I don't want to be rude, but this appears to be just shoddy overuse of rationalist lingo in the name of shoehorning a myopic and empirically unsupported political agenda into the consequentialist framework.
What observed empirical effects? You link to a very strange post saying, concretely, that
This person has had a falling-out with their friends who believe HBD, apparently because they have come to harbor other right-wing ideas poorly compatible with aspects of this person's identity and lifesyle.
Those friends had drifted to the right because they felt persecuted "by people on the left or center-left" due to them believing HBD.
This person had concluded that HBD is pseudoscientific, by virtue of right-wingers being nasty to trans people and vegans.
Pardon me, what? Is this your evidence base?
№1-2 might as well be considered arguments for lesser demonization of HBD. There is nothing inherently political about thinking one way or another about sources of cognitive differences; the political valence is imposed on such hypotheses by external forces. If smart people independently arrive at HBD as a morally neutral explanation for generally available observations, then it's not very prudent on part of "the left or center-left" to baselessly label them racists, supporters of genocidal far-right ideologies, insane cranks and such and leave them no choice except break their own minds into an Orwellian mold, learn to live in falsehood, or go rightward. When they say, like Bostrom, that they are motivated by humanitarian impulses, they can be taken at their word.
You, however, seem to conclude that the only problem is insufficient intensity of vilification of HBD, now as a "cause area" unto itself; that these people can be intimidated into not believing what they see, through pure peer pressure and pushing the topic to the fringe instead of rational persuasion.
№3 is honestly horrifying in terms of epistemic integrity. You seem to be dismissive of truth as a terminal value, so let's put it like this: a person who sees nothing wrong with such pseudoreasoning – and, given the score, that's normal on EA forum –can delude oneself into excusing arbitrary atrocities; or less dangerously, draining resources into arbitrarily ineffective causes just to feel good about oneself.
I haven't previously heard anyone in EA say that it's vital for our epistemic integrity to freely discuss infohazards
Far-right ideas have created enormous suffering over the past few centuries. As far as I know, we don't have a great theory for how this happened. But it seems fairly clear that it has something to do with memetics
We don't have a good theory, in part, because there's no meaningful way to lump together "far-right ideas" over "the past few centuries", or indeed seriously analyze anything prior to the 20th century through these lens. Do you mean Jacobites or Bourbons by far-right? Why not address la Terreur as an archetypal case of the idea of egalitarianism causing mass death and suffering in the characteristic manner of an infohazard? Should this make us suspicious of egalitarian ideation in general?
Here's a honest thought: the notion of "memetics" or "infohazards" is an infohazard in its own right. It's bad philosophy, and it offers zero explanatory power over traditional terms like "undeservedly popular idea", "misleading idea" or "dangerous idea" but it gives the false impression of such adjectives having been substantiated. It's just a way of whitewashing a classical illiberal and, indeed, totalitarian belief that some ideas must be kept away from the plebeians because they are akin to a plague. In illiberal societies those are "democracy" and "independent thought"; we have a consensus that theories justifying restriction of access to those are vacuous and evil, but those theories at least had some substance, unlike equivocation here about suffering caused by "far right" and, by an entirely frivolous extension, HBD.
In sum, analogizing ideas and their bearers to infectious agents invading and spreading within the body politic is a staple of far-right sociology that exploits deep-seated reactions of disease-associated disgust, fear and distrust of outsiders, and that's all there is to "memetics" in such colloquial use. Perhaps you could do without resorting to such tools for thought.
It would be much better if we had a robust, principled method to guard against harms from far-right ideas.
Perhaps there is, and it's called "law" and "democracy", and you need to argue in a principled way for your cost-benefit analysis that concludes that extant legal and political checks against far-right threats are insufficient, and concludes with embracing of some of the worst totalitarian legacies to ostracize an apparent scientific truth.
But can you be trusted to actually think that, given what you say about utility of public admission of opinions in question? For an external observer, it's a coin toss. And the same for the entirety of your reasoning. As an aside, I'd be terrified of a person who can willfully come to believe – or go through the motions of believing – what he or she believes to be morally prudent but epistemically wrong. Who knows what else can get embedded in one's mind in this manner.
