Friday, May 24, 2013

The rhetorical approach to selling eugenics

... is not to emphasize positive over negative iterations of the idea, or even, as I've done in the past, to focus on the consequent egalitarianism that these approaches, if put into practice, will presumably foster. No matter how delicately broached and amenably angled, it's inevitably received as being too harshly comparative, too judgmental, and we simply can't have that! Arguing that this guy should have five kids while this guy shouldn't have any is socially and politically intolerable in the contemporary Western world, irrespective of the merits of such an argument.

It needs to be about individual choice and self-improvement. Forget pushing (to a broad audience, anyway) for one group to do more procreating and another to do less of it. In vitro fertilization has come a long way in the last five decades, and embryo selection has progressed from being all about viability to include avoidance of undesirable traits, the most conspicuous of which is Down syndrome. And the inclusion of putatively desirable traits is somewhere just around the corner. Eugenics need not be about telling the best to breed more and the worst to breed less, but instead marketing it as a game anyone can play. If everyone begins systematically making the most of what they have, we get a rising tide raising all boats outcome.

Since we're on the subject of selling the idea of eugenics for public consumption, Razib Khan recently pointed out the seemingly obvious point (which I nonetheless have never explicitly acknowledged until he spelled it out for me) that eugenics, rather than being a way for the patricians to further distance themselves from the plebes, is actually a means of moving towards that good old American ideal of everyone having an equal--or at least less significantly unequal--shot:
Steve [Hsu] has much to lose in a selfish zero sum sense because he’s already rather assured of intelligent offspring. He’s smart. His wife is smart. Standard quantitative genetics implies that even if they regress to the mean his offspring will be quite bright. There may not be much more juice to squeeze out of that genetic background. It may be very different for a couple with more average endowments. So sorry to turn this upside down, but personal eugenics may in fact be a boon for the ugly, stupid, and psychologically unstable, because it gives them a opportunity to close much of the gap with those who were lucky in the genetic lottery. Some of you may object to terms such as “ugly,” “stupid,” or “psychological unstable.” But people with these issues have to deal with them in their day to day. One can make all the platitudes one wants to make about “inner beauty,” but very few people live by this ideal.
Anyhow, these rhetorical considerations are probably going to become moot in the future. Prospective parents are going to take advantage of technologies that give their children an edge in life regardless of braying from opinion makers. Elites can crow all they want about the need for racial and socioeconomic integration in schools; couples are still going to do their damnedest to get their own kids into the iciest, most affluent schools they are able to.

There is already good deal of conceptual acceptance of eugenic practices on the individual level, especially on the political left. From 2004, the most recent year in which the dichotomous question was posed, the percentages of GSS respondents who say they'd have an abortion or want their partners to have one if a test revealed "the baby has a serious genetic defect", by political orientation (n = 2,451):

PoliticsYes abort
Liberal49.8%
Moderate36.0%
Conservative21.5%

Though the major media regularly treats eugenics as being under the aegis of the right, I suspect the subset of the right to which it actually applies is about as large as the population for which "secular right" is an accurate descriptor.

GSS variables used: GENEABRT, YEAR(2004), POLVIEWS(1-2)(3-5)(6-7)

20 comments:

JayMan said...

I believe this was Richard Lynn's conclusion.

Anonymous said...

Assuming that this is all "just around the corner" may be a bit optimistic. Prof. Hsu thinks it may take 20 years to fully sort out the genetic basis of intelligence, and even that could be optimistic depending on what the early results show. Then there is the issue that pre-implantation embryo selection limits the parents to embryos with the best combination of alleles from the set of alleles shared between them, not the objectively best set of alleles. One would expect that two less intelligent parents would not share an abundance of the superior alleles. More intrusive genetic engineering would be necessary, and we aren't there yet. It could be many decades before we are truly capable of churning out smart adults from embryos that otherwise would have produced dullards.

Audacious Epigone said...

Jayman is referring to Lynn's 2001 book Eugenics: A Reassessment (the Kindle version of which goes for over $100, wow).

Thanks for pointing that out.

John said...

There is a difference between selective breeding and molecular genetic engineering. Telling Jennifer to have 5 kids and Lisa to have none is not going to sound good to Lisa. This type of eugenics is never going to be popular.

However, the ability to alter individual genes has the potential to be popular for all the reasons listed. It could be an equalizer for ability, but might lead to islands of warring peoples with very different values. (The Nietzscheans from the TV show Andromeda is a pretty good speculation as to what might happen). In the short term, though, I don't think there is any stopping it. Once the Chinese decide to all make themselves 30 IQ points smarter, the rest of the world will have no choice to follow.

