do is create islands where people can think (and say!) the unthinkable, at least for a while.

@desert lady: speak not ill of idiots. It is hard to see the obvious and say it aloud, like the boy observing the emperor's clothes. After years of effort and training, I am still only able to be simple half the time, hence my continuing assistant status.

I do aspire to becoming a journeyman village idiot someday. Ah well, I am off to argue philosophy with the Town Drunk.

Anonymous said...

Couldn't being against discrimination be more about self-interest, whether conscious or unconscious, than the difficulty of statistical reasoning?

If you start legitimizing even just a little discrimination, it could lead to people with all sorts of preferences wondering why they can't discriminate according to their own preferences.

Lucius said...

I fear this is almost a kabbalistic thing with David Brooks. It's mirrors within mirrors.

It is intriguing to think he reads you: there's an interesting frisson in that notion.

But what would it avail? if he only works to turn your ideas on their heads?

And though he is famous and carries weight (or is that just water?) with/for someone, what sort of person takes him seriously? Can one imagine what Edmund Burke would do with this court eunuch who abuses his name?

Anyway: what you're saying about statistical reasoning is, I suppose, clear enough in its thrust; yet it's also unfleshed-out.

I think sometimes about how B. Russell poo-poos all logic before him. The Anglo-Analytics sound always very confident with this sort of thing, until I remember that all the questions they find propositional logic inadequate for sound awfully like imaginary problems, or at best, purely mathematical ones.

We must remember that many of the premonitions about the world of "race,class, gender" statistics can help us search into were once the acknowledged and unchallenged collective currency of the world-- two centuries ago. "Moral reasoning", as known in a Dr. Johnson, Shaftesbury, Kant, or Madison is surely more enlightening than today's jesuitical casuistry in the service of things for which the word "license" is too tame--all too often produced with a tip of the hat, however spurious, toward Statistics.

Anonymous said...

Look, if ever there were a more apt time for the cowardly elite like Brooks to use stats to write about government policy, it's now, in the gun debate.

Have you once heard a network commentator or writer like Brooks say, "You know, gun violence committed by the non-mentally ill, the majority of all gun violence, occurs over and over throughout the decades in certain areas of the country and among certain groups...and those groups ain't whites, Asians, or women."

Harry Baldwin said...

desert lady said... I have come to the conclusion that over 95% of all internet political comments are posted either by idiots or by paid shills.

And I take it you place yourself in the remaining five percent. Care to put it to a vote?

Anonymous said...

"In general, I suspect that our moral reasoning methods just haven't caught up to the vast advances in statistical reasoning over the last couple of centuries."

I think it's the other way around: our "crude" unconscious statistical methods (i.e. stereotyping, both justified and not) override the moral reasoning (specious or not), when self-interest over non-trivial stakes is involved.

Aaron Gross said...

Maybe statistical reasoning is too difficult for Steve Sailer. I'm sure that Brooks is more or less familiar with the probabilities involving race and crime, IQ, etc.

In a Bayesian framework, which has been used by race-realist commentators in these problems since at least the 1970s, decisions are made to minimize risk. Risk is a function of two things: probabilities of the outcomes, and the losses of the outcomes. The losses (like the probabilities) are subjective. So two people, Steve Sailer and David Brooks, can agree on the probability measure but make very different decisions, because of their different loss functions.

Rev. Right said...
This comment has been removed by the author.
Rev. Right said...

"Both blacks and whites subtly try to get a white partner when asked to team up to do an intellectually difficult task"

Somewhat more subtly than they would try to get a black partner if they were asked to team up for a competetive two-on-two basketball tournament.

Subtly is the key. Still gotta win.

commonwealth contrarian said...

Most people seem fine with statistical reasoning as long as and they don't have to engage their emotions.

However, as soon as you start talking about statistics in relation to hot-button issues like race, intelligence, and poverty, the red mist soon overwhelms the frontal lobes and statistical reasoning becomes impossible.

This is why thousands of left liberal women with psychology degrees can pass statistics l01, but are totally hopeless when it comes to having a real world debate about the Bell Curve or race and crime statistics.

Auntie Analogue said...


Men reason - sometimes deliberately, sometimes quite unconcsiously, instinctually - with statistics all the time:

36-24-26.

Add up those numbers and you get 86. Which is what a man's rational brain content is when he's got those stats before his eyes in the flesh.


(Winks & Grins, Sailermates!)

Simon in London said...

>>Auntie Analogue said...

Men reason - sometimes deliberately, sometimes quite unconcsiously, instinctually - with statistics all the time:

36-24-26.<<

Should be 36-24-34 - the 7:10 waist:hip ratio is a hardcoded index of female desirability in the male brain. Which adds up to a much more reasonable 94. :)

Anonymous said...

I think an enormous amount of effort, on the part of many educators over many decades, has been put into teaching everyone that discrimination, prejudice and stereotyping are bad. That effort has succeeded. Of course by "discrimination" I mean being able to distinguish the differences between different individuals and categories of people, by "prejudice" I mean learning to make good decisions based on past experience and by "stereotyping" I mean recognizing patterns and probabilities.

