31 March 2011

The Case Against The Label "Borderline Personality Disorder"

There is an established mental health disorder with the name "Borderline Personality Disorder."  This label has its critics.  See also here.

First, the category itself is questionable, because it is so often co-morbid with other conditions, doesn't have a fixed core of symptoms found in every person given that diagnosis, and is often confused with conditions like bipolar disorder, which can present similarly.  It is a muddy diagnosis that may not have a common cause or a common suitable treatment that fudges inadequacies in the diagnostic system.

Arguably, this is a feature and not a flaw.  Nobody in the clinical community is claiming seriously that people who are diagnosed with Borderline Personality Disorder by qualified psychiatrists who are doing their job properly do not have a mental health condition.   Once one is diagnosed with "something," the specific treatment plan may be rather ad hoc in any case, tailored to what works and the individualized presentation in the patient.  Ambiguous muddy diagnoses prevent someone who clearly has something wrong from falling through the cracks because they don't fit in a specific box.

But, the lack of coherence in the category does illustrate that the modern psychiatric medical establishment isn't really sure what is wrong and doesn't really know what to do about it.

Second, the label has pejorative connotations, seems to be applied to women when men with the same symptoms would be given another diagnosis (often "psychopathy" or in DSM-IV language, a particular subtype of "anti-social personality disorder"), may capture clinician bias about a particular patient rather than objective symptoms, and  it isn't really clear what metaphor is being invoked when the "borderline" label is used - borderline between what and what?  Critics argue that the label essentially consists of being both crazy and difficult, rather than merely crazy, or merely difficult.

Now one can argue that any classification that carries as diagnostic criteria more of the folk definitions of "evil" than all but one or two other diagnoses is inevitably going to develop prejorative connotations, just as labels for people with low IQ have to be re-invented every decade or two because they inevitably develop those connotations and cease to have exclusively clinical meaning.

30 March 2011

New Yorker Magazine Still Grammar Challenged

Language log has this ongoing story covered.

Yemen Falls Apart Again

For most of my life, Yemen was two countries, North Yemen and South Yemen.  Then it unified.  Then, Tunsia's revolution spread and its generals decided to back the protesters after a massacre by snipers of men, women and children who were protesting for change, forcing the local dictator to make major concessions.  Now, six of its eighteen provinces are no longer under central government control with some Northern and some Southern provinces operating autonomously.

Republicans For Pollution

Republicans are insisting in budget negotiations on continuing to add to global warming, allowing cement makers to add mercury to the air, keeping the Chesapeake Bay polluted, and allowing coal companies to conduct mountaintop removal mining. 

Who knew that the Tea Party movement was really a mass outpouring of support for more pollution?  They want to shut down the government over these issues?  I don't think that they'll win any political wars in a government shutdown by going to the barricades for utility companies, air pollution and the right of corporations to create Appalachian moonscapes.

Fines As An Alternative To Incarceration

Colorado, like almost all American jurisdictions, makes heavy use of small fines for petty criminal offenses, but includes fines almost as an afterthought, if at all, as the sanction for more serious offenses and has the option to impose many months or years of incarceration as a sentence even in the case of offenses that are usually disposed of only with a fine such as municipal ordinance violations and more serious traffic offenses.

As the Denver Daily News explains:

The fine for using a prostitute would be raised up to $10,000 under a bill that passed out of the Senate yesterday. The minimum fine for prostitution-related offenses would be raised to $5,000. The current fine for prostitution-related offenses can currently be under $100. Additionally, the bill would create “john schools” that first-time offenders could attend to get a deferred judgment. The scared straight-type program would educate first-time offenders on the harrowing effects of prostitution. The bill passed on a 32-2 vote and now goes to the House.

While sentences of incarceration produce big costs for taxpayers, fines raise revenues. Moreover, the maximum sentence of incarceration for an offense is generally much more severe in terms of economic impact than the related fine.

For example, a judge can impose a sentence of eighteen months of imprisonment for a class one misdemeanor, or a fine of $5,000. But, an eighteen month sentence of imprisonment would deprive the person sentenced of eighteen months of income in a state where the median income in a single person household is $52,430 in Colorado in 2010, depriving that median individual of more than $75,000 of income, in addition to considerable lost future income because the incarcerated individual is very likely to be unemployed for a substantial period of time upon release, and economic harm associated with involuntarily breaking their residential or automotive lease or default on their mortgage and car payments, etc. Even for someone earning only a far below median wage and netting out costs of room and board that are provided in jail, eighteen months of incarceration has far more economic impact than a $5,000 fine for a large share of all convicted criminal defendants.

Similarly, a judge can impose a sentence of a year in jail for a class two misdemeanor, but can impose of fine of no more than $1,000, and can impose a sentence of six months in jail for a class three misdemeanor, but can impose a fine of only $750.

The amounts of the fine somewhat understate the situation, because a variety of courts costs and fees apply when an individual is convicted of a crime, and there are also costs for a privately retained criminal defense attorney (if any), and the possibility of a restitution award. But, the general observation holds true and restitution awards are generally far narrower than the compensatory element of a civil tort judgment.

Of course, one of the reasons that we ended up with this kind of statutory framework is that a large share of criminal defendants in the United States are indigent or near indigent (something that can be estimated by the large percentage of criminal defendants who avail themselves of the services of the public defender), and that when imposed together with incarceration, the defendant has little or no means of generating income to pay it. Imposing fines that can never be paid is pointless and makes it harder to reintegrate the defendant into the community after the defendant is released. Indeed, one of the rules of thumb in the law is that legislators and judges and regulators tend to punish offenses committed by defendants without the financial wherewithal to make those they harm whole with incarceration and tend to punish offenses committed by defendants with an ability to pay with civil judgments and fines.

But, not all criminal defendants are indigent. One of the reasons that a large fine for people who purchase the services of prostitutes is attractive to legislators is that this is a class of criminal defendants who often do have the financial means to pay these fines. Many traffic, vice and white collar criminal defendants, generally, have a substantial ability to pay criminal fines. And, some minority of everyday crimes, especially misdemeanors and less serious felonies, are likewise committed by individuals with an ability to pay substantial fines for whom a sentence of incarceration would be an intense economic burden.  Large fines also make the option of prosecuting criminal cases against corporate defendants more attractive.

The predominant punishment in Germany and many Scandinavian countries for what would be serious misdemeanors or minor felonies in the United States is a "day fine" equal to one day's average income in some recent time period for the defendant times the number of days of fines imposed, in lieu of a maximum term of incarceration of the same length. Defaults result in imprisonment for the number of day-fine days of fines not paid. This keeps the defendant in the community, employed, with an incentive to work more to pay off the fine and shed the burden sooner, and without imposing significant administration costs on the public and without disproportionately punishing low income defendants. Part of what makes that system work is the fact t hat these countries don't have the kind of "underclass" of persistently poor people who regularly get caught up in their criminal justice system that we do in the United States. But, there is still surely a class of offenses in the United States for which a day-fine would be a useful criminal justice option.

This isn't to say that I support this Johns bill just passed by the State Senate. The policy case for criminalizing prostitution between consenting adults at all is weak, and many our world peers tolerate some form of legalized prostitution. In my view, taxing and regulating vice is generally preferrable to using the criminal law to prohibit it and trying to enforce that law.

But, while I don't support increasing penalties for adult prostitution, period, the approach of imposing a high fine for an offense in a situation where there is a realistic possibility that criminal defendants will be able to pay it, rather than incarceration, is an option that has much wider applicability.

Similarly, in cases where the existing fines are the predominant form of punishment for a crime and are adequate to the task, for example, for many traffic offenses that carry criminal penalties, ordinance violations, petty offenses and minor misdemeanors, removing the possibility of a sentence of incarceration would be desirable, because it would lower the stakes of proceeding, reduce the possibility of abuses of judicial discretion in sentencing, elminate the constitutional requirement that counsel be provided at public expense in the proceeding, and have little impact on the day to day outcomes of the criminal justice system in these areas. Where an offense is such that arrest power and brief periods of incarceration are helpful in enforcing the law (e.g. public drunkeness or disturbing the peace), a maximum period of incarceration of a few days, as opposed to many months, might be attached to a significant fine.

