Saturday, December 5, 2015

Increasingly it seem that there is nothing straight-forward under the sun

Cowen remarks in Best movies of 2015 by Tyler Cowen that
I thought this was the worst year for movies since I have been watching them. In fact I think you could multiply this year’s good films by two and still have the worst year for movies in a long, long time. Maybe by three.
He lists the ones he has thought were adequate. Onf of which is
Red Army, a documentary about the hockey team of the Soviet Red Army, its rise and fall. Chock full of social science and public choice, I loved this movie, philosophical too, even though I am not especially interested in hockey. One of my favorite documentaries.
I draw attention to this as an example of the complexity of what attracts attention. As the barriers to accessing information, knowledge, and experiences decline, we are faced with the dilemma of increased access and static receptors. We can only absorb so much. So how do we filter out the extraneous and limit ourselves to that which is most relevant to us?

This challenge is further exacerbated by the fact that we, as individuals, are not static systems. Our interests change all the time. You can't simply design filters and walk away. The filters have to evolve. Further, we function at a concrete level (facts-are-facts) as well as at a metaphorical/abstract level. What we see on the surface is not necessarily what is going on at a deeper level.

Cowen's comment is an example of this dynamism. He has no interest in ice hockey per se but he is intensely interested in social science and public choice. The disinterest in the superficial topic of ice-hockey is superseded by the intense interest in the topics of social science and public choice which the documentary illustrate.

Not only is there nothing new under the sun, increasingly it seem that there is nothing straight-forward under the sun.

The Pigeonhole Principle

A frigid Saturday morning, linking around the internet, catching up on stored articles set aside during a busy week. Hit on an article on a couple scientists bridging biological neurology and artificial intelligence, which leads to questions about higher mathematics and logic which in turn lead to this delightful principle of which I had not been formally aware but which is seductively obvious, the Pigeonhole Principle. From Wikipedia:
In mathematics, the pigeonhole principle states that if n items are put into m containers, with n > m, then at least one container must contain more than one item.[1] This theorem is exemplified in real-life by truisms like "there must be at least two left gloves or two right gloves in a group of three gloves". It is an example of a counting argument, and despite seeming intuitive it can be used to demonstrate possibly unexpected results; for example, that two people in London have the same number of hairs on their heads (see below).
I find it fascinating how there are principles lurking behind knowledge, ideas which were on the cognitive horizon which, once pointed out, are obvious. This is one of those.

Some thirty or forty years ago I learned that if there are 23 people in a room, there is a greater than 50% probability that two of them have the same birthday. When I first heard this, I sort of lazily muscled through the abstract ideas behind the forecast in order to verify to my own satisfaction that the claim was likely true. But once satisfied, I have never really thought about it again. I certainly never sat down and worked out the specifics.

Now, from an article about modelling the brain, by many linking ideas, I discover that it is the Pigeonhole principle which underpins the forecast that there is a 50% chance that, in a room of 23 people, two of them will have the same birthday. Again from Wikipedia:
The birthday problem asks, for a set of n randomly chosen people, what is the probability that some pair of them will have the same birthday? By the pigeonhole principle, if there are 367 people in the room, we know that there is at least one pair who share the same birthday, as there are only 366 possible birthdays to choose from (including February 29, if present). The birthday "paradox" refers to the result that even if the group is as small as 23 individuals, there will still be a pair of people with the same birthday with a 50% probability. While at first glance this may seem surprising, it intuitively makes sense when considering that a comparison will actually be made between every possible pair of people rather than fixing one individual and comparing them solely to the rest of the group.
Wonderful.


Friday, December 4, 2015

Climate change summary

From What Should We Do About Climate Change? by David Siegel. A reasonable summary of the global climate change argument and the weaknesses and fallacies. Currently a lot of government and NGOs with a financial interest in there being alarm about climate change hold the field but their position has been weakening in the face of evidence and rising non-insider interest and resistance.

Here are Siegel's ten general conclusions from a scan of the current field:
1. Weather is not climate. There are no studies showing a conclusive link between global warming and increased frequency or intensity of storms, droughts, floods, cold or heat waves.

