Eugenics has a bad name. Over the past 70 years, its opponents have linked it to racism and even genocide, and some Christians call it a blasphemous attempt to improve on God’s creation. In its heyday in the 1920s, however, eugenics was developed by leading biologists, including the founders of modern genetics. Their work was promoted and often put into policy by eminent statesmen and intellectuals. Eugenics was supported by a broad coalition,including progressives and socialists. On the Right, the Nazis supported it as well, but it is an absurd caricature to call it “Nazi science.”
The basic principles of heredity on which eugenics was based have not changed. It is our approach to science–and our rejection of it–that have supposedly discredited eugenics. In fact, dramatic advances in genetic engineering are constantly opening up new possibilities for modern eugenics.
People are eugenicists by instinct: They want to mate with healthy partners who show signs of intelligence and other desirable traits. Legal prohibitions against incest, meant to reduce the chance of inbreeding, go back at least as far as the Code of Hammurabi–c. 1750 BC–and the Old Testament. Plato and Aristotle promoted eugenic measures, and every society that domesticated livestock discovered the importance of selective breeding.
Even in today’s harshly anti-scientific atmosphere, parents frequently practice eugenics even though they never use the word. People with certain conditions seek genetic counseling in the hope of avoiding passing on genetic diseases to their children. Amniocentesis, or genetic sampling of the fetus, is a common procedure that has the same goal. When infertile couples go to sperm banks or seek eggs, they look for positive qualities in donors.
Not even the most militant political correctness has been able to root out popular understanding of certain basic truths: Parents pass on their traits to their children, and it is foolish to pretend they don’t.
It was Francis Galton who coined the Greek-derived term “eugenics,” meaning well born or good breeding. In Hereditary Genius (1869), he argued that mental abilities, just like physical features, are heritable. Through an analysis of the pedigrees of eminent families in England, he concluded that talent does not occur randomly, but instead runs in the families. Therefore, it would be possible, through arranged marriages, to breed people with desirable traits such as good health, intelligence, and noble character. A society of such individuals would far surpass the average abilities of the original population.
Until the Industrial Revolution, healthy and eugenic fertility was the rule in Europe, with the most capable people having the most children. In England in the 1620s, for example, the middle classes had 4.4 children per woman compared to 2.1 for the working class. There were no antibiotics or advanced lifesaving medicine, so less healthy people often died before they could reproduce.
Mass production of rubber condoms by Goodyear in 1860 started the era of affordable and efficient birth control. Middle and upper classes quickly adopted this novelty, while lower classes did not, thus ending the age-old correlation between high social class and larger families.
The First World War was a horrific slaughter of some of Europe’s best men. Indeed, IQ testing was introduced in the army in part to decrease the chances of using men of ability as cannon fodder. After the Second World War, eugenics became widely associated with Nazi atrocities, and by the 1960s and ’70s it was almost universally rejected.
However, it is dangerous to reject eugenics. Without it we have dysgenics, or the spread of less desirable traits, and the ability to maintain civilization declines. Here are some of the most sobering trends.
It is generally accepted that for a population to maintain its numbers, every woman must have an average of 2.2 children. The industrialized nations of Asia and Europe, which have the highest-IQ populations, have the lowest birth rates and are not reproducing themselves: 1.4 births per woman in Japan, 1.25 in South Korea, 1.6 in Russia and Canada, 1.44 in Germany and Italy, etc. The highest birth rates are in the most impoverished countries of Africa: Niger–6.76, Mali, Burundi, and Somalia–6, Burkina Faso–5.86, Angola–5.37, Ethiopia–5.15, Zambia–5.72, Uganda–5.98, etc.
At the same time, in industrialized countries modern medicine have greatly relaxed the environmental pressure that winnows out defects, which increases the frequency of heritable diseases. Also, parents are having children at increasingly later ages, which raises the number of spontaneous mutations–especially in sperm but also in eggs–which are then introduced into a child’s genome. Chemicals used in consumer products, radiation, and radio-wave-emitting devices could also contribute to higher mutation rates.
