Monday, June 30, 2014

Its poor record on keeping people alive

From NHS is the world's best healthcare system, report says by Denis Campbell and Nicholas Watt.
In the Commonwealth Fund study the UK came first out of the 11 countries in eight of the 11 measures of care the authors looked at. It got top place on measures including providing effective care, safe care, co-ordinated care and patient-centred care. The fund also rated the NHS as the best for giving access to care and for efficient use of resources.
Sounds good? But, wait, that's not all.
The only serious black mark against the NHS was its poor record on keeping people alive.
Well, yeah, other than that. And how bad is it doing at keeping people alive?
On a composite "healthy lives" score, which includes deaths among infants and patients who would have survived had they received timely and effective healthcare, the UK came 10th.
That's 10th out of 11.

To be fair to the NHS, health system comparisons within countries are notoriously difficult let alone between countries. In addition, most these type of reports are written by industry insiders with some sort of policy axe to grind and budgets to be enhanced.

So the real issue is perhaps with the report writers (and the media reporting it) rather than the NHS. What kind of sane health system looks so bureaucratically good but allows its patients to die at greater rates than just about everyone else. Something is amiss and it may or may not be the NHS.

Sunday, June 29, 2014

Considering these events in isolation can lead to policies that are far from optimal

An interesting issue which is too rarely engaged with and an interesting discussion. How much risk mitigation can we afford? From How many catastrophes can we avert? by Tyler Cowen. from the originally cited paper.
How should we evaluate public policies or projects to avert or reduce the likelihood of a catastrophic event? Examples might include a greenhouse gas abatement policy to avert a climate change catastrophe, investments in vaccine technologies that would help respond to a “mega-virus,” or the construction of levees to avert major flooding. A policy to avert a particular catastrophe considered in isolation might be evaluated in a cost-bene fit framework. But because society faces multiple potential catastrophes, simple cost-bene fit analysis breaks down: Even if the benefi t of averting each one exceeds the cost, we should not avert all of them. We explore the policy interdependence of catastrophic events, and show that considering these events in isolation can lead to policies that are far from optimal. We develop a rule for determining which events should be averted and which should not.
Risk identification, risk quantification, uncertainty, risk mitigation, present vs. future trade-offs, consumption vs. investment trade-offs, tactical vs. strategic trade-offs. It's all there, bundled up.

There has been, among advocacy groups, an orientation towards the precautionary principle, a quite flawed risk approach, over the past couple of decades. While at its heart, there is a commonsense nugget that you shouldn't do anything rash and without some degree of confidence on the outcomes, it is in practice a disgraceful exercise in reductio ad absurdum whereby no action can take place without the authorization of some preferred group, just in case the action might have an unanticipated negative consequence.

Cowen's post and the original work begin to map some of the boundaries to how one can and ought to make communal risk/uncertainty decisions. Only so much of a society's surplus can be diverted into crony capitalism, non-productive consumption, and moral hazard before the productivity of the system is undermined. So how do you assess real but remote events that have large magnitude consequences. How do you make trade-off decisions?

These are the sorts of issues that Nassim Nicholas Taleb is wrestling with in Antifragile: Things That Gain from Disorder. Part of his answer is that there are some events that you cannot actually plan for, all you can do is prepare your system in such a way that if the highly unlikely event does occur, you can recover as painlessly as possible.

Saturday, June 28, 2014

Silent, upon a peak in Darien

On first looking into Chapman's Homer
by John Keats

Much have I travell'd in the realms of gold,
And many goodly states and kingdoms seen;
Round many western islands have I been
Which bards in fealty to Apollo hold.

Oft of one wide expanse had I been told
That deep-brow'd Homer ruled as his demesne:
Yet did I never breathe its pure serene
Till I heard Chapman speak out loud and bold:

Then felt I like some watcher of the skies
When a new planet swims into his ken;
Or like stout Cortez, when with eagle eyes
He stared at the Pacific—and all his men
Look'd at each other with a wild surmise—
Silent, upon a peak in Darien.

Friday, June 27, 2014

Fiction and Nonfiction, boys and girls

From a recent discussion about fiction and nonfiction and the stereotypes of gendered reading. Here is some information that was collected and shared.

Because we define terms and concepts differently (ex. Are YA books the books that young adults read or are they a marketing category of literary fiction we choose to call YA?) we often end up talking at cross-purposes. When we speak about reading are we including books and non-books (magazines, newspapers, comics, text messages)? When we talk about reading are we talking about voluntary elective reading or are we including reading related to work and study? Regrettably these distinctions are important and either are elided or not addressed in a lot of studies.

For example, there is the stereotype that boys don’t read and when they do read, they primarily prefer nonfiction. Almost as a derivative of this first stereotype, there is then also the corollary stereotype that girls don’t read nonfiction.

Both of these are true with very particular definitions and wrong when loosely stated.

While relying primarily on US data, I have included data from a couple of robust recent studies from the UK and Australia. Their reading patterns at the macro level are close to those in the US and there is granular detail in those recent studies that is likely reflective of the US as well (or is consistent with smaller, less robust studies from the US)

Summary: Boys read less than girls, starting roughly equal in grade school and the gap widening through adulthood. Boys that read only one or the other, read about as much fiction as nonfiction but girls read a good deal more fiction than nonfiction. The plurality/near majority of both girls and boys read both fiction and nonfiction. Girls tend to be more catholic in the range of their reading (by f/nf as well as by subject, protagonist, etc.) Boys demonstrate a significant focus in their preferences. While boys and girls have a majority overlap in the titles they elect to read, the overlap is largest in K and least in 12th grade. The books that they elect to read that do not overlap conform to traditional stereotypes with boys preferring adventure/action and girls preferring relationships/romance.

Stipulated that anybody can read anything they wish and that there is no normative basis for prescribing what people ought to be reading. Any and all recommendations to a young reader are ideally based on a deep knowledge of the child’s reading ability, context, and interests.

Do males read less than females?

Yes, particularly if we are focusing on elective reading of books (as opposed to reading required for work or study). It appears to start out in equal proportions as an infant but as they grow older the pattern of females spending more time reading electively grows.
64% of adult women read at least one book in 2012 (and 56 percent read at least one literary book), compared to only 45% of men (only 37 percent read at least one literary book). Source.

While women spend more time reading than males, the gap is not quite as wide as might be assumed, particularly given differences in labor force participation rates. A study from Australia provides some illumination of the impact of labor force participation rates on reading patterns. Just as in the US, males read less than females, however, from that study, 77% of females read for pleasure but only 66% of males. In contrast, 51% of males read for work or study but only 36% of females. Source.

In total, in the US, men spend about 95 hours a year reading (elective and non-elective) and women spend 139 hours, nearly 50% more. Males spend 0.25 hours per day and Females 0.36 hours per day on weekdays reading and 0.29 and 0.44 on weekends and holidays. Source.

Interestingly, while they do spend more time reading to children, women spend only 33% more time reading to children than do men, Males spend 0.03 hours per day and Females 0.04 hours per day, [Source] [0.05 and 0.07 respectively when children are under 6 years of age]

There is a material enthusiasm gap. 69% of females “really like reading” whereas only 41% of males respond so, a 68% gap. Source.

I have not been able to find robust empirical data for gender differentiated reading volumes for younger ages but the patchy small studies that I have come across are consistent with the adult patterns, particularly in high school.

Do males prefer nonfiction over fiction?

No. There are about an equal number of males (27%) who read only fiction as there are who read only nonfiction (28%) and 45% read both fiction and nonfiction. The appearance of males preferring nonfiction probably arises in part from comparison with females where there is a marked preference for fiction with 29% of females reading only fiction while only 17% read only nonfiction.
Approximately 50% of the population does no elective reading in a given year.

Among the 50% that do electively read, 23% read only fiction, 19% read only nonfiction, and 58% read both fiction and nonfiction. Source.

Among males, 27% read only fiction, 28% read only non-fiction and 45% read both. Among females, 29% read only fiction, 17% read only non-fiction and 53% read both. Source.

In terms of fiction and nonfiction, 80% of readers of fiction are female, 20% are male. Source 1 and Source 2.

Males read for discovery and knowledge acquisition about 25% more than females. Females read for escape and relaxation about 30% more than males. Source.

Are there gender patterns in reading form/genre/topic?

The widest gap occurs in the history genre, which 40% of male readers but only 23% of female readers say they have read in the past year. There is also a decided male bias in the non-fiction genres of political (25% compared to 10%), current affairs (20% compared to 9%), and business (16% to 12%). Source. Other categories with 25% or greater male interest over female interest include sport, science, technology and atlas/dictionary. In all four categories, male interest is actually 2-4 times that of female. Source.

The most significantly female-oriented non-fiction genre is self-help, which 19% of female readers and 12% of male readers have read in the past year. Source. Other categories where female reading interest is 25% or greater than male interest include biographies, gardening, cooking, and hobbies/crafts. Source.

Are there gender patterns in title preferences among youth readers?