Anonymous said...

If you want lynns books. I can send them.

Lynn doesn't mind. I asked him once.

Anonymous said...

The Chinese don't yet have the ability to make all of themselves 30 IQ points smarter. The BGI hasn't even published a "first draft" of candidate loci affecting intelligence. When it does, there will still be a lot of work to do to ascertain the genes involved and to ensure that they aren't false positives. Prof. Hsu thinks it will take 20 years to sort everything out. Just knowing the genes doesn't allow manipulation of those genes in vivo. For the near term, the intelligence benefit will have to come from pre-implantation embryo selection. If only near-optimization at hundreds or thousands of loci can raise individuals' IQ's by 30 points, this technology can't get the Chinese there in one generation. I don't doubt that eventually genetic engineering will eventually come, but it may be many decades before a birthed huma can pop in to the lab for a global genetic intelligence enhancement.

Anonymous said...

Anonymous 5/25/13, 11:07 AM:
"pre-implantation embryo selection limits the parents to embryos with the best combination of alleles from the set of alleles shared between them, not the objectively best set of alleles. One would expect that two less intelligent parents would not share an abundance of the superior alleles."


True, but if I understand the math correctly, two "slightly dumb" White parents (IQ=85) would create children with mean IQ~92 and SD~15. Not too bright. But if they created, say, twenty zygotes and chose the smartest one, now you're looking at IQ~117 (the 95th percentile of that distribution), a gain of 32 points.

It's like Maxwell's Daemon applied to genetics: give the little guy a statistically broad sample, and let him choose only what you want out of the process.

--Daemon

Anonymous said...

How do you know that among the twenty zygotes there will be one predisposed to an IQ of 117? The zygotes will have inherited variable combinations of the parents' alleles, and the likelihood of any random zygote having the all of the parents' best alleles at 1000 independent loci will be much lower than 1 in 20. I think you're too optimistic.

Anonymous said...

Based on the regression to the mean equation Steve Hsu posted to his blog I think it would work out to be more like 115 because the residual SD is around 12.5 rather than 15 (somewhat less than the whole population SD), but that seems correct to me.

How do you know that among the twenty zygotes there will be one predisposed to an IQ of 117? The zygotes will have inherited variable combinations of the parents' alleles, and the likelihood of any random zygote having the all of the parents' best alleles at 1000 independent loci will be much lower than 1 in 20. I think you're too optimistic.

They don't need all the parents best alleles.

IQ of 115 is around the 95% percentile for children of two parents with a mean IQ 85. That's just how it is from what has been measured, regardless of whether they have all the parents best alleles.

Most likely, the so improbable they never actually happen in reality children who have anything close all the parents best alleles would much, much, much more different from their parents than mere IQ 115 (and a generation of such impossible babies would be far more different from us than mere Blacks and Jews are different from one another).

Anonymous said...

Based on the regression to the mean equation Steve Hsu posted to his blog I think it would work out to be more like 115 because the residual SD is around 12.5 rather than 15 (somewhat less than the whole population SD), but that seems correct to me.

Sorry, that should be 105 for the 95% percentile of the offspring of two 85 IQ persons.

Anonymous said...

We're sort of arguing past each other here. I'm trying to build up from the genetics standpoint, and it's not clear to me how you can just assume that zygote selection is going to guarantee a 2-SD improvement. Nearly all of the genes implicated are likely to be of very low effect. It's quite possible that in a batch of 20 zygotes, the best candidate will have relatively optimized only a few hundred loci out of thousands -- only relatively optimized because it is possible that the dullard parents possess optimal alleles at only a few loci. So I think you're too optimistic for what this technique can reproducibly accomplish in one generation of breeding from two low-IQ parents.

Anonymous said...

Forgot to mention that genetics may not even explain the whole picture of intelligence. I'm a pretty strong hereditarian, but we're setting ourselves up for disappointment if we think we will understand every aspect of this complex trait from the BGI. There are issues of protein-protein interaction to consider (synergistic interactions of proteins produced by specific combinations of genes with optimal alleles) and genes that don't appear to have an impact that will subsequently be found to moderate the activity of the proteins produced by the identified genes. Those are only some of the obvious wild cards that could come into play. This stuff is not all cut and dried, but that isn't a reason not to pursue it.

Audacious Epigone said...

Gentlemen, use handles! It's tough for laymen to follow Anonymous. Thanks for the discussion.

Dan said...

It's not even clear to me that IQ by itself is a measure you want to be singularly pursuing.

It seems to me that high victorian civilization resulted from selection of a collection of qualities, including intelligence but also including health, ability to delay gratification, ability to and willingness to work hard above and beyond subsistence and more.