Contraception continues its dysgenic effect, not only because high-IQ groups have fewer children but because they have children at later ages, thus further decreasing their relative contribution to the gene pool. Because IQ is estimated to be 80 percent heritable, this skew lowers the average IQ in each successive generation. Welfare also encourages lower-IQ groups to have more children.
Finally, emigration from poor to rich countries probably lowers the average intelligence of both the sending and receiving countries. African emigrants, for example, are often of above-average ability in their home countries, but still bring down the average in Europe or America.
One of the key concepts in evolution is genetic load. This is the number of damaging and even potentially lethal alleles (gene variants) in the gene pool. Spontaneous–or de novo–mutations contribute to genetic load. In order to keep the load stable, the rate of such mutations must not exceed the rate at which they are eliminated–which happens when the person carrying them fails to reproduce. De novo mutations are not alleles passed on in the usual manner from parent to child; they are simply genetic copying errors.
Every child usually gets 60 to 80 de novo mutations. The majority are neutral; only two or three can potentially disrupt gene functions. Most of the deleterious variants are recessive, which means that they cause damage only in unusual cases, in which a child inherits the same mutation from both parents. Still, a de novo mutation at the wrong spot can be devastating.
Favorable mutations are exceedingly rare. The human genome has evolved through chance mutation for millions of years; at this point, a copying error is much more likely to upset a carefully evolved structure rather than improve it. These errors come in various types with specialized names–copy number variations, chromosomal deletions, microsatellite expansion or contraction, aberrant methylation, etc.–and their cumulative effect compromises fitness.
The number of de novo mutations a child gets depends largely on the father’s age; each additional year of father’s age contributes an average of two to three more such mutations. Although older mothers are known to contribute chromosomal abnormalities, such as trisomy 21 (Down syndrome), 80 percent of de novo mutations come from the father. These mutations are passed down to succeeding generations. There will be a significant accumulation of genetic load as more and more generations live under conditions of relaxed genetic pressure in which a high percentage of the population succeeds in reproducing.
Medicine is one of the greatest contributors to genetic load, because it blocks purifying selection. The left-hand graph shows how much modern medicine and public health has reduced death rates (in Australia), especially in the first few years of life. Although it may seem harsh to say so, early deaths kept genetic load in balance by removing deleterious alleles from the gene pool. According to one study, if approximately 16 percent of each generation fails to reproduce, that removes enough de novo mutations to maintain balance.
However, now that so many more people are surviving to child-bearing age, many unfavorable conditions are increasing. Gene pool deterioration is most obvious in heritable conditions that were once fatal but now are not. For example, before the development of insulin treatment, babies born with Type I diabetes died before reaching maturity, whereas now they can survive and have children. In many countries, Type I diabetes is increasing at roughly 5 percent a year, and in Finland, it increased 338 percent over a 32-year period in children ages one to four. These rates are clearly far more rapid than genetic change in the population, so such increases appear to be the result of a combination of increased genetic load and little-understood environmental factors.
In the distant past, deafness could have been a lethal condition if it meant an inability to hear an approaching predator. More recently, it made it difficult to find a spouse. Now, advanced countries have special schools for deaf people where they meet and marry other deaf people. This further propagates undesirable alleles.
Asthma and allergies do not usually kill people, but before the development of modern treatments, they undoubtedly reduced reproductive success. Now, they need not interfere with reproduction, and their incidence is rising. Asthma has at least tripled over the past 25 years, and now affects more than 22 million Americans. Allergies are also increasing, with such things as peanut allergies–virtually unheard of 50 years ago–appearing in day care centers and schools. Both asthma and allergies are heritable.
A rigorous investigation by the Mayo Clinic found that the incidence of celiac disease (CD) has increased 450 percent in the United States since 1950. CD is highly heritable, and now affects one in 133 Americans (0.75 percent). Its incidence continues to grow, and people with the condition must avoid food with gluten. The Mayo Clinic found that if they were unaware of their condition they were four times more likely than those without it to die over a 45-year period.
