9 ms·

I was using "heritable" as a synonym for "genetically determined." Thanks for the clarification. To be a bit more specific, it is obviously possible to environm

by sfblah 4y ago

I was using "heritable" as a synonym for "genetically determined." Thanks for the clarification. To be a bit more specific, it is obviously possible to environmentally lower someone's IQ from their genetic potential. Examples include lead poisoning and denial of basic nutrition. Similarly, you can environmentally prevent someone from being able to run as fast as they might genetically have been able to by injuring their legs in various ways in childhood. in both of those cases, as in all things genetic, expanding the ceiling is a far different matter. There is no known treatment I could have been given to enable me to run a 2 hour marathon. Similarly, there is no known way to cause a 100-IQ person to reliably test at a 140 IQ. That said, progress is being made on artificial intelligence and genetics, so that could all change.

runarberg 4y ago
Now that we agree on what you mean by heritable, can you clarify what you mean by intelligence? In traditional psychometrics intelligence is usually operationally defined in terms of the tests them selves. These tests correlate broadly with SAT scores and we can probably agree that disabilities and environments exists which gives people obvious disadvantage when taking the SATs, and that those people should be accommodated accordingly. Given that you accept the operationally defined term for intelligence, how should we treat group difference in IQ any differently than how we treat (or aspire to treat) group difference in SATs? In other words, why should we care about IQ at all?
- sfblah 4y ago
  Good questions. I believe it is worth studying and debating the definition of intelligence. I believe modern IQ tests do a reasonable job of crystallizing a value which is both meaningful and useful. In general, the goal is to quantify an individual's mental potential, which might be understood as their processor speed, amount of RAM in their brain, pre-programmed learning algorithm, and so forth. As with any concept in social sciences it's difficult to define it perfectly, but I believe it's worth the effort. As for why we should care? 1. Western societies spend incredible amounts of resources on programs to equalize opportunity for people. Large fractions of this money in most western nations are allocated based on what I view to be a misunderstanding of what enables a person to prosper financially or otherwise. Examples include universal preschool and the like. Accepting the reality of IQ would assist in eliminating ineffective programs and allocating capital more effectively. 2. The quasi-religious doctrine of universal racism is an existential risk to modern liberal democracy, and it is built fundamentally on the false premise that IQ is environmental or doesn't exist. Eliminating that premise would allow consideration of other explanations for differential outcomes between groups that don't rely on accusations of hatred and bigotry. 3. Major research institutions across America and abroad, as well as large corporations, risk decreasing the rate of technological and economic progress by implementing programs built on a quasi-religious premise that people are treated vastly unequally for reasons having nothing to do with their capability. Accepting IQ as a valid science and explanation would not only de-taboo important potential areas of research, it would eliminate costly and wasteful efforts which will never bear fruit, as they seek to address problems which don't even exist. 4. An incorrect understanding of the causes behind achievement gaps provides onramps to corrupt political participants who take advantage of the basic logical fallacy that "false statements can imply anything" to redirect funding and attention away from pressing issues. Concretely, this results in death and violence resulting from failure to properly apply the law and failure to differentiate between what is true and what is false. "Without vision, a people perish." 5. Acceptance of IQ enables creation of tiered magnet schools, trade schools and other opportunities which benefit all of society, particularly those who are at the bottom of the economic ladder but who test high. Instead, at least in America, the trend is toward the elimination of all magnet programs because they are "racist." The harms here are obvious, and they all stem from an unwillingness to even consider that IQ exists and might be genetic. 6. Refusal to discuss IQ leads to it being something that is whispered about behind closed doors, but cannot be discussed openly. This directly harms those who are often (and often wrongly) associated with groups that might be thought to have lower IQs. Bringing the reality of IQ into the open enables a real conversation about it, and it allows confronting racists on much more solid ground, since the debate is no longer undergirded with taboos and untruths. 7. Acceptance of IQ allows targeting social programs toward the truly needy, accounting for the reality that in modern human life, IQ is the variable that best correlates with material success. This would eliminate injustices like affirmative action programs going to wealthy immigrants from other nations instead of those truly in need. It levels the playing field and could serve as a basis for rational policy such as targeted universal basic income. 8. AI research will ultimately force us to accept IQ as largely genetic. Accepting IQ now will avoid some sort of schism in the coming decades over this question, and allow a more rational discourse over AI's role in our lives and how to best harness it and avoid its risks. Is that sufficient?
  - runarberg 4y ago
