Supplemental questions that I don’t know the answers to: how significant are the effects of inbreeding? Are they often so bad that it would be better for an inbred child to never have existed? How does that compare to having non-inbred children with known high risks of genetic defects? To what extent can the effects be tested for (both before and during pregnancy)?
I’d be very surprised if inbreeding was so bad that careful consensual incestuous sex, with the intent of getting an abortion if pregnancy does occur, wasn’t worth the risk.
Edit: Okay, inbreeding seems to be much worse than I’d anticipated.
A sibling-incest child looses 28 IQ points. The risks of all genetic disorders rises massively, from p^2/4 to about p/8, so if the prevalence of carriers of a recessive disease (p) is 1%, then the disease probability would rise by a factor of 50.
That’s almost two standard deviations and looks iffy to me. Yes, I followed the link, the main study resulting in this number is behind the paywall, but I suspect that the sample wasn’t very representative.
Basically, for a sub-population with a recessive trait that leads to mental retardation the outcomes are going to be massively different from the outcomes for a sub-population without such a recessive trait.
I’m impressed she managed to get that big a sample in just Czechoslovakia.
Her sample covers 37 years (1933-1970) and, looking at the sources, she utilized official reports including maternity homes, district courts, etc. so the sample is all Czechoslovakia had officially.
What makes you think there are any human populations none of whose recessive and mutation loads affect cognition?
I don’t think this. Intelligence is strongly polygenic, as far as I know, so for everyone it’s a mix of something good and something bad. But that gene mix is uneven in populations, so some people (low-IQ) get more bad and less good genes, while some people (high-IQ) get more good and less bad.
The thing is, I would expect incest to strongly correlate with very low IQ (your basic drives are still there, but social norms are… less binding). Besides, it’s easier for smarter people to not be caught. So if you select a population which engages in incest (and is detected), you are co-selecting for low IQ and for a larger proportion of bad-for-IQ genes. And the larger that proportion, the worse are the chances (growing superlinearly, too) for the child to have normal IQ.
Do note that in the study sample only 4 females and 2 males among parents attended secondary school, the rest didn’t have any education beyond elementary school (out of 141 mothers and 138 fathers).
Do note that in the study sample only 4 females and 2 males among parents attended secondary school, the rest didn’t have any education beyond elementary school (out of 141 mothers and 138 fathers).
Poor countries are like that. The people in the Indian studies and elsewhere won’t be too highly educated either.
So if you select a population which engages in incest, you are co-selecting for low IQ and so for a larger proportion of bad-for-IQ genes. And the larger that proportion, the worse are the chances (growing superlinearly, too) for the child to have normal IQ.
Shouldn’t affect within-population comparisons… Although since prevalence of cousin-marriage differs drastically from country to country, the inbreeding effect could be driving a nontrivial amount of between-population differences in intelligence. (And of course, it’s not like intelligence is unrelated to national wealth either.)
This is Europe, though, and Communist governments tend to be big on education.
Shouldn’t affect within-population comparisons.
Within which population? The control group involves one parent “from the outside”, so regression to the mean kicks in and the chance of the recessives finding a pair falls dramatically.
I am not arguing that incest has no significant consequences. I am arguing that if you take children of incestuous unions where both parents have reasonable IQ (say, >85), the mean IQ of children would NOT drop by 28 points.
This is Europe, though, and Communist governments tend to be big on education.
It is one of the poorest parts of Europe, and Communist governments tended to be big on a lot of things they couldn’t deliver.
Within which population? The control group involves one parent “from the outside”, so regression to the mean kicks in and the chance of the recessives finding a pair falls dramatically.
If you’re comparing within an Indian population, then the much higher rates of inbreeding aren’t the confound; all you have is the remaining selection effect, and that’s must be small because anything else would drastically contradict the animal and other breeding experiments, and the estimates from genomic methods.
I am arguing that if you take children of incestuous unions where both parents have reasonable IQ (say, >85), the mean IQ of children would NOT drop by 28 points.