Intensive Supervision Adult Probation in Colorado

A Denver Post story last Sunday noted that ten adult men on probation in Colorado have been charged with murder or attempted murder in the last nine months. Five of them were in adult intensive supervision probation, one was in sex offender intensive supervision probation, three were on ordinary adult probation (one for burglary, one for gang related burglary, and one for domestic violence and in each case producing an attempted murder rather than a successful one). Many of them were in violation of prrobation conditions but faced no consequences for their violations. The probation department is at 90% of authorized strength.

because they had been recognized as relatively high risk (and at least one of the ten who was not on intensive probation was guilty of attempted murder, but was not successful in completing his crime).

Is the program worth it? As their editorial today notes, this program is much less expensive than incarceration. But, the highest risk individuals on probation may belong in prison.

Felonies By Probationers

From a public safety perspective, the biggest concern is that probationers commit new felonies, particuarly serious new felonies. How common is this?

The number of felonies in 2010 by program participants divided by the program size is as follows:

Intensive Juvenile 44/452 (9.7%)
Intensive Adult 79/1,408 (5.6%)
Intensive Female 10/262 (3.8%)
Juvenile (ordinary) 135/5,946 (2.3%)
Intensive Sex Offender 22/1,301 (1.7%)
Adult (ordinary) 570/41,107 (1.4%)
Private Non-Drunk Driving 65/9,067 (0.7%)
State Monitored Drunk Driving 41/11,448 (0.4%)
Private Drunk Driving 34/14,126 (0.2%)

Total: 1,000/85,142 (1.2%)

People on probation commit about 3% of all felonies in the State of Colorado and make up about 1.7% of the state's population including children. Thus, overall, probationers (who are much more likely to be adults and much more likely to be non-elderly males) aren't particularly high risk as a whole compared to demographically similar non-probationers, particularly when the offense by high risk probationers are removed from the total. But, a subset of probationers, adult men in intensive supervision probation who make up about 3% of the total probation caseload, do appear to pose a considerably high risk to the general public.

Offenders who commit felonies while on probation are subject to sentences of at least the midpoint of the presumption range for the offense, and up to twice as much as the presumptive maximum for that offense, Section 18-1.3-401(8)(a)(III), Colorado Revised Statutes, in addition to having their probation revoked and facing a sentence of incarceration on the original crime. So, these offenses are committed despite an enhanced deterrent incentive not to commit them relative to an ordinary individual.

All of the murders and many of the attempted murders were committed by adults in the intensive adult and intensive sex offender program (6 out of 101 felonies compared to 4 out of the 570 felonies committed by ordinary adult probationers), suggesting that in addition to having a much higher felony termination rate than other probationers, that these offenders also commit, on average, more serious felonies than other probationers who have their probation revoked for felonies.

Policy Implications

This may mean that these high risk individuals simply need to receive even more intensive supervision than they already receive, or it may mean that they would be better incarcerated. Given the very high likelihood that an incarcerated felon will commit a new felony after being released, it isn't obvious that the public is safer with incarceration than it is with probation in intensive supervision cases. But, it may very well be that the public would be better protected by incarceration in this subset of cases.

More generally, if the reason to prefer probation to incarceration is that it reduces the risk that a convicted felon is less likely to commit a felony that harms a member of the general public, and to reduce the severity of the felonies that are committed by such people, it may make sense to put more of the high risk men who receive intensive supervision when on probation in prison, and to allow more low risk individuals to receive probation sentences.

In particular, many women, white collar offenders, and vice offenders in our prisons who aren't part of criminal gangs or organized crime enterprises may pose a lower risk to the general public than some of the more serious male "blue collar crime" offenders who receive probation in the current system. Even if the lower risk offenders do commit new felonies, these offenders are unlikely to commit the violent felonies that the public is most concerned about preventing.

It is also worth examining whether probation is too often imposed for minor offenses where the benefits resulting from the supervision element of the sentence is slight, because probation officers are spread so thin for low risk offenders. For example, it isn't obvious that the system has sufficient resources to adequately sanction the large number of offenders who abscond or commit technical violations in a consistent manner, or that even the minor probationers who do abscond or commit technical violations really do present a serious threat.

Perhaps the state needs an option in many cases that produce probation sentences today of imposing a significant fine and a sentence of "unsupervised probation" as all or part of the probation period for minor offenders, which is revoked only for new felonies or misdemeanors or non-payment of fines, which qualifies the individual for an enhanced sentence on the new offense and a sentence of incarceration on the old offense, if a new offense is committed within a certain time period, in addition to a fine, while not actually having the relatively elaborate conditions of probation with many opportunities for technical violations and absconding through mere flakiness, and the supervision costs, that are present in the typical probatioon case.

How common is probation?

At the end of the 2010 fiscal year, there were 3,423 people in an intensive supervision probation program in Colorado's state courts: 1,408 in the program for adult men, 1,301 in the program for sex offenders, 262 in a program for women, and 452 in a program for juveniles.

Other kinds of probation are much more common. There were 41,107 adults on regular probation, 5,946 on regular juvenile probation, 11,448 on state monitored probation for drunk driving offenses, 9,067 on private probation for non-drunk driving offenses, and 14,126 on private probation for drunk driving offenses.

The 41,107 adults on regular probation were further broken down as: 901 new probationers awaiting risk classification, 4,166 maximum risk, 10,173 medium risk, 8,911 miniumum risk, 5,758 adminstrative, 3,328 community corrections, 4,875 domestic violence, 1,137 sex offender, and 1,858 interstate transfers.

The 5,946 juveniles on regular probation were further broken down as: 176 new probationers awaiting risk classification, 920 (15%) maximum risk, 1,930 (32%) medium risk, 1,685 (28%) miniumum risk, 779 (13%) administrative, 20 in community corrections, 342 sex offenders, and 94 interstate transfers.

In all, there were 85,142 people on probation in Colorado on June 30, 2010. Probation includes community corrections.

Probation is a common sanction following a criminal conviction, indeed, it is the modal punishment for many offenses. There were 10,460 new felony probation sentences, 14,851 new misdemeanor probation sentences, 228 new petty misdemeanor probation sentences, 103 new traffic probation sentences, and 694 other new probation sentences (ordinance violations or lack of coding information) in the 2010 fiscal year. In all 26,336 new probation sentences were imposed in the 2010 fiscal year. Of completed probation sentences, 38% were for a year or less, 36% were for one to two years, 15% were for two to three years, and 11% were for more than three years.

There were 36,993 new felony offense prosecutions in 2010 in Colorado, 11,640 new juvenile delinquency prosecutions,, 69,695 new misdemeanor prosecutions (excluding traffic cases and drunk driving), and 28,429 drunk driving prosecutions.

How Often Is Probation Successfully Completed And Why Do Offenders Fail?

Of all people whose probation ended in 2010, 67% of adult regular probation sentences were completed successfully (including the 11% of the total who complete a community corrections sentence successfully), 15% of those on probation had their probation revoked for technical violation or a new crime while on probation, 13% absconded ("Absconded refers to probationers who became fugitives and are no longer compliant with probation supervision."), 3% were deported, 1% died, and 1% of cases were cloused out for administrative reasons. Of the 3,144 adult probation revocations in 2010, 570 (18%) were for new felonies, 756 (24%) were for new misdemeanors, and 1,818 (58%) were for technical violations of probation terms.

Of the 5,906 non-drunk driving private probation sentences terminated each year (a lower risk population), 4,590 are successful (78%), 614 are revoked (10%) (65 new felonies (11%), 146 new misdemeanors (24%), 66% technical violations) and 702 abscond (12%). For private probation related to drunk driving offenses, 82% complete their sentence successfully, 9% have their probation revoked (34 new felonies (4%), 288 new misdemeanors (31%), and 65% technical violations), and 9% abscond.

Among state monitored drunk driving probation cases, 73% complete successfully, 16% have their probation revoked (1,411) (41 which is 4% for new felonies, 136 which is 10% for new misdemeanors, and 87% for technical violations), and 11% abscond.

For regular juvenile probation, 73% successfully complete their sentences, 22% have probation revoked, and 5% abscond. Of the 989 juvenile revocations, 135 (14%) are for a new felony, 183 (18%) are for a new misdemeanor, and the remaining 68% are for technical violations.

In the adult intensive supervision program, 66% are successful terminations, 26% (317) are revoked, and 9% (108) abscond. Of the 317 revocations, 79 (25%) are for new felonies, 51 (16%) are for new misdemeanors, and 59% are for technical violations.

For sex offenders in intensive supervision, there are 22 new felonies (12% of revocations), 8 new misdemeanors (4% of revocations), and 158 revocations for technical violations (84%). Only 39% of sex offenders in intensive supervision successfully complete the program, while 53% have their probation revoked and 8% abscond.