I don't understand how this can happen on the EA Forum. Why would anyone believing in this and wanting to do good promote this?
Well, consider that, as it tends to happen in debates, people on the other side may be as perfectly sure about you being misguided and promoting harmful beliefs as you are about them; and that your proud obliviousness with regard to their rationale doesn't do your attempt at persuasion any more good than your unwillingness to debate the object level does.
Consider, further, that your entire model of this problem space really could be wrong and founded on entirely dishonest indoctrination, both about the scholarly object level and about social dynamics and relative effects of different beliefs.
Finally, consider that some people just have a very strong aversion to the idea that a third party can have the moral and intellectual authority to tell them which thoughts are infohazards. If nothing else, that could help you understand how this can happen.
If you want to do good, why would you ever, in our world, spread these views?
Personally – because I do, in fact, believe that you are profoundly wrong, that even historically these views did not contribute to much harm (despite much misinformation concocted by partisans: policies we know to be harmful are attributable to different systems of views); that, in general, any thesis about systematic relation in the pattern {views I don't like}=>{atrocities} is highly suspect and should be scrutinized (e.g. with theists who attribute Stalin's brutality to atheism, or derive all of morality from their particular religion); and that my views offer a reliable way to reduce the amount of suffering humans are subjected to, in many ways from optimizing allocation of funds to unlocking advances in medical and educational research to mitigating slander and gaslighting heaped upon hundreds of millions of innocent people.
Crucially, because I believe that, all that medium-term cost-benefit analysis aside, the process of maintaining views you assume are beneficial constitutes an X-risk (actually a family of different X-risks, in Bostrom's own classification), by comprehensively corrupting the institution of science and many other institutions. In other words: I think there is no plausible scenario where we achieve substantially more human flourishing in a hundred years – or ever – while deluding ourselves about the blank slate; that it's you who is infecting others with the "Basilisk" thought virus. And that, say, arguments about the terrible history of some tens of thousands of people whom Americans have tortured under the banner of eugenics – after abusing and murdering millions of people whilst being first ignorant, then in denial about natural selection – miss the point entirely, both the point of effective altruism and of rational debate.
If the impact of spreading these views is more tragedies happening, more suffering, and more people dying early, please consider these views an infohazard and don't even talk about them unless you're absolutely sure your views are not going to spread to people who'll become more intolerant- or more violent.
This is an impossible standard and you probably know it. Risks of a given strategy must be assessed in the context of the full universe of its alternatives; else the party that gets to cherrypick which risks are worth bringing up can insist on arbitrary measures. By the way, I could provide nontrivial evidence that your views have contributed to making a great number of people more intolerant and more violent, and have caused thousands of excess deaths over the last three years; but, unlike your wholly hypothetical fearmongering, it's likely to get me banned.
Indeed, I could ask in the same spirit: what makes people upvote you? If your logic of cherrypicking risks and demonizing comparative debate is sound, then why don't they just disregard GiveWell and donate all of their savings to the first local pet shelter that gets to pester them with heart-rending imagery of suffering puppies? Maybe they like puppies to suffer?! This is not just manipulation: rising above such manipulation is the whole conceit of this movement, yet you commit it freely and to popular applause.
To make me or anyone like me change my mind, strong and honest empirical and consequentialist arguments addressing these points are required. But that's exactly what you say is "much less relevant" than just demanding compliance. Well. I beg to differ.
For my part, I do not particularly hope to persuade you or anyone here, and guidelines say we should strive to limit ourselves to explaining the issue. Honestly it's just interesting at this point, can you contemplate the idea of being wrong, not just about "HBD" but about its consequences, or are you the definition of a mindkilled fanatic who can't take a detached view at his own sermon and see that it's heavy on affirmation, light on evidence?
First, that depends on what you mean by "this stuff"; Bird does not study intelligence nor behavioral genetics for a living, he's a plant geneticist. Skewed though the survey may be, it's probably more representative than a single non-expert.
Second, why do you suppose the non-response rate is so high and so skewed? And might it have something in common with your own refusal to continue our conversation on merits of your list?