Yes. Renaissance Learning 2014 report, What Kids Are Reading has data on 10m school age children and their (supposedly) elective reading. I recognize the shortcomings of that program but it does cover 10m of 50m students and it is the only source of which I am aware that offers any sort of insight as what kids actually read versus what others think they ought to read. Source.
Taking Grade 12 top 20 most read nonfiction books, there are only 13 that are common between boys and girls and those thirteen titles are differently ranked. For example, Blind Side: Evolution of a Game is number 7 on the boy’s list but 19 on the girl’s.

Of the 7 different books on each list, the seven boy books are overwhelmingly sports/military. The seven girl books are overwhelmingly relationship/dependency/dysfunction.

Taking the Grade 12 top 20 most read books (i.e. fiction - only one nonfiction cracks the top 20 list, Night by Eli Wiesel on the boys list), thirteen are common between boys and girls.

Of the seven different books on each list, for boys, one is nonfiction, and four are adventure. For girls, five of the seven books are romance/relationship.

So basically a 60-65% overlap in interests but the differences conform to stereotype.

Are these gaps and patterns unique to the US or are they common among other countries?

While the numbers vary a bit by country, the patterns and gaps are common among OECD countries. The reading performance gaps as measured by educational tests between male and female students exists at just about all ages in all OECD countries and has been in evidence since the 1970s.

Are there other gender based differences?

Yes probably, but I cannot document other than direct experience and conversations with librarians. I have the strong impression that the YA category has muddied the reading waters. The overwhelming majority of YA is not age bracketed but rather the marketing category of literary fiction which is in turn read primarily by adult women. Among 12-18 year olds, the YA that is read is read substantially by females. The upshot is that for girls, with a documented predilection for fiction, there is a logical transition from picture books to chapter books to YA to adult. For boys, with a greater (proportionally) interest in nonfiction, there is a much larger percentage that go from chapter books straight to adult books. I think this is the blind spot in many conversations. Is Hot Zone a YA book? Clearly not. Is it read by YA (12-18)? Very much so. Is it read more by male young adults than female young adults? It appears so to me and those with whom I have discussed the issue (generically around nonfiction titles). How big is the population of books that are read by young adults which are not designated as YA books? I don’t know but my sense is that it is large.

In the Renaissance data, of the 27 titles among the 20 top boys and girls books read by 12th graders, only 13 might be considered YA (for example on ambiguity – Twilight; yes YA per publisher, no per author. I included it as yes in the count). This implies that about 50% of what young adults read are actually adult books.

There has been a thread that we need more YA nonfiction. But I wonder if that is the case. If boys are transitioning straight to adult nonfiction, is that in any way a bad thing? It is not clear to me that it is. Probably virtually everyone of us on this list serv (assuming a certain age) grew up in that environment, i.e. before the marketing category that became YA. Might there be a benefit to easier YA nonfiction books being available? Perhaps, but it is not clear to me that there is and Sully’s publisher letter would indicate that the benefit isn’t seen to be there commercially. Why would you go for a YA version of Kon-tiki, The Great Escape, A Night to Remember, 60 Seconds Over Tokyo, Krakatoa, etc. when the adult versions are accessible and gripping? Alternatively, are there any YA nonfiction books that come close to the caliber (and likely longevity) of those titles?

Are there agreed reasons for the differences in gender reading patterns?

No. The UK just completed a study on male reading gaps. Their conclusions follow. Source.

The home and family environment, where girls are more likely to be bought books and taken to the library, and where mothers are more likely to support and role model reading;

The school environment, where teachers may have a limited knowledge of contemporary and attractive texts for boys and where boys may not be given the opportunity to develop their identity as a reader through experiencing reading for enjoyment;

Male gender identities which do not value learning and reading as a mark of success.

Reading for pleasure needs to be an integral element in a school’s teaching and learning strategy and teachers need to be supported in their knowledge of relevant quality texts that will engage all pupils. There is a specific danger that a predominantly female workforce will unconsciously privilege texts that are more attractive to girls.

However, that is but one among many reports. There is no strong consensus. The issue of female dominated education environments possibly creating a mismatch with young male reader interests is a pretty commonly identified issue but the evidence is substantially anecdotal and logical. There aren’t many empirical studies that have been conducted to confirm or refute.

Are there institutional barriers to greater access to nonfiction?

I suspect so. My experience is that in general librarians tend to be much more knowledgeable and aware of front-list books and literary fiction than they are of back-list and non-fiction. This makes sense.

If boys are going straight to adult nonfiction, that is outside the youth librarian bailiwick (obviously might not be true for many particular librarians).

Literary writing in nonfiction exists in large quantity but it is overwhelmingly in the backlist which is a pretty big beast. There might be a hundred adult titles on WWII published in a given year but the number of them which are youth accessible and which are also written in literary fashion is quite a bit smaller.

Recommending quality nonfiction books requires as a predicate not only literary awareness (how well is it written) but also domain knowledge of the subject (how accurately is it written). With literary fiction, domain knowledge is less critical. The consequence is that nonfiction books are necessarily less well served on average. The population that have Variable X (literary awareness) AND Variable Y (Domain Knowledge) must necessarily be smaller than the population that only has Variable X.

The divisibility of nonfiction is greater than fiction. I have not figured out a way to define or measure this but see it a lot. Perhaps it is better said that the fungiblity of nonfiction is less than that of fiction. If you want a good literary nonfiction narrative based on WWII in the Pacific there are a couple of dozen books to choose from. Once you have exhausted those, there is little else. You can recommend WWII nonfiction literary treatments of Europe but the capacity to port the reader’s interest from one theater to the other is often surprisingly low. This may have less to do with the nature of nonfiction and more to do with the doggedness of interest of male readers but my sense is that there is an issue of fungibility. You can see this in the low penetration rates of nonfiction books compared to fiction in the Renaissance Learning 2014 report, What Kids Are Reading. Source.

For example, the top three nonfiction titles among 11th and 12th graders were Black Boy by Wright, Walden by Thoreau, and 1776 by McCullough read by 0.08%, 0.04% and 0.02% of 11th and 12th graders respectively. In contrast, the top three fiction titles were The Great Gatsby, The Scarlett Letter and The Cask of Amontillado read by 6.82%, 3.06%, and 1.91% respectively. So the top three fiction books were read by roughly 85 times as many 11th & 12th graders as the top three nonfiction. In other words, the reading population for a given nonfiction title is much smaller than a comparable given fiction title.

It should be noted that the differentials between nonfiction and fiction intensity are much smaller (still there, just smaller) at the K-1 level than the 11-12 level. At K-1 the ratios are more like 10 times the readership for fiction over nonfiction (by ranking) versus the 85 times at the older grades. This can be read that as children age their nonfiction interests become much more particular versus their fiction interests.

Thursday, June 26, 2014

Government failure UK edition

A rather interesting discussion at Restoring Anatole France by Tyler Cowen.
Situation: A large grocery chain in the UK, TESCO (something like our Walmart), has a problem as reported here. "Customers told us they were intimidated by antisocial behaviour outside our Regent Street store." Their solution was to "put studs in place to try to stop it." Homeless advocates identified this as an anti-homelessness strategy and mustered considerable, outrage, protests, and criminal mischief (one group poured concrete slurry over the inch high spikes. City government as represented by the mayor, took umbrage at TESCO's actions. TESCO immediately committed to removing the spikes and finding some other solution.

But this is interesting. TESCO here is the victim of illegal behavior (sleeping on TESCO property). They are a food related business so hygiene is obviously a significant policy issue. Customer safety is obviously a significant issue. The police are obviously unwilling or unable to address the problem.

So why is TESCO the bad guy, victim of demonstrations by homelessness advocates, berated by the government? They broke no law, they were seeking to comply with hygiene regulations, they were looking after the interests of their customers.

Well you can understand why advocates would exploit this. It is what advocates do. While there are some who are well intentioned and even effective, there are very many who are simply interested in being seen to be altruistic, though usually at someone else's expense. Protests like these do wonders for raising contributions.

I can't even begin to explain the Mayor's stance other than diversion. It is the failure of the state to enforce its own laws which is the nominal cause (the root cause being homelessness itself) so blaming the victim is a convenient strategy, particularly when the scapegoat is a large corporation.

Cowen looks at this in terms of economic principles, casting TESCO's initial response as a means of raising the cost (by raising inconvenience) to the homeless of sleeping on the TESCO property. As is often the case, a lot of variance in the comments as well as insight.
Yes, it seems to me that the issue is not about raising the costs of being homeless to deter homelessness. The issue is that the cost of dealing with homelessness is being borne disproportionately by Tesco instead of society as a whole.
There’s a cognitive/behavioral anomaly at work here. If we were to levy taxes on a small subset of randomly chosen people, or even corporations, to pay for homeless shelters, most of us would consider that to be unfair. It would be considered similarly unfair if the government were to randomly choose households that would be required to allow homeless to camp on their lawns or front porches. However, when homeless people “naturally” cluster around Tesco’s stores, it is considered “inhumane” for Tesco to not want the burden of dealing with the homeless to fall solely on their shoulders. I suspect that very few of those criticizing Tesco are volunteering their own homes to house the homeless.
As is so often in these cases, the advocates create a small sensation around a symptom (sleeping on TESCOs property by the homeless) without choosing to address the disease (homelessness itself). Homelessness (addiction, mental health, behaviors, etc.) is a notoriously difficult issue to effectively address. Better by far, in terms of advocates, to focus on the symptoms than the disease. Here, take an aspirin

Wednesday, June 25, 2014

How much are children reading?