In the female intensive supervision program, 69% of cases are successful, 22% (32) are revoked (10 which is 31% for new felonies, 3 which is 9% for new misdemeanors and 59% for technical violations), and 8% abscond.

For juveniles under intensive supervision, 46% are successful, 49% (232) have probation revoked, and 5% abscond. 44 juvenile intensive supervision revocations (19%) are for new felonies, 48 (21%) are for new misdemeanors and 60% are for technical violations.

After probation terms are completed there is considerable recidivism, although not nearly as much as that for offenders who complete prison terms.

Miscellaneous Background

When you are convicted of a crime in Colorado, a judge can fine you, a judge can sentence you to a period of incarceration, or a judge can put you on probation subject to a wide variety of conditions (there are a few other options as well, like community corrections and deferred judgments).

There are three main classes of people who are out in the community while involved in the criminal justice process. People on bond awaiting conviction, people on probation, and people on parole. The first two are supervised by the judicial branch, the last is supervised by the executive branch's parole system. (Incarceration prior to trial or after conviction for a misdemeanor is handled by the executive branch of county government, while incarceration after conviction for a felony is handled by the executive branch of state government together with parole). The probation department also does pre-sentence investigation of criminal defendants.

Investment Linked To Unemployment

Investment as a share of GDP empirically shows astrong, inverse statistical relationship with unemployment. Unemployment falls when investment is high, and rises when investment is low (or visa versa).

29 March 2011

Childhood Psychological Tendencies Persistent

Democratic socialists, including those in Britain, are much better at conducting very large scale, multi-purpose longitudinal studies than Americans, and those studies are starting to bear fruit. The British National Child Development Study tracks 17,634 children born in March 1958 for fifty years, checking in on the subject's progress at nine regular intervals from birth to the present and gathering detailed and accurate informaton each time from multiple sources in multiple institutions. What did they find?

In this article we assess and compare long-term adult socioeconomic status impacts from having experienced psychological and physical health problems in childhood. . . . Large effects are found due to childhood psychological problems on the ability of affected children to work and earn as adults and on intergenerational and within-generation social mobility. Adult family incomes are reduced by 28% by age 50, with sustained impacts on labor supply, marriage stability, and the conscientiousness and agreeableness components of the “Big Five” personality traits. Effects of psychological health disorders during childhood are far more important over a lifetime than physical health problems.

From Alissa Goodman, Robert Joyce, and James P. Smith, "The long shadow cast by childhood physical and mental problems on adult life", PNSA 2010.

The abstract poses the question as one of a particular childhood "problem" having a long term effect, but the results are equally consistent with the notion that psychological health has a large congential component, with much of the rest of the environmental component been based on the quality of one's early childhood environment, rather than later environment. In this view, the "problems" that a child experiences are simply manifestations of who a child is that persist throughout life rather than being the cause of later in life difficulties. In our mental health and IQ focused world, in which capacity to labor physically is largely unimportant socio-economically, it also isn't surprising that fitness that is physical, or the lack of that fitness is comparatively unimportant.

Alas, the early childhood window may be surprisingly narrow, and social class effects tend to be much more powerful than instructional effects. For example, according to another very large longitudinal study (n=13,776), the beneficial effects of full day kindergarten programs relative to partial day kindergarten programs in low income children tend to vanish by the third grade:

[T]he study found that the reading and math skills of children in full-day kindergarten grew faster from the fall to the spring of their kindergarten year, compared to the academic skills of children in part-day kindergarten.

However, the study also found that the full-day kindergarteners' gains in reading and math did not last far beyond the kindergarten year. In fact, from the spring of their kindergarten year through fifth grade, the academic skills of children in part-day kindergarten grew faster than those of children in full-day kindergarten, with the advantage of full-day versus part-day programs fading by the spring of third grade. The fade-out can be explained, in part, by the fact that the children in part-day kindergarten were less poor and had more stimulating home environments than those in full-day programs[.]

Social class linked intellectual achievement gaps are already great at age two.


The good news, for the parents of older children, I suppose, is that whatever good or bad your parenting has had on their lives, the older your children get, the less the quality of your parenting seems to matter. So, the pressure isn't as intense. A child's academic progress and character as a late tween is immensely predictive of the rest of the course of that child's life.

Then again, this may simply be a product of the fact that whatever parenting and educational quality was present early on continues to be present and having the same effect, with interruptions of these patterns being too rare to be statistically significant. Unless you are comparing kids who experience late childhood misfortune against their otherwise equal pre-misfortunte peers, you can really accurately measure the impact of those misfortunes on a child's life chances.

On the other hand, these studies are really terribly discouraging for those whose job it is to try to get people on track in life once their early childhood years have passed. The school system, for example, frequently becomes involved in a child's life only starting in kindergarten at age five or six years old, and the second study seems to suggest that it is already extremely hard to help kids who are developmentally behind at that point catch up to their more academically ready peers.

Needless to say, if your job is to turn around the lives of kids who have already started high school, fifteen years of accumulated inertia is just that much harder to reverse. If your job is to turn around the lives of adult prison inmates, the burden you must overcome is surely even more profoundly difficult to surmount.

As previous posts at this blog have noted, you may be born with a gene which is a powerful predicter of your capacity to sustain a long term marriage, and people who commit violent crimes as adults overwhelmingly are already continuing to have discipline issues that their peers have grown out of by the time that they are in elementary school: "it is extremely unlikely that an adolescent who has not been highly physically aggressive in the past will suddenly manifest significant problems with physical aggression." Such seemingly learned and social traits as succeptibility to bullying and one's likelihood of being a teenaged crime victim likewise show powerful genetic predispositions. The non-IQ factors that play an important part in life success seem to manifest just as early in life as academic ability does: Your ability to defer gratification at age four turns out to be a powerful discriminator for your life success a decade and a half later. Children likely to grow up to be psychopaths already start acting differently at age three (also here). A group of three non-IQ related genes can account to up to a 0.5 point difference in your GPA on top of any IQ effect.

We may live in the post-nature v. nuture synthesis of genes x enviroment, but the more we know, the more predictable our lives seem to be. For us, like characters in Greek tragedies, it seems that knowing our fates offers us surprisingly little help in changing our fates. Our genes are seemingly immutable, and most mental health genetic conditions are incurable, even if they are treatable in some cases. Many psychological predispositions have no established treatment. Even if you know you have a gene that makes you likely to have short lived relationships, there is no drug you can take to fix that and no solidly evidence based therapy that will solve your predisposition. Gene therapy hasn't reached that point yet, and given the implications it would have for personal identity and neurodiversity in society as a whole, perhaps this is a good thing.

Our enviroments, while theoretically mutable, aren't easy to change. And they seem to matter. People in the top fifth of the income distribution are twenty-eight times as likely to attend a selective college as those in the bottom fifth of the income distribution, for example. We don't choose our parents, and it seems as if those of us who are raised by someone other than our parents are usually worse off in the bargain - although perhaps starting off in a spot like that leaves one off poorly even with great improvement. And, of course, unlike J.K. Rowling or Eminem, most of us who find ourselves in poverty don't readily emerge from it to great riches by virtue of our intellectual talents. Parents who know that their poverty and lack of education is dragging down their children have little capacity to do anything about it. Even the heart wrenching option once available of sending tween or teen children into domestic service for a more prosperous family, or a scholarship funded tenure at a boarding school, in hopes that some of the benefits of that social class will rub off on them has largely vanished as options. Assortive marriage trends are reducing the availability of marriage as a means of class advancement, and American social class lines are deeper than those in contemporary Europe or earlier eras of American history. Just about the only thing a parent can do to improve the environment that the parent provides his or her children is to transform him or herself, which is a profoundly difficult matter.

Anecdotal evidence suggests that people do indeed change sometimes. But, the overwhelming weight of the statistics and studies tell us that this is the exception and not the norm, and becomes ever more exceptional as we get older. Kids who are well behind grade level in reading in the third grade are almost certainly not going to go to Harvard, are exceedingly unlikely to go to medical school or law school, and will have to struggle mightily and have considerable good fortune to earn a college degree of any kind. Kids who are perfect angels in fifth grade are exceedingly unlikely to have criminal justice problems or violence issues in the remainder of their lives unless they have a family history of a late manifesting mental health condition like bipolar disorder, or schizophrenia, or substance abuse, or experience a brain injury, or sustain PTSD.

I don't want the world to be this way. I'm a liberal. I believe in progress and opportunity and social mobility. I value education. Environment does matter, especially for the less well off, even if they can't change that environment themselves.

