Hatrack River
Hatrack.com   The Internet  
Home   |   About Orson Scott Card   |   News & Reviews   |   OSC Library   |   Forums   |   Contact   |   Links
Research Area   |   Writing Lessons   |   Writers Workshops   |   OSC at SVU   |   Calendar   |   Store
Print this page E-mail this page RSS FeedsRSS Feeds
What's New?

Uncle Orson Reviews Everything
April 6, 2017

First appeared in print in The Rhino Times, Greensboro, NC.

Stats, Men Not At Work, Eggs

These past few weeks, I've read three books that consisted mostly of statistics -- both methodology and results.

Even if you're a statistician -- which I am not -- no one can call such analyses and lists "thrilling." Yet when good writers (which invariably overlaps with the group called "good thinkers") give an account of careful, rigorous statistical analysis, the results are both fascinating and reasonably reliable.

Remember that many of the statistics we're fed on the news are complete nonsense, either because the analysis is faulty or because there was never any data to begin with.

For instance, you remember that ludicrous statistic that Super Bowl Sunday was the busiest day of the year for shelters for battered women. It was based on, you guessed it, nothing at all. It was just a mean thing to say about men, which always makes politically correct people feel happy and proud.

The three books I'm talking about are:

Men Without Work: America's Invisible Crisis, by Nicholas Eberstadt

Graphing Jane Austen: The Evolutionary Basis of Literary Meaning, by Joseph Carroll, Jonathan Gottschall, John A. Johnson, and Daniel J. Kruger

The Bestseller Code: Anatomy of the Blockbuster Novel, by Jodie Archer and Matthew L. Jockers.

The second two are of great interest to a fiction writer like me, though let me tell any novelists reading this review that The Bestseller Code does not get very specific about the list of elements that mark bestsellers in their computer algorithms.

While such a list would be fascinating, too many writers would be tempted to deform their stories by trying to insert "bestseller elements" that don't belong there. That's why I also haven't recommended the book to my writing students this semester; I don't want to read the results when student writers subvert their own vision and talent by trying to fit a formula.

(By the way, I wish you could read some of the wonderful work my students have done this semester. And, in all likelihood, someday you will.)

The scientific project in The Bestseller Code was fascinating. The authors used data-mining to extract from thousands of recent fiction books -- both bestsellers and non-bestsellers -- the kinds of events in their storylines that make it possible for the computer to predict which books will be bestsellers and which will not.

Remember, this has nothing to do with the "quality" of the books -- just the sales, as reported by the New York Times list.

Getting a computer to "understand," from the texts of the novels alone, what is "happening" in the stories, who the heroes and heroines are, and so on, requires extraordinarily clever program design and execution, but the authors ended up with a powerful instrument that was successful, at rates consistently above 80 percent, to predict which novel texts would be bestsellers.

Remember that even 90 percent accuracy means that ten percent of the would-be bestsellers were missed -- or that ten percent of those predicted to be bestsellers were not.

I really appreciated the time the authors took to describe their methodology -- how they kept their samples from being biased, and how their instrument, having been perfect before 2005, was later tested on the very different bestsellers and non-bestsellers between 2005 and 2011. The results -- even though they now included the Shades of Grey series and the Girl Who books -- were a solid demonstration of the accuracy of their instrument.

But they refuse all requests by authors to let them run their novels through that instrument. The real interest should be centered on what we learn about human beings from the kinds of stories that do and don't pique their interest.

As for Graphing Jane Austen, the project is even more esoteric than The Bestseller Code. Here, the authors (who include Jonathan Gottschall, who gave us the wonderful and useful book The Storytelling Animal) surveyed hundreds of readers, asking them to give numerical ratings to 2,000 characters from 202 British novels -- including all of Jane Austen's books.

The questions about these characters were carefully designed to create usable data. Naturally, the characters in Jane Austen's books were far more familiar to most participants than the characters in more obscure novels, so the resulting data were adjusted for sample size.

What emerged was a fascinating and, in many cases, fairly clear idea of the "agonistic" (dealing with protagonists, antagonists, and minor characters) attributes and structures in these novels.

