Eugenics is an ancient idea with a terrible modern reputation. Is that reputation now irrrelevant?
[A discussion paper for a talk I gave in 2015: reproduced here for my records, and perhaps your interest]
- Nineteenth century liberals —- contemporaries of Darwin —- revived the idea, which comprises two related objectives, one negative, one positive. They are: the elimination of defects and diseases in the population and the improvement of desirable population characteristics such as intellectual or physical capacity.
Although enthusiastically subscribed by leading thinkers and adopted by governments in the first few decades of the 20th century, eugenic ideas, then, had no plausible program. Nazi theories of racial hygeine and their murderous pursuit by Hitler’s government perverted the ‘negative’ eugenic objective.The idea of eugenics, positive or negative was apparently tainted forever.
Then, at the end of the twentieth century, advances in cellular biology and genomic science provided some hope of a program for (‘negative’) eugenic improvements to which there are few oblections: the elimination of genetic defects and diseases. But the mechanisms — already in selective use — are expensive, unreliable and do not necessarily eliminate defective genes.
In the past year or two [mid-2015] biological laboratories around the world have rushed to acquire a new, simple and effective genome editing tool, based on “natural” mechanisms used by bacteria to stave off infection. Last month, Chinese researchers announced they used CRISPR-CAS9 to edit the genome of human embryos to delete genes responsible for ß-thalassemia.
The rapid spread of a comparatively simple mechanism that could finally achieve eugenic objectives suggests it is high time to restart the discussion (lest governments again rush into unwarranted action).
A brief historical review
Eugenics has a reputation no better than racism or slavery.
It was an ancient idea (approved by Plato in The Republic) when revived in the UK in the 19th century. An early, and distinguished, advocate was the Victorian polymath (cousin of Charles Darwin), Francis Dalton, who coined the term ‘eugenics’.1
Dalton began the debate about “nature versus nurture”, and gave us the phrase. He devised innovative experiments to investigate the thesis that intelligence (‘Genius’) was heritable. In studies of twins (now a stand-by of psychological technique) he concluded that it was nature, rather than nurture, that accounted for superior intellectual ability. Dalton, like his cousin, was also a thoughtful conttributor to evolutionary theory and an effective opponent of Lamarck’s proposals — to which Darwin had been somewhat attracted — for the inheritance of acquired traits.2 Although his eugenic theories had no biological foundation (Gregor Mendel’s genetic experiments were not rediscovered until the turn of the 20th century), Dalton believed he had established a sound basis for eugenic practice.
The enthusiastic embrace of empirical discovery was what made 18th and 19th century Britain great. Eugenics became an academic discipline. Intellectual and political leaders formed learned societies to promote eugenic ideas and programs. In keeping with internationalist spirit of the first age of ‘globalization’, an International Federation of Eugenics Organizations coordinated scientific research. The Carnegie foundation established a Eugenics Records Office. Governments in many countries adopted eugenic policies and programmes including birth control and abortions for classes of intellectually or physically disabled people; they promoted marriages and births for classes of ‘desirable’ individuals.
But these practices were, like many government programmes, often free of good supporting evidence, not monitored for success, and easily subverted by the prejudices, hunches and even malice of the parties in power. The Nazi theories of racial hygeine and the genocide, murder and mutilations carried out against Jews, (overt) homosexuals, “gypsies” and the disabled or indigent represented the deepest corruption of egugenic ideas. A series of post-war UN conventions on human rights and genocide condemned and prohibted actions taken in the name of eugenics. Still, government eugenic programs have never quite disappeared; witness the eugenic objectives of the baby bonuses for graduates that were a feature of Lee Kuan Yew’s Singapore population policy.
Eugenic programs have often been described as either “positive” or “negative”. The positive goal is to favour the reproduction of individuals whose desirable traits are considered heritable. The goal of a “negative” programme is to eliminate the undesirable traits observed in a population by preventing their reproduction. The negative programme has been the focus of condemnation because, until the end of the 20th century, the most direct mechanism was murder or sterilisation of individuals whose undesirable traits were (thought to be) heritable. But there were also more benign social and technological trends that qualified. For example, the renowned biologist J.B.S. Haldane observed that the arrival of widespread bus transport in rural england in the 1930s “by breaking up inbred village communities, was a powerful eugenic agent.”