Following up on yesterday's about children actually read, another source is from Renaissance Learning who administer the Accelerated Reading program designed to monitor and improve children's reading abilities. The program is widely used (by perhaps 37% of children in lower grades) it is also widely criticized. A number of studies indicate that it is effective in improving reading test scores. On the other hand it is criticized for not improving critical thinking. There is among teachers, particularly especially good ones, a visceral dislike for AR.

All that said, it does provide an interest annual report, What Kids Are Reading with a lot of data, constrained by the circumstances of the program as it might be.

In grades 2 - 5 there are about a million and a half reading participants in each grade, about 37% of all kids in each of those age ranges. RL measures the number of books a child reads as well as the number of words in each of those books thus allowing an overall measure of volume of reading in terms of number of words per year.
Sheer volume of book exposure peaks in second grade at 55 books per reader.
The average student in high school is reading just over 5 books a year (but about 300,000 words per reader).
Reading volume (words per reader) rises from first grade to sixth grade when it peaks at 419,000 words per year.
Reading then falls 25% till ninth grade when it stabilizes at around 300,000 words per year for the rest of high school.
The gender balance starts out in balance from first to sixth grade at about 52% of reading done by girls and 48% by boys.
From sixth grade to ninth, the ratio widens dramatically. From ninth grade on girls do about 58% of all reading and boys about 42%.
There are only about 8% as many participants in 12th grade as there are in first grade/
Nonfiction only shows up once among the 20 most popular titles for each grade.

Tuesday, June 24, 2014

What the clerisy think children ought to read versus what children actually read

On the reading related list_servs to which I belong, there is a strong orientation to talk about what advocates want children to be reading and very little or no discussion as to what children are actually reading. Part of this is explainable by the presence of many authors on those list_servs. It is not unexpected that their interest is in children reading what they are writing. But the interests of authors and the interests of children often do not coincide.

It is hard to address this because there is so little publicly available robust data on reading, sales, etc. You have to make do with snippets here and there, knowing that it is not the whole story but that it might possibly be indicative.

List_serv conversations tend to be dominated by YA, by the front list (books published within the past year), by literary fiction (as opposed to series and genre), and by social justice issues (gender, race, LGBT, individual and familial disadvantage, etc.). The publisher surveys show the readers of this small slice of children's books to be primarily middle-aged women (80% according to PW). What is missing is nonfiction, series, action, plot, etc.

My estimation of what is discussed would be something like:
* Frontlist 80%, Backlist 20%
* Social Justice issues 75%, all other issues 25%
* YA 65%, Middle Grade books 25%, Picturebooks 10%
So what are children actually reading? Hard to tell. There are a couple of sources I looked at just to put at least a smidgen of data on the table.

One source is Publisher Weekly's annual list of bestsellers. These are books that sold more than 100,000 copies in the past year. PW tracks this information by hardback (HB), paperback (PB), frontlist (FL) and backlist (BL), and e-books (which I have omitted because the numbers are so small). There were 461 children's best sellers in 2013.
* 461 best seller titles
* 87,038,495 individual books sold
* These 461 titles out of some 25,000 new titles constitute approximately 50% of the revenue for the children's books section of publishing
* There is an incredibly steep declination gradient. Just 35 titles, 8% of all bestsellers, are responsible for 31% of all best seller sales.
* Front list is 41% of the market (25% hardback and 16% paperback)
* Back list is 59% of the market (33% hardback and 26% paperback)
* YA is 25% of the market (by unit sales), Middle Grade 35%, Picture Books 40%
* If you assume the that publisher survey information is correct and remove adult YA readers from the counts, and assume that adults do not constitute a material population of MG and PB readers, you get substantially different numbers for what children are reading.
* YA 7%
* MG 43%
* Picture books 50%
* Only 16 of the 461 titles are nonfiction (NF). This constitutes 3.5% of the titles and 2.4% of the units sold.
* The distribution of NF is different from fiction with 56% being frontlist and 44% being backlist.
* The average bestseller title sells 219,000 units. The average social justice related titles struggle to achieve 5,000 unit sales in a year.
* The top five back list titles are from within the past five years.
* Number 6 on the backlist is Green Eggs and Ham, selling some 950,000 copies a year, 54 years after first being published.
* Other old time favorites include Dr. Seuss's ABC at Number 10, and Goodnight Moon at Number 11 with 770,000 copies being sold each year.
So there is something of a mismatch between what list_servs are interested in discussing versus what children appear interested in solely based on bestsellers.
* List_servs tend to focus on the frontlist 80% of the time whereas readers are about half that rate at 41% of their reading being front list
* List_servs tend to be dominated by social justice concerns 75% of the time whereas only about 5% of what children actually read (among bestsellers) have a social justice element.
* List_servs focus on YA 65% of the time whereas YA is only 7% of what children read, Middle Grade books get discussed 25% of the time when they are about 43% of the market, and picturebooks are discussed 10% of the time when they are 50% of what children read.

Monday, June 23, 2014

Paging Thomas Piketty

In a letter to Thomas Jefferson (17 October 1788) from James Madison.
Wherever the real power in a Government lies, there is the danger of oppression. In our Governments, the real power lies in the majority of the Community, and the invasion of private rights is chiefly to be apprehended, not from the acts of Government contrary to the sense of its constituents, but from acts in which the Government is the mere instrument of the major number of the constituents.

Sunday, June 22, 2014

Word counts and reading speeds

Useful information. From Word Count for Famous Novels by Steph Mineart. Here's a site for testing your reading speed, Speed Reading Test Online. From that site,
If top readers read at speeds of above 1000 words per minute (wpm) with near 85% comprehension, they only represent 1% of readers. Average readers are the majority and only reach around 200 wpm with a typical comprehension of 60%. This seems surprising since most readers, actively reading work documents, newspapers, magazines, books or the contents of a computer display are practicing daily for at least one hour. With such an intense training everyone should be close to top performances.

Unfortunately, this is far from the real situation. The average reader is five times slower than the good reader. Things are even worse if we consider reading efficiency as well as speed. Reading efficiency is reading speed weighted by comprehension rate and it amounts to 200 x 60% or 120 efficient words per minute (ewpm) for the average reader and to 1000 x 85% or 850 ewpm for top readers. Thus, an efficiency ratio of seven divides these two categories.
On one site, a commenter observes that there are "different 'averages' depending on the sort of reading one did. For 'pleasure' reading, or for comprehension, the speed was considered to be around 200 words per minute. For learning or memorizing it was around 100 wpm."

So if the average novel is about 100,000 words and the average reading speed is 200 words per minute, then the amount of time required to read an average novel is 8.3 hours. If the circumstances of your job and family life constrain you to reading only half an hour a day (a supposition), then you would read a novel every 17 days, or 21 books a year. That's for someone in the habit of reading continuously.

The BLS time use surveys indicate that the average person spends about 6 minutes a day reading. On that basis, then, we would expect the average person to take 83 days to complete an average book at an average speed. Alternatively, the average person would complete 4.5 books a year.

All very rough and ready, but usefully indicative of the bounds.

An example of the number of words in particular books:

Alice's Adventures in Wonderland 26,432
Lion the Witch and the Wardrob 36, 363
Peter Pan 47,232
Wind in the Willows 58,428
Secret Garden 80,398

Philosopher’s Stone – 77,325
Chamber of Secrets – 84,799
Prisoner of Azkaban – 106,821
Goblet of Fire – 190,858
Order of the Phoenix – 257,154
Half Blood Prince – 169,441
Deathly Hallows – 198,227

The Hobbit – 95,022
The Lord of the Rings – 455,125
The Two Towers – 143,436
The Return of the King – 134,462

Saturday, June 21, 2014

An unctuously superior tone on the basis of very imperfect understanding

I did not realize how conservative a cast of mind had T.S. Eliot. Here he is on friend Stephen Spender as quoted in Liberalism: The Life of an Idea by Edmund Fawcett.
He is a liberal and therefore tends to intolerance and judging others and he tends to take an unctuously superior tone on the basis of very imperfect understanding.