It would probably drop by more like 25 points, looking at the weighted averages. (For the surviving children, that is.)
It is one of the poorest parts of Europe, and Communist governments tended to be big on a lot of things they couldn’t deliver.
Communist governments delivered on that one.
Take a look here, specifically pages 21 and 23. The secondary education in Eastern Europe was more prevalent than in Mediterranean countries and Great Britain + Ireland (but less than in Nordic countries and Central Europe). And Czechoslovakia was one of the better Eastern European countries.
If you’re comparing within an Indian population
I went and looked at the Jammu & Kashmir study and it is more convincing than the Czechoslovak study. Hm. It seems my scepticism about the 20+ IQ points drop was unfounded, changing my mind… :-)
But why did my intuition didn’t like the large magnitude of IQ drop? I think because it implies that intelligence is very fragile and very easy to genetically screw up. But if the IQ drop is valid, then intelligence is very fragile. Hmm...
There are two separate questions here. One is: for a given pair of closely related people who very much want to have sex with one another, is doing so (carefully) worth the risk? The other is: should we adjust our societal norms to make things easier for people in that situation?
It seems quite plausible to me that the answers might be “yes, sure” and “heck no”, respectively because, as you say, if lots of siblings or other closely related people have sex then some of them will have children.
Slightly-parallel question: if someone is addicted to heroin and can procure some, should they take it? The answer might be yes, at least some of the time, but we probably still want norms that discourage people from getting addicted in the first place.
(I wonder whether laws and other norms against incest provide some protection against abuse by parents and elder siblings. That shouldn’t be necessary—they should be protected by laws against abuse and against sex with people almost certainly too young for properly informed consent—but maybe there’s some extra deterrent effect.)
Professor Alan Bittles, director for the centre for human genetics in Perth, Australia has collated data on infant mortality in children born within first-cousin marriages from around the world and found that the extra increased risk of death is 1.2%.
In terms of birth defects, he says, the risks rise from about 2% in the general population to 4% when the parents are closely related.
However, cousin marriages might typically be in cultures which accept or promote cousin marriages. This might not work out the same way for pairings from non-cousin marriage cultures.
Supplemental questions that I don’t know the answers to: how significant are the effects of inbreeding? Are they often so bad that it would be better for an inbred child to never have existed? How does that compare to having non-inbred children with known high risks of genetic defects? To what extent can the effects be tested for (both before and during pregnancy)?
I’d be very surprised if inbreeding was so bad that careful consensual incestuous sex, with the intent of getting an abortion if pregnancy does occur, wasn’t worth the risk.
Edit: Okay, inbreeding seems to be much worse than I’d anticipated.
According to this:
http://dare.uva.nl/cgi/arno/show.cgi?fid=152307
A sibling-incest child looses 28 IQ points. The risks of all genetic disorders rises massively, from p^2/4 to about p/8, so if the prevalence of carriers of a recessive disease (p) is 1%, then the disease probability would rise by a factor of 50.
That’s almost two standard deviations and looks iffy to me. Yes, I followed the link, the main study resulting in this number is behind the paywall, but I suspect that the sample wasn’t very representative.
Basically, for a sub-population with a recessive trait that leads to mental retardation the outcomes are going to be massively different from the outcomes for a sub-population without such a recessive trait.
Available the usual places: https://www.dropbox.com/s/8gv0el0anfyymed/1971-seemanova.pdf / http://moscow.sci-hub.bz/092ee3d082e9cffd04b7064c36ba808a/10.1159%40000152391.pdf (Personally, I’m impressed she managed to get that big a sample in just Czechoslovakia. Creepy. Also, note that’s only one of the studies being meta-analyzed.)
It’s very clear that inbreeding is really bad: to give some examples, it drives species extinct within generations, its effects are long-term and underestimated, inbreeding can result in 20+ IQ points loss in other populations such as in India and of course we all know about the Habsburgs (which was so extreme that “Charles II was moderately more inbred than the average among the offspring from brother-sister matings”).