Still, there have been potentially desirable targets for negative eugenics for some decades since the recognition that some diseases and disorders have a genetic basis (which is not to say ’cause’). As of 2013, there were 3788 diseases or disorders associated with a fault in a known gene. They include achondroplasia (dwafism) and autism, some breast, colon and prostate cancers, Crohn’s disease, cystic fibrosis, Down syndrome, Duchenne muscular dystrophy, hemochromatosis, hemophilia, Marfan syntrome, myotonic distrophy, Parkinson’s, retinitis pigmentosa, sickle cell disease, thalassemia and many, many others. Some of these — such as dwarfism — are also due to spontaneous chromosomal abnormalities, but all in this list are also heritable.
So, despite its poor reputation, most of us would propbably subscribe to a negative eugenic programme if it eliminated the heritable risk of these diseases and disorders and could be achieved without doing other harm.3 Indeed, eugenic manipulation of the ‘gene pool’ has been taking place at the margins for a decade without any controversy. In all high-income economies and in many emerging economies too (China, India), pre-implantation genetic testing of IVF-initiated pregancies is available for prospective parents with known genetic risks. Technicians remove one or two cells from the blastomere some three to five days after fertilization of the ovum. They test these cells for the target genetic defect and only those embryos that, thanks to the Mendelian lottery, are free of the disease are implanted.
But this procedure does not necessarily eliminate genetic risk from the germ line. On the contrary, in the case of an autosomal recessive (single-) gene defect affecting both parents, any embryo has a 50% chance of having one defective gene (but no disease) and only a 25% chance of having no defective gene in its germ cells. Three-quarters of all embryos will be ‘successful’ from the viewpoint of the parents (and the child), but only a quarter of them will succeed from an eugenist’s viewpoint. In any case, the incentives of pre-implantation selection are all wrong for the eugenicist’s purposes. The rewards the parents seek are at strong odds, while the rewards the eugenist seeks are unlikely by comparison. As for positive eugenics, until recently it seemed at best dotty, at worst a repulsive racist fantasy. In any case, it lacked credibility. The length of a human generation — say, 16 to 20 years — was too long to allow repeated experiment based on Mendelian genetic combinatorics.
Pre-implantation testing, for example, would be of little use in a “positive” eugenic program. Assuming the genetic basis of desirable traits such as skill in mathematics were known (an heroic assumption) a rigorous program of pre-implant selection might ensure that every infant carried the optimum mix of alleles from two “fit” parents. If the odds of positive outcomes were better than in the case of recessive autosomal disease, there would be no runt in any litter. If the desirable characteristic were more uniformly heritable than recessive genetic disease (and not especially vulnerable to variable epigenetic expression), pre-implantation procedures might actually change the composition of the gene pool. But those are demanding pre-conditions. It might be more effective simply to subsidise the natural reproduction of the fittest individuals and submit to the Mendelian lottery on the theory that, among a large number of offspring, the proportion of less-fit would be small.4
Nor is ‘cloning’ an effective mechanism of positive eugenics. It is certainly feasible and happens both naturally (monozygotic twins) and, potentially, by intervention in an IVF pregnancy to divide the blastomere and implant both embryos. But a positive eugenic program would seek to clone individuals whose fitness was already demonstrated; adults, in other words. The available cloning mechanism for adults is Somatic Cell Nuclear Transfer (SCNT), which extracts the full (‘double’) set of chromosomes necessary for the viability of the organism from a somatic (usually ‘stem’) cell of the adult subject and implants it as a ‘replacememnt nucleus’ in the cytoplasm of an ovum from another individual. This is the mechanism that gave us “Dolly” the cloned sheep. But a decade of experience with cloning of large mammals (dairy cattle, racehorses) shows that clones are generally less robust than other animals and that pregnancies with cloned embryos are more difficult and have a lower success rate.