Friday, June 20, 2014

Scottish independence and the narcissism of small differences

From Networks and Hierarchies by Niall Ferguson.
In the networked world, the danger is not popular insurrection but indifference; the political challenge is not to withstand popular anger but to transmit any kind of signal through the noise. What can focus us, albeit briefly, on the tiresome business of how we are governed or, at least, by whom? When we speak of “populism” today, we mean simply a politics that is audible as well as intelligible to the man in the street. Not that the man in the street is actually in the street. Far more likely, he is the man slumped on his sofa, his attention skipping fitfully from television to laptop to tablet to smartphone and back to television. And what gets his attention? The end of history? The clash of civilizations? The answer turns out to be the narcissism of small differences.
Tyler Cowen has a post, Charlie Stross unintentionally explains why Scottish independence is a bad idea by Tyler Cowen in which he observes the shallowness of the Scottish Independence case and it reminds me of Fergusson's narcissim of small differences.
But my view remains pretty straightforward: when dislike of the policy choices of the electorate leads to a serious movement for secession, something has gone deeply wrong with the preconditions for democratic attachment. The UK is hardly the Third Reich, it has a long tradition of honest elections, and for left-leaning individuals the share of British government in gdp is likely to stay well over 40% in all plausible futures and furthermore most of the conservatives are relatively liberal on social questions. For those who favor independence for the Scots, what kind of general principle might you lay out for when other peoples also should seek secession? Do they think that the strongly red states in America also should consider secession? How about Vermont? I understand the libertarian case for such secessions, but most supporters of Scottish independence are not arguing from libertarian premises. How much secession do they think should be happening? Or do they hold particularist views which do not admit of any generalization at all? Either way, I consider this a true crisis of governance.

Thursday, June 19, 2014

Confidence, Motivation and Effort

From Why do children of Indian immigrants dominate American spelling bees? by Eugene Volokh.
The last eight national champions and thirteen of the last seventeen have been of Indian descent, a string of victories that began in 1999 with Nupur Lala’s win.

Given the extremely difficult and competitive nature of the spelling bees, it is impossible that the continuing domination of the spelling bees by Indian students over such an extended period is a fluke or an accident. There must be a rational explanation for their success, some secret to their prowess. Consequently, I would like to reopen the debate on this subject and welcome suggestions for making sense of this astonishing phenomenon.
Here are some of the contributions from comments.
Parental ambition

For the same reason Koreans own half the dry cleaning business in the US. It's easy for a minority group to find a niche and shape in-group norms in order to dominate in it.

North South Foundation bees as a feeder system

"Most American kids look up to sports figures," he says. "Indian kids are more interested in education, and they finally have a role model."

Spellbound, the 2002 documentary that featured Indian-American Nupur Lala's run to the 1999 Scripps title

Many first-generation South Asian parents saw NSF as a way for their children to assimilate—the best way to understand a culture, after all, is to learn its language.

Predecessor exemplars

Because their parents speak good English at home (thereby excluding quite a few other immigrant groups), combined with no desire to see their boys "waste" time playing sport (so excluding other US minorities who focus on that).

My experience of Indian immigrants (admittedly not in the US and obviously hardly exhaustive) is that many of them live vicariously through their boys, and conflate their own worth with the academic success of their boys.

Model minority myth

Relatively high parental educational attainment

Relatively dominant focus on education as a factor for economic national and self-improvement in India

Focus and effort on a competitive academic pursuit that promotes self-improvement, memory, and competition

Media attention

Spelling is also very individualistic.

There is a large volume of very high IQ tech industry immigration to the U.S. from South Asia and those who excel in spelling bees are mostly very high IQ children of these very high IQ parents

Success begets imitation. For every South Asian spelling bee winner, ten other contestants are encouraged in their belief that they can do the same.

Immigrants, in general, tend to be high in grit and inclined to devote great efforts to advancing academically.

English, of a sort, is a native language or highly fluently spoken second language for most South Asian immigrants that gives their children an edge over comparable immigrants from places like China or Russia where this is not the case.

The key to success is simply preparation

Specifically, an Indian-American cultural organization called the North South Foundation sponsors a spelling bee league. It is not just the best spelling bee league but almost the only one. Therefore its participants dominate the Scripps National Spelling Bee. Those participants are drawn almost entirely from children of Indian immigrants, who after all are the people who participate in Indian-American cultural organizations.

Why some Indian-Americans put so much effort into spelling bees and practically no one else does is an interesting question, but the fact that they do it adequately explains their dominance.

Indian immigrant children are doing well at this because their mothers value the activity.

I think it's just as simple as the fact that Indian-American families (or some of them) have made this a priority, whereas other ethnic groups have not.

Indians are exposed to different languages. They typically know at least 2 (an Indian language + English), and come into contact with at least a couple more.

Maybe they don't find American sports as interesting so competitive Indians find an outlet in intellectual sport.

Another factor could be overcompensation. Indians typically have English as a first language, but most Americans will probably interpret the accent as evidence of poor English skills.

It's not about emphasizing education broadly; it's likely that Indian-Americans strongly emphasize the spelling bee specifically, highly out of proportion compared to other groups, and even other groups with equally high emphasis on education generally.

First, most of the kids come from families with parents of above average intelligence.

Second, the Indian culture focuses on education...yes, more so than many other (but NOT all other) cultures.

Third, Indians come from an 'English first' culture...which gives them a leg up on many other immigrant groups, such as the Chinese.

But fourth: Indian parents (especially mothers) are really, really driven.

The Indian education system emphasizes rote memorization, as does English spelling. This pre-selects American immigrants who will train their children rigorously.

A commitment to education and a rigid enforcement of "There is only one right answer", none of this whole math / good effort nonsense?
The consensus across several sites seems to be that the results arise because of 1) High IQ homesettings, 2) Parental prioritization, 3) Institutional infrastructure (NSF), 4) Community awareness (documentary), 5) Immigrant desire for excelling, 6) Comparative advantage for Indians of multilingualism and native English compared to other immigrant groups and 7) Effort.

All reasonable hypotheses. Consistent with The Triple Package by Amy Chua and Jeb Rubenfeld - Confidence, Motivation and Effort.

Wednesday, June 18, 2014

Sails, like a meteor kindling in its flight

To the Autumnal Moon
by Samuel Taylor Coleridge

Mild splendour of the various-vested Night!
Mother of wildly-working visions! hail!
I watch thy gliding, while with watery light
Thy weak eye glimmers through a fleecy veil;
And when thou lovest thy pale orb to shroud
Behind the gathered blackness lost on high;
And when thou dartest from the wind-rent cloud
Thy placid lightning o'er the awakened sky.
Ah such is Hope! as changeful and as fair!
Now dimly peering on the wistful sight;
Now hid behind the dragon-wing'd Despair:
But soon emerging in her radiant might
She o'er the sorrow-clouded breast of Care
Sails, like a meteor kindling in its flight.

Tuesday, June 17, 2014

The clerisy and the rest

An interesting observation from The New Right by David Brooks.
The nanny state may have drained civil society, but simply removing the nanny state will not restore it. There have to be programs that encourage local paternalism: early education programs with wraparound services to reinforce parenting skills, social entrepreneurship funds to reweave community, paternalistic welfare rules to encourage work.

Second, conservatives should not be naïve about sin. We are moving from a world dominated by big cross-class organizations, like public bureaucracies, corporations and unions, toward a world dominated by clusters of networked power. These clusters — Wall Street, Washington, big agriculture, big energy, big universities — are dominated by interlocking elites who create self-serving arrangements for themselves. Society is split between those bred into these networks and those who are not. Moreover, the U.S. economy is increasingly competing against autocratic economies, which play by their own self-serving rules.

Monday, June 16, 2014

Just as long as the distance between two dates on a headstone

From The Return by Hakan Nesser. Attributed to fictional poet W.F. Mahler.
You ask me how long life is,
and I shall tell you like it is.
Just as long as the distance
between two dates on a headstone.

Dissatisfaction with fair processes yielding undesirable outcomes

A week or so ago, I mocked an outraged Social Justice Warrior (Privileged classist advocate with no contextual knowledge - Math is Hard edition). The advocate was outraged that half, yes half, of all children were growing up in households below the national median in income. That was simply unacceptable to the advocate.

It might be easy to similarly mock Nate Cohn for his post, Why Hispanics Don’t Have a Larger Political Voice?. Playing it for laughs and relying on the information in the post, the alternate headline could be styled "Hispanics Voters Disadvantaged: Illegal and Underage".

But that would miss that Cohn is actually asking an interesting question and one that so many advocacy groups skate across without wishing to acknowledge. Hispanic in itself is a made up concept of dubious validity grouping, as it does, all sorts of people from widely different circumstance based on, essentially, their or one of their immediate family speaking Spanish as a mother tongue.

Granted that people of a Spanish speaking heritage are 17% of the population.

Cohn introduces his post in advocate's language. He is raising the question in the context of the apparent demise of any probable movement on immigration reform, an initiative assumed to benefit Democrats.
Hispanic-Americans are growing in number, coveted by the nation’s political parties and deeply in favor of an overhaul of the nation’s immigration laws. Given this combination, why does such an overhaul still seem to be such a long shot in Washington?