It's not surprising that well-educated people are bad at all these skills when they've been culturally bred out of us for at least six decades.

Grant C. said...

Soon, K-Thug will be telling us that we are not only racists, but stupid, too, because Africans were the first people so they have accumulated all the knowledge of the ages in the black collective consciousness, so they are naturally more talented at intellectually rigorous tasks.

Henry Canaday said...

Fun talk last night at the Local Lefty Boosktore by a Dartmouth professor named Wheelan, who has written “Naked Statistics.” Like Steve, he believes probability and statistics should be taught before calculus because 1) they are easier and 2) they are more useful.

Wheelan entertained the audience by citing some examples of practical statistical use: 1) Schlitz conducting a high-profile test of beer preferences at a 1980s Superbowl, which proved that all 1980s beers were pretty much the same; 2) Target marketers using shopping data to discover a teenage girl was pregnant before her father did; and 3) the growing use of value-added statistics to measure and reward teacher performance.

True to LLB tradition, a retired teacher complained about any approach, statistical or student survey, to measure teacher performance.

ironrailsironweights said...

In short, the Scoth-Irish even Anglicized ones like Brooks

He is Jewish.

Peter

Joe D. said...

Modern life is really like something out of Stalinist Russia or 1984. The fact that blacks are less intelligent than whites is obvious, even to young children.

Where I grew up, about 30% of the kids were black, and I distinctly remember in elementary school observing that the black kids tended to be a year or two older (they were being held back for failing grades), were more boisterous, more likely to get in trouble, tended to be poorer, and while generally friendly, they could blow up in a fit of rage at the drop of a hat. I expect most people who grow up around blacks also experience these things too, assuming they don't have some liberal nitwit parent re-interpreting things for them with horseshit excuses.

Anonymous said...

most humans have a problem understanding it's statistically LUDICROUS to assume one population's (say, whites) mean on a characteristic (say, ability) is exactly = to another population (say, blacks). that means everytime a smart black dies, a smart white has to die for those population means to always be exactly equal. it's ridiculous to assume the null hypothesis will be true for such things. - panjoomby

David James said...

"It's possible that Brooks is being satirical here..."

I think that's wishful thinking on your part. If we look at his comments in the larger context of the clip from the article (listed in your article below this one), he seems serious, even though in reality his complaint borders on insanity to the point of pure comedy gold.

Anonymous said...

In general, I suspect that our moral reasoning methods just haven't caught up to the vast advances in statistical reasoning over the last couple of centuries.

Define "our".

When you say "moral reasoning", you're talking about the Scots-Irish and their Frankfurt School agenda of cultural and institutional destruction.

Since when is theirs "ours"?

Problems arise with "the vast advances in statistical reasoning" only when those advances get in the way of The Narrative.

In the very rare instances when those advances might be used to further The Narrative [e.g. in micro-analyzing low-information-voters and their voting proclivities], then you can be damned certain that those advances will be seized upon to further The Narrative.

Matt said...

The losses (like the probabilities) are subjective. So two people, Steve Sailer and David Brooks, can agree on the probability measure but make very different decisions, because of their different loss functions.

True enough, but the point of "rigging" society would be the recognition that the vast majority of people have loss functions more like Steve Sailer than like David Brooks.

Anonymous said...

"Whenever I have to lift something really heavy, I subtly seek out a strong person to help me."

Better analogy. I subtly seek out large males when I need physical help, since large and male correlate with strength. David Brooks thinks you're supposed to select someone at random, even if they are tiny, since that tiny person could be a powerlifter.

Reg Cæsar said...

36-24-26
--Auntie

She's gonna tip over!
For better examples of 36-24-26, try this and this.

Snippet said...

I think Brooks' problem is not a failure to get statistical reasoning, which, I think he gets most of the time. It is a failure to acknowledge uncomfortable truths.

This failure is the mother of a million smaller failure-children

rob said...

Brooks does read Sailer. At least he used to. He cited the population vote-Republican county correlation that Sailer found. Brooks and some other guy did a talk at my school. When I got my book signed, not having many social skills, I asked Brooks why he cited the weaker overall pop growth-Republican when he could have cited a stronger one: white pop growth-Republican county. He seemed sheepish and said that he had been leery of citing Sailer at all, and yet felt he couldn't pretend the discovery was his. Quite a minimally-decent chap.

The only questions are whether Brooks still reads Sailer, and does he comment. Anonymous, I'm lookin' at choo.

pat said...

Statistic is actually a pretty static field itself.

Most people learn correlation in their first stat class. First they learn the Person Product Moment and if they are taking a modern course they learn Spearman's Rank Order non-parametric correlation. This is just about the most modern stat concept they will acquire in school. And indeed Spearman was born a decade or so after Pearson. But Spearmsn's correlation was first published in 1904.

When I taught statistics I never remember teaching anything based of anything that anyone had learned since the fifties.

Then when I went into teaching computer science things were much different. Almost everything I taught in my Data Commincations classes is now completely obsolete. That includes Novell software and hardware like an RS-232 port.