Then in 2012 the outlook for both positive and negative eugenic programmes changed, dramatically. Researchers in the USA and France discovered a rapid, uncomplicated, ‘natural’ way to manipulate genes in both germ and somatic cells that works in vivo.5 The breakthrough came from studying a technique used by bacteria to fight viral infections. Just a decade earlier, researchers began to suspect that certain repeated sequences (so-called CRISPR sequences) of DNA code in bacterial DNA were really “libraries” of viral DNA code, stored by the bacteria in their own DNA. The bacteria used these libraries to recognize and then to destroy invading viral infections using a set of enzyme ‘scissors’ (themselves coded by CAS genes or ‘CRISPR-associated’ genes) to chop-up and disrupt viral DNA. In bacteria, the CRISPR-CAS system is a sort of immune memory because bactirial insert the CRISPR sequences into their own DNA (using the same CAS scissors). When it reproduces, the bacterium passes-on its ‘library’ to the next generation along with copies of the rest of its DNA.6
It turned out that the CRISPR-CAS mechanism for disrupting viral DNA was particularly simple and efficient because it relied only on two RNA ‘messenger’ strands created from the CRISPR ‘library’ templates to program the CAS enzyme gene-scissors. The CAS enzyme ‘cradling’ these two CRISPR RNA sequences locks onto to the start of a segment of the viral DNA identified as a sequence of bases matching its RNA guides. The CAS ‘scissors’ then cuts both strands of the DNA molecule at the end of the matching sequence. In 2012, Jennifer Doudna and her colleagues showed that they could further simplify and improve this natural system, turning it from a find-and-destroy to a find-and-replace gene editor. They synthesised a single strand of RNA to use for the ‘find’ half of the operation. They implanted their RNA strand in a CAS enzyme that cut the double strand of DNA at the ‘found’ locations (more than one location, if necessary, clipping out a length of the strand). They also demonstrated that if they introduced a short sequence of DNA with the appropriate form into the nuclear ‘soup’ the cell’s repair mechanisms would use this sequence to ‘repair’ the DNA they had just cut. This is the ‘replace’ side of the operation. The end result is a revised ‘double-helix’ DNA molecule that can take its place with other such molecules in the genetic material of the cell.
The simplicity of this system, borrowed from an ancient, universal bacterial survival mechanism was immediately attractive to biologists interested in bacterial, plant and animal genome editing. Even before Doudna’s lab had demonstrated the artificial RNA system for directed editing, food technologists at Danisco had begun to manipulate CRSIPR sequences in bacteria found in yoghurts and cheeses to improve their resistance to viral infection (Eaten yoghurt or cheese rcently? You’ve almost certainly eaten CRISPRized cells). After the publication of the Doudna paper, literally hundreds of biology research and industrial laboratories rushed to replicate and apply the CRISPR-CAS method.
Soon groups of scientists were writing portentious warnings in the New York Times, as well as the journals Nature and Science, demanding that scientists everywhere abjure the use of this technology to alter the human genome.7 They acknowlged the value of genome editing technologies of somatic cells as a defense against disease. They warned, reasonably enough, of unintended consequences if researchers applied the techniques to germ cells. They warned against attempts to develop of “non-therapeutic genetic enhancement”, but they did not explain why this would be a bad thing.
Then, in April, a team of researchers at a university in Guangzhou reported the ambiguous and mostly disappointing results of an attempt to modify the gene responsible for β-thalassaemia in human embryos. The journals Nature and Science refused to publish the research (although they did not say why): the online journal Protein and Cell carried the paper.8
Discussion and questions
Governments have been repsonible for the worst (the only?) excesses committed in the name of eugenics. They alone have had the coercive power, in the past, to enforce a eugenic program. But is the genome a “public good”? Is it a sort of “commons” whose exploitation is likely to give rise to the “market failures” that justify government regulation?
Do public vaccination programs (so far, not eugenic because they do not impact the genome) suggest that the genome must be a public good? If there were a genomic change that, say, would give individuals a heritable resistance to HIV (an infectious disease) would governments have the right to enforce genomic treatment on all individuals on the basis that the genome is a public good ? Or, is the genome merely an abstract collective term for a private good — the genetic heritage of individuals — where government has no positive role (and will only make a mess of things as it has in the past)?