One reason is that no demographic group is more marginalized in American elections than Hispanics. Many are ineligible to vote, while those who can vote often do not or are concentrated in noncompetitive districts and states. The dynamic will be particularly strong in this year’s midterms in November, when Hispanic voters will represent a tiny fraction of the electorate in the states and districts critical to the battle for control of Congress.
Marginalized implies an intentional effort to restrict. Is that what is happening? Well, not according to Cohn. As becomes clear, Hispanics are not marginalized, they are underrepresented. Isn't that the same thing? No. Why are they underrepresented? Because a large percentage are not citizens (only 69% are citizens). Oh, and also because a disproportionate number are below the age of 18 (28% of Hispanics are under 18 versus 22% for the population at large). These two factors alone mean that
As a result, only 49 percent of Hispanics are eligible to vote, compared with 74 percent of non-Hispanics. Hispanics make up just 11 percent of the voting-eligible population.
So from 17 to 11% based simply on the laws of the land regarding age and citizenship.

Impact is further reduced by low voter turnout. As Cohn points out, the Hispanic population skews younger than the population at large and young people across all demographics vote in lower concentrations than do older people. The consequence is that
In 2012, the turnout rate for potential Hispanic voters was 48 percent, compared with 66.2 percent among blacks and 64.1 percent among whites.
Finally there is the issue of historical concentration (three states have large Hispanic presence, Florida, Texas and California). This concentration is exacerbated by the Voting Rights Act which encourages the concentration of minorities to create minority majority districts.
As a result, half of all Hispanics live in just 65 of the nation’s 435 congressional districts. In districts held by House Republicans, Hispanics represent only 6.7 percent of eligible voters. There are a handful of competitive House districts with a large number of Hispanic voters, making the Hispanic population share in the House battlegrounds, at 7.4 percent, slightly higher than the Senate share. But with a healthy edge in the House, the G.O.P. can afford to lose the handful of competitive seats where Hispanics represent a meaningful share of eligible voters.
Note that this is a trade-off decision that not everyone is comfortable with. By creating majority minority districts, you virtually ensure that there is ethnic diversity in the House of Representatives which is a good outcome. On the other hand, by concentrating populations, you substantially decrease their influence in all other districts. If you had randomly assigned districts, you would likely have less visible diversity in the House but likely more Hispanic influence on legislation. For philosophical reasons you might prefer one outcome over the other but there is nothing nefarious about the issue. It is a choice.

So even though couched in advocate's language Cohn is actually dealing with a very relevant issue that centers on definitions and trade-offs in complex systems. Low Hispanic influence is not a product of nefariousness, xenophobia, or racism. It is the product of universally applied laws, personal choices, and trade-off decisions.

Sunday, June 15, 2014

Two channels and a few hours

Maj Sjöwall and Per Wahlöö were pioneers of the Scandinavian mystery. Writing between 1965 and 1975 they produced ten novels featuring Martin Beck, a police detective in Stockholm, Sweden. I was living in Stockholm at the time these were written and read at least three or four of them then. Over the years I have come across another two or three and read them. For some reason, I thought I had read them all. But recently in a used book store, while waiting for one of the kids, I found Murder at the Savoy by Maj Sjöwall and Per Wahlöö, previously unknown to me. The Beck series is not particularly high literature, but it is set in a city I knew well and of which I have fond memories.

The book I found is the 1972 Bantam paperback edition (originally published in 1970 in Sweden). The cover and pages are yellowed and brittle, the spine cracked in two places. In fact, gingerly as I held the book, the cover became separated in the reading of it. Maj Sjöwall and Per Wahlöö were both Marxists, (yes that long ago) and it shows in the text. There is a very jaundiced view of capitalists, the aristocracy, the government, the security forces and the police. Not dissimilar to a later writer, Henning Mankell whom I also enjoy.

But oddly, in both cases, their Marxism shows up in their books, not so much as a commitment to a (failed) ideology but rather in the form of a deep aversion to crony capitalism, rent seeking businesses and regulatory capture of the government bureaucracy. I think those concerns tend to transcend political doctrine.

So reading this book some forty years after it was published, how does it bear up? Not too badly, though I recollect enjoying others in the series more.

An observation of city life. The detective is reviewing a long list of crimes that have been committed over the weekend.
In almost all the cases, alcohol or drugs were of decisive importance. It may have been partly due to the heat, but more basic was the system itself, the relentless logic of the big city, which wore down the weak-willed and the maladjusted and drove them to senseless actions.
At some time in the past, I have commented on Clay Sharkey's speculation that drugs and alcohol are the coping mechanism people resort to during periods of rapid change.

There is this chillingly prescient passage.
He didn't have the vaguest idea of how to organize the search for Viktor Palmgren's killer. Assassinations hardly ever occurred in Sweden - he couldn't remember any political murder occurring in modern times. He wished that the information he had to go on wasn't so vague and that he knew a little better where to start looking.
Innocuous save with the knowledge of hindsight. Sixteen years after this was written, the Prime Minister of Sweden, Olof Palme was assassinated walking home with his wife from a movie. The assassin was never caught and Palme's murder remains a mystery.

There are a couple of passages which are striking for their contrast between the characterization of the city by Sjöwall and Wahlöö and the city that I remember. As a young adolescent riding public transportation between school and the suburbs, you couldn't help but be aware of the drunks and the drug addicts and the petty crime. Sjöwall and Wahlöö paint a much darker picture of the city than I recall though.

Finally there was this little line which might go unnoticed by anyone not having lived in Sweden at that time. Detective Beck has returned to his hotel room in the early evening, at loose ends.
Martin Beck turned on the bedside lamp and glanced at the TV set. He had no desire to turn it on and, besides, the programs were probably all over by now.
In that period of time, 1970-75, there were two TV channels, both government owned. There were plenty of Bulgarian and Polish cartoons, some British dramas and lots of black and white movies from 1915-1930s from America. I am the only person in my generation of whom I am aware here in Atlanta, with a wide viewing knowledge of Charlie Chaplin, Laurel & Hardy, Rudolph Valentino, Buster Keaton (one of my favorites), Harold Lloyd and others.

I recollect both channels being broadcast only in black and white. They were imaginatively, and as befitting a government enterprise, called Channel One and Channel Two. I remember practicing my Swedish by watching the subtitles scrolling by on English language shows.

The significance of the quoted passage is that, even with only two channels, they broadcast only a few hours. Basically from about 3 in the afternoon to somewhere around 9pm or a little later (if the final show was a movie).

Today, you check into a modern hotel and you have 30-50-100 channels with 24 hour content from all over the world. Back then, two channels a few hours.

The enemy of the productive class

From The Actual Politics of Professors by Neil Gross from a couple of years ago.

An interesting article that highlights the importance of definitions, the corrosiveness of advocacy and the paradox of academia. On the latter point, the paradox, as I see it, is that while most surveys show the bulk of the professoriate to lean left, whether by voter registration, self-identification, or dollar contributions to campaigns, and while most extreme policies seem to have their tap root in academia, on the other hand much of the most solid research seems to lean right in the sense of classical liberalism, liberty, freedom. The Niall Ferguson's, Angela Duckworth's, Thomas Sowell's of the world are all university professors. I don't know whether they identify with the right side of politics but they are certainly not of the left.

In the article Gross is clearly playing the advocacy role. He seems concerned that in a period of budget cuts, the reputation of the Academy as being left-leaning might have some undesirable repercussions and so he is clearly trying to downplay how left the Academy might be.
To answer this question, among others, I analyzed data from surveys and interviews with professors, including a nationally-representative survey of the American professoriate, conducted in 2006 with the sociologist Solon Simmons. My research shows that only about 9 percent of professors are political radicals on the far left, on the basis of their opinions about a wide range of social and political matters, and their self-descriptions (for example, whether they describe themselves as radicals). More common in the professoriate—a left-leaning occupation, to be sure—are progressives, who account for roughly a third of the faculty (and whose redistributionism is more limited in scope), and academics in the center left, who make up an additional 14 percent of professors.
From his research, we know that the majority (56%) are left of center. Gross does not provide the rest of the breakdown which is where the corruption of advocacy comes in. Omission of pivotal evidence is usually indicative that the omitted information does not support the hypothesis. Following Gross's taxonomy, I am guessing (because I have to owing to Gross's omission) that the unstated categories are: Center/Moderate, Center Right, Conservative, Radical Right. Assuming that there is no NA category, and based on other studies, I am guessing that the corresponding numbers that have been omitted are something like:
Center/Moderate - 30%
Center Right - 10%
Conservative - 5%
Radical Right - 0%
The skew becomes much clearer when you consider the omitted data. Radical Left:Radical Right - 9:0; Progressive:Conservative - 33:5; Center Left:Center Right - 14:10. That's the picture Gross is trying to obscure, hence the corruption of advocacy.

Playing the role of advocate doesn't mean that Gross is either wrong or uninformative, just that you have to be careful, you aren't dealing with an honest broker in pursuit of the truth. You are dealing with someone who is making a preferred point.