Part of their purpose was to examine the treatment of female characters vis-a-vis males, but they were surprised to discover that, regardless of the sex of the respondents, there was a far greater distinction between protagonists and antagonists than between male and female characters, at least in the eyes of these relatively expert readers.

No matter how much we might like -- or love -- these novels' protagonists, the strongest emotional responses were generated by the antagonists. We may not all love the same people, but boy oh boy do we hate the bad guys!

Now, there were some books and some characters that were hard to classify. Is Becky Sharp in Thackeray's Vanity Fair a heroine or a villain? (Correct answer: both, if you're paying attention.)

The characters were also sorted by other categories, including their motives, and the result of this study was, in my opinion, far more useful for a writer than the results of The Bestseller Code, primarily because Graphing Jane Austen doesn't give you any formulas and it's not about sales.

It's simply a way of examining and thinking about characters -- and writers will do better if they understand their own characters better.

However, since Graphing Jane Austen is a scholarly book with a small audience, it is priced accordingly. The cheapest copies are about $80, and even the Kindle version costs more than $78. You can "rent" the Kindle edition for thirty bucks -- whatever that means.

To me, it was worth the cost to buy the Kindle version. But I'm a specialist in precisely the area they studied, so this was, to me, like a chef buying really expensive knives. The Bestseller Code is priced far more rationally ($16 on Amazon for the hardcover, $13 for the Kindle edition).

Or you can download for free a Kindle version of the second chapter of the book, under the title "Becoming Bestsellers: John Grisham and Danielle Steele." Kind of a free sample, to convince you that you're going to learn more than you actually are from the full book.

Perhaps writers wishing to learn their craft would gain more from buying and reading Lisa Cron's book Wired for Story: The Writer's Guide to Using Brain Science to Hook Readers from the Very First Sentence. It's based on some of the science in The Storytelling Animal, and even though it leans a bit too much toward formulas, it provides a lot of useful and potentially helpful information.

Both Graphing Jane Austen and The Bestseller Code spend a lot of time doing an excellent job of describing their methodology and showing why it is as reliable as it is, without achieving perfection. If you don't understand how statistics work, and why some stats are reliable and some are utter trash, both these books will demonstrate how careful scientists think about, design, and analyze stats.


Which brings me to the third statistical book I read this past fortnight: Men Without Work: America's Invisible Crisis, by Nicholas Eberstadt. It may be the most mind-numbingly statistical of these three books -- but it's also the one that should be read and studied by anybody who is making public policy at every level of government -- from school boards on up to congresswights and White House staffers.

(Not even his biggest fans think that President Trump can or will actually read a whole book, least of all a hard one.)

Men Without Work deals with those ten or twelve million American men of prime working age who have opted out of employment ... completely. They aren't looking for work. They aren't working.

Eberstadt mines the available data to try to figure out what they're doing with their days (mostly watching screens; we can only guess if it's movies, documentaries, TV shows, or porn) and how they're managing to live (many are on some kind of government assistance, but most are dependent on the generosity and patience of women).

These are not baby boomers entering retirement. These are prime-age men who are Neither Employed nor in Education and Training (NEETs). And Eberstadt's quest is to figure out (1) why they are Not In the Labor Force (NILF) and (2) what, if anything, we can do about it as a society, whether by government action or other means.

And we do have to do something, because the increase in their numbers has been pretty continuous since 1965. Other first-world nations simply don't have this problem. They deal with cycles of unemployment, but they don't have a problem with huge numbers of able-bodied working-age men who have opted out of "adult responsibilities not only as breadwinners but as parents, family members, community members, and citizens."

Because that's what Eberstadt has learned. These prime-age NILFs have a lot of things in common. They are far more likely than working or unemployed-but-seeking-work men to be:



Less educated

Former prisoners

Not likely to be employed even if they apply

We're not talking about homeless guys who can't get an interview because they're dirty and don't own a suit. But because they have been "un-workers" for years, employers are going to look at their applications, see a long stretch of unemployment, and reject their application out of hand.

This is not irrational. Being out of the workforce changes you. Eberstadt says, "The data here suggest that something like infantilization besets some un-working men."

