Monday, November 26, 2018

The Eugenic Bogeyman

It looks as though some scientists in China have produced two infants from ova edited using Crispr. The objective was to disable a gene associated with vulnerability to AIDS. A good deal of the commentary on the report is negative, with talk of eugenics and risking the human gene pool and such. I find it hard to see much basis for such concern. There are obviously risks which the parents should have been, it is claimed were, informed of, but then there are risks to producing a child by the usual technology as well. 

I am particularly unsympathetic with the way in which “eugenics” is used as a bogey word, since it confuses two quite different things. Eugenics in the sense of some people deciding what children other people will have is a bad thing, especially when it involves some people deciding that other people will not be permitted to have children. Eugenics in the sense of couples trying to improve the quality of the children they have seems like a reasonable and unobjectionable activity. At the individual level it happens every time someone includes, in the choice of whom to marry, the consideration of what sort of children the proposed spouse will produce.

My favorite version of eugenics is the one described in an early Heinlein novel (Beyond This Horizon). It was a technology that let a couple select on both egg and sperm, thus choosing, among the children they could have, which ones they did have.

Comments? Any readers inclined to defend the arguments against this sort of technology?

17 comments:

David Lubkin said...

Your first variety of eugenics divides into two. I agree that people preventing or punishing other people from having children is a bad thing. Certainly if it's by State action. However, why is there anything wrong with privately encouraging or rewarding people who have children?

As the Jewish Agency did when I lived in Israel, giving a cash gift to Jewish mothers when they gave birth. Or in Heinlein's Howard Foundation.

Anonymous said...

We have eugenics already, in that we test fetuses for genetic abnormalities of various sorts, and conditionally abort.

For example, in Europe Down's Syndrome has been nearly eliminated in the lastest birth cohorts in many countries.

Perry E. Metzger said...

Presciently, Ilya Somin posted a great essay on this only two weeks ago. See: https://reason.com/volokh/2018/11/11/in-defense-of-designer-babies

Well dot dot dot said...

David, I think I once saw you either personally defend, or reference in agreement someone else who defended, the position that humans should maximize our ability to reach our natural age limit (about 120 years) but not use technological interventions that would allow us to age past that.

I think the same ethos of moderation (or however you'd characterize it...phronesis, perhaps?) could be applied here: improving the quality of children via mate selection, screening for serious diseases, etc. is in-bounds; going into the genetic code and switching things on and off is out-of-bounds. (Maybe that's essentially the same as what Heinlein was saying? I haven't read him.)

SB said...

I haven't spent a lot of mental energy on the ethics of genetic manipulation, but here's one concern I can see people having:

Individual opportunity is a Good Thing, in particular the ability for individuals to rise and fall according to their own abilities, while large and persistent socioeconomic inequalities are a Bad Thing, in that they tend to create instability, hopelessness, unrest, and violence. One of the complaints leveled against U.S. society in recent decades is that its socioeconomic mobility is relatively low among developed nations, and lower than it was fifty or sixty years ago, i.e. your parents' socioeconomic level (wealth, educational attainment, neighborhood of residence, etc.) is a better predictor of your socioeconomic level than it is in most other developed nations, or was in this nation two generations ago. This puts a positive-feedback loop (or at least not as much of a negative-feedback loop as might be desired) around socioeconomic inequality: the children of the rich and powerful are likely to be rich and powerful regardless of their own personal qualities, and the children of the poor are likely to be poor regardless of their own personal qualities. By contrast, in a hypothetical egalitarian utopia, each person would have the same opportunities from birth, regardless of parentage (except in the inevitable sense that genetic traits are heritable).

There's already a physical component to this positive-feedback loop: the children of the rich and powerful are likely to have better nutrition and better health care than those of the poor, which gives the former another advantage over the latter in becoming rich and powerful themselves. And to the extent that rich and powerful people tend to associate with, and choose their mates from, other rich and powerful people, any genetic differences between rich and poor are likely to be maintained or increased over time. It could reasonably be argued that individual-level genetic modification (assuming that it'll be expensive, at least at first) further accelerates the positive-feedback loop: the children of the rich and powerful will have not only better nutrition and better health care, but "better" genes (taller, more conventionally attractive, less disease-susceptible, better at particular academic skills) than those of the poor.

The dystopian vision raised by all this is a society divided along genetic, educational, medical, and socioeconomic lines into permanent underclasses and permanent ruler classes, a la "Brave New World". There are already tendencies moving us in this direction, but individual, a la carte genetic modification accelerates it. One can reasonably ask whether such a society is objectively worse than a society in which each human has more-or-less-equal opportunities from birth, but at the very least, it would be a change from the ideals espoused by most human societies in the past few centuries.

And the farther such genetic bifurcation proceeds, the blurrier the definition of "human". Maybe that's OK: maybe the definition of "human" should and inevitably will become blurrier, as the definitions of "male" and "female" have become blurrier.

Perry E. Metzger said...

The worry that only the rich will have access to genetic improvements is ridiculous. It's much like the worry that only the rich will have access to computers or cellular phones.

Furthermore, productivity is not zero sum. If the whole world has much higher average intelligence, then everyone's standard of living goes up, and that's true even if it's only higher on average.

Well dot dot dot said...

@Perry: But would you argue against the idea that genetic improvements would likely be accessible only to the rich at first? So then the question is, does this give them a "head start" of some kind that would create (or widen) lasting inequality even after the technology is widely accessible?

I don't think such a thing can be analogized to personal computers and cell phones either, since most people utilize those technologies for what are at the end of the day mostly trivial purposes. (In other words, the gap between not having a cell phone and what the average person does with his cell phone is not as wide as the gap between what the average person does with his cell phone and what that cell phone is capable of allowing a person to do.)