But, one can't simply ignore what study after study after study tells you either. One can look for loopholes and exceptions and ways to beat the system. There clearly are some and they should be leveraged for all that they are worth. The more I read these studies, the more that I am forced to conclude that policies designed to secure social justice have to focus more on developing meaningful productive paths for the people who we have in our society, and less on trying to transform people into something that they are not and cannot reasonably aspire to become -- especially for older youths and adults for whom the die is more firmly cast.

28 March 2011

Fukishima Risk

Japan did some good things to respond to the Fukishima reactor disaster. Most importantly, they evacuated people close to the reactor promptly. Radiation intensity declines as a function of distance squared from the source. Harm from radiation is roughly proportional to duration time intensity. Thus, the sooner you evacuate people who are closest to the scene, the more you mitigate harm.

At 50 km away from the source, the intensity is 1% of intensity at 5 km away, which is in turn 1% of the intensity at 500 meters away. The intensity at 500km away is about one millionth of the intesity at 500 meters away. The intensity at 5000 km away (the continental U.S. perhaps) is about one hundred millionth as great.

How Great Was The Exposure?

The risk at 50 km away in this case for twenty-four hours on the peak days (March 16/17), 3.6 microSv was at an intensity that was exceed unsafe levels if sustained for a week or two, and about three times as much as the highest exposure of anyone outside the plant received for the entire duration of the Three Mile Island incident and about the same as a half an hour at a typical spot at the Chernobyl plant in 2010. The risk at 300 km away would be safe almost indefinitely.

This intensity of exposure hasn't been sustained, but radiation levels have been elevated.

The exposure at 5km away for the same twenty-four hours (360 microSV), is close to the amount necessary to induce immediate radiation poisoning, would be about seven times safe levels for an entire year in that day alone, is about three hundred times as much as the highest exposure of anyone outside the plant received for the entire duration of the Three Mile Island incident, and is about the same as fifty hours at a typical spot at the Chernobyl plant in 2010.

At 500 meters away from the source in that time period, a four and a half hour exposure is almost inevitably fatal, and thirteen and a half minutes would be enough to induce immediate radiation sickness. At 50 meters away from the source in that time period, a three minute exposure is almost inevitably fatal and eight seconds would be enough to induce immediate radiation sickness. At 5 meters away from the source in that time period, a two second exposure is almost inevitably fatal.

Some plant employees and accident response workers may face deadly doses of radiation (and they were probably acutely aware of that risk and heroically did their jobs anyway), but the general population of Japan isn't at nearly that risk.

Also in the good news department is that the quite low levels of radioactivity found at greater distances from the plant have been mostly in form of radioactive iodine that has a quite rapid half life and won't be a continuing source of high level radioactivity exposure.


The Fukishima accident is still a big deal. It is at least the second worst civilian nuclear power plant disaster in the history of the world. It is far worse that Three Mile Island. Time will tell if this is worse or not as bad as Chernobyl, although it seems likely that it will be not as bad, and a swift response should help as well to mitigate the impact. On the other hand, the harm can only be mitigated if a sustainable way to limit ongoing radiation exposure can be found fairly swiftly. Progress is being made on this front, but it isn't clear that a sustainable solution has been found yet.

While low level nuclear waste rapidly decays to non-dangerous radioactivity levels, high level nuclear waste from the core of reactors continues to be dangerous for much longer time periods.

Even if a solution is found to stop intense radiation from escaping the plant, unless it can be almost perfectly contained, there is a real likelihood that the plant will have to be abandoned and that some radius around them will have to be put off limits as a hot zone. The proximity of the plant to the ocean (it is about 200 meters from the ocean), while good for short term cooling resources, also makes keeping radioactivty materials confined to the plant in the long term harder.

The evacuation has disrupted the lives of hundreds of thousand of people for weeks, the disaster will probably lead to permanent loss of large amounts of real estate and tangible person property for tens of thousands of people (in excess of the losses directly caused by the tsunami and earthquake) probably on the order of a billion dollars worth, will probably kill tens of people from radiation exposure, will probably cause something on the order of a billion dollars of damage to the plant, and will seriously inconvenience hundreds of millions of people. The harm is on the order of magnitude of several billions of dollars, or even tens of billions of dollars depending on remediation costs, etc.

This may still be an acceptable cost if very infrequent compared to coal that kills 30,000 people a year in the United States alone (about the same as the entire toll of the tsunami and earthquake scaled to Japan's population), but it isn't nominal either.

Houston Looks To Drunk Tank To Save Money

Opening a sobering center — a drunk tank where the 19,000 people picked up for public intoxication in the city every year could be taken in lieu of jail — could save Houston money and help connect chronic alcoholics with the assistance they need, according to police, mental health advocates and a City Council member.

While there is no formal proposal, proponents envision a center staffed by substance abuse professionals who could counsel alcoholics and connect them with the housing, treatment and other services.

"It's a more humane way of dealing with alcoholics," said Houston police Lt. Mike Lee. He envisions a scenario in which those arrested for public intoxication would spend several hours sobering up, after which they would be free to go with no arrest record, no court appearances and no extended stay requiring meals, medical attention and officer supervision.

From here.

While Texas is famous for the abuses of its criminal justice system, it has also produced a number of sensible ideas for reforms, such as this one.

The Clan of the Hippo Hunters

A trip to the Denver Zoo this weekend brought to my attention an underused tool for making sense of pre-history, along with population genetics, linguistic family relationships, archaeological evidence, the population genetics of domesticated plants and animals, cultural similarities, and ancient climate evidence. This is ancient wild plant and animal ranges.

Hippos and Rhinos

My attention was particularly focused on the historical range of the hippo and the rhino. In about 1942, both had a range that can be roughly summed up as the Sahel, East Africa, and Southern Africa to the south of the Congo Basin jungle and (at least for the hippo) the Nile River Basin. In the fifty years since then, the range of the rhino has contracted dramatically to scattered islands of wild populations, although the range of the hippo has contracted much less. Pygmy hippos are also found in the Congo basin. All of the non-pygmy hunter-gatherer societies of early Africa would have co-existed with the ancient rhinoceros.

I also found particularly interesting the ancient range of the hippo, which extended beyond Africa to essentially that part of Europe which was part of the Cardium Pottery Neolithic (about 14,000+ years after the mainland hippo went extinct) as well as additional swaths of Greece, Anatolia, Southern Persia, and the Western Coast of India right down to its Southern tip, although apparently, not into Eastern India. European hippos mostly went extinct before the last glacial maximum ca. 20,000 years ago, quite possibly as a result of modern human over hunting. But, dwarf hippos were found in Crete, Cyprus, Malta and Sicily until later dates, with the Cyprus Dwarf Hippopotamus surviving until the end of the Pleistocene or early Holocene (about ten thousand years after mainland European hippos went extinct).

It is customary to visualize early hunter-gatherer populations is mammoth hunters. But, at first, in addition to ancient members of the elephant family, there were also ancient relatives of hippos and rhinos for early hunter-gatherers to hunt. The woolly rhinoceros, for example, survived through about 8,000 B.C.E. in Western Siberia and had a range that reached as far east as Southern England in the Paleolithic. An Elasmotherium, which was an ancient rhinoceros that was the largest land animal to ever live, was dragged into a pair of caves in Southern Siberia approximately 50,000 years ago by some predator, human or otherwise.

It is easy to fall into the trap of thinking of the ancient world as similar to that of the current one and draw conclusions about human pre-history accordingly, but one has to remember that in the Out of Africa era that much of the territory that modern humans would have encountered was teaming with a wide variety of giant wild animals not adapted to human hunting techniques that would have been abundant sources of meat (although the giant predators who could have been a threat to early modern humans were more abundant as well).

For them, megafauna may have been a bit like oil is today, a resource that when used unsustainable would have eased human transition into a new technological era and wider range until exhausted.


The ancient range of the Tapir was also interesting. This pig-like relative of the hippo has New World varieties, found in Latin America, and is also found in Malaysia, probably because it arose prior to split of the Americas from the Old World due to continental drift. Its ancient range includes not only Southeast Asia (including all of the Sunda area that extends into Indonesia) and continues into India in a territory roughly conterminous with the region where Munda languages are spoken. The span is notable because it crosses the boundary between India and Burma has been a major barrier to East to West gene flow, roughly marking the divided between West Eurasian population genetics and East Eurasian population genetics, just as the Munda languages are an example of the Austroasiatic languages crossing this boundary.