Now, there is a huge group of American workers who are likely to drop out of all employment for years -- for decades -- but when they return to the workforce, they are readily given jobs in which they are highly likely to succeed and be remarkably productive.

That group is called "women."

When women leave the workforce to stay home and care for children, they are probably busier than they were when employed-but-childless. Being a stay-at-home mother means fulfilling responsibilities around the clock, and any employer who doesn't recognize that active motherhood is an excellent predictor of workplace productivity doesn't deserve to succeed.

Being an un-working man who is not a stay-at-home dad (a very, very tiny portion of the NILFs) gives no such preparation for future employment. On the contrary, spending days in absolute idleness is about as debilitating, in terms of employment, as spending those days in a marijuana or alcoholic haze. Watching screens for hours and hours is not good job preparation.

What put these men into this situation? Well, this is where things get tricky, and this book is remarkable in the fact that it includes, within its covers, criticisms of Eberstadt's conclusions by two other respected economists. Eberstadt knows that the data just don't exist to completely explain why men leave the workforce.

Partial explanations are pretty plain. The portion of NILFs who are ex-convicts have run into the nasty problem that while our society talks about rehabilitation, many employers are reluctant to hire ex-cons. They conjure up images from prison movies and storylines from TV cop shows, in which ex-cons routinely return to a life of crime.

And, of course, some of these NILFs might be criminals right now -- that wouldn't show up in the stats.

But Eberstadt dispels that conjecture pretty effectively by pointing out that NILFhood is not statistically tied with the huge rise in crime that began in the 1960s -- nor with the huge drop in violent crime that began in the 1980s.

In other words, at the very time when the number of criminals who had served their time and got out of prison increased greatly, violent crime nationwide went down sharply. So release from prison did not result in a crime wave, but the opposite.

Yet ex-cons have such a hard time finding work, it's no wonder so many of them become prime-age un-workers.

One of Eberstadt's critics pointed out that while Eberhardt emphasizes the continuity of men leaving the workforce in relatively steady numbers since 1965, he glosses over the fact that each recession, with its spike in unemployment, worsens the problem sharply.

Close examination of the data shows that recessions put men (and women) out of work, and what increases the number of un-working men is that fewer men go back to work when the recession ends.

But this doesn't actually help us identify the causes of un-working among men. Because women and European men do go back to work after recessions; it's only here in the USA that such a large number of men stay out of work and stop looking.

Nor can it be blamed on deindustrialization, because European men have faced the identical problems and they're still working!

Maybe, for some reason, America has stopped stigmatizing un-working men, while European societies still exert strong social pressure. Men there go back to work because they would be ashamed not to.

But why aren't American men ashamed to be un-workers? Remember that these are to a large extent the same men who do not educate themselves, do not marry, and survive by accepting the financial or practical support of women they are not married to. For some reason, in America there isn't enough of a social stigma attached to this situation to provoke these men to prepare themselves and get back to work.

Oh, by the way: Eberstadt sorts out this problem by race and national origin, and finds that hispanic immigrants are simply not among the un-working. But they aren't "taking" jobs away from the American un-workers. How could they? The un-workers aren't even looking for work.

As for race, of course American-born non-hispanic blacks are over-represented among the NILFs -- after all, they're also overrepresented among ex-cons and high school dropouts. But they aren't even close to being a majority of the un-working men.

Eberstadt, who holds the Henry Wendt Chair in Political Economy at the American Enterprise Institute, is aware of his own pre-formed beliefs, so he is self-skeptical about his tendency to think that government handouts are a significant cause of prime-age men remaining out of the labor force.

And his critics point out that even though applications for government disability status greatly increase with each recession, the number of approvals of such applications do not. Men don't leave the workforce because government doles are such a bargain -- in the vast majority of cases, any kind of work will be better, financially, than any kind of government support. Our welfare system, in short, is not an incentive, though it is, as intended, a safety net.

I would like to add my own two cents' worth to the discussion of causes by pointing out the stats that Eberstadt repeatedly cites but does not explore: The role of low education levels in predicting whether a man will become a NILF.