Thus, it seems the best bet is to sort of let the chips fall where they may but have one simple rule:

****If you want to go on state aid you will be taken care of but you must use birth control. If you can manage without any help then you are free to procreate at will.****

This also seems like the only thing that can achieve what are probably the three most important parts of the solution:

(1) Long-term sustainability

(2) Fairness and the possibility of freedom

(3) Mercy for the poor

A solution *must* have all three elements for it to be tolerable. The Malthusian world of the past offered 1 and 2 only. This was tolerable in a world of scarcity but it is not tolerable today.

I suspect we must either arrive at this solution or return to a Malthusian world where natural law takes over. In the long run it seems inevitable. An equilibrium must be found.

Anonymous said...

Assume you harvest 20 eggs from a woman and fertilize them with the husband's sperm. Grow them to roughly the 128 cell stage, grab off a couple of cells, then freeze the zygote. Now, examine your cells for, say 500 traits, then choose the optimal zygote. This is selecting for fitness, not necessarily intelligence. Howeve, general fitness and intelligence are correlated. You now have the fittest child if they had twenty in a row. Do it twice, and the parents' chances of having two children at this level of fitness is one in 400.

Audacious Epigone said...

Dan,

Unfortunately, sterilization for welfare is intolerable in today's political climate as well.

Last anon,

Pithily put, thanks.

Dan said...

"Unfortunately, sterilization for welfare is intolerable in today's political climate as well."

I would never say sterilization; I would say birth control until you can get on your feet, a kind of prepared parenthood.

Yes, you are quite right that is not politically feasible now. But I think the realm of what is politically feasible is subject to change. In the mid 1990s, DOMA passed congress with about 7/8 of both houses, but all of Hollywood, most of the media and the entire political left focused for 20 years on reeducating the public toward normalizating gay marriage, something which had never been accepted as an institution in human history (even in places like ancient Greece).

Surely linking financial independence with reproduction would be an easier sell, if you could promise generous benefits to those who want it, and solvency for the nation?

In any case, the alternative (likelier?) solution is what we see in India or Brazil today. These are socialist regimes that promise everything to everybody but have little to actually deliver.

Will the left ever act rationally to save the system? China has a 1 child policy, so such things are not unheard of. And the above solution would be far gentler than China because it promises (a) generous welfare and (b) total childbearing freedom, if you just become independent.

In an American where greater than 50% of babies now depend on WIC, it seems the trajectory has to change one way or another. If America continues in its present trajectory, Obama's historical legacy will be overwhelmingly negative.

Anonymous said...

Assume you harvest 20 eggs from a woman and fertilize them with the husband's sperm. Grow them to roughly the 128 cell stage, grab off a couple of cells, then freeze the zygote. Now, examine your cells for, say 500 traits, then choose the optimal zygote.

The "optimal zygote" in your scenario is optimized only within the constraints of the parents' suite of alleles. If they both possess below average global phenotypes, one would expect that they have fewer optimized loci than individuals with above average global phenotypes. So, yes, one could improve their offspring in this manner, but it wouldn't necessarily result in drastic improvement. There will still be many loci at which the "enhanced" offspring are not optimized.

Anonymous said...

Sorry to drop my pearl of wisdom ;-) then cut out for several days...

I know little about genetics. I simply offer the purely empirical observation that regression to the mean exists for IQ. I don’t understand the details, and I doubt anyone does at this time. I'm not sure it matters for this discussion (though the more complex it is, the more the central limit theorem will apply).

If I understand the data correctly, what we observe empirically is that parents of IQs {gM,gF} will produce children with IQ that's a random variable gC with mean of

(A(gM+gF)/2 + B x populationmean) / (A+B)
where A is perhaps a bit larger than B

and a standard deviation of about 12.5. (I had just assumed it was 15, but apparently I was wrong.)

I agree with the anon geneticist. I'm not saying that two dullards will produce an Aristotle, any more than I'm saying that you could roll two dice and get a 20. But if you roll two dice twenty times, it is likely that the same dice that I rolled for a 4 will give you at least one 11. And one is all it takes.

--Daemon

Anonymous said...

" agree with the anon geneticist. I'm not saying that two dullards will produce an Aristotle, any more than I'm saying that you could roll two dice and get a 20. But if you roll two dice twenty times, it is likely that the same dice that I rolled for a 4 will give you at least one 11. And one is all it takes."

Quite.

If you accept that at least practically speaking IQ measures ability to function in the modern world then getting 85s to 95 is a very big deal in terms of overall social prosperity.