The implication is that this apparent stark population genetic divide, rather than being a product of a natural geographic barriers (like the divide between Melanesia and Southeast Asia, the divide between Southern Europe and North Africa, the divide between North Africa and Subsaharan Africa, or the divide between South Asia and Siberia) may be of fairly recent origins, perhaps dating to the marginalization of Munda language populations during the expansion of Dravidian language speaking populations in the last few thousand years. Prior to that time, India might have had a more East Eurasian biased population genetic mix.

Link List Revised

I've added a few links and removed a few from my link list. Those removed are either because despite the excellence of the blog involved, I simply don't write much about that topic (e.g. Coyote Gulch with his comprehensive coverage of Colorado water law), because the blog has not been updated lately (George in Denver), or because the blog is no longer aimed primarily at a general audience (e.g. Evil Mommy). The links added are ones that I have found myself looking at a on a regular basis lately: Dienekes' Anthropology Blog, Language Log, Neuroskeptic and Resonaances (from which I have omitted the accent over the first "e" because this causes the system not to accept it in the link list).

25 March 2011

Homo Erectus Era In India Dated UPDATED

Current research in India therefore indicates that the Acheulean industry ranges from 1.5 million years ago to 120,000 years[.]

From here.

Generally speaking, Archeulean industry (which basically amounts to a certain kind of ancient stone tools), is associated with archaic hominins and most prominently, with Homo Erectus. It follows "the more primitive Oldowan technology some 1.8 million years ago" associated with Homo habilis, and is found in a period often called the Lower Paleolithic.

Acheulean tools were not made by fully modern humans that is, Homo sapiens although the early or non-modern (transitional) Homo sapiens idaltu did use Late Acheulean tools as did proto-Neanderthal species. Most notably however it is Homo ergaster (sometimes called early Homo erectus), whose assemblages are almost exclusively Acheulean, who used the technique. Later, the related species Homo heidelbergensis also used it extensively.

The 1.5 million years ago date suggests that Homo Erectus, or a similarly sophisticated hominin arrived in India within 300,000 years after this species of hominin evolved in Africa.

If the dating is upheld, the implication is that ancestors, such as Homo erectus, reached India at an early stage. . . . we opine that these Late Acheulean industries were probably made by an archaic, but somewhat bigger brained ancestor, such as Homo heidelbergensis.

While there is evidence of overlap between archaic hominin populations and modern humans, and even genetic evidence of admixture between Neanderthals and modern humans and between Denisovians and modern humans, the evidence also seems to suggest in the Neanderthal case that the period over overlap was not very long, probably not more than a few thousand years, or perhaps ten thousands years at the outside with a moving boundary in which the hominin species overlapped. There is also no good reason to think that overlaps of different human species, the Denisovians perhaps in Southeast Asia, or Homo floresiensis found in Flores about 20,000 years ago, was much longer. Likewise, there is no evidence of a long period of overlap between Neanderthals and the archaic hominins that preceded them in Europe.

The implication is that Neanderthals and/or modern humans first took hold in South Asia and replaced the hominins associated with Archeulean industry at that point in time sometime in the vicinity of 120,000 to 140,000 years ago, give or take ten or twenty thousand years. This provides an independent line of evidence to support an oldest possible date for an Out of Africa event for modern humans. The presence of Archeulean industry in South Asia 140,000 years ago makes the presence of modern humans in India or anywhere to the East of this point on the coastal route more than 150,000 years ago.

In fact, we don't see that. While we see Java man which might be Homo Eretus or something similar about 1.9 million years ago in Indonesia, the oldest skeletons claimed to be modern human anywhere outside of Africa are about 100,000 years old (including skeletons in China and the Levant), and there are direct signs of modern humans in India around 75,000 years ago. This coincides reasonably closely with the timing of the demise of Archeulean industry in South Asia. Indeed, at this point, it isn't even entirely clear that there was an overlap from the archaeology, although it is hard to see what could have wiped out archaic hominins in Asia other than a new species of hominin that replaced them in their ecological niche.

In Europe, the Archeulean industry was followed by the Mousterian industry of the Middle Stone Age associated mostly, but not entirely, with Neanderthals:

Mousterian tools that have been found in Europe were made by Neanderthals and date from between 300,000 BP and 30,000 BP (from Layer 2A dated 330 ± 5 ka, (OIS) 9 at Pradayrol, France). In Northern Africa and the Near East they were also produced by anatomically modern humans. In the Levant for example, assemblages produced by Neanderthals are indistinguishable from those produced by Qafzeh type modern humans. It may be an example of acculturation of modern humans by Neanderthals because the culture after 130,000 years reached the Levant from Europe (the first Mousterian industry appears there 200,000 BP) and the modern Qafzeh type humans appear in the Levant another 100,000 years later. It was superseded by the Châtelperronian industry around 35,000-29,000 BP.

One of the oldest stone industries associated with modern humans is the Aterian from about 82,000 years ago "in the region around the Atlas Mountains and the northern Sahara. The industry was probably created by modern humans (Homo sapiens), albeit of an early type, as shown by the few skeletal remains known so far from sites on the Moroccan Atlantic coast extending to Egypt."

There is no evidence, however, that the Neanderthal range range reached as far as South Asia:

It appears incorrect, based on present research and known fossil finds, to refer to any fossil outside Europe or Western and Central Asia as a true Neanderthal. True Neanderthals had a known range that possibly extended as far east as the Altai Mountains, but not farther to the east or south, and apparently not into Africa. At any rate, in Africa the land immediately south of the Neanderthal range was possessed by "modern" H. sap., since at least 160,000 years before the present.

The Shanidar site in Iran (also here) is one of the most easternmost Neanderthal sites yet discovered, although this could be simply a product of preservation conditions. A 2010 academic paper argues that there is "evidence of human occupation in Lower Sindh (Pakistan) during the Middle Pleistocene, which is demonstrated by the recovery of chipped stone assemblages with evident Levallois characteristics." This could have been either early modern humans or Neanderthals, and is the part of India most closely adjacent to the known Neanderthal range.

According to the few absolute dates so far available, Middle Palaeolithic complexes are represented in the region since roughly 150 ky, while the Late (Upper) Palaeolithic ones make their appearance at least just after 40 ky from the present, although the dispersal of modern individuals, following a coastal route, is suggested to have taken place some 10 ky before. The problem related to the makers of the Middle Palaeolithic tools is still debated, mainly because of the absence of fossil human remains of this period in the entire Subcontinent.

One of the most important issues consists of the south-easternmost spread of the Neanderthal Levalloisian assemblages that is so far badly defined. Although typical Levalloisian Mousterian industries are known from south-eastern Arabia, Iran, Afghanistan, and former Soviet Central Asia, characteristic Levalloisian assemblages are almost unknown in the Indian Subcontinent, except for a few surface sites in Lower Sindh and the Indus Valley, which have been discussed in a recent paper. Furthermore the more recent studies seem to support the impression that “the early Middle Palaeolithic (or Middle Stone Age) of India and Nepal probably developed indigenously”, which suggests the existence of a distinctive boundary between the west and the east marked by the axis of Indus river valley.

Thus, the archaeology simultaneous puts what might very well have been a Neanderthal presence in the Indus River Valley and seems to argue against their presence elsewhere in South Asia. Very crudely speaking, there were probably Neanderthals at some point in Pakistan, but not in India, Bangladesh, or Nepal.

The implication seems to be that in most of India the first modern humans would have encountered Homo heidelbergensis, or some similar species of hominin, when they first arrived, and that this South Asian hominin was sufficiently successful to prevent Neanderthals from encroaching upon their range.

This matters a lot as we try to decipher the meaning of the fact that modern humans have an average of about 1-4% Neanderthal ancestry, as determined by comparing ancient Neanderthal DNA with modern Eurasian DNA and identifying genes found in Eurasians and Neanderthals but not in Africans as Neanderthal source genes. The plot thickens as we learn that Melanesians apparently have about 4% Denisovian genes (based on a similar comparison with ancient DNA found in a Central Asian cave) on top of their Neanderthal inheritance, and that the Neanderthal genes found in Chinese people are mostly different ones than the Neanderthal genes found in Europeans.

Of course, a wild card is that some of what appear to be Neanderthal genes in some or all modern humans may be genes from some other kind of archaic hominin that shared those genes with Neanderthals. Indeed, it isn't even clear if the distinct pattern of Neanderthal genes in China and Europe respectively are due to founder effects after admiture while part of a common population, or separate admixture events.