I suspect that professionals who prefer not to respond, rather than respond in the negative about genetic contributions to the IQ gap, are driven by contradictory impulses: they believe that the evidence doesn't allow for a confident "100% environmental" response and, being scientists, have problem with outright lying, but they also don't want to give the impression of supporting socially unapproved beliefs or "validating" the very inquiry into this topic. So they'd rather wash their hands of the whole issue, and allow their less squeamish colleagues to give the impression of moderate consensus in favor of genetic contribution.
HBD is a hypothesis for how the world works, so the burden of proof is on HBD and giving a bad reason not to believe in HBD is not evidence for HBD.
This logic is only applicable to contrived scenarios where there is no prior knowledge at all – but you need some worldly knowledge to understand what both these hypotheses are about.
Crucially, there is the zero-sum nature of public debate. People deliberately publicizing reasons to not believe some politically laden hypothesis are not random sources of data found via unbiased search: they are expected to cherrypick damning weaknesses. They are also communicating standards of the intellectual tradition that stands by the opposing hypothesis. A rational layman starts with equal uncertainty about truth values of competing hypotheses, but learning that one side makes use of arguments that are blatantly unconvincing on grounds of mundane common sense can be taken as provisional evidence against their thesis even before increasing object-level certainty: poor epistemology is evidence against ability to discover truth, and low-quality cherrypicked arguments point to a comprehensively weak case. Again, consider beliefs generally known to be kooky, and what they bring to bear on the opposition. Their standard of rigor alone is discrediting to what they believe in.
Moreover, I've established that, upon checking, some of your links positively provide evidence in favor of HBD, rather than against – at least by the standard of evidence implicit in the phrasing of the list. Returning again to the Identical Ancestors Point, is presented as an Anti-HBD finding in the first place because it implies a very low prior for genetic divergence of populations, migrations somehow averaging it all out: Human migrations over the last thousand years have been such that literally everyone on Earth is a descendant of literally everyone that lived 7000 years ago whose offspring didn't die out. (Is this the wrong takeaway? What, then, did you mean to say by adding it?) Looking into the actual paper, we see: ...For example, a present-day Norwegian generally owes the majority of his or her ancestry to people living in northern Europe at the IA point, and a very small portion to people living throughout the rest of the world. Furthermore, because DNA is inherited in relatively large segments from ancestors, an individual will receive little or no actual genetic inheritance from the vast majority of the ancestors living at the IA point. Not only does this make the original argument invalid (even in a strong absolute sense – there can be zero common inheritance!) – it directly reinforces the HBD conjecture that long-term (i.e. pre-IAP) divergent local adaptation is relevant to current genetic (and trait) differences.
LW has an empiricist tradition, the term 'rationalist' is a misnomer.
I agree that this is improper and irritating terminology, because doctrinally, LW asserts its allegiance to empiricism, with all the talk about Bayes-updating on evidence and how rationalists must "win". But in practice this isn't so clear-cut: LW is fascinated with armchair thought experiments (that routinely count as evidence to update on), and all the attention devoted to infohazards, Pascal mugging, one-boxing, AI scenarios etc. suggests that they, as a living tradition, are not resilient to speculation the way pure empiricists – say, regular natural scientists – would be. So, not necessarily a misnomer.
the major ethical theories don't have any inconsistencies as far as I can tell
They are internally consistent, but I think the point of ethical theory is to clarify the intuitively knowable essence of moral action for purposes of nontrivial decisionmaking, not to assert what morality is and derive an arbitrary decision rule from there. Utilitarianism is often criticized for things like the repugnant conclusion, yet non-utilitarian ethical theories routinely produce more grating outputs, because they fail to capture the most significant part of intuitive ethics, which is mostly about harm reduction under conditions of resource scarcity. They are less consistent with ethics given to us in lived experience, so to speak.
even if you don't know which way the arrow of causality points, that's still an unnecessarily big risk. It's not particularly altruistic to make statements that have that big a chance of helping racists. You could also spend your time... not doing that.