2. Natural variation in weather and climate is tremendous. Most of what people call “global warming” is natural, not man-made. The earth is warming, but not quickly, not much, and not lately.

3. There is tremendous uncertainty as to how the climate really works. Climate models are not yet skillful; predictions are unresolved.

4. New research shows fluctuations in energy from the sun correlate very strongly with changes in earth’s temperature, better than CO2 levels.

5. CO2 has very little to do with it. CO2 continues its relentless rise, yet our planet hasn’t warmed in 18 years now. (Despite talk of 2015 as the “hottest year on record,” there has been no average global mean temperature increase since 1997.) All the decarbonization we can do isn’t going to change the climate much.

6. There is no such thing as “carbon pollution.” Carbon dioxide is coming out of your nose right now; it is not a poisonous gas. CO2 concentrations in previous eras have been many times higher than they are today.

7. Sea level will probably continue to rise — but not quickly, and not much. Researchers have found no link between CO2 and sea level.

8. The Arctic experiences natural variation as well, with some years warmer earlier than others. Polar bear numbers are up, not down. That has more to do with hunting permits than CO2. Antarctic ice is growing, not shrinking.

9. No one has demonstrated any unnatural damage to reef or marine systems. Additional man-made CO2 will not likely harm oceans, reef systems, or marine life. Fish are mostly threatened by people, who eat them. Reefs are more threatened by sunscreen than by CO2.

10. The Intergovernmental Panel on Climate Change and others are pursuing a political agenda and a PR campaign, not scientific inquiry. There’s a tremendous amount of trickery going on under the surface.
All of this is consistent with the preponderance of the evidence I have seen save only item 9. I am still unclear about both the status of reefs worldwide as well as the causal mechanisms where there has been clear bleaching. Siegel's number 9 is likely true but I haven't seen enough to have confidence one way or another.

Thursday, December 3, 2015

Measurement forces us to confront decisions we don't wish to make.

From The effects of the medical marijuana market on substance abuse by Tyler Cowen.

The argument is about the possible consequences of drug legalization, looking at medical marijuana as a particular case. What happens to drug use when marijuana for medical purposes is legalized? From the abstract of the original study Cowen is discussing:
I find that growth in the legal medical marijuana market significantly increases recreational use among all age groups. Increased consumption among older adults has positive consequences in the form of an 11% reduction in alcohol- and opioid-poisoning deaths. However, increased consumption among youths leads to negative externalities. Raising the share of adults registered as medical marijuana patients by one percentage point increases the prevalence of recent marijuana use among adolescents and young adults by 5-6% and generates negative externalities in the form of increased traffic fatalities (7%) and alcohol poisoning deaths (4%).
With complex, dynamic multi-causal systems, outcomes are hard to forecast, particularly across disparate populations. Cowen sees this as evidence supporting his assumption that:
When it comes to “those who already are screwed up,” namely the older generation, it is best to shunt them off into pot, compared to the relevant alternatives. But when it comes to the younger generation, the new norm that “pot is OK” may in fact not be best in the longer run. So in sum,while I (TC, not the author necessarily) favor marijuana decriminalization, we should hold mixed moods towards its practical effects.
Well, not just mixed moods. If we accept at face value the outcome of this study, we are explicitly sacrificing the young to improve the circumstances of the older generations. Whose lives do we value and to what extent? Measurement forces us to confront decisions we don't wish to make.

Wednesday, December 2, 2015

I don't think that meme means what you think that meme means

John Althouse Cohen comments on this picture which has been making the rounds.

#Perspective #YoureAnImmigrantToo

Posted by Lastrealindians on Friday, November 20, 2015

To paraphrase Inigo Montoya, "I don't think that meme means what you think that meme means."

Cohen observes:
The number of Facebook posts I've seen like this — a picture of Native Americans encountering European settlers, suggesting an analogy to America's current issues with refugees or immigrants — has me concerned for this country's educational system.
First, history: When Europeans arrived in America, this was, on the whole, a very bad thing for Native Americans. That's why if you look around at the United States, you see very little left of Native American culture, which was almost entirely replaced with European culture.