David Friedman said...

@Well dot dot:

I have never held that position. I am in favor of technologies to stop, ideally to reverse, aging. I can see no good reason to stop at 120.

You may be confusing me with someone else.

Perry E. Metzger said...

"But would you argue against the idea that genetic improvements would likely be accessible only to the rich at first?" — probably I would argue against this, yes, as almost all countries have (for good or ill, probably for ill) removed healthcare from the marketplace. Even in the United States, it would largely be employers paying via health insurance for genetic modification of embryos to remove genetic diseases, not individuals. Even off-coverage care will probably be accessible quickly, though. I've seen quite poor individuals buy IVF services; it is in the interest of the people selling the services to maximize their total profit, not their cost per customer.

"I don't think such a thing can be analogized to personal computers and cell phones either, since most people utilize those technologies for what are at the end of the day mostly trivial purposes." — I think that's an unreasonable claim, but it also ignores the fact that modern mobile phones involve technologies vastly more complicated to deliver than CRISPR is. If you don't know how complicated and vast the engineering network required to deliver a single 7nm chip is, let alone a whole cellphone, let alone the network it's connected to, you have no appreciation for how remarkable it is that someone can buy the results of hundreds of billions of dollars of capital investment for a couple hundred dollars. This points to the fact that incredibly complicated technologies involving huge capital expenditures can ultimately be purchased for very few labor hours of even a median income worker.

David Friedman said...

@Stephen Bloch:

A couple of points:

"the children of the rich and powerful are likely to be rich and powerful regardless of their own personal qualities, and the children of the poor are likely to be poor regardless of their own personal qualities."

That might be true, but it doesn't follow from the evidence you offer because of the role of heritable characteristics. A decrease in social mobility could be due to a decrease in random factors that changed social status. For an extreme case, imagine a society where everyone plays the lottery and the prizes are large. Ending that would reduce social mobility not because outcomes were becoming less dependent on personal qualities but because they were becoming more dependent on personal qualities.

Getting back to gene editing, it could have the opposite of the effect you describe. Suppose existing inequalities are largely due to innate genetic differences. Gene editing would let individuals with heritable characteristics that led to low status produce children with characteristics that led to high status, by copying genes that high status people already had. The worse the genotype you start with, the greater the opportunities to improve it.

On the whole, assortative mating seems more likely to lead the sort of problem you suggest, a point that I think was made in The Bell Curve, whose authors were worried about the same problem you are worried about. They argued that the society had become more meritocratic, and that tended to produce a greater problem of social inequality, both because it led to more assortative mating and hence a wider distribution of characteristics and because it meant that richer people not only thought they were smarter than poorer people, they were usually right, with unfortunate social consequences.

Perry E. Metzger said...

Indeed. Gene editing offers us not only a way out of the assortative mating conundrum, and is also a way to get to a society where there are far more people capable of doing much more interesting things than assembly line work or serving fast food. Higher productivity will benefit just about everyone. The scarcest and most important resource is smart people; having more smart people means more hard problems will be fixed.

Further, people seem to care a great deal about having children with the greatest possible advantages. If you want your children to be well educated, why wouldn't you also logically want them to be healthier and smarter? And this is good for everyone, since the world is not a zero sum game.

The only people who obviously lose are those who have a religious or quasi-religious attachment to doing things "the natural way".

Nate said...

I think conflating mate choice with individual gene selection is wrong.

Also, the phrases "most of our clients choose" and "we recommend" from your Zygote Selection Specialist will result in de facto selection pressures / eugenics, despite the illusion of individual couples making choices.

David Friedman said...

Nate:

Auto salesmen try to pressure their customers too, but we don't observe everyone buying the same model, or brand, or color, or size, or ... .

Do you regard human choice in general as an illusion? Hard to see why it is any more such in this case.

Nate said...

David,

I don't think the salesman analogy is as useful as it could be, for several reasons, although it would be interesting to consider the dystopian possibilities of "premium upgrades" on zygotes. I'm trying to be charitable and assume that whatever edits are possible are available to everyone.

A better way to see the pattern i anticipate would be to look at baby names and how they self-organize into trends without anyone at the wheel. That's fine for names, not so good for genetic distributions frequencies.

I don't regard human choice as an illusion, but I think we are often deluded about how much agency we are actually expressing in our choices.

RKN said...

Any readers inclined to defend the arguments against this sort of technology?

As appeared in the article (not the comments), the arguments raised by scientists who know what they're talking about (e.g., Drs. Topol and Church) aren't arguments against the technology, they are arguments against the premature application of this technology in humans, and the way He carried out the "experiment." I don't have a problem defending any one of those arguments. Did Dr He explain to his clients the long-term risks associated with a homozygous knockout of CCR5? I doubt it, since I don't think anybody really knows. This is a highly conserved receptor expressed on many different cell types with varying molecular function. Hard for me to believe its only role is mediating passage of HIV into the cell. Time may tell.

Richard Nikoley said...

Hey David. “Nicholas Rich” here. USENET days.

Was just scanning your blog and shared your favorite jokes.

Eugenia and Dysgenia is a rather new specific interest and it so happens I just did an interview about it, with someone who has actually read MoF and seems to characterize you as “least wrong,” which is a great complement in my book.

https://youtu.be/Mu4wkkqJpmE

Perry E. Metzger said...

"A better way to see the pattern i anticipate would be to look at baby names and how they self-organize into trends without anyone at the wheel. That's fine for names, not so good for genetic distributions frequencies."

Even the most popular baby names are not adopted by the bulk of the population. Indeed, there are probably more baby names in use than there are genetic variants of the average allele.