Twin studies have shown that autism is highly heritable, and its frequency in the United States has increased 600 percent since mid-’70s. Half of this growth is attributed to better diagnosis and awareness of the condition; the other half remains a mystery. People have proposed various environmental causes but there is strong evidence that accumulated genetic load is partly responsible: One study suggests that 10 percent of cases are due to de novo variants.
Similarly, recent research has found that the number of genes expressed in the brain is compromised by de novo mutations in people with autism, and that older parents are more likely to have autistic children. It may be that autism is increasing more rapidly than the accumulation of genetic load because of the highly interconnected nature of the genes that affect the brain. It may be that the number of mutations above a certain threshold can cause the entire neuronal network to function abnormally, thus leading to autism and possibly other neurological conditions.
The average IQ of autistic individuals is around 70, which means that, in varying degree, half suffer from what is considered intellectual disability (ID). More than 3.5 million Americans have some form of autism spectrum disorder. The average cost of lifelong support for an autistic person with ID is reported to be $2.4 million, and $1.4 million for those with IQs over 70. The annual cost of autism in the US is reported to be $236-262 billion.
Genetic load could build up to the point that medicine cannot cope with the associated health compromises. This is especially likely for neurological conditions, such as autism, for which there is no known treatment. Likewise, the appearance of “superbugs” that have evolved immunity to all common antibiotics is a serious potential threat. If drug-resistant bacteria become common, our immune systems will have to combat these pathogens unaided, which was the rule before the discovery of antibiotics.
Accumulation of genetic load could have similar results. If medicine is unable to keep up with the rising genetic load, the number of people who die before reproductive age will start rising again. This would happen even with the medicine functioning at a very high level. In other words, purifying selection will resume so as to prevent further load accumulation.
If, in the future, there is a failure of the medical system or in the welfare policies that make medicine available to people who cannot afford it, there could be a devastating jump in the number of people of all ages who die.
In China and Russia in the 19th century, as many as half of all babies died before adolescence. Imagine genetic load accumulating to the point that death rates returned to that level (16 to 20 percent would be a theoretical minimum) despite modern medicine. Then imagine a societal collapse that eliminates modern medicine or sharply limits its availability.
This nightmare vision is not new. Nobel Prize winner Hermann J. Muller is his famous 1950 paper “Our Load of Mutations” argued that a steady accumulation of genetic load will lead to immense suffering for future generations. In his view, the only way to eliminate mutations was through “rationally directed guidance of reproduction.” He thought “abstention from reproduction” was far more desirable than the eventual alternative: “failure in a struggle for existence.”
The term ‘debtor’ is appropriate for such generations because, by instituting for their own immediate benefit ameliorative procedures which delay the attainment of equilibrium and raise the equilibrium level of mutant gene frequency, they transfer to their descendants a price of detriment which the latter must eventually pay in full.
Without a reduction in the germline transmission [passing on to succeeding generations] of deleterious mutations, the mean phenotypes of the residents of industrialized nations are likely to be rather different in just two or three centuries, with significant incapacitation at the morphological, physiological, and neurobiological levels.
Ever since the widespread use of the condom, there has unquestionably been dysgenic fertility for intelligence in advanced countries. In one recent study, Michael Woodley argued that Victorians were, on average, 13 IQ points smarter than we are. Such studies appear to conflict with what is known as the Flynn Effect, or the well-documented fact that actual performance on IQ tests has been rising in recent decades. However, Philippe Rushton and Arthur Jensen have argued that the IQ sub-tests that improved most were measuring mental processes not highly correlated with general intelligence. It is widely assumed that tested IQ has not risen due to wider distribution of alleles associated with high intelligence, but because of better nutrition, greater familiarity with testing procedures, and perhaps the effect of early use of computers and smart devices.