Still, these facts, taken together, greatly narrow the scenarios in which Neanderthal admixture which seems clear to have taken place, happened. Let's review them, all of which seem to be a pretty solid footing:

* Neanderthals never had a range further east than Pakistan.
* Neanderthals never made it into Africa.
* Neanderthals were present in the Levant and Europe and Persia and Central Asia when modern human proto-Eurasians left Africa starting around 100,000 years ago.
* Neanderthals went extinct 22,000 to 30,000 years ago (the younger date is the most recent Mousterian find, the older date is the youngest skeletal remains).
* Modern human started to migrate along the Southern Route around 100,000 years ago or so.
* The Chinese have their roots overwhelmingly peoples who took a Southern route out of Africa, as the genetic evidence overwhelmingly indicates, although this may have involved multiple waves of migration.
* Modern humans did not reach Europe any earlier than about 50,000 years ago, and redating of the oldest modern human remains in Europe suggest that it could have been less than 30,000 years ago.
* Modern humans appear to have been absent from the Levant from 75,000 years ago to 50,000 years ago.
* Papuans and Australian aboriginal ancestors reached their respective lands around 45,000 to 50,000 years ago and had about 4% Neanderthal admixture at that point.
* A significant portion of modern European genetics has its sources in waves of migration to Europe from outside Europe after the Last Glacial Maximum ca. 20,000 years ago, after the Neanderthals were extinct.

The overlap between Neanderthal range and the Southern route of migration out of Africa is pretty narrow, fixing a time and place where Neanderthal admixture with proto-East Asians could have happened to a region limited to the Arabian Pennisula, Anatolia, Persia, or the Indus River Valley sometime between 50,000 and 100,000 years ago. While this is somewhat vague and subject to caveats, it does narrow the scope of what could have happened a great deal.

Moreover, given that we know modern humans were present in India at the time of the Toba explosion around 75,000 years ago, and if we can assume continuity between those modern humans and subsequent ones in terms of levels of Neanderthal admixture (this is "the big if"), then the time frame for Neanderthal admixture may be 75,000 years ago to 100,000 years ago, in the same region, before rather than during or after the time period in which modern humans appear to have retreated from or gone extinct in the Levant.

In contrast, the admixture events that gave rise to Neanderthal admixture in Europeans are not nearly so constrained. This could have happened as late as about 30,000 years ago or as early as about 100,000 years ago in either Europe or any of the places it could have happened for the East Asian populations.

It would be interesting to see which Neanderthal genes are most common in South Asians, and if there is a cline similar to the ANI/ASI cline in those genes.

UPDATE (3-28-2011):

John Hawks has more to add to this story:

The "Homo heidelbergensis" model is in such utter disarray right now, I'm not sure many paleoanthropologists have realized the full extent of the problems. You should know that I don't believe in Homo heidelbergensis, never have. A couple of months ago, I was discussing some of the issues about mutation rate estimation with a very prominent geneticist, and the conversation turned to Homo heidelbergensis. What a shock the Denisova sequence should have been to those itching to see a H. heidelbergensis incursion into Asia!

Pre-Clovis Strata In New World Definitively Dated

A Texas archaeological dig has cleanly established the existence of a human settlement from around 15,500 to 13,200 years ago. This is about two thousand years earlier than the distinctive Clovis culture strata that is found across North America, and about a thousand years older than the oldest archaeological remains found in South America. This age is consistent with an entry of the earliest hominins to the Americas via the Beringian land bridge (where Alaska and its Aleutian Islands are today) and an expansion beyond Beringia when glacial melt would have opened a passage to the continental interior.

This site is notable because its dating is particularly "clean", because it clearly pre-dates the Clovis sites, and because it is in the interior of North America. It had seemed possible before this find, that earlier pre-Clovis sites fit a pattern that could have arguably been confined to the Pacific Coast of the Americas and their near vicinity, rather than also including the interior of the Americas and Eastern Coasts.

Open Question: Why do the earliest skeletons in the Americas seem to look different from later skeletons in the Americas?

The new find does not have bones, so ancient DNA and physical anthropology describing the people who inhabited the earliest strata of settlements in the Americas is not possible from this site. Some of the oldest skeletons in the Americas are from people who would have looked physically very different from modern Native Americans in all eras starting a few thousand years later, and somewhat similar to the very oldest East Asian skeletons. It isn't clear if the differences between the oldest skeletons and the later ones in physical type are a case of population replacement, human evolution as a result of new conditions, or factors like disease that could produce a few fluke skeletons that have been misinterpreted as seemingly belonging to a different racial type.

It is clear from genetic evidence that the vast majority of Native American genes found today have their origin in a pretty small population that expanded almost everywhere at about the same time when humans first emerged onto the continent from Beringia. The biggest divide in Native American genetics seems to suggest, however, that the genetic mix in the Native American groups that settled the North American interior is somewhat different than the genetic mix in the Native American groups that settled the Pacific Coast. It isn't clear if this was a product of random founder effects, or if there was some degree of population structure in the Beringian source population for the Americas.

It also isn't clear if there were one or more subsequent waves of people who arrived from Asia, perhaps by boat or perhaps in separate waves of migration that were closely spaced in time (the Na-Dene are often seen as a candidate population that could have given rise to the genetic difference between North America and South America), outside of the circumpolar region, that had a minor but detectable impact of Native American population genetics.

Finally, we can't rule out the possibility that the wave of humans arriving in the New World via the Beringian land bridge initially included modern humans from Asia of both an "archaic race" (paleoindians) and a more "modern race" and that the archaic modern humans vanished due to some combination of failure to adapt to the New World, conflict between the two populations that they lost, and admixture to the point where they were diluted into invisibility in modern Native American gene pools.

The plausiblity of a scenario with a "lost race" that inhabited the Americas as the same time as the ancestors of modern Native Americans is enhanced by the fact that we know that at least two other human populations, the Vikings, and a paleo-Eskimo population that arrived after the closing of the Beringian land bridge and then were replaced by the Eskimos, arrived in the Americans only the disappear without leaving a notable genetic trace in contemporary Native American populations.

Open Question: When did human first arrive in America?

The new find does, however, help us to bette calibrate the amount of genetic mutation that is observed in the Americas in indigeneous populations to a time of divergence.

That task is complicated by the fact that we don't know how much time the proto-Native Americans spent as a genetically distinct population in Beringia and Northest Asia before entering into the interior of the Americas. In particular, it isn't entirely clear if the proto-Native American population was drawn from pre-Last Glacial Maximum populations of Northeast Asia, post-Last Glacial Maximum populations of Northeast Asia, or a mix of the two. In one scenario, there was a population that entered Beringia, was trapped their by ice in both directions for many thousands of years, and then expanded out afterwards (Beringia's weather would have been mild for its lattitude due to ocean currents and prevailing winds). In another scenario, the stayover of the proto-Native Americans in Beringia en route to the Americas would have been relatively brief.

But, in all of the scenarios not ruled out by contrary genetic evidence or other evidence, the proto-Native American population would have been in a bottleneck situation in Beringia for some period of time before rapidly expanding out into virgin territory across all parts of North America and South America. So, timing of the "breakout" into the Americas can still be inferred by the amount of genetic diversity in different regions of the Americas, without knowing too much about how long that population was in the Americas. And, in theory, one can look at the amount of genetic differences between all American populations and plausible source populations in Asia, to estimate the age of the most recent common ancestor of the two populations. By subtracting the apparent expansion date of the Native American popluations of about 15,000 years ago, from the apparent age of the most recent common ancestor of the Asia souce populations and the American populations, it ought to be possible to estimate how long a distinct proto-Native American population was genetically isolated either in Berginia itself or in some territory in Northeast Asia.

In practice, this isn't so easy, because the margin of error in dating the oldest human habitation sites discovered (roughly plus or minus 900 years) and the margin of error in estimating divergence times based on genetic evidence (roughly plus or minus a few thousand years) in large in relationship to the probable length of the time that this population was isolated pre-expansion, which is a few hundred years at the low end of theoretical expectation (i.e. about 14,500 years ago) and about seven thousand years at the high end of theoretical expectation (i.e. about 22,000 years ago).

Open Question: Where there pre-land bridge or post-land bridge waves of human migration to the Americas?