Next, analogies: Who is being analogized to whom? If you're analogizing current-day American citizens to Native Americans in the time of Columbus, then how would you expect Americans to react to that? Since the historical experience for Native Americans was a very bad one, you should expect American citizens who take your analogy seriously to have a negative reaction to whatever you're analogizing to European colonialists. If that isn't the reaction you're hoping to provoke, then it's time to stop and think about whether this is something you really want to post to the internet.

We are Olympic champions in the Jumping to Conclusions contest



It does not bode well

Given the preponderance of Islamic inspired terrorism in recent years, there has been, in some quarters, a perhaps well-intended effort to introduce perspective and balance to the picture. This has led to claims that the bulk of terrorist acts are committed by Christians. But truth and reality don't require balance. They are what they are. So what is the truth rather than the simple assertion?

There are reasons to call into question the good faith bona fides of those making these claims. The polite reading of their motivation is that they are concerned that Americans will form blind prejudices based on exceptional cases and that they might turn on Muslim Americans. George Bush did a reasonably good job of drawing the distinction between community and perpetrators and conveying the message that all Americans are Americans and should not be hived off as groups.

This concern about the susceptibility of Americans to bigotry, however well intended, is markedly prejudicial and bigoted in itself and goes against all the measurable evidence. Americans are a distinctly tolerant and receptive culture and clashes almost always happen over personal actions and behaviors rather than prejudices. In a country of 320 million, there are always going to be aberrations but they are few and far between. A few years ago, I looked at the FBI hate crimes statistics for 2000-2010 and the numbers were very low. Anti-muslim hate crimes were tiny and dwarfed by the on-going and persistent infection of anti-semitic hate crimes.

The claim that most terrorist acts in the US are committed by Christians is absurd on its face but insight and perspective come from discovering what you thought you knew might not be the case. It is worthwhile investigating the claim at least at a cursory level. David French has a whack at addressing this in No, America Doesn’t Have a Christian Terrorism Problem. French makes some good points but I don't think it is an airtight case. He has a handful of stories, narratives, anecdotes, instances that are consistent, but ultimately you want something numerically robust. How do we define a terrorist act (distinct from a crime)? This isn't a pedantic issue. Notoriously, the Obama administration classified the Fort Hood attack by Major Nidal Hasan as an incident of workplace violence rather than terrorism and yet the popular vernacular understanding is much clearer. It was a terroristic attack for religiously motivated purposes. What makes the Chattanooga shootout a terrorist event as opposed to a gang shooting? Given the definition we choose to use, how many terrorist acts have there been in the past decade? Of those, how many have been committed for avowedly "Christian" reasons? I cannot come up with any. Islamic inspired attacks? Sure. Fort Hood, Boston, Chattanooga, etc. Lots of attacks, lots of deaths.

As is often the case, the claim rests on some definitional issues, which French touches on.
We’ve seen this movie before. Following the dreadful Oklahoma City bombing, many on the left tried to pin Timothy McVeigh on Rush Limbaugh, and despite the fact that McVeigh called himself an agnostic, some leftists still refer to him a “Christian terrorist.”
Separate from the claim about Christian terrorism, there is another trope going around, something along the lines that we are not at risk from foreign terrorists but from homegrown terrorism. Again, a claim that could be true but when you dig into it you find definitional issues. This is illustrated by a recent report released by George Washington University, ISIS in America. They are looking at the 71 people arrested in the US since March 2014 for ISIS-related activities. The claim in the executive summary is that:
The profiles of individuals involved in ISIS-related activities in the U.S. differ widely in race, age, social class, education, and family background. Their motivations are equally diverse and defy easy analysis.
They seem to be conveying the idea that anyone could be an ISIS terrorist, that there are no predictive patterns among the 71. But that doesn't appear to be true. They don't provide a summary of the metrics to support their statement so I eyeballed the detailed individual narratives they provide on the 71. That's where definitions become important. For example, does a white Bosnian Muslim immigrant to the US who has become a naturalized citizen count as homegrown terror or international terror?

