4 ms·

Also surprising to many westerners is the extent to which Chine encourages and provides the opportunity for folks to volunteer for a eugenics programs: https://

by andreyf 10y ago

Also surprising to many westerners is the extent to which Chine encourages and provides the opportunity for folks to volunteer for a eugenics programs: https://www.edge.org/response-detail/23838 https://www.edge.org/response-detail/23838

gwern 10y ago
Miller overstates his case. Aside from the total failure of the mentioned BGI project and it being beaten to the punch by the first IQ GWAS hits being released by the Western SSGAC consortium that very year (Rietveld et al 2013), Westerners practice a good deal of that 'Chinese eugenics' (look at Down rates, assortative mating, fertility rates by education level, and the practices of sperm & egg banks), and his (and Lynn's) estimate of the benefits of embryo selection of being 5-15 IQ points is considerably off because the upper bound set by heritability & average number of embryos in IVF is ~9 points and then half of them are lost to various steps, so even universal use of IVF would not deliver more than maybe 4.5 IQ points (see https://www.gwern.net/Embryo%20selection https://www.gwern.net/Embryo%20selection for my more detailed analysis of limits). I don't expect Chinese women to use IVF much more than Western women - there are limits to the tiger mother thing. (CRISPR, likewise, is greatly overrated.) The real discontinuous changes would come from technologies further down the pipeline: induced pluripotency to allow creation of hundreds of embryos to select from, iterated embryo selection, and genome synthesis.
- defen 10y ago
  > even universal use of IVF would not deliver more than maybe 4.5 IQ points That still doubles the number of people in your population with a 130+ IQ. Assuming baseline 100 mean and 15 std dev.
  - gwern 10y ago
    Sure (I include a table of the consequences on the tail for various scenarios), but consider that the Chinese population is already more than twice the size of the US population and available measurements suggest it may have a higher starting mean as well, but that has not had a huge impact. Why would another doubling be expected to? (The US could probably easily double its 130+ numbers if it wanted to just by changing immigration policies.) At the very least, it can't be expected to have the impact of anything like an IES/synthesis scenario where the gains would be anywhere upwards of +100 points.
    - defen 10y ago
      Oh yeah, a gain of +100 points is just insane and it would be difficult to predict what that society would even be like. I was just trying to suggest that +4.5 to the mean might also have potentially big effects. As for why China isn't farther ahead - IQ isn't everything. They modernized pretty late in the game, and operated under a profoundly stupid economic system for a good part of the 20th century. I believe that their current government is pretty corrupt, but I'm far from an expert.

by andreyf 10y ago

gwern 10y ago

Miller overstates his case. Aside from the total failure of the mentioned BGI project and it being beaten to the punch by the first IQ GWAS hits being released by the Western SSGAC consortium that very year (Rietveld et al 2013), Westerners practice a good deal of that 'Chinese eugenics' (look at Down rates, assortative mating, fertility rates by education level, and the practices of sperm & egg banks), and his (and Lynn's) estimate of the benefits of embryo selection of being 5-15 IQ points is considerably off because the upper bound set by heritability & average number of embryos in IVF is ~9 points and then half of them are lost to various steps, so even universal use of IVF would not deliver more than maybe 4.5 IQ points (see https://www.gwern.net/Embryo%20selection https://www.gwern.net/Embryo%20selection for my more detailed analysis of limits). I don't expect Chinese women to use IVF much more than Western women - there are limits to the tiger mother thing. (CRISPR, likewise, is greatly overrated.) The real discontinuous changes would come from technologies further down the pipeline: induced pluripotency to allow creation of hundreds of embryos to select from, iterated embryo selection, and genome synthesis.

defen 10y ago
> even universal use of IVF would not deliver more than maybe 4.5 IQ points That still doubles the number of people in your population with a 130+ IQ. Assuming baseline 100 mean and 15 std dev.
- gwern 10y ago
  Sure (I include a table of the consequences on the tail for various scenarios), but consider that the Chinese population is already more than twice the size of the US population and available measurements suggest it may have a higher starting mean as well, but that has not had a huge impact. Why would another doubling be expected to? (The US could probably easily double its 130+ numbers if it wanted to just by changing immigration policies.) At the very least, it can't be expected to have the impact of anything like an IES/synthesis scenario where the gains would be anywhere upwards of +100 points.
  - defen 10y ago
    Oh yeah, a gain of +100 points is just insane and it would be difficult to predict what that society would even be like. I was just trying to suggest that +4.5 to the mean might also have potentially big effects. As for why China isn't farther ahead - IQ isn't everything. They modernized pretty late in the game, and operated under a profoundly stupid economic system for a good part of the 20th century. I believe that their current government is pretty corrupt, but I'm far from an expert.