OECD jurisdictions strictly protect an individual’s genetic information under privacy laws. Does this suggest that the genome is merely a collective term for a private property? What is the appropriate response to the “unintended consequences” threat? A “ban”? A UN convention? Should it cover the editing of both somatic cells and germ cells? > Would a government ban — say, co-ordinated by an UN agency — on germ-line editing work? Would it be universally respected? Or would it be like a ‘cartel’ restriction; open to covert cheating? Is there a ‘first mover’ advantage in eugenic intervention that would favor such cheating? Developing superior warriors?
- Considering the risks of editing the genome, would a UN-coordinated restriction on germ-line editing even be necessary? Consider that none of the dozens of states that have nuclear weapons has used one in the past 70 years or even placed itself in a position where that choice might be imminent. Is there a problem in principle with “non-therapeutic” intervention (“positive” eugenics when heritable) as the scientists writing in Nature claim?
Suppose it became possible to select for certain desirable characteristics (not merely the absence of disease), whether on a heritable basis or not. Suppose parents could choose to have children that had a high chance of being a “super athlete” or a high chance of being a prodigal mathematician or musician, would parents take the option?
Suppose, further, (more realistically?) that the outcome of such a non-therapeutic action entailed mixed-risks. Say, due to unpredictable epigenetic factors, there were a 90 percent chance the child born would be a genius and a ten percent chance the child would become disturbed recluse or, possibly, psychotic. What would parents choose?
What would guide parents in the “design” of their children, should that prove possible? Would the considerations be different in the case of a heritable (eugenic) intervention? PWG 30 April, 2015
The Guangzhou researchers set up their experiment with care. They used only ‘non-viable’ embryos (doubly-fertilized ova with three sets of chromosomes that cannot result in a live birth). But the experiment did not succeed. The programmed ‘find and replace’ had a low hit rate and that there were ‘off-target’ mutations apparently due to the CRISPR-CAS9 enzymes attaching to the wrong genes. The opponents of any human genome editing acclaimed the messy result as a confirmation of their warnings. But, it might be more reasonable to see the Chinese result as a “fruitful failure” that shows where research should now focus to achieve the “100%” success that an acutal intervention would need.
Dalton is better known today as the inventor of fundamental statistical concepts (the ‘standard deviation’ as a measure of variance; the bi-variate normal distribution and regression analysis, and; “reversion to the mean” that we now call the central-limit theorem). He devised anthropometric and physical metrics such as a system for the classification of fingerprints and the first mapping of weather phenomena including the discovery that wind-direction can reveal an anti-cyclonic depression. He proposd the psychometric “lexical hypothesis”: the idea that small characteristic variances in language use, measured statistically, can reveal personality traits. Among his more whimsical discoveries: he contributed a letter to Nature in the 1870s on the best method of cutting a round cake “according to scientific principles”, and; he invented one of the most intriguing scientific toys ever, the Quincunx. ↩
To anticipate the final section of this paper, it turns out that CRISPR is a “Lamarckian” mechanism that for aeons has permitted bacteria to inherit acquired characteristics (specifically, a defence against viral DNA). So Lamarck, and the ‘early’ Darwin were not so far off the mark, after all. ↩
Genetic manipulation is risky because (among other problems) collateral impacts may be important. For example, there is a link between susceptibility to sickle-cell disease and resistance to malaria that is well-attested in Africa where both diseases are prevalent. Should a genetic modification that eliminated genes associated with sickle cell disease become widespread in future generations, it may have consequences for the prevalence (perhaps, at that stage, the ‘return’) of malaria that is a much bigger killer than sickle cell disease. ↩
This strategy, too, is defective since it works better for favoring the reproduction of the fittest males than females (for obvious reasons) ↩
See http://quantamagazine.org/20150206-crispr-dna-editor-bacteria/ for an account of the history of the discovery of the CRISPR mechanism. ↩
Of course, humans too have an ‘adaptive’ immune mechanism with ‘membory’. But it is a much more complex mechanism involving antigen proteins and special immune cells. And it works only during the lifetime of the individual; it is not integrated into the germ cells and passed on to the next geneation. Hence, every generation needs a new round of vaccinations. ↩
Liang, P. et al. Protein & Cell::http://dx.doi.org/10.1007/s13238-015‑0153-5 ↩