No, the extent of the purported risk matters. You are just falling back on the unsupported prior about cost-benefit ratio because you have preemptively excluded all factors that may change the qualitative conclusion of "not doing that". To give a specific example: under the assumption that HBD is wrong, we must consider disparate outcomes to be a result of some discrimination and devote resources to alleviate it; but if HBD is actually right, this'd necessarily mean that our costly attempts to help low-performing groups are suboptimal or futile (as in, not effective), and that we will have unfairly attributed blame, harming other groups psychologically and materially. Then there are knock-on effects of harming science: for starters, fears of enabling racists can hold back genomic medicine (and population-specific treatment) by increasing hurdles to data collection and access. We do not have a priori knowledge as to which costs are negligible. On a more meta level, Scott Alexander's parable comes to mind.
IMO it's a weak argument because for all the racism, black Americans still report the highest self-esteem of all racial groups; and theories of stereotype threat are apparently unsupported by high-quality data; so it isn't clear what the odds are that some HBD research or whatever would harm people substantially. But even before that – there are laws against hate speech and discrimination, and they can be strengthened if needed; it seems very suboptimal to focus on not developing neutral knowledge only to deny hateful ideologies rhetorical ammo, instead of dealing with them directly. By the way, cannot racists point to censorshipas sufficient evidence of their correctness, if their intent is to spin available facts to their benefit? Actually, doesn't this enable them to – convincingly – claim that facts are much worse than they are, that the genomic gap in cognitive ability is bigger than the non-zero gap we'd have found (and, I believe, have partially found) with proper research (which is currently prohibited)? And in any case, you have to put racism-driven harms in the context of costs of pretending that HBD is certainly false – that is, under the assumption that we are "just not doing that" and have no clue whether it is or isn't true.
As an aside, I am personally puzzled by the strong conviction of many that HBD becoming common knowledge could lead to normalization of racial discrimination. This is a normative, not scientific question. Societies with Social Darwinist values do not need HBD to embrace and exacerbate the status quo of disparate power. Societies with ethnocentric values opportunistically oppose and exploit ethnic outgroups regardless of relative merit. Mainstream modern value system depends on the premise of human rights, not equality of capability. We do not hold that it is normal to oppress individuals who are known to be below average in some morally neutral trait (except maybe for an expansive definition of "oppress" and clinical issues having to do with lack of legal capacity), we have a strong revulsion to identity-based discrimination, and we understand the unreasonableness of treating individuals on the basis of average values.
Uncharitably, in the case of EA, this concern may have to do with the strain (common to EA and LW) of conspiratorial elitism and distrust in the democratic process, and with the unconscious belief that intelligence does define moral worth. That's ...not a very popular belief. I would deeply hate it if my cognitive betters acted like they have greater moral worth than myself, and therefore, to be fair, I cannot deny equal moral worth to people of lower ability. Most people correctly believe that they aren't brilliant, but they're not so dull as to not arrive at this logic. There are some contingent factors that complicate the picture, but not fundamentally.
Without getting into the weeds of stereotype scholarship, the extent of claimed harms, and the irrational denial of the role of merit in achievement (reasoning in that wiki page doesn't even begin to address what would happen in a "proper" meritocratic society after a few generations, because it is premised on genes not contributing to achievement; this is a typical case of an unexamined anti-HBD prior leading to policy errors)... I'll just say that in my opinion both those issues, insofar as they harm anyone, have to do with beliefs about moral qualities. If "Protestant ethic" is alive and prescribes vilification of people of lower morally neutral ability, then that is a problem in its own right and beyond the scope of this conversation. Luckily, Protestant ethic also encourages treating people on a case by case basis.
What global utility?
Crucially, the search for interventions that actually close the IQ gap. As it stands, we have picked low-hanging fruit like lead exposure, malnutrition, iodine deficiency, parasites and such (in developed nations; I expect EA efforts in Africa to keep delivering on this front), and are left with pursuing dead ends of addressing iniquity like the "food deserts" nonsense, or doubling down on stuff like school spending, that has long ran into diminishing or zero returns and is only popular because to point out its inefficacy means to risk being labeled racist. As Nathan Cofnas argues,
...But the reason that these programs, which Kourany rightly says ought to exist, have never been created is not because of racism but because of the taboo on talking about genetic differences among policy makers. No mainstream politician can acknowledge that there are differences that might call for the creation of a program to “work with the strengths and work on the weaknesses of every [ethnic] group to help make them the very best they can be.” It is hereditarians who have advocated these programs and environmentalists who have resisted them.