Gross offers this information, which I suspect is the source of the above mentioned paradox.
Radical academics, it turns out, are overrepresented not at elite research universities, like Harvard, but at small liberal-arts colleges. Most are concentrated in a handful of social sciences and humanities fields, like mine, sociology (in which radicals are nevertheless in the distinct minority), and in tiny interdisciplinary programs like women’s studies and African-American studies.
This rings true. A couple of months ago, there was the incident at UC Santa Barbra when Professor of feminist studies, Mireille Miller-Young was videotaped assaulting a 16 year old anti-abortion protestor and stealing her sign. Non-tier Tier 1 university/feminist studies. Then there was Laura Curry, adjunct professor of media studies and University of Buffalo, verbally assaulting protesters and creating a disturbance at an anti-abortion demonstration. Non-tier Tier 1 university/media studies.

I suspect that Gross is on to something. That radical professors in non-Tier 1 universities and in a very restricted set of fields, make a lot of news by the intensity of their anger and both help shape the impression of Academia as a hotbed of radicalism and lend the appearance that all the Academy might be overwhelmingly Hard Left.

The commenters on the article also have some interesting observations.
Thus, one reliable generalization that can be made about leftist academics is that they are the enemy of the productive class. A second would be that they will defend to the death the most privileged group, the tenured class. - Commenter mxb22
I suspect that mxb22 is on to something with his/her first observation. Professors with their tenure are much more able to focus narrowly and make the argument that for a particular issue, the best is more desirable than the good. "We can do better." To people living in the real world having to constantly make decisions with partial knowledge and make trade-offs between two (or more) nearly equally valued goals, they are always having to settle on the most optimal outcome among several competing goals and sacrifice the best outcome for any one of those goals. To denizens of the real world, the clerissy's pursuit of the very best outcome on a single goal does have the appearance of being the enemy of real world productivity.

The commenters call Gross out on not defining what he means by radical. Gross seems to accept that radical is something like a self-professed Marxist. In fact he makes it much worse by setting the bar ridiculously high by accepting the definition of a right wing politician.
In the course of seven years of research, I never encountered any radical professors who advocated “overthrowing the United States government.”
If that is your definition of radical, then clearly the Academy largely passes that bar. But that clearly is a fairly antiquated bar. Even accepting Marxism as the definition of a radical is not particularly enlightening. I doubt that Elizabeth Warren is or has ever identified as a Marxist but she is certainly a radical.

Even more importantly, and not much addressed in the article, how important are these pigeon holes? What makes someone conservative or progressive? What do we mean by that. A lot of libertarians are socially liberal and financially conservative; where does that fit. Many conservatives are exercised about inequality but ascribe its origins to different causes than might a progressive; where do they fit?

Finally, Gross does make a couple of points that I think are really quite important and which do not get enough attention.

Every system needs variance in it in order to evolve. Uniformity can only exist in a stable environment. If the context and environment change, then there has to be enough play in the system so that selective pressure can works its magic and evolve the system along with the context. Who wants to be the best flint knapper in a silicon chip context?

Gross, as an advocate, couches his point in terms of university funding.
Is it a problem for American higher education that 9 percent of faculty members are political radicals? The answer is that far-left academic radicalism is both a weakness and a strength. Were there no radical professors for conservatives to fulminate against—or had radical academics done more to keep their politics and their work separate—there might well be fewer political attacks on higher education today, and greater public support for colleges and universities. Radical professors in the post-1960s period overestimated how much tolerance there would be for them, and how far the idea of academic freedom could be stretched. Also, some academic radicals, privileging politics over scholarship, have waged unproductive battles against the scientific aspirations of their colleagues.

At the same time, academic radicals in the social sciences and humanities have given us novel and important ways of thinking about society and culture. They have alerted scholars and students to previously unrecognized dynamics of inequality around race, class, gender, and sexuality.
Forget the issue of funding. Is it important to have variance in the system? Absolutely. Is it worthwhile for their to be intelligent people pursuing dead-end speculations and fossilized ideologies. Sure, within the limits of available resources. That's how you evolve. Pursuing the obvious is only marginally rewarding. It is people on the frontier of the absurd and ridiculous who open up new vistas. Granted, only one in a hundred is likely to be a stout Cortez staring at the Pacific. The other ninety nine will be staring into a dark hole of meaninglessness and vociferously arguing for its relevance. Gross is right that "the social sciences and humanities have given us novel" ways of looking at society and culture. But for all the sturm and drang of the past three or four decades, it is not clear to me that these have indeed been materially important. Or at least not yet.

Yes, we need variance. Yes, much of that variance will be unproductive.

Gross approaches the real issue but then does not address it, mostly because of his advocacy around funding. It is not whether there should be radicals in the academic system pursuing non-productive or even destructive research and policy advocacy. The issue is how much and of what sort. Hard Right and Hard Left are equally noxious and potentially destructive. Gross tries to minimize that there are 9% Hard Left. He chooses to hide that there are (likely) 0% Hard Right. Shame on him for that omission. I think the real issue though is something more like, how many extremists do we want (0%, 9%, or other), does it make a difference whether they are Right or Left (i.e. logically wouldn't we want more balance?), is it important that they are isolated within certain departments?

I suspect we would be better off with slightly fewer radicals (say 5% rather than 9%). That we would be better off with a balance between the extremes (2.5% on the Hard Left and 2.5% on the Hard Right). Finally, I suspect that having all your radicals concentrated in victim studies departments is probably not healthy. In an environment of increased resource constraints, I suspect it is those departments that most likely are going to be on the chopping block (a la University of South Carolina). Diversity is important but needs to be balanced, dispersed and limited to a level which is affordable.

Saturday, June 14, 2014

I cannot undertake to lay my finger on that article of the Constitution which granted a right to Congress of expending, on objects of benevolence, the money of their constituents.

From Annals of Congress, House of Representatives, 3rd Congress, 1st Session, page 170 (1794-01-10). On Congress being petitioned for funds to extend relief to French refugees from the genocide occurring in Haiti.
Mr. Madison wished to relieve the sufferers, but was afraid of establishing a dangerous precedent, which might hereafter be perverted to the countenance of purposes very different from those of charity. He acknowledged, for his own part, that he could not undertake to lay his finger on that article in the Federal Constitution which granted a right of Congress of expending, on objects of benevolence, the money of their constituents.
This is usually, in quotation, rendered as
I cannot undertake to lay my finger on that article of the Constitution which granted a right to Congress of expending, on objects of benevolence, the money of their constituents.

The narcissism of small differences

From Networks and Hierarchies by Niall Ferguson. Very thought provoking. Read the whole thing. Below are snippets which give you the gist of his argument stripped of the supporting examples, data, and evidence.
In today’s terms, the hierarchy is not a single city but the state itself, the vertically structured super-polity that evolved out of the republics and monarchies of early modern Europe. Though not the most populous nation in the world, the United States is certainly the world’s most powerful state, despite the limits imposed by checks (to lobbyists) and balances (as in bank). Its nearest rival, the People’s Republic of China, is usually seen as a profoundly different kind of state, for while the United States has two major parties and a gaggle of tiny ones, the People’s Republic has one and only one. American government is founded on the separation of powers, not least the independence of its judiciary; the PRC subordinates law, such as it has evolved in China over the centuries, to the dictates of the Communist Party.

Yet both states are republics, with roughly comparable vertical structures of administration and not wholly dissimilar concentrations of power in the hands of the central government. Economically, the two systems are certainly converging, with China looking ever more to market signals and incentives, while the United States keeps increasing the statutory and regulatory power of government over producers and consumers. And, to an extent that disturbs civil libertarians on both Left and Right, the U.S. government exerts control and practices surveillance over its citizens in ways that are functionally closer to contemporary China than to the America of the Founding Fathers.
Clashes between hierarchies and networks are not new in history; on the contrary, there is a sense in which they are history.
Hierarchies seek to exploit the positive externalities of networks. States need networks, for no political hierarchy, no matter how powerful, can plan all the clever things that networks spontaneously generate. But if the hierarchy comes to control the networks so much as to compromise their benign self-organizing capacities, then innovation is bound to wane.
The mid 20th century was the zenith of hierarchy. Although World War I ended with the collapse of no fewer than four of the great dynastic empires—the Romanov, Habsburg, Hohenzollern, and Ottoman—they were replaced with astonishing swiftness by new and stronger states based on the normative paradigm of the nation-state, the ethno-linguistically defined anti-imperium.

Not only did the period after 1918 witness the rise of the most centrally controlled states of all time (Stalin’s Soviet Union, Hitler’s Third Reich and Mao’s People’s Republic); it was also an era in which hierarchies flourished in the economic, social and cultural spheres. Central planners ruled, whether they worked for governments, armies or large corporations. In Aldous Huxley’s Brave New World (1932), the Fordist World State controls everything from eugenics to narcotics and euthanasia; the fate of the non-conformist Bernard Marx is banishment. In Orwell’s Nineteen Eighty-Four (1949) there is not the slightest chance that Winston Smith will be able to challenge Big Brother’s rule over Airstrip One; his fate is to be tortured and brainwashed. A remarkable number of the literary heroes of the high Cold War era were crushed by one system or the other: from Heller’s John Yossarian to le Carré’s Alec Leamas to Solzhenytsin’s Ivan Denisovich.