Back when the Ophelia Syndrome was all the rage, we were schooled in the idea that women are often demeaned or overlooked at school. There were no stats to support this view, because, of course, it isn't true.

The overwhelming evidence is that men drop out of school and fail in school at rates shockingly higher than women.

Women are far likelier to graduate from high school, enter college, get a bachelor's degree, go on to do graduate work, and leave college prepared to step into gainful employment at a far higher level than anyone who did not graduate from high school, enter college, get a degree, or go to grad school.

America's educational system isn't "failing" boys -- it is actively driving them out. Then, when they become "prime-age" men looking for work, these less-educated men are far less likely to be hired.

Of course we need to change our schools to be more boy-friendly -- but in all candor, this is not likely to happen as long as the overwhelming majority of schoolteachers and school administrators are women, or are committed to treating natural girl-behavior as "good" and natural boy-behavior as "bad."

Boys who are pounded with disapproval for being nine-year-old boys will reach age fifteen already at a severe educational disadvantage. Every behavior that is natural to them is constantly disapproved, while girls thrive in precisely the culture that schools foster.

Here's where the problem of boys' disadvantages in American schools really gets bad: Since 1965, the number of jobs that "require" a college degree has exploded, even though almost none of these jobs actually require that their workers know anything they would have learned in college.

My mother used to be head of the College Advisement Center in one of America's leading business schools at a major university, and she told me repeatedly that when students come back to talk to her after graduation, they usually reported that it took them six months on the job to unlearn everything they were taught in business school that was simply wrong.

It's not as if the actual value of college graduates to most businesses that hire them is higher than the value of high school grads -- or high school dropouts -- who are willing to work hard, follow instructions, and show initiative and responsibility.

But the personnel officers at these corporations will say, The very fact that they dropped out of high school or didn't go to college shows that they aren't willing to work hard, follow instructions, or show initiative or responsibility.

To which I would answer, The educational system already proved to them that working hard, following instructions, and showing initiative and responsibility did not lead to success in high school or, far too often, in college.

Eberstadt is clear about one thing: As long as ten or twelve million prime-age able-bodied men remain out of the work force, it is costing all of us a lot of money.

If they were working, at average levels of productivity, and earning average or even below-average salaries, it would be a huge boost to the whole economy. There would be more jobs; everybody would sell more goods; and just imagine the relief and happiness it would provide to the women who are at present subsidizing their idle lifestyle.

There is no government action at levels above school boards and school administrations that is likely to improve the problem of American un-working men.

But American businesses can make a huge difference right now by dropping the college-degree requirement for jobs that don't really require a college education -- which is most of them outside the professions. Obviously, to become a doctor, lawyer, or engineer, you have to have high-level training.

But to get a job as a newspaper reporter, a store manager, a fireman, a videogame designer, an office worker, or a salesman, it's absurd to think that any college training will improve on simply getting hired and learning the job from people who are already doing it.

Do college grads really have a better learning curve than other people when it comes to learning on the job?

We already know how many entrepreneurs either dropped out of or never went to college before they started making their millions (or billions). I personally know many non-college-educated people who are doing just fine in fields that have many college grads in them.

But because they lack that completely irrelevant college degree, they are still being punished -- they often start at a lower salary, they are given fewer promotions, and bosses are still likely to explain, "I'd give you a better raise, but you are already receiving the highest salary I can give to an employee who doesn't have a college degree."

This is, of course, both foolish and unfair. If somebody's doing the job well, then who cares if he had a college degree?

Since dropping out of school or not going to college are closely linked to hireability in our absurd era, how can we ignore the probability that when school systems punish and drive out boys for the crime of boyishness, while girls thrive in school, we are guaranteeing that men are somewhat more likely to resort to crime, and much more likely to give up on getting any kind of job?

Stop requiring that your employees have college degrees to be hired or receive raises, unless a specific college degree is actually needed. Don't use "college degree" as a cheap way of screening applicants -- it's discriminatory against men who have already been mistreated by the schools.