    I’m gonna merge the thread from https://news.ycombinator.com/item?id=35518975 https://news.ycombinator.com/item?id=35518975 where you said: > * Encouraging higher IQ parents to have more children through subsidies, thereby increasing the likelihood of future technological innovations assisting all mankind. This is a schoolbook example of a eugenicist argument. Are you sure you want to go there? I’m gonna assume that you meant something different than what I took from this, and you are not a eugenicists. However I do want to speak a little about why eugenicists are wrong. The problem with trying to select for IQ is that there is no broad consensus about what constitutes a good intelligent trait. There are thousands of traits that work differently in thousands of situations. When there is no consensus, somebody must take authority. And if that authority has nefarious reasons, than we are pretty screwed. More likely no such trait, nor sets of traits exist. As Charles Darwin him self demonstrated, there is value in diversity. By not selecting for a single trait, and allowing our cognitive abilities to vary as much as possible, there is way more chances of something beneficial to our species to develop and be selected for. Now this doesn’t mean we shouldn’t accommodate for disabilities. However it isn’t 1908 any more. We have better tests and metrics to detect disabilities, and we have better means of accommodating. IQ has run its course. We don’t need it anymore. The science has advanced, and the theory is obsolete. And this ties to the above post, and other posts you’ve done in this tree. You seem to think there is some scientific is pretty sound and there is some suppression going on: > 6. Refusal to discuss IQ leads to it being something that is whispered about behind closed doors, but cannot be discussed openly. This directly harms those who are often (and often wrongly) associated with groups that might be thought to have lower IQs. Bringing the reality of IQ into the open enables a real conversation about it, and it allows confronting racists on much more solid ground, since the debate is no longer undergirded with taboos and untruths. If you think that, you’d be wrong. Modern psychology looks for behavior, if a model predicts behavior, your pretty good. IQ does no such thing. So perhaps it is sociology, not psychology and is looking for population statistics, well... IQ is pretty bat at that too. There are number of other metrics which does that better. The simplest one being school grades. And to apply Occam’s razor, why invent an entire new construct intelligence when grades suffice. We don’t not discuss IQ. It has been done extensively for well over a century now. and IQ has simply lost to better theories. There is no taboo, just sore loosers that keep on and on about a theory that holds no relevance any more. Just look at The Bell Curve it’s been almost 30 years and we are still talking about it, despite it being a schoolbook example of how to do bad science. We finished talking about N-rays and phrenology a long time ago, it is time we do so with IQ as well. IQ is dead, it was never anything more than a bad theory which took way too long formally die.
tptacek 4y ago
I don't understand how your logic holds together once you concede that heritability doesn't mean "genetically determined". This is the problem with lots of arguments about "heritability" of IQ; heritability is simply the ratio of genetic effects to the total variance in a population. If we're asking whether IQ is genetic, then saying that something is "highly heritable" is simply restating the question. (I find it helpful to remember that lipstick-wearing is highly heritable despite zero genetic determination, and number of toes isn't very heritable at all despite total genetic determination.) In particular, you can't convincingly go from discussing the seemingly profound effects acknowledging IQ heritability will have on racism, in order to avoid "demon-haunted" religious arguments, to a shrug and a handwave about genotypic (and epigenetic) effects vs. environmental effects. With due respect to your good-faith attempt to establish a logical baseline to this whole situation, the demons are haunting your argument, not those of people who'd push back on it. It's an earlier rebuttal and you can find more precise and current ones now, but Ned Block's heritability piece is a good starting point for this stuff (you can just Google for "ned block heritable", the SERP will be dozens of links to it.)
- sfblah 4y ago
  Your first sentence is a complete misread of my comment, so I'm not going to reply. I said exactly the opposite of what you claim I said. Please re-read what I wrote and try again. Edit: I read a bit of Ned Block's piece. I believe he more or less immediately dismisses my view as "Extreme geneticism". While I don't agree that different groups in a given country grow up in the same circumstances, I also don't believe this matters much, if at all. If life teaches anything, it's that it's incredibly difficult to construct a machine that absolutely prevents the success of one group or another. Life is diverse and adaptable and finds a way. There are plenty of examples of groups historically facing terrible environments who nonetheless prospered. Block also complains about a lack of sufficient data on the question of IQ. Fine, but one must immediately ask why there is such a lack of data. And that returns us to the demon-haunted world of quasi-religious opposition to such research. Bottom line is, if you want to argue for the Rube Goldberg machine of -isms that produce differential outcomes in groups, prove it. I believe that, just as in other areas of human endeavor such as athletics, the prior should be that genetics are the most important input.