Kraus was right: The information technology of mid-century overwhelmingly favored the hierarchies.
Today, by contrast, the hierarchies seem to be in much more trouble. The most obvious challenge to established hierarchies is the flow of information unleashed by the advent of the personal computer, email, and the internet, which have allowed ordinary citizens to organize themselves into much larger and more dispersed networks than has ever been possible before.
The challenge these new networks pose to established hierarchies is threefold. First, they vastly increase the volume of information to which citizens can have access, as well as the speed with which they can have access to it. Second, they empower individual citizens to publicize things that might otherwise remain secret or known only to a few. Edward Snowden and Daniel Ellsberg did the same thing by making public classified documents, but Snowden has already revealed much more than Ellsberg and to vastly more people, while Julian Assange, the founder of WikiLeaks, has far out-scooped Carl Bernstein and Bob Woodward (even if he has not yet helped to bring down an American President). Third, and perhaps most importantly, the networks expose by their very performance the inefficiency of hierarchical government.
Yet it would be naive to assume that we are witnessing the dawn of a new era of free and equal netizens, all empowered by technology to speak truth to (and about) power, just as it would be naive to assume that the hierarchical state is doomed, if not to revolutionary downfall then at least to a permanent diminution of its capacity for social control.

Modern networks have prospered, paradoxically, in ways that are profoundly inegalitarian. That is because ownership of the information infrastructure and the rents from it are so concentrated.
In the networked world, the danger is not popular insurrection but indifference; the political challenge is not to withstand popular anger but to transmit any kind of signal through the noise. What can focus us, albeit briefly, on the tiresome business of how we are governed or, at least, by whom? When we speak of “populism” today, we mean simply a politics that is audible as well as intelligible to the man in the street. Not that the man in the street is actually in the street. Far more likely, he is the man slumped on his sofa, his attention skipping fitfully from television to laptop to tablet to smartphone and back to television. And what gets his attention? The end of history? The clash of civilizations? The answer turns out to be the narcissism of small differences.

Liberals denounce conservatives with astonishing vituperation; Republicans inveigh against Democrats. But to the rest of the world what is striking are the strange things nearly all Americans agree about (for example, that children should be packed off to camps in the summer). Many English people are outraged about immigrant Romanians. But to East Asian eyes the English are scarcely distinguishable from Romanians. . . .

It is no accident that most of the world’s conflicts today are not between civilizations, as Samuel Huntington foresaw, but between neighbors. That, after all, is what is really going on in Syria, Iraq, and the Central African Republic, not to mention Ukraine. Can anyone other than a Russian or a Ukrainian tell a Russian and a Ukrainian apart? And yet how readily one is pitted against the other, and how distractingly.
Yet our own time is profoundly different from the mid-20th century. The near-autarkic, commanding and controlling states that emerged from the Depression, World War II, and the early Cold War exist only as pale shadows of their former selves. Today, the combination of technological innovation and international economic integration has created entirely new forms of organization—vast, privately owned networks—that were scarcely dreamt of by Keynes and Kennan. We must ask ourselves: Are these new networks really emancipating us from the tyranny of the hierarchical empire-states? Or will the hierarchies ultimately take over the networks as they did a century ago, in 1914, successfully subordinating them to the priorities of the national security state?

A libertarian utopia of free and equal netizens—all networked together, sharing all available data with maximum transparency and minimal privacy settings—has a certain appeal, especially to the young. It is romantic to picture these netizens, like the workers in Lang’s Metropolis, spontaneously rising up against the world’s corrupt hierarchies. Yet the suspicion cannot be dismissed that, despite all the hype of the Information Age and all the brouhaha about Messrs. Snowden and Assange, the old hierarchies and new networks are in the process of reaching a quiet accommodation with one another, much as thrones and telephones did a century ago.

Friday, June 13, 2014

Wittgenstein on accurate, precise and useful

From Philosophical Investigations by Ludwig Wittgenstein
To repeat, we can draw a boundary - for a special purpose. Does it take that to make the concept usable? Not at all! (Except for that special purpose.) No more than it took the definition: 1 pace = 75 cm. to make the measure of length ‘one pace’ usable. And if you want to say “But still, before that it wasn’t an exact measure”, then I reply: very well, it was an exact one. – Though you still owe me a definition of exactness. . . . Is it even always an advantage to replace an indistinct picture by a sharp one? Isn’t the indistinct one often exactly what we need?
In many discussions we sling around terms that have some implication of a boundary but, when you investigate, has not so much a boundary as a loose, undefined and unguarded frontier, subject to change, and including some things people don't expect and excluding others they were certain were within the boundary.

Precision, accuracy and usefulness are important concepts that are often at the crux of many disagreements.

Thursday, June 12, 2014

Privileged classist advocate with no contextual knowledge - Math is Hard edition

An example of contextual myopia when people become so focused on the importance of their cause that they lose sight of the foolishness of their argument. Doesn't matter how good the cause might be. This has nothing to do with the value of the cause, motivation or effectiveness. Just the argument. From Inequality Begins at Birth in America by Jeff Madrick, a tautological headline if there ever was one. I clicked through because I thought it might have some evidentiary information, but no. Just an advocate's polemic.

I quit reading eleven paragraphs in when I got to:
In addition, nearly one in two children under 18 live below double the poverty line (about $47,000 today), an income just barely adequate to meet basic needs, and completely inadequate in many localities.
What a great example of contextual myopia. Almost Lake Woebegonian in its simplicity.

Median household income is at just about $50,000. Ignoring all the various nuances, you would expect nearly one in two children under 18 to live in a household with an income below $47,000. Basically the advocate is saying that about half the children live in households below the median. Well, yeah. I am going out on a limb, but I have to guess that Madrick is in that half of the class that makes the top half possible. In the catalogue of compelling arguments, I think this one is probably pretty close to the bottom.

And it is interesting, to me, how this plays out. Childhood poverty is a complex and real issue. It warrants attention and careful action. However, if its vocal advocates make a rookie mistake like this, why would you trust them to make any important decision? You end up with someone passionately believing something that others are likely to believe as well but undermining their position by letting their passion blind them to the context and thereby sabotaging the argument that others are already willing to believe.

You don't have to scratch much further below the argument to get to an inference that is insulting. Madrick apparently believes that the majority of Americans, existing on the median household income are "existing on an income just barely adequate to meet basic needs". Really? Just how rich and privileged do you have to be to believe that to be the case. There is such a stench of disengaged privileged altruism that it is enough to make you gag. More to the point, what is the household income level that Madrick thinks makes you safe from bare adequacy? And even more pertinently, how on earth would you achieve that?

The US is the most productive nation in the world and yet according to Madrick, the median income is just barely adequate to make ends meet. I wonder what he would think if he travelled outside his privileged cocoon and saw how the rest of the world lived.

Wednesday, June 11, 2014

Intolerance, savage attacks, and argument incapacity

I am often concerned about the rarely acknowledged classism embedded in many people's behavior and in some aspects of the political parties and that seems reinforced by two recent articles and a recent event.

A Case Study in Lifting College Attendance by David Leonhardt highlights an innovative approach in Delaware to ensure that those with the ability and desire to attend competitive universities do in fact attend such universities.
Leslie Carlson, a guidance counselor at Mount Pleasant High School, asked Ms. Nye where she would go if money were no object. She replied that money was an object. Mrs. Carlson said: Yes, but what if it weren’t? Ms. Nye eventually named Stanford, before adding that the application fee of $90 was more than she could afford. The counselor then told her about the fee waivers she would soon receive.

This report is an encouraging example of what might make things work better but it is also a reminder of how often factors that seem trivial to more affluent families can keep low-income students from college. Before that conversation, the only college to which Ms. Nye had applied was the Colorado School of Mines, because the application was free. Ultimately, Ms. Nye received a scholarship to Stanford that requires no payment from her parents and no loans for either her or her parents. Her contribution consists of $5,000 a year she is supposed to earn through a campus job.
No matter how well intentioned we are, it is challenging to understand the reality that others face who lead lives substantially different from our own. A $90 application fee is the difference between attending Stanford and not? Inconceivable. And yet for some, that is a reality.

Well intentioned gentry are prone to designing elaborate programs to make better the lives of those less fortunate. Most often these programs fail because they are based on gentry assumptions and are not grounded in the experienced reality of the truly poor. Classism that is blind to its own privileges and that is also willing to ignore the negative consequences arising from such blindness (most people are satisfied by measuring success against the goodness of intentions rather than achieved outcomes) is hidden in plain sight.

And then there is this study, Slut-shaming has little to do with sex, study finds: Sociologists say affluent university women use slut-shaming to show poorer women they are ‘trashy’ and don’t belong by Marisa Taylor.
A new study of college women and their attitudes about so-called sluttiness found that slut-shaming — calling out a woman for her supposedly promiscuous sexual behavior — actually had more to do with a woman’s social class than it did with sexual activity.