Instead, go to the trouble of creating tests to show how quickly and well your applicants -- with or without degrees -- are able to learn the kinds of tasks they will actually do if they get the job.

They give typing tests to secretaries, and nobody has any problem preferring the faster, more accurate typist -- regardless of formal education.

And, above all, don't just panic at the sight of a "Yes" after "Have you ever been convicted of a felony?" Take the time and trouble to meet the applicant if he is otherwise acceptable. Find out who he is, not who he was. By the early 30s, falling testosterone levels move many men out of the category of "likely to be involved in violence," so why not give them a chance to show who they are now?

Remember: Mean girls don't get kicked out of school. They get rushed for sororities -- while their male equivalents, having used their fists instead of their mouths, serve time.

Prime-age men who are not working should not be ignored. They are our sons, our brothers, our fellow-citizens -- even if they're likely not to be our husbands, our fathers, or our co-workers.

Their lives are not happy, but they've given up on trying to find a place in our society. Let's open up some places by removing -- or at least lowering -- high walls that they can never get over.


The Most Perfect Thing, by Tim Birkhead, is a well-written and obsessively detailed examination of the science of birds' eggs -- and the sad history of "scientific" egg collecting.

Mostly, the book deals with the eggs of a seabird called guillemot (pronounced GILL-uh-mott), which lays its eggs, without a nest, on ledges on the windswept Bempton Cliffs in Yorkshire. Its eggs are varied in size, shape, and color, and there has been a lot of speculation about how the pear-like shape might have evolved.

The most popular theory has been that the shape makes the egg roll safely in place, so that when the parent bird flies away, the egg won't roll off the ledge. This has even been demonstrated on television, with an "expert" showing how the egg simply rolls in place. But Birkhead debunks this by pointing out that these demonstrations of the eggs spinning in place are always done with empty eggshells.

As soon you try the same demonstration with an egg that contains yolk, albumen (the "white"), and an embryo, everything changes.

Besides, as Birkhead points out, if the egg evolved its shape to prevent rolling off the ledges, why is it so completely ineffective? Because if something loud (like a gunshot) startles all the guillemots on the Bempton Cliffs, you get an avalanche of eggs falling and breaking as the adult birds fling themselves into the sky in a panic.

Another myth about egg shape is that eggs have an elongated shape so that the skinny end can lead the way through its passage out of the mother bird. A nice idea, except that actual science shows that it's the blunt end that exits the hen first.

Some eggs seem to have evolved their shape so that they can lie most efficiently in the nest. With the skinny ends all pointing inward, they make a more compact mass, allowing the brooding parent to cover and warm them all more efficiently.

This has nothing to do with the guillemot egg's shape, though, because they aren't in nests and they're only laid one at a time. As for rolling, the birds know their eggs are going to roll off the ledge -- so when the mother lays the egg, she does a little bit of acrobatic maneuvering to keep it from rolling as she positions herself to keep it warm.

Did you know that most birds that sit on eggs to incubate them develop a "brood patch" during nesting season? This is a region of skin on their underside that loses its feathers and gets a great deal of blood near the surface of the skin, so that this warmer-than-usual patch can rest directly on the eggs without any insulating feathers in between.

And what about that bubble of air inside the egg, that gives hard-boiled eggs a dent in (usually) the blunt end? When the chick is ready to hatch, that's the air it breathes while it's breaking out of the shell. There isn't much air there, but it's enough for a couple of breaths, and then it's up to the chick to get that shell cracked so more air can come in.

What about before the chick is ready to hatch? I had never wondered how embryos are supplied with air inside the egg, but of course they need oxygen just like embryonic and fetal human babies in the womb. Here's an answer I found absolutely amazing. Bird embryos attach to the eggshell in a similar way to the human baby's attachment to the placenta.

The embryo then grows tiny blood vessels that completely line the inside of the eggshell. These pass right next to the pores in the shell -- tiny air tubes that you often can't see with the naked eye. Through these holes in the eggshell's surface, the embryo gets oxygen and discharges other gases, so that the seemingly impermeable shell is actually a working lung, so to speak.

The egg is an active organ of embryonic bird metabolism.