What makes you think there are any human populations none of whose recessive and mutation loads affect cognition?
Her sample covers 37 years (1933-1970) and, looking at the sources, she utilized official reports including maternity homes, district courts, etc. so the sample is all Czechoslovakia had officially.
I don’t think this. Intelligence is strongly polygenic, as far as I know, so for everyone it’s a mix of something good and something bad. But that gene mix is uneven in populations, so some people (low-IQ) get more bad and less good genes, while some people (high-IQ) get more good and less bad.
The thing is, I would expect incest to strongly correlate with very low IQ (your basic drives are still there, but social norms are… less binding). Besides, it’s easier for smarter people to not be caught. So if you select a population which engages in incest (and is detected), you are co-selecting for low IQ and for a larger proportion of bad-for-IQ genes. And the larger that proportion, the worse are the chances (growing superlinearly, too) for the child to have normal IQ.
Do note that in the study sample only 4 females and 2 males among parents attended secondary school, the rest didn’t have any education beyond elementary school (out of 141 mothers and 138 fathers).
P.S. Thanks for the link to the study.
Poor countries are like that. The people in the Indian studies and elsewhere won’t be too highly educated either.
Shouldn’t affect within-population comparisons… Although since prevalence of cousin-marriage differs drastically from country to country, the inbreeding effect could be driving a nontrivial amount of between-population differences in intelligence. (And of course, it’s not like intelligence is unrelated to national wealth either.)
This is Europe, though, and Communist governments tend to be big on education.
Within which population? The control group involves one parent “from the outside”, so regression to the mean kicks in and the chance of the recessives finding a pair falls dramatically.
I am not arguing that incest has no significant consequences. I am arguing that if you take children of incestuous unions where both parents have reasonable IQ (say, >85), the mean IQ of children would NOT drop by 28 points.
It is one of the poorest parts of Europe, and Communist governments tended to be big on a lot of things they couldn’t deliver.
If you’re comparing within an Indian population, then the much higher rates of inbreeding aren’t the confound; all you have is the remaining selection effect, and that’s must be small because anything else would drastically contradict the animal and other breeding experiments, and the estimates from genomic methods.
It would probably drop by more like 25 points, looking at the weighted averages. (For the surviving children, that is.)
Communist governments delivered on that one.
Take a look here, specifically pages 21 and 23. The secondary education in Eastern Europe was more prevalent than in Mediterranean countries and Great Britain + Ireland (but less than in Nordic countries and Central Europe). And Czechoslovakia was one of the better Eastern European countries.
I went and looked at the Jammu & Kashmir study and it is more convincing than the Czechoslovak study. Hm. It seems my scepticism about the 20+ IQ points drop was unfounded, changing my mind… :-)
But why did my intuition didn’t like the large magnitude of IQ drop? I think because it implies that intelligence is very fragile and very easy to genetically screw up. But if the IQ drop is valid, then intelligence is very fragile. Hmm...
There are two separate questions here. One is: for a given pair of closely related people who very much want to have sex with one another, is doing so (carefully) worth the risk? The other is: should we adjust our societal norms to make things easier for people in that situation?
It seems quite plausible to me that the answers might be “yes, sure” and “heck no”, respectively because, as you say, if lots of siblings or other closely related people have sex then some of them will have children.
Slightly-parallel question: if someone is addicted to heroin and can procure some, should they take it? The answer might be yes, at least some of the time, but we probably still want norms that discourage people from getting addicted in the first place.
(I wonder whether laws and other norms against incest provide some protection against abuse by parents and elder siblings. That shouldn’t be necessary—they should be protected by laws against abuse and against sex with people almost certainly too young for properly informed consent—but maybe there’s some extra deterrent effect.)
http://news.bbc.co.uk/2/hi/health/7404730.stm
However, cousin marriages might typically be in cultures which accept or promote cousin marriages. This might not work out the same way for pairings from non-cousin marriage cultures.