Ironically, Cofnas got in trouble for this. If the suppression starts this far upstream from the object level, how can our priors be trusted?
there are other correlations that are both on firmer ground, have more global utility and aren't fulfilling the desires of racists
I sincerely doubt you can prove 1 or 2 (given that your critiques of relevant methodology weren't persuasive), and it looks like assigning any value to 3, on its own, is pure spite that is best left out of effective altruism. Making racists mad is not, in fact, a positive good, fun as it may be.
If you had read my comments
Have read some. I explicitly say I'm addressing the state of discourse here, more than just your comment. I respond to you in particular when I quote specific passages. Sorry if that was unclear.
did support the environmental explantation of the IQ gap
Again, this is not HBD Central, and it is sufficient to establish that there is legitimate uncertainty, so we cannot fall back on the comfortable prior that costs of repudiating HBD are negligible.
Very civil. It will not surprise you to learn that this does not motivate me to keep reading.
Well, I believe that misleading people, and even wasting people's time on true but irrelevant, misleadingly phrased, blatantly misinterpreting and patently false claims is a form of rudeness that's extra obnoxious, because it craftily avoids opprobrium one could earn with trivial show of disrespect. It's not fair to act indignant about an unflattering comparison after doing that. Even so, I've made peace with Brandolini's law, and kept addressing those claims on the object level, to substantiate my "very civil" summary and so that "EAs were to become a little more reserved on this topic, and at least stopped turning off potential recruits with irrefutable displays of irrationality." To be honest your reaction isn't wholly unexpected, but I did hope that I've been polite enough to merit some tolerance.
Yeah I'm out.
OK but please think of your stated desire is to persuade those who happen to believe in HBD to disbelieve it. Obviously you've failed in my case, but I maintain that flaming out like that is detrimental even as far as fence-sitters are concerned. I believe I've provided sufficient receipts for the purpose of showing how your list is inadequate.
It is really not hard to showboat on this topic, by citing from very clearly argued stuff like this or "authoritative" sources like that review or very technical recent papers or just by gesturing in the general direction of environmentalist rhetoric that is... the way I've described, and evident in, e.g., this condemnation of Cofnas, mired in (what I hope is obvious after my initial comment) logical fallacies and half-truths and raw indignation. Or one can just say that if this guy is challenged not by rational and empirical arguments but by being repeatedly called a pseudoscientist and getting a page full of personal attacks on him to the top of search results for his name (a page he responds to with an even pettier page), then he may get a lot of uncomfortable stuff right.
My point is not to showboat but to argue that people who pursue this anti-HBD rhetorical strategy, including you, are probably not succeeding, and are doing the community no favors.
This looks like retconning of history. EA and rationalism go way back, and the entire premise of EA is that determining what makes more good through "rationalist", or more precisely, consequentialist lens is moral. There is no conflict of principles.
The quality of discussion on the value of tolerating Bostrom's (or anyone else's ) opinions on race&IQ is incredibly low, and the discussion is informed by emotion rather than even trivial consequentialist analysis. The failure to approach this issue analytically is a failure both by Rationalist and by old-school EA standards.
This list is a good example of the sort of arguments that look persuasive to those already opposed to HBD, but can push people on the fence towards accepting it, so it may be net-negative from your perspective. This is what has happened to me, and I'll elaborate on why – so that you may rethink your approach, if nothing else.
Disclaimer: I am a non-Western person with few traits worth mentioning. I identify with the rationalist tradition as established on LW, feel sympathy for the ideal of effective altruism, respect Bostrom despite some disagreements, have donated to GiveWell charities on EA advice, but I have not participated more directly. Seeing the drama, people expressing disappointment and threatening to leave the community, and the volume of meta-discussion, I feel like clarifying a few details that may be hard to notice from within your current culture, and hopefully helping you mend the fracture that is currently getting filled with the race-iq stuff.