When glaciers melted, the land bridge between Asia and the Americas was submerged, effectively isolating the Americas from the rest of the world until the arrival of Columbus in 1492, with a handful of very limited exceptions:
* some circumpolar Eskimo and paleo-Eskimo interchange between North America and Siberia,
* a brief and failed interlude of Viking efforts to colonize a few places on the North Atlantic coast of North America about a thousand years ago, and
* a probable brief episode of contact (but not population exchange) between the Austronesians and the South Americans on the Pacific Coast that brough the kumara (a kind of South American origin yam) and some associated names for it that are used in South American languages, to the Polynesians sometime in the last two thousand years or so.

It isn't clear what lasting cultural impact, if any, this brief episode of Polynesian-South American contact had on the South Americans. There are musical similarites between South American music and the musics of some cultures in Southeast Asia and Melanesia, and this is one time that this musical culture could have arrived, but the music of Polynesian cultures does not share the traits found in both Southeast Asia and Melanesia on one hand, and South America on the other, so music as a cultural impact seems unlikely.

The circumpolar contacts are the only known contracts between the Americas and the rest of the world between the closing of the Beringian land bridge and 1492 CE, that are known to have had a genetic impact on modern Native American populations.

There are a couple of New World sites that claim much older ages (ca. 30,000 years ago), but there is considerable dispute over the methodology used to establish those dates from cinders that may or may not have had human origins and whose dating may or may not be reliable. If there were populations in the Americas that old, one also has to explain why modern humans expanding into the virgin territories of the Americas did not show the signs found in all other cases where that happened: megafauna extinction, widespread population expansion leaving archaeological traces, etc., yet did leave isolated traces at very far flung sites. And, one also has to presume a level of sea faring expertise greater than was known to be available in any candidate populations around that time. And, one has to explain the lack of the distinct population genetic signatures you would expect from that population.

Obviously, new discoveries can only push earliest Native American populations further back in time, but the knowledge we have of ancient climate and the timing of megafauna extinctions widely believed to be related to the arrival of humans in the Americas strongly disfavors a scenario in which humans arrived very much earlier. Even the sites with archaic modern human physical anthropology are quite close in time (within a few thousand years) of the sites with modern human physical anthropology quite simmilar to modern populations.

24 March 2011

Class, Culture, Toilets and Time

I heard two recent stories about toilets in Asia in the last twenty-four hours.

HADENYA, Japan — The colossal wave that swept away this tiny fishing hamlet also washed out nearby bridges, phone lines and cellphone service, leaving survivors shivering and dazed and completely cut off at a hilltop community center. . . . With no time to mourn for their missing loved ones, they were immediately thrust into the struggle to stay alive in the frigid winter cold, amid a hushed, apocalyptic landscape of wrecked homes, crushed vehicles and stranded boats. They had scant food and fuel and no news from the outside world — not even the scope of the devastation.

On Wednesday, after the Japanese military finally reached them for the first time since the tsunami struck 12 days ago, by erecting makeshift bridges and cutting roads through the debris, they told a remarkable tale of survival that drew uniquely on the tight bonds of their once-tidy village, having quickly reorganized themselves roughly along the lines of their original community: choosing leaders, assigning tasks and helping the young and the weak. . . .

Almost as soon as the waters receded, those rescued here said, they began dividing tasks along gender lines, with women boiling water and preparing food, while men went scavenging for firewood and gasoline. Within days, they said, they had re-established a complex community, with a hierarchy and division of labor, in which members were assigned daily tasks. . . .

Refugee centers like this one in Hadenya exhibit a proud cooperative spirit, and also a keen desire to maintain Japan’s tidy perfectionism. Along the hallways, boxes of supplies lie stacked in orderly rows. The toilets are immaculate, with cups and soap neatly lined up. At the entrance, sheets of paper list names and assigned tasks for the day, like chopping firewood, carrying supplies and cooking.

From here.

How can you not admire a society that is organized enough to keep its toilets immaculate and its people well organized, even in the most dire circumstances?

Yesterday, I heard a story on the radio about the state of rail infrastructure in India (probably NPR or the BBC although I can't find a link to the story), where construction surges ahead in places like Mumbai, while the existing infrastructure is aging and underfunded. Particularly striking was an account of a train called the "Super" which was anything but. It's toilets, typical of passenger rail in India, were noxious. As another recent story on the subject sums it up:

For most visitors, rail travel in India is an indispensable part of any holiday, although an ability to overlook the often filthy toilets and deal with basic comfort and crowded carriages is required.

India has no shortage of cheap labor. Indeed, it has a government program that guarantees every person in India who shows up wanting to work a minimum wage job a certain number of days per year, usually on infrastructure projects. It also has trains full of people who could volunteer to do the work if organized properly. But, it doesn't. It has nausea inducing cesspools used only by the most desperate instead that make life miserable for train passengers and present a poor image of the country to every traveler domestic and foreign.

Meanwhile, Japan manages to keep its toilets pristine, even an isolated village devastated by a 9.0 magnitude earthquake and a tsunami that hasn't had contact with the outside world for a week and a half.

It is hard to chalk this difference to religion. Both Hindu and Islamic religious precepts give considerable attention to ritual purity and cleanliness; at least as much as the Shinto, Buddhist and Confucian syncretism found in Japan. Nor can one look to differences in climate - if anything, toilet cleanliness is more important from a public health perspective in sultry India than it is in frigid Northern Japan. Nor, must toilets on conveyances necessarily be putrid. Amtrak and most commercial airlines manage to have at least marginally clean toilets most of the time. Back in the era when trains were a leading means of intercity transportation in the United States, toilets on trains were cleaner than they are on Amtrak today.

Instead, it is a product of culture, and perhaps, in a not unrelated way, of economic development. Japanese communities place a premium on having decently clean toilets and are willing and able to organize themselves to secure them. India's rail culture continually drifts on the edge of mayhem and chaos, and the communities that use and manage it lack the will to decide that nasty toilets on trains don't have to be tolerated.

It isn't that South Asians don't prefer clean toilets to dirty ones. An ability to provide indoor plumbing and running water are now central to a young man's hopes of securing a good marriage match in India these days, very much as they were in the Reconstruction Era American South. But, for whatever reason, today's India is having trouble organizing itself as a society to secure this desirable thing.

American culture is somewhere in between. There are places - upper floors of office buildings, university classroom buildings, McDonald's restaurants, airports, the Cherry Creek Mall in Denver, and country clubs, where toilets are routinely tidy. There are other places - small rural gas stations, the Burger King restaurant at Cherry Creek North in Denver, public parks, the first floor of inner city libraries, outhouses at construction sites, and bus stations, where nasty bathrooms are the norm.

In America, toilet cleanliness is a pretty reliable social class marker. The higher the social class of the people who usually use it, the cleaner they tend to be, although this is also strongly a function in the institutional culture of the toilet operator. Toilet cleanliness is also something that is typically greater in private settings than in public ones in the United States, although there isn't a firm general rule.

There are other fascinating things one can learn from toilet culture as well. Americans, despite their strong ideological commitment to market approaches to economics, very rarely charge directly for using a toilet, although many establishments try to limit them to customers only, especially in areas with large numbers of transients. The decidedly socialist leaning French, by comparison, routinely charge a small fee to use a toilet, or at least expect users to tip an attendant. French fee for service toilets tend to be quite clean.

Toilets are interesting as a cultural indicator because they are impossible not to observe if you spend any amount of time someplace, because it is a universal indicator since every society has to have some way of handling the issue, and because it is pretty much impossible to hide their condition.

To the extent that toilet conditions reflect some version of economic development and social class, how "civilized" a culture is to use an out of fashion term and way of thinking, and there seems to be reason to think that this is something of a general trend, one also has to ask how much this is an effect of affluence, and how much it is a result of some attitude about how communal affairs are managed that gives rise to affluence. Are clean toilets a result of a shared high value afforded to a tidy and well ordered public sphere? Of shared social norms? Of minimal civic obligation? And, these these same values translate into affluence in the economy?

People like Mustafa Kemal Atatürk, the founder of the modern Turkish State, were convinced that this kind of thing was crucial to economic development. His agenda for reform included not just changes to the legal system and political system, but such cultural level reforms as the adoption of a national, Western style dress code designed to root out social and religious markers of the Ottoman era, despite the lack of an obvious connection between wearing suits and ties and conducting the affairs of a modern society. After all, as a former employer of mine enjoyed observing, the British conquered the world in shorts. But, it also seems that there really is something cultural that has to transform in the course of economic development.