I loved this and many other wonderful facts about birds' eggs. I also loved Birkhead's strict insistence on scientific rigor. He clearly sets out the scientific method:

1. Gather data by close observation, keeping careful, detailed records of everything.

2. Make guesses about why the behavior, processes, or structures you observe are the way they are.

3. Then set out to disprove your hypotheses through rigorous experimentation.

The only "proof" of causality that is ever possible is to consistently fail to disprove your guesses.

He specifically warns of the danger of taking the opposite approach -- trying to find evidence to support your guesses. In the history of egg-and-bird research, he has more than a few examples of false conclusions that were reached by naturalists who tried to "prove" their guesses instead of trying to disprove them.

Of course, there are also examples of naturalists who didn't try to prove or disprove anything -- they just published their guesses as "findings," which were then repeated by others, sometimes for centuries, before actual science was applied to the question and those guesses were shown to be wrong.

Much -- perhaps most -- of The Most Perfect Thing is devoted to the history of egg-collecting in the name of science. The collection of these eggs has been both destructive and nearly useless over the centuries, since many of the most prominent collectors kept no records.

Birkhead has still been able to learn a few things from those long-hoarded eggshells, but it's not as if you can revive the dodo from a fragment of its eggshell.

Do you know what determines the thickness of an eggshell? The weight of the parent bird that is going to sit on it. Eggshells that break under the parent's weight are not going to yield many living offspring.

Yet the young bird must be able to break the eggshell in order to get out of it. There are a few species in which the parents help the babies break free, but in most cases they're on their own. That's why so many birds have an "egg tooth" when they hatch -- a bony growth on the beak that helps them break through. They lose the egg tooth within a few days or weeks, either wearing it off or absorbing it back into the body.

I can't promise every page will be thrilling, because so much of the book is devoted to the antics of rather dull egg collectors. My favorite humans in the book were the Yorkshire "climmers" (climbers), who are lowered over the cliffs to collect eggs. Oh, and the Russian scientist who used his knowledge of birds and their eggs to help a northern Russian city survive a foodless wartime winter.

There are enough fascinating facts, and Birkhead is such a clear and engaging writer, that I was never tempted to stop listening to Gareth Armstrong's excellent reading of The Most Perfect Thing, even during the less-interesting bits.

Now if only we could get the global warming hoaxters to read this book and find out how actual science is done, since all we have from them is the guess that human activity is causing climate change -- and absolutely no effort to test or disprove that guess. Instead, they try to use the force of law, social pressure, and denial of grants and publication in order to foist this guess on the public as if it were "science."

These clowns really need to try actual science. It yields such wonderful results wherever it has been tried.

Of course, they aren't always the results you want, but tough noogies, kids, if you can't deal with disappointment, you shouldn't be in the science biz.


I told you last week that I didn't see how David Baldacci could keep his series about detective Amos Decker series going after Memory Man. The first volume is so deeply involved with the pivotal events in Decker's private life that I worried the later volumes would be formulaic and nowhere near as moving.

My worries were pointless. I've now read the first sequel novel, The Last Mile, in which Decker gets involved in the case of a Texas death row inmate who had once been a hot NFL prospect before he was convicted of murdering his parents. A confession by a death row inmate in another state has set this man free, but Decker, who played against him in college, is determined to find out who really killed the man's parents -- and why they set him up to take the fall for it.

Because we really get to know the ins and outs of the falsely convicted man, this novel is just as personal and surprising and moving as Memory Man. Arguably, because there is no serial killer involved, it's even better.


This Saturday and Sunday, April 8 and 9, there'll be a performance of an Easter oratorio, Lamb of God, retelling the last week of the life of Christ.

Free of charge, it begins at 6:30 p.m. at the meetinghouse of the Church of Jesus Christ of Latter-day Saints at 3619 Pinetop Road.

The music is good, the message is moving, the performers are excellent, and I invite anyone who wants to experience a new musical retelling of the story of Jesus' death and resurrection.

E-mail this page
Copyright © 2017 Hatrack River Enterprises Inc. All rights reserved.
Reproduction in whole or in part without permission is prohibited.