by sfblah 4y ago

runarberg 4y ago

Now that we agree on what you mean by heritable, can you clarify what you mean by intelligence? In traditional psychometrics intelligence is usually operationally defined in terms of the tests them selves. These tests correlate broadly with SAT scores and we can probably agree that disabilities and environments exists which gives people obvious disadvantage when taking the SATs, and that those people should be accommodated accordingly. Given that you accept the operationally defined term for intelligence, how should we treat group difference in IQ any differently than how we treat (or aspire to treat) group difference in SATs? In other words, why should we care about IQ at all?

sfblah 4y ago
Good questions. I believe it is worth studying and debating the definition of intelligence. I believe modern IQ tests do a reasonable job of crystallizing a value which is both meaningful and useful. In general, the goal is to quantify an individual's mental potential, which might be understood as their processor speed, amount of RAM in their brain, pre-programmed learning algorithm, and so forth. As with any concept in social sciences it's difficult to define it perfectly, but I believe it's worth the effort. As for why we should care? 1. Western societies spend incredible amounts of resources on programs to equalize opportunity for people. Large fractions of this money in most western nations are allocated based on what I view to be a misunderstanding of what enables a person to prosper financially or otherwise. Examples include universal preschool and the like. Accepting the reality of IQ would assist in eliminating ineffective programs and allocating capital more effectively. 2. The quasi-religious doctrine of universal racism is an existential risk to modern liberal democracy, and it is built fundamentally on the false premise that IQ is environmental or doesn't exist. Eliminating that premise would allow consideration of other explanations for differential outcomes between groups that don't rely on accusations of hatred and bigotry. 3. Major research institutions across America and abroad, as well as large corporations, risk decreasing the rate of technological and economic progress by implementing programs built on a quasi-religious premise that people are treated vastly unequally for reasons having nothing to do with their capability. Accepting IQ as a valid science and explanation would not only de-taboo important potential areas of research, it would eliminate costly and wasteful efforts which will never bear fruit, as they seek to address problems which don't even exist. 4. An incorrect understanding of the causes behind achievement gaps provides onramps to corrupt political participants who take advantage of the basic logical fallacy that "false statements can imply anything" to redirect funding and attention away from pressing issues. Concretely, this results in death and violence resulting from failure to properly apply the law and failure to differentiate between what is true and what is false. "Without vision, a people perish." 5. Acceptance of IQ enables creation of tiered magnet schools, trade schools and other opportunities which benefit all of society, particularly those who are at the bottom of the economic ladder but who test high. Instead, at least in America, the trend is toward the elimination of all magnet programs because they are "racist." The harms here are obvious, and they all stem from an unwillingness to even consider that IQ exists and might be genetic. 6. Refusal to discuss IQ leads to it being something that is whispered about behind closed doors, but cannot be discussed openly. This directly harms those who are often (and often wrongly) associated with groups that might be thought to have lower IQs. Bringing the reality of IQ into the open enables a real conversation about it, and it allows confronting racists on much more solid ground, since the debate is no longer undergirded with taboos and untruths. 7. Acceptance of IQ allows targeting social programs toward the truly needy, accounting for the reality that in modern human life, IQ is the variable that best correlates with material success. This would eliminate injustices like affirmative action programs going to wealthy immigrants from other nations instead of those truly in need. It levels the playing field and could serve as a basis for rational policy such as targeted universal basic income. 8. AI research will ultimately force us to accept IQ as largely genetic. Accepting IQ now will avoid some sort of schism in the coming decades over this question, and allow a more rational discourse over AI's role in our lives and how to best harness it and avoid its risks. Is that sufficient?
- runarberg 4y ago
  I’m gonna merge the thread from https://news.ycombinator.com/item?id=35518975 https://news.ycombinator.com/item?id=35518975 where you said: > * Encouraging higher IQ parents to have more children through subsidies, thereby increasing the likelihood of future technological innovations assisting all mankind. This is a schoolbook example of a eugenicist argument. Are you sure you want to go there? I’m gonna assume that you meant something different than what I took from this, and you are not a eugenicists. However I do want to speak a little about why eugenicists are wrong. The problem with trying to select for IQ is that there is no broad consensus about what constitutes a good intelligent trait. There are thousands of traits that work differently in thousands of situations. When there is no consensus, somebody must take authority. And if that authority has nefarious reasons, than we are pretty screwed. More likely no such trait, nor sets of traits exist. As Charles Darwin him self demonstrated, there is value in diversity. By not selecting for a single trait, and allowing our cognitive abilities to vary as much as possible, there is way more chances of something beneficial to our species to develop and be selected for. Now this doesn’t mean we shouldn’t accommodate for disabilities. However it isn’t 1908 any more. We have better tests and metrics to detect disabilities, and we have better means of accommodating. IQ has run its course. We don’t need it anymore. The science has advanced, and the theory is obsolete. And this ties to the above post, and other posts you’ve done in this tree. You seem to think there is some scientific is pretty sound and there is some suppression going on: > 6. Refusal to discuss IQ leads to it being something that is whispered about behind closed doors, but cannot be discussed openly. This directly harms those who are often (and often wrongly) associated with groups that might be thought to have lower IQs. Bringing the reality of IQ into the open enables a real conversation about it, and it allows confronting racists on much more solid ground, since the debate is no longer undergirded with taboos and untruths. If you think that, you’d be wrong. Modern psychology looks for behavior, if a model predicts behavior, your pretty good. IQ does no such thing. So perhaps it is sociology, not psychology and is looking for population statistics, well... IQ is pretty bat at that too. There are number of other metrics which does that better. The simplest one being school grades. And to apply Occam’s razor, why invent an entire new construct intelligence when grades suffice. We don’t not discuss IQ. It has been done extensively for well over a century now. and IQ has simply lost to better theories. There is no taboo, just sore loosers that keep on and on about a theory that holds no relevance any more. Just look at The Bell Curve it’s been almost 30 years and we are still talking about it, despite it being a schoolbook example of how to do bad science. We finished talking about N-rays and phrenology a long time ago, it is time we do so with IQ as well. IQ is dead, it was never anything more than a bad theory which took way too long formally die.