Sociologists from the University of Michigan and the University of California at Merced occupied a dorm room in a large Midwestern university, regularly interacting with and interviewing 53 women about their attitudes on school, friends, partying and sexuality from the time they moved in as freshman and following up for the next five years.

The researchers discovered that definitions of "slutty" behavior and the act of slut-shaming was largely determined along class lines rather than based on actual sexual behavior. What's more, they found the more affluent women were able to engage in more sexual experimentation without being slut-shamed, while the less-affluent women were ridiculed as sluts for being “trashy” or “not classy,” even though they engaged in less sexual behavior.
Again, the usually unacknowledged role of class. And while the blame for many social ills is set at the feet of the patriarchy, there is a lot of evidence that many of the social ills arise from both intended and unintended exclusionary actions undertaken by women.

The following article highlights the extraordinary complexity of some of these issues. Yes we want people to enjoy maximum freedom. But that is set against the reality that freedom to make choices means some people will make bad choices and that those with the least resources are least able to cope with the consequences of those bad choices. The unpleasant trade-offs implied are captured in this article, One way to end violence against women? Married dads. by W. Bradford Wilcox and Robin Fretwell Wilson in which they observe the correlation between marriage and desirable life outcomes for women.

Marriage is not actually the answer. It is the values and behaviors that permit a marriage to succeed that are the real determinants. Everyone getting married won't, per se, solve any problems. Coaching and equipping people with the values and behaviors that might make them eligible for marriage probably would.
But obscured in the public conversation about the violence against women is the fact that some other men are more likely to protect women, directly and indirectly, from the threat of male violence: married biological fathers. The bottom line is this: Married women are notably safer than their unmarried peers, and girls raised in a home with their married father are markedly less likely to be abused or assaulted than children living without their own father.


Marriage is no panacea when it comes to male violence. But married fathers are much less likely to resort to violence than men who are not tied by marriage or biology to a female. And, most fundamentally, for the girls and women in their lives, married fathers provide direct protection by watching out for the physical welfare of their wives and daughters, and indirect protection by increasing the odds they live in safe homes and are not exposed to men likely to pose a threat.
The classism here is that of various segments of the feminist movement who want to argue that women can make any lifestyle choice they want (which they should be able to) without any trade-offs or consequences (which is obviously not true.) Marriage may be an undue constriction on women of a privileged background and resources but it is a more than rational choice for other women. The issue is that those movements (dominated by the privileged) are trying to advocate policies that can be made to work for them (because of their income, or education or familial resources) but which are dangerous for those without the resources of their class.

Finally there is this spat that occurred in the past week. Ruth Graham published an opinion piece in The Slate, Against YA subtitled "Read whatever you want. But you should feel embarrassed when what you’re reading was written for children." Her argument is that Young Adult (YA) literary fiction is written for an immature audience that lacks mature judgment and that that lack of maturity is reflected in the simplicity of the YA stories. Graham's position is only a variant of the age old view that there is a hierarchy of bad, better and good literature and that one should aspire to reading good literature. Graham acknowledges the value of YA for its intended audience, i.e. she is not condemning the genre per se. And while her argument is aimed at YA, her underlying model of good literature versus everything else is not restricted to YA. She alludes to the same issue with mystery writing. She believes that once you acquire the cognitive capacity for more complex, nuanced, and ambiguous literature, you should not sully yourself with simpler, clearer and more didactic books.
But the very ways that YA is pleasurable are at odds with the way that adult fiction is pleasurable. There’s of course no shame in writing about teenagers; think Shakespeare or the Brontë sisters or Megan Abbott. But crucially, YA books present the teenage perspective in a fundamentally uncritical way. It’s not simply that YA readers are asked to immerse themselves in a character’s emotional life—that’s the trick of so much great fiction—but that they are asked to abandon the mature insights into that perspective that they (supposedly) have acquired as adults. When chapter after chapter in Eleanor & Park ends with some version of “He’d never get enough of her,” the reader seems to be expected to swoon. But how can a grown-up, even one happy to be reminded of the shivers of first love, not also roll her eyes?

Most importantly, these books consistently indulge in the kind of endings that teenagers want to see, but which adult readers ought to reject as far too simple. YA endings are uniformly satisfying, whether that satisfaction comes through weeping or cheering. These endings are emblematic of the fact that the emotional and moral ambiguity of adult fiction—of the real world—is nowhere in evidence in YA fiction. These endings are for readers who prefer things to be wrapped up neatly, our heroes married or dead or happily grasping hands, looking to the future. But wanting endings like this is no more ambitious than only wanting to read books with “likable” protagonists.

Fellow grown-ups, at the risk of sounding snobbish and joyless and old, we are better than this. I know, I know: Live and let read. Far be it from me to disrupt the “everyone should just read/watch/listen to whatever they like” ethos of our era. There’s room for pleasure, escapism, juicy plots, and satisfying endings on the shelves of the serious reader. And if people are reading Eleanor & Park instead of watching Nashville or reading detective novels, so be it, I suppose. But if they are substituting maudlin teen dramas for the complexity of great adult literature, then they are missing something.
This old gradgrindian, puritanical model of ratcheted reading (you can read up but you shouldn't read down) is a normative position on the part of Graham. She doesn't attempt to compare the advantages and costs of complex reading with the advantages and costs of simple reading. She merely asserts that the former is better than the latter.

So an opinion piece with a reasonable and by no means uncommon argument. For those wanting to reject the argument, that path is reasonably clear; either refute her details (not particularly productive to drag the reading audience into minutia of definitions, details and nuance) or demonstrate why the gradgrindian model of reading (purposeful reading for cognitive improvement) is neither how people actually do read nor is it associated with superior reading outcomes. People actually take a portfolio approach of reading: some simple stuff, some complex; some purposeful, some random; some reading for pleasure, some for achievement; some reading because it is required, some because it is elected; some contemporary, some some classic; some fiction, some nonfiction. Not only is there a portfolio of reading attributes but that portfolio and the balance of elements is constantly shifting and changing over time. The normative argument that people OUGHT to read the most complex nuanced works they can at all times is refuted by how people actually do read and by the fact that the most successful people in any field all similarly demonstrate the portfolio affect.

A straightforward normative argument is the instigating event and there is a straightforward counterargument available.

But how did the clerisy actually respond on reading-related list_servs, blogs, twitter, and comments? The response tended to affirm Graham's argument that YA literature appeals to immature minds. The response might harshly, but not particularly inaccurately, be characterized as classic middle school mean girl. Who is this person, Graham,? Why did she say mean things about YA? Doesn't she know there is some good YA? What are her credentials? How can we punish her (let's write to the editor)? The overwhelming majority of responses were classic Us/Them ostracism. She is not one of us and therefore we can launch unconstrained ad hominem attacks on her credentials, intelligence, ignorance, motives, create strawman arguments that were not part of her opinion piece and savagely knock down those strawmen arguments, derision, mockery and even satire. What was largely missing was anyone actually civilly engaging with Graham's argument and making an affirmative case for the value of YA literary fiction.

Part of the issue was undoubtedly that YA is an ambiguous category. It is of recent vintage and was initially essentially a publisher marketing ploy for increased market segmentation. One that has been largely successful it should be added. There are three challenges. 1) Is YA literary fiction that which is written by the author with a 12-18 year old audience in mind? OR 2) Is YA a style of writing characterized by emotion, dysfunction, relationships, OR 3) Is YA what 12-17 year olds actually read? According to publisher surveys, between 55-80% of readers are older than 18. The median reader is a 40 year old woman. Basically, most published YA literature is not read by young adults AND most of what young adults read is not YA.

Finally, publisher surveys seem also to indicate that the bulk of literary reader fiction books are middle and up in terms of income and to be predominantly college educated.

Whether intentionally or not, Graham skates very close to a stereotype of YA readers as being 40 year old college educated privileged women reading simplistic, predictable, and emotionally incontinent YA "literary" fiction. Graham finds that unseemly. While that is not directly the argument that Graham made, I think that is the implication to which most commenters and bloggers are responding.

But it is the response that is interesting. These blogs and list_servs are overwhelmingly of the clerisy. Urban, educated, well-off professionals, creatives (authors), ostensibly liberal, firmly committed to multiculturalism, inclusion, respect, etc. And yet, when something near and dear is challenged, their response is like everyone elses, primal tribalism - destroy the evil intruder. Like William Golding's Lord of the Flies (often characterized as YA though written for adults decades before the marketing push for a YA category), the mask is slipped.

But it is not so much about the hypocrisy. We are all more or less hypocritical; there is always a gap between how we aspire to behave and how we actually behave. What strikes me is the blindness to the classism. These are by and large very privileged people by income, by education, by status. Their intolerance of anyone else having a different opinion is troubling, the savagery of their attacks disconcerting, and their incapacity to respond to an argument in dispassionate terms (even when they have the better argument) is striking.