I could still teach statistics today with no catch up. But I, who was once a technical editor for Novell books and a cracker jack communications teacher, could not get a Data Communications teaching job today. I'd have to study up first on Wi-Fi. When I left computer science teaching not so long ago, Apple was a small marginal company.


Albertosaurus

Jack Bolling said...

Better analogy. I subtly seek out large males when I need physical help, since large and male correlate with strength. David Brooks thinks you're supposed to select someone at random, even if they are tiny, since that tiny person could be a powerlifter.

By assuming that the 6'3" man is stronger than the 5'2" woman, not only are you being prejudiced -- quite literally, you're making an advance judgment without 100% certainty -- but you're discriminating -- you're making a distinction between different categories in reality. The conscious mind, it would seem, is racist.

But for all you know, the hulking man might have sciatica and the diminutive woman might be a long-lost Williams sister.

In the movie Fearless Jeff Bridges' character walks across a busy highway without getting hit by a car. He should be commended for not only not judging their danger in advance, which would cause them offense and perhaps prevent them from slowing down, but for overlooking the difference (quite superficial really) between the place where cars are moving really fast and where they're not.

Svigor said...

I think an enormous amount of effort, on the part of many educators over many decades, has been put into teaching everyone that discrimination, prejudice and stereotyping are bad. That effort has succeeded. Of course by "discrimination" I mean being able to distinguish the differences between different individuals and categories of people, by "prejudice" I mean learning to make good decisions based on past experience and by "stereotyping" I mean recognizing patterns and probabilities.

It's not surprising that well-educated people are bad at all these skills when they've been culturally bred out of us for at least six decades.


You're in the ball park at least, but in my experience this is spurious. It falls apart on examination, owing to liberals' myriad, and dearly-held, prejudices and stereotypes (southerners, whites, males, blonds, conservatives, rural people, right-wingers, etc.). No, like most supposed principles attributed to liberals, this one is iron pyrite.

Liberals have few principles. Instead, they have a narrative. Their narrative is a lot like Abrahamic religion, in that they have to check the manual for an opinion, and there's no way to boil that down to a set of principles because the orthodoxy doesn't conform to principles. The only way to properly train a liberal is with years of institutional and mass media indoctrination; they learn it by rote.

Anonymous said...

My employer's HR department assures us, though diversity training, that Mr. Brook's causal assumptions are a rock solid fact and that we should all be polishing our talents at spotting such subtle acts of racism. For instance, did you know that much of what you witness in everyday life has no credible explanation aside from racism. And did you further realize that any attempt to establish a credible explanation is denialism and as such, itself an expression of institutional racism. You learn a lot in diversity training.

Jack Bolling said...

Liberals have few principles. Instead, they have a narrative. Their narrative is a lot like Abrahamic religion, in that they have to check the manual for an opinion, and there's no way to boil that down to a set of principles because the orthodoxy doesn't conform to principles. The only way to properly train a liberal is with years of institutional and mass media indoctrination; they learn it by rote.

The French writer Celine said it best: it's very easy to talk like a leftist, it's very hard to actually be a leftist. Talking like a leftist means that you get to take the other guy's stuff. Acting like a leftist means that you have to give up your stuff too.

Celine said he'd be first on board the communist bandwagon if it were sincere. Because of his literary fame, in 1936 he was granted a red-carpet tour of the Soviet Union. Whereupon any and all of his previous leftist sympathies evaporated--and keep in mind, he was on a guided tour.

Upon returning to France, he began speaking truth to the powerfully powerless... I think you can fill in the blanks from here...

Anonymous said...

"Upon returning to France, he began speaking truth to the powerfully powerless... I think you can fill in the blanks from here..."

Yup.

Svigor said...

Jack, you keep mentioning Celene and I'm going to have to read him. The point you attribute to him reminds me of another point; again like Abrahamic religion, leftism is hegemonic and evangelical. It too knows damned well better than to tolerate competition, its doom.

Svigor said...

And did you further realize that any attempt to establish a credible explanation is denialism and as such, itself an expression of institutional racism. You learn a lot in diversity training.

And the more facts anyone has at his command in contravention to the phantom racism explanation, the bigger a "RACISS!" he is. Duh.

Who else but a "RACISS!" would have at his command a plethora of facts contravening the phantom racism schtick? Reminds one of the Inquisition, no?

Anonymous said...

Liberals have few principles. Instead, they have a narrative. Their narrative is a lot like Abrahamic religion, in that they have to check the manual for an opinion, and there's no way to boil that down to a set of principles because the orthodoxy doesn't conform to principles. The only way to properly train a liberal is with years of institutional and mass media indoctrination; they learn it by rote.

Someone understands.

Garland said...

How much of it is an issue of politics though? Arent people pretty good about statistical reasoning when sensitive issues arent on the line? As you say, reality is stochastical (is that even an adj?). People function off of statistical impressions in every thing they do, even if they dont realize it. I cant think of any evidence that the problem is actual math-understanding rather than the age-old willful resistance of what they dont want to, or feel they should not, know.