tptacek 4y ago

I don't understand how your logic holds together once you concede that heritability doesn't mean "genetically determined". This is the problem with lots of arguments about "heritability" of IQ; heritability is simply the ratio of genetic effects to the total variance in a population. If we're asking whether IQ is genetic, then saying that something is "highly heritable" is simply restating the question. (I find it helpful to remember that lipstick-wearing is highly heritable despite zero genetic determination, and number of toes isn't very heritable at all despite total genetic determination.) In particular, you can't convincingly go from discussing the seemingly profound effects acknowledging IQ heritability will have on racism, in order to avoid "demon-haunted" religious arguments, to a shrug and a handwave about genotypic (and epigenetic) effects vs. environmental effects. With due respect to your good-faith attempt to establish a logical baseline to this whole situation, the demons are haunting your argument, not those of people who'd push back on it. It's an earlier rebuttal and you can find more precise and current ones now, but Ned Block's heritability piece is a good starting point for this stuff (you can just Google for "ned block heritable", the SERP will be dozens of links to it.)

sfblah 4y ago
Your first sentence is a complete misread of my comment, so I'm not going to reply. I said exactly the opposite of what you claim I said. Please re-read what I wrote and try again. Edit: I read a bit of Ned Block's piece. I believe he more or less immediately dismisses my view as "Extreme geneticism". While I don't agree that different groups in a given country grow up in the same circumstances, I also don't believe this matters much, if at all. If life teaches anything, it's that it's incredibly difficult to construct a machine that absolutely prevents the success of one group or another. Life is diverse and adaptable and finds a way. There are plenty of examples of groups historically facing terrible environments who nonetheless prospered. Block also complains about a lack of sufficient data on the question of IQ. Fine, but one must immediately ask why there is such a lack of data. And that returns us to the demon-haunted world of quasi-religious opposition to such research. Bottom line is, if you want to argue for the Rube Goldberg machine of -isms that produce differential outcomes in groups, prove it. I believe that, just as in other areas of human endeavor such as athletics, the prior should be that genetics are the most important input.