All else being equal, people who hang around such communities prefer consistent models (indeed, utilitarianism itself is a radical solution to inconsistencies in other ethical theories). This discourse is suffused with intellectual inconsistency, on many levels of varying contentiousness.
On the faint level of moral intuitions, there's the strange beeline from the poorly supported prior that normalization of beliefs like Bostrom's will lead to bad effects like discrimination, to the consequentialist decision against entertaining them. It is not clear that Bostrom's beliefs are harmful in this way, or more likely to encourage a net increase in discrimination than their negation. Arguments from historical precedent have big problems with them: they do not address the direction of causality, or the fact that different cultures can have different reactions to the same information. As it is not considered normal in the modern culture to equate moral worth and ability for individuals of any group, it can be expected that the same will hold should the difference in ability between groups be acknowledged. Arguments from personal distress of users are valid points with regard to community health, but obviously (I hope) incommensurate with the question of global utility, and do not directly weigh on it. So the consequentialist case for not taking Bostrom's belief in good faith is already suspect.
Perhaps the most obvious level is that specific failings Bostrom is credibly accused of (racist attitude, belief in the racial IQ difference, belief in the validity of IQ measurement) do not depend on HBD. (He has done himself no favors by bringing up eugenics). So it's bizarre to see many people denounce his beliefs in toto, but support this denunciation with environmentalist explanations of the IQ gap – in effect, conceding the specific factual claim in Bostrom's old email, or at least demonstrating that it is not beyond the pale by their own standard. To be clear: it is not in doubt that the IQ gap between Black and White Americans exists; and that it is as predictive of outcomes associated with cognitive capacity as IQ measurement is (which is to say, highly predictive – and this, too, is mainstream consensus). People who act indignant about such statements send a huge red flag, demonstrating either general unwillingness to educate themselves or irrational ideological bias on this specific matter. People who bring up irrelevant anti-HBD talking points demonstrate confused reasoning.
Less obviously, the problem is portraying this as an open-and-shut case – a portrayal which doesn't really survive scrutiny. I don't know how to put this nicely, but what your list most reminds me of is polemics of sophisticated Creationists in the heyday of New Atheism. It's a mix of true but irrelevant, misleadingly phrased, blatantly misinterpreting and patently false claims. Instrumentally they are gotchas; structurally, opening moves aimed at people who are not familiar with the debate and are not aware that all those issues had long been answered, and the debate is incredibly mature. Of course, in all such debates both sides can assert that they've solved every vulnerability, and this forum isn't some HBD Central. So I won't compete in citations, and will just address things a total layman, provided he's minimally erudite, napkin-numerate, capable of critical thought and aware of basic logical fallacies, could spot, if he were so inclined. You way "Now you might want to attack one of these (and feel free to send me a message), but even if you're right, that would still leave more than enough reasons to stay away from HBD." What if we go through every one of these?
Human biodiversity is actually pretty low. Homo sapiens has been through a number of bottlenecks. – maybe true but vacuous. "Pretty" low relative to what baseline? How would we even tell – do we have anything like IQ for other species? Does this genetic fact establish some prior for the magnitude in differences in measurable phenotypic traits between groups? What about individuals? What we do know that people with a priori negligible "biodiversity" – as in, children in ethnically endogamous marriages, even in isolated villages – routinely have large differences in all traits of interest. So how much diversity is needed, really, to introduce some measurable population-level divergence? Likewise for the point about bottlenecks, what of it? Should our layman just conclude that this is an authoritative-sounding technical term?