This rise in IQ scores seems to have come to an end and gone into reverse in Finland,Denmark and Sweden. The IQ-boosting effect of improved nutrition and more stimulating environments may have reached the limit of their effectiveness, and deterioration in the genetic makeup of Europeans may have begun to drag down actual IQ test scores. In other words, a rise in phenotypic IQ (actual performance on IQ tests) caused by environmental factors was masking an ongoing decline in genotypic IQ (the distribution of high-IQ-related alleles).
Fertility by race within the United States is certainly dysgenic for intelligence. Hispanic women have the highest lifetime fertility at 2.82, with black women next at 2.02. Whites are clearly at sub-replacement fertility at 1.85. As the percentages of blacks and Hispanics increases, the average IQ will fall (the decline will be slowed by Asians who, though only 5 percent of the population, have a higher average IQ than whites). Eventually, the United States could have the IQ profile of a Third-World nation.
Assuming an initial average IQ of 100 points and a standard deviation of 15, this graph shows the effects of shifting the average to the left by five and then 15 points. With an initial average of 100 points (shown in red), the number of mentally retarded (IQ less than 70) and highly gifted (IQ over 130) is each about 2.2 percent of the total, as indicated by the shaded areas. Each shift to the left greatly changes the area under the curve at both extremes.
A 15-point decline in the average, to the black American average IQ of 85, reduces the over-130 population to almost zero, while greatly expanding the under-70 population. At that point, only 16 percent of the population is at or above an IQ of 100. Professor Linda Gottfredson of University of Delaware has vividly described the outcomes in daily life of different levels of IQ. Everything about life in the United States will change as the average IQ falls.
There have, in fact, been examples of extremely rapid declines in population IQ. The city of Detroit was founded in 1701 by a party of French colonists led by Antoine de la Mothe Cadillac. By 1900 it was known as Paris of the West for its grand boulevards–and it had a 98.55 percent-white population of 285,704.
With the growth of the automobile industry, blacks moved to the city from the South, and by the end of the 1960s they accounted for about half the population. After one of America’s worst race riots in 1967, whites and skilled blacks left for the suburbs and the productive capacity of the city was critically compromised.
The result of this dramatic shift in population–and IQ–is the blasted shell of a city we see today. The destruction was speeded by systematic arson, especially the pre-Halloween orgy of fire and destruction known as Devil’s Night. At its worst, 810 fires were set in a three-day span. The destruction of Detroit is comparable to the results of a nuclear strike. Entire blocks of former residential communities have been obliterated, as can be clearly seen here. Detroit is now a popular destination for photographers who record the astonishing ruins of a once-great city that finally went bankrupt in 2013 with $20 billion in debt.
Few commentators see the pillage of Detroit through a prism of diminishing IQ. They would nevertheless be forced to acknowledge that half of all city residents are functionally illiterate, and that National Assessment of Educational Progress (NAEP) scores are at shocking lows. As Michael Casserly, executive director of the Council on Great City Schools, said of Detroit:
The truth here is that no jurisdiction of any kind in the history of NAEP has ever registered such low numbers. They are just above what one would expect by chance alone–as if the kids simply guessed at the answers. These numbers–to our minds–are shocking, appalling, and outrageous . . . .
Mr. Casserly blamed the public schools, and it is true that half of Detroit schools have closed since 1950. However, the remaining schools have been sufficiently funded, with annual expenditures of $13,825 per child for the 2012-2013 school year. Ninety-five percent of the students in Detroit public schools are black, so the problem is more likely to be the students, not the schools.
Standardized tests such as NAEP and the SAT are highly correlated with IQ. The correlation between SAT and IQ is so close–in the 0.72 to 0.88 range–that SAT scores are often used to estimate IQ. The following graphs show national SAT scores by race from 1987 to 2013.
It should be clear that what destroyed Detroit was a rapid decline in average IQ–one almost exactly equivalent to the theoretical 15-point decline plotted on the bell-curves above. The consequences of this decline would have been even more devastating if Detroit had not been part of the richest country in the world and surrounded by an infrastructure that greatly cushioned the collapse.