One of the fascinating examples comes from British history in the early Industrial Era, one of the first places in the world anywhere that was organized socially and technologically on an industrial basis (continental Europe followed later, often heavily driven by government action as opposed to private investment). At that time and place was a cultural upheaval that included a greatly increased attention to punctuality and uniform time measurements. Clock towers went up all over England to keep communities synchronized. Many farm workers in the first waves converting to the industrial regime had great difficulty getting used to ideas like showing up on time and getting paid by the hour. While not many economists study international toilet cleanliness levels, there is an entire subfield of studies on the link between punctuality and time consciousness and economic development (see e.g. here and here and here and here and here and here and here). Economists study how fast people walk, how fast people talk, how timely trains are, and more. The aphorism about the Italian fascists is that at least they managed to get the trains to run on time. Standardized time zones have their origins in the efforts of train operators to coordinate their time tables.

Certainly, suits and ties are any more necessary to economic development than kimonos. Likewise, toilet cleanliness is not a cause of the rise and fall of great empires. But, one doesn't have to believe such absurdities to conclude that economic development and cultural transformation are inextricably intertwined, and that the development of some set of social virtues is as a critical a technology as the railroad or the internal combustion engine to attaining the economic scale and social coordination that is necessary for an affluent society.

23 March 2011

Blogger Weirdness

My two blogs at blogger (this one and Wash Park Poet) were removed for about a half an hour to an hour today for no apparent reason and then just as mysteriously reappeared. I have no idea why this happened.

Fannie Mae and Freddie Mac Bailout Most Costly

As I noted yesterday, most of the bailouts in the wake of the financial crisis were cheap. In all, they cost less than $50 billion of taxpayer money, and a third of that went to job saving and federal government revenue preserving/federal government expense reducing bailouts of General Motors and Chrysler.

There were a pair of bailouts, however, that were much more expensive: Fannie Mae and Freddie Mac, the government chartered, but privately owned firms that repurchased mortgages and mortgage securities meeting their guidelines from mortgage originators. According to a Los Angeles Times story from September 2010:

The bailouts of the two former government-sponsored enterprises, which continue to keep the mortgage financing market afloat almost single-handedly, already have reached $148.2 billion as bad loans they purchased during the real estate boom continue to fail.

Concerns were raised about the ultimate price tag when the Obama administration in December lifted a $400-billion cap on the federal commitment to Fannie and Freddie through 2012. Officials at the time said they did so to provide certainty to the real estate market as the White House and Congress wrestle with the future of the entities.

[Federal Housing Finance Agency acting director Edward J.] DeMarco told a House Financial Services subcommittee Thursday that the total cost of the bailout "appeared to be less than $400 billion." That figure would hold even under most scenarios analyzed by Fannie and Freddie in which the economy suffers another "severe stress." . . . Altogether, Fannie and Freddie hold $1.6 trillion worth of mortgage loans.

The federal government guaranteed the obligations of Fannie Mae and Freddie Mac in exchange for a 79.9% ownership interest in each of them, and there are in receivership. There was not a legally binding guarantee of either corporation's debts by the federal government prior to the financial crisis, but many investors had invested on the assumption that these institutions would be considered to big to fail and to tied to the federal government (despite its lack of ownership) to permit their obligations to default.

Some of the loans Fannie and Freddie own include "buy back" obligations from the sellers if they were improperly unwritten (e.g. if they knowingly accepted false appraisals or false borrower information). As a result, "Lenders repurchased $8.7 billion worth of single-family mortgage loans last year, DeMarco said. And as of June 30, an additional $11.1 billion in repurchase requests from Fannie and Freddie were pending." But, that even if all of the requested repurchases were made, only a drop in the bucket of the total losses sustained. Several dozen subpeonas are outstanding to determine if there are more cases where repurchases can be requested.

Record low new home sales and falling median home sale prices in February suggest that the market to which Fannie Mae and Fredie Mac are exposed won't be recovering any time soon.

Lets recap:

Financial industry bailout (x Fannie and Freddie): Under $33 billion

Auto Industry bailout: About $17 billion (but saved Feds $25 billion)

Fannie and Freddie bailout: $148 billion+ (probably under $400 billion). Another estimate suggests a $154 billion taxpayer cost for the bailout.

Stimulus spending: $690 billion including:
*$158 billion for Bush stimulus package
*$150 billion in public works projects for transportation, energy and technology in Obama Stimulus package
*$87 billion to help states meet rising Medicaid costs.
*$3 billion for Cars for Clunkers
*$267 billion for other Obama stimulus package spending

Stimulus tax cuts: $563 billion including:
* $268 billion for Bush stimulus tax cuts
* $16 billion for Homebuyer's Tax Cut
* $116 billion for Obama's tax cut to 95% of Americans
* $166 billion for Obama's other tax cuts
* $13 billion for Obama's payroll tax cut holiday

Still unknown: Recoveries in lawsuits against bad loan originators and investment banks that lied to sell mortgage backed securities.

Selected private sector losses from the financial crisis:

The public cost of the financial crisis was accompanied by massive private sector losses. Stockholders of failed financial institutions lost almost everything, whether or not their shares were ultimately cancelled or merely diluted. Bondholders in General Motors, Chrysler, and many bailed out financial institutions also experienced great losses -- although some bondholders in failed financial institutions (most notably Fannie Mae and Freddie Mac) appear to have been held largely harmless if they had the intestinal fortitude not to sell out to speculators.

Thus, ownership of failed institutions largely got their just deserts, while senior management was often sacked but rewarded handsomely upon their departure, and bondholders (particularly speculators) sometimes benefitted a great deal from bailouts.

Trade creditors, on average, did very well compared to long term financial creditors.


* "fraud-related losses to AIG shareholders totaled $1.2 billion to $1.4 billion. Another methodology from the expert put the losses at around $543 million to $598 million" Shareholders in AIG (collectively) weren't completely wiped out, but lost almost everything: "AIG's share prices had fallen over 95% to just $1.25 by September 16, 2008, from a 52-week high of $70.13. At the stock market's opening on September 16, 2008, AIG's stock dropped 60 percent."


* Lehman Brothers shareholders, who were wiped out when it went bankrupt, lost about $41 billion in the year before it collapsed.

* As of March 16, 2001, "Lehman's revised plan would repay [Lehman] creditors roughly $60.1 billion, equal to about 18.6 cents on the dollar based on an estimated $322 billion of valid claims." It hopes to obtain approval for the plan by November.


* General Motors creditor losses: "Before entering bankruptcy on June 1, 2009, GM had $54.4 billion in debt and owed an additional $20 billion to a retiree health-care trust managed by the United Auto Workers. . . . GM now owes $15.6 billion in debt and preferred stock and $9.4 billion in underfunded retiree obligations." The retirees also received common stock, so may not be as underfunded as it appears looking only at liabilities. GM creditors lost about $40 billion.

* General Motors stockholder losses since 2007: $15 billion. GM shareholders were wiped out.


* Chrysler secured bondholders received 29% of the face value of their investment. Their losses were approximately $16 billion.

* Chrysler equity holders via Cerberus Capital Management purchased an 80.1% stake in the company from DaimlerChrysler AG on May 14, 2007 for $7.4 billion. Chrysler equity holders were wiped out. On October 23, 2008, Daimler announced that its stake in Chrysler had a book value of zero dollars after write offs and charges. Daimler paid $36 billion for all of Chrysler in 1997.

Fannie Mae and Freddie Mac

* Fannie Mae shares were trading at $69.49 on June 18, 2007 and were trading at 27 cents a share on September 30, 2010. Fannie Mae shareholders lost about $64 billion.

* Freddie Mac shares were trading at $34 a share at the end of 2007. By February 2010, its shares were selling at $1.21, a decline of about 96%. The loss to shareholders was about $17 billion.


A summary of market capitalization changes for some major players from October 9, 2007 and September 12, 2008 is found here:

Citigroup: $236.7 billion to $97.8 billion.
Bank of America: $236.5 billion to $150.2 billion.
AIG: $179.8 billion to $32.3 billion
Goldman Sachs: $97.7 billion to $61.3 billion
American Express: $74.8 billion to $45 billion.
Morgan Stanley: $73.1 billion to $41.1 billion.
Fannie Mae: $64.8 billion to $700 million.
Merrill Lynch: $63.9 billion to $24.2 billion
Freddie Mac: $41.5 billion to $300 million.
Lehman Brothers: $34.4 billion to $2.5 billion.
Washington Mutual: $31.1 billion to $2.9 billion

Washington Mutual would ultimately be wiped out in bankruptcy.

In that time period: "4 trillion has been wiped off the total market capitalization of the U.S. stock market . . . Of that, nearly $1 trillion is from the decline in the financial sector alone."