Human migrations over the last thousand years have been such that literally everyone on Earth is a descendant of literally everyone that lived 7000 years ago whose offspring didn't die out. This is known as the Identical Ancestors Point. – grossly misleading/false, and doesn't pass basic sanity check. Is every single unadmixed Indigenous Australian really a descendant of "literally everyone" 7000 years ago, same as every single Han Chinese? But, looking it up, National Geographic says that "Aboriginal Australians are all related to a common ancestor who was a member of a distinct population that emerged on the mainland about 50,000 years ago", which implies people of other populations are not all related to him. Aha, here's where your figure comes from: "Rohde, Olson and Chang showed through simulations that, given the false assumption of random mate choice without geographic barriers, the Identical Ancestors Point for all humans would be surprisingly recent, on the order of 5,000-15,000 years ago." But it is indeed false, there were barriers for the entire history of our species, such as oceans; and how do migrations of the last millenium negate it? More importantly, it's a quantitative issue. Your link goes on to say: "Thus, even though the Norwegian and Japanese person share the same set of ancestors, these ancestors appear in their family tree in dramatically different proportions. A Japanese person in 5000 BC with present-day descendants will likely appear trillions of times in a modern-day Japanese person's family tree, but might appear only one time in a Norwegian person's family tree." Seeing as every specimen can have novel genetic variants, this should allow for arbitrary magnitude of divergence, no?
"The very-diverse are hunter-gatherer populations who live in clusters scattered across Central and Southern Africa, in Namibia, Botswana, the Congo and a dozen other nations. They include the Khoi, the San, the Mbuti, the Mbenga, the Twa and the Hadza. [...] Save for that isolated half-million very-diverse hunter-gatherer minority, all Sub-Saharan Africans constitute the genetically in-between group of 1.14 billion not-very-diverse humans. [...] The founder population of today’s genetically not-very-diverse Sub-Saharan Africans shows signatures of a population crash, probably with a toll of 50% and probably from an effective breeding population of something like 50,000 people down to 25,000. So out of a founding population on the scale of a very small city 60,000-120,000 years ago, today’s 1.14 billion Sub-Saharan Africans have gone on to people every habitable space of that hugely varied continent. They are significantly more genetically diverse than their relatives on five other continents, but nowhere near as diverse as the neighbors in their midst. [...] 93-98.5% of the ancestry of humans outside of Sub-Saharan Africa (among those with no recent Sub-Saharan African ancestry, obviously) derives from a breeding population of 1,000 to 10,000, which expanded rapidly 60,000 years ago." So this tells us that a) this measure of diversity is mostly sensitive to minimal sizes of ancient populations, b) people descending from most of those diverse clusters constitute a tiny fraction of African population, and are thus non-central cases of colloquial "Africans", c) the overwhelming majority of people of SSA ancestry descend from a single breeding population, which can be distinguished from other populations – including the one that had left Africa. The story is more complex, but the point is that this diversity appears not to be a good reason to not recognize [the majority of] Africans as a specific separate group.
4. Race isn't a valid construct, genetically speaking. It's not well defined". – but aren't we already talking of genomic ancestry? So this is a true but irrelevant objection. Now, people are of course free to believe that conventional self-reported "races", which are, as is often correctly said, social constructs, do not correspond to continental-level ancestry – although noisily in many cases. I think this is pretty absurd on its face, but anyway, Googling tells us"In mothers self-identified as Black and White, the imputed ancestry proportions were 77.6% African and 75.1% European respectively" in a "diverse" NYC sample, and I'd expect less cosmopolitan groups to show higher figures. However unfit race is for purposes of cutting-edge research, in the aggregate data it is robustly aligned with ancestry, which is well-defined.