There is a 0.757 correlation between national IQ and Gross Domestic Product (GDP) per capita, and with many other desirable outcomes such as low crime rates, rule of law, freedom of the press, etc. We can anticipate a similar–but slower–collapse at the national level if average IQ continues to fall.
To return to Hermann Muller’s paper, “Our Load of Mutations,” he argued that if people were properly educated about the importance of heredity and were guided by medical committees, eugenic goals could be achieved by “freely exercised volition of the individuals concerned.” If those least likely to contribute positively voluntarily remained childless it would mimic natural, purifying selection and would keep genetic load in check.
Of all the possible eugenic methods, this would be the least controversial, but its success would require a fundamental change in thinking. Today, it is virtually impossible to base public policy or even public information on the obvious fact that mating choices can damage the gene pool. And even in a radically changed climate, in which it were generally accepted that it was not desirable for poorly endowed people to reproduce, they are not likely to care very much about the common good.
William Shockley, who invented the transistor and who later took a great interest in eugenics, rarely took positions on public policy, but he did propose “thought experiments” about ways to encourage the less fit to refrain from reproduction. The best known was the $1,000 Bonus Proposal. Anyone of childbearing age would be offered $1,000 for every IQ point under 100 if he agreed to be sterilized. Shockley even suggested that “bounty hunters” could be rewarded for finding such people if they were too stupid to learn about the program on their own. Opponents have argued that anyone who would qualify for a substantial payment would not have the capacity to give informed consent to be sterilized. If that is true, it seems hard to argue that they would have the capacity to be competent parents.
At a 1963 Ciba Foundation conference in London called “Man and His Future,” three distinguished biologists and Nobel laureates–Hermann Muller, Joshua Lederberg, and Francis Crick–took a different approach. They discussed many ideas about the future of mankind, specifically Muller’s proposal of establishing a sperm bank to which only the best men would contribute.
Inventor and industrialist Robert Graham later worked with Muller to establish what became the Repository for Germinal Choice–a name chosen by Muller. Graham discusses the project here, here, and here.
The repository operated according to two eugenic principles. The most obvious was raising the average IQ by increasing the number of children fathered by gifted men. However, it also decreased genetic load by storing sperm from earlier decades–and thought to have a lower load–to make it available for later use. The result was the births of 229 children from gifted donors with IQs of at least 140. Unfortunately, shortly after Robert Graham’s death in 1997 the repository was closed and all specimens were destroyed.
The repository’s approach was similar to one suggested by Michael Lynch of Indiana University. He has proposed transgenerational cryogenic storage for future use of gametes and/or embryos. The goal would be to decrease the number of de novo mutations by using germinal material that was harvested generations ago and stored in liquid nitrogen. Using material from the 1950s, for example, would skip decades’ worth of load and probably greatly reduce the chances of transmitting celiac disease, asthma, diabetes, etc.
Another potential approach involves editing out the genetic load from the genome using a technique called CRISPR/Cas9. CRISPR stands for “clustered regularly interspaced short palindromic repeats,” and Cas9 is the name of a specialized enzyme used to cut DNA. When it is perfected it could be used to snip out deleterious alleles that add to the load, with the objective of returning genes to their ancestral state. This technology has been tried in China and is still too inaccurate to use in humans but it shows considerable promise.
Richard Lynn–who has unquestionably done more than anyone recently to rehabilitate serious thinking about eugenics–has proposed embryo selection as a eugenic technique. This would involve in vitro fertilization of a large number of a woman’s eggs, letting the embryos develop slightly, and then scanning their genomes and implanting only the most promising ones.
This would not solve the problem of genetic load, and it is likely to remain impractical for a long time. The problem is that it is notoriously difficult to screen for traits that are influenced by more than just a few genes. We can easily screen for known “Mendelian” disorders, such as Tay-Sachs, cystic fibrosis, Marfan syndrome, Rett syndrome, etc., because a defect in a single gene causes a single disease. We know where to look for the defective genes.