Based on the quick review, 60% of the 71 are first or second generation Muslim immigrants. 40% are American converts, most of whom are African-American. I see three native born white converts. One Hispanic convert. No East Asian or South Asian. No Native Americans.

Despite the claim that they differ widely by race, age, etc. there is actually a quite defined profile. All the 71 are Muslim. Most are recent immigrants from the Middle East or Muslim countries and virtually all the rest are African-American. Virtually all are between 20-35 years old.

The issue to me is not what the actual profile might be. The issue is the reluctance to speak factually. One of our greatest assets as a nation is a trust and confidence in one another as well as trust and confidence in our system of government. Trust among citizens is high on a global scale and I suspect higher than it has long been. There has been a sustained Gramscian effort to cultivate racial and other identities and encourage people to think of themselves as groups rather than individuals. Blessedly, outside a few academic disciplines, Hollywood, the media and a few other postmodern Frankfurt School redoubts, this appeal to racial and group identity has largely failed to take root. We are Americans and we trust one another.

Our second pillar, trust in our form of government has taken a lot of hits in the past two or three decades, deservedly so. The regrettable thing is that it is not the form of government at fault but the mixed bag of characters we have had populating it. The worst political class ever, as some put it. None-the-less, the distrust in government is there.

Claiming that most terrorist attacks in the US are Christian, and that most of the terrorist attacks are homegrown can both be argued on tendentious grounds with twisted definitions and deliberate obfuscations. It can be argued, but it is not correct. As long as politicians and their fellow travellers are peddling these myths, it tells the citizenry that the political class think A) our citizens are idiots, and B) our citizens are easily swayed bigots lacking all form of self-control.

No wonder there is so little trust in government. It does not bode well.

Tuesday, December 1, 2015

TULIP

From an article comparing the fanatical extremism on campus to the extremes of Calvinism. The author points out TULIP from the Five Points of Calvinism. The article is Politically Correct Holy Rollers: The New Campus Revival by Helen Andrews.

TULIP, from Wikipedia:

"Total depravity," also called "total inability," asserts that as a consequence of the fall of man into sin, every person is enslaved to sin.

"Unconditional election" asserts that God has chosen from eternity those whom he will bring to himself not based on foreseen virtue, merit, or faith in those people; rather, his choice is unconditionally grounded in his mercy alone.

"Limited atonement," also called "particular redemption" or "definite atonement", asserts that Jesus's substitutionary atonement was definite and certain in its purpose and in what it accomplished.

"Irresistible grace," also called "efficacious grace", asserts that the saving grace of God is effectually applied to those whom he has determined to save (that is, the elect) and overcomes their resistance to obeying the call of the gospel, bringing them to a saving faith.

"Perseverance of the saints" (or perseverance of God with the saints) (the word "saints" is used to refer to all who are set apart by God, and not of those who are exceptionally holy, canonized, or in heaven) asserts that since God is sovereign and his will cannot be frustrated by humans or anything else, those whom God has called into communion with himself will continue in faith until the end.
Not a perfect match but it certainly echoes the strident assertions about privilege, institutional racism, etc.

A woman dives into the confusing vacuum created by an unanswered text

Another example of Poe's Law at work. The article is The Five Stages of Ghosting Grief: A woman dives into the confusing vacuum created by an unanswered text by Rachel Fields. I saw it on the front page and, through sheer horror, had to click through. Was this click bait for the shallow, self-absorbed, vapid, wealthy, female New York Times reader? A vein of their content which is widely reviled. Or was I misunderstanding the headline and there was actually something cognitively respectable in the piece?

Having clicked through and read the piece, I cannot decide whether it is pure parody of that for which the NYT is reviled, or is it a serious piece? Thinking I must be obtuse, I click into the comments to see how readers are responding. There are a good number who are taking it at face value as, essentially, an advice column. There are others who see it as the drivellings of an unhealthy mind and unbalanced personality. And there are others who are trying to put the best face on it and are assuming it must be a humorous parody.

Such confusion among readers reflects poorly on both the writer and the editors.

Monday, November 30, 2015