However, consider a seemingly simple trait: physical height. We know that variations in height are approximately 95 percent heritable, so screening embryos for the right genes should make it possible to predict an embryo’s height as an adult. In fact, the best available mathematical model for predicting height uses 1,900 different alleles but achieves a correlation of only 0.41 with actual height.
Intelligence is even harder to predict. A study of more than 100,000 people uncovered three genetic variants for IQ, but their impact was minimal: Each accounted for an average of only 0.3 points on an IQ test. That means that a person with two copies of each of the variants (for a total of six) would score only 1.8 points higher on an intelligence test than a person with none of them.
A study of the genomes of 17 supercentenarians (people who lived past the age of 110) detected no genetic variants associate with longevity, though high intelligence appears to be a strong predictor of good health and longevity.
At present, it is simply not possible to screen an embryo–or a child–to predict polygenic traits (ones caused by many genes), such as height and intelligence, though it is an interesting theoretical possibility. Great efforts have already been made to screen for these traits, and as soon as they succeed, there will probably be tests available on the market.
One exotic method that could be used to promote good health and intelligence is time-separated twinning. It involves collecting and fertilizing human eggs, which can be artificially caused to divide into identical twins. If preimplantation genetic screening showed no detectable disorders, one of the twin embryos could be implanted in the womb and the rest frozen. If the resulting child proved to be healthy and intelligent, the other twins could be implanted with high confidence that they, too, would be intelligent and healthy. Parents would be getting a child who was a perfect genetic copy of someone who was already living.
How long would you have to observe the first twin to be sure the frozen embryos were of high quality? Predictions of mental retardation can be made by age two, with 85 percent accuracy. Reasonably good assessments of intelligence are possible at ages five to seven, but waiting until 11 to 13 gives better results. For late-onset diseases such as diabetes or Alzheimer’s one should wait 50 years or even longer. The predictive power of the frozen embryos would increase over time.
Some people would be disturbed by the idea of bringing into the world the identical twin of someone who was already an adult–or even already dead–but time-separated twinning would be a very accurate way of predicting what a child would be like and would lower rates of complex hereditary disorders. This paper explains how time-separated twinning could prevent complex diseases.
SIRM Las Vegas is a company that is already investigating the use of artificial twinning to improve the odds of helping women get pregnant. When there is only a small number of embryos available for implantation, creating twins increases their number and gives a woman that many more chances to have a child.
The next logical step would be to establish a bank of demi-embryos, that is, frozen embryos that were identical twins of highly capable people who had already been born–some may be adults or even deceased. A childless couple could conceivably meet and directly evaluate the person whose identical twin they were considering for implantation. Most people have a strong desire to be the parents of their own genetic offspring. Choosing to implant a demi-embryo might be more attractive if it were the twin of a particularly capable relative. Implanting identical twins may seem futuristic and disconcerting, but if such options were available, some people would certainly use them.
A certain number of people would be happy to implant the twin of a famous beauty queen, intellectual, or sports star. A fertility clinic would have to make sure that a single desirable person was not copied too many times, with a limit of, say, no more than 20 per generation. This, along with good record-keeping, would limit the chances of someone in the next generation unwittingly marrying someone who was, genetically, a half sibling.
If it is not possible to use such techniques as CRISPR to snip out genetic load, the accumulation of load could be greatly slowed simply by freezing genetic material from the present and using it in the future. It is entirely possible that people 100 years from now will look back at today’s genetic material as healthy and highly desirable. This is by no means an absurd idea for anyone who thinks about the long-term prospects for our species.
At present, low birth rates among the most capable mean there is a tremendous waste of good germinal material. What we take for granted and discard today could become an extremely valuable commodity in the next century. Future generations could benefit greatly from sperm and egg donations from the brightest and fittest, and augment these advantages with artificial twinning and trans-generational cryoconservation. It remains to be seen whether our species has the wisdom and foresight even to think in these terms.