5. Intelligence is not well defined. There's no single definition of intelligence on which people from different fields can agree. – blatantly misinterpreting. The cited paper states: "...Nevertheless, some definitions are clearly more concise, precise and general than others. Furthermore, it is clear that many of the definitions listed above are strongly related to each other and share many common features" and goes on to propose a unified definition: “Intelligence measures an agent’s ability to achieve goals in a wide range of environments.” Features such as the ability to learn and adapt, or to understand, are implicit in the above definition as these capacities enable an agent to succeed in a wide range of environments. Also it is not clear why we'd even need people from different fields (in this case, psychology and AI research!) to agree on a definition of intelligence to have a useful measurement of human smarts. And this is what has happened with IQ:
6. IQ has a number of flaws. It is by definition Gaussian without having appeared empirically first and the g construct itself has almost certainly no neurological basis and is purely an artifact of factor analysis. – this is just some Gish Gallop. To begin with, I don't see your link supporting your summarization – except the vague "number of flaws". If I may, where have you taken this list from? In any case, anything but God has a number of flaws; your link says that "According to Weiten, "IQ tests are valid measures of the kind of intelligence necessary to do well in academic work."" and "clinical psychologists generally regard IQ scores as having sufficient statistical validity for many clinical purposes". It doesn't seem like there's any scientific objection as to the validity of IQ as a measurement of what's casually called smarts and understood to be smarts in the context of this discussion – even though there are weird attempts to drown this fact in caveats. Why is the part about assumed Gaussian even relevant? What would it mean for g to have a neurological basis, and why would that matter in the discussion of HBD? ...And the part about g being "purely an artifact of factor analysis" is plain false, far as I can tell. It comes from Cosma Shalizi's essay that misstates the reason for the existence of positive manifold. "If I take any group of variables which are positively correlated, there will, as a matter of algebraic necessity, be a single dominant general factor... Since intelligence tests are made to correlate with each other, it follows trivially that there must appear to be a general factor of intelligence." This is just a lie: a great deal of effort has been devoted to making cognitive tests comprehensive and diverse assessments of ability, but positive correlations pop out on their own, even in research informed by Shalizi's assumptions, e.g. "The WJ-R was developed based on the idea that the g factor is a statistical artifact with no psychological relevance. Nevertheless, all of its subtests are intercorrelated and, when factor analyzed, it reveals a general factor that is no less prominent than those of more traditional IQ tests". And "...All 861 correlations are positive. Subtests of each IQ battery correlate positively not only with each other but also with the subtests of the other IQ batteries. This is, of course, something that the developers of the three different batteries could not have planned – and even if they could have, they would not have had any reason to do so, given their different theoretical presuppositions."
7. Twin studies are flawed in methodology. Twins, even identical twins, simply do not have exactly the same DNA. – again, misinterpreting; there are flaws but the method is not summarily "flawed" just because a section about flaws exists. The first link is a list of objections but in no way does it show or argue that they are decisive, or even apply at all to current methods (there are "responses to critiques" subsections). The second is apparently irrelevant, and was already addressed by another user.
8. Evolution isn't just mutations and natural selection. Not every trait is an adaptation. (a link to Wiki on "Evolution – Evolutionary processes") – ...okay but how does this even support your case? I'm honestly unsure what the idea here is. Taken literally, your summary suggests that evolution can produce maladaptive changes, so we cannot assume that all (or any) populations will be maximally fit (for their environment). This is a pro-eugenicist take, if anything. Whereas the page itself discusses mechanisms of change in allele frequency and does not have any clear impact on the validity of HBD one way or another.
9. Heritability does not imply genetic determinism. Many things are heritable and do not involve genes.These include epigenetic mechanisms, microbiota, or even environmental stress on germinal cells. – irrelevant/false. The link is to "Heritability – Controversies" with some nitpicks of unclear truth value. The second is a general overview of possible issues with heritability estimates. It does not weigh in on HBD and accepts the premise of variable genetic contributions to human intelligence: As a case in point, consider that both genes and environment have the potential to influence intelligence. Heritability could increase if genetic variation increases, causing individuals to show more phenotypic variation, like showing different levels of intelligence. On the other hand, heritability might also increase if the environmental variation decreases, causing individuals to show less phenotypic variation. This says, concretely, that in more equal environments we will observe more true genetic effects on variation in intelligence, so whatever differences in genetic effects on this trait there are between groups, they will become more pronounced. By the way this is terrible for the anti-HBD position because it means that the state of perfect environmental equality – one could say equality of opportunity – will collapse into genetic determinism (modulo random noise). Your own idea seems to be that non-genetic mechanisms of apparent heritability can be interrupted by a positive environmental intervention. What share of "heritable" variance can it explain, at a maximum? Like, concretely, to what extent do you think the racial IQ gap is explained by microbiota, epigenetic mechanisms and environmental stress on germinal cells? Those are all quantifiable and falsifiable claims, but you just gesture at them. At this point, a dedicated layman looks it up and sees that they can explain very little indeed.
10. We don't mate randomly, which is an assumption in many genetics studies. – irrelevant applause lights, "genetics bad". Which studies, and does this matter for HBD? I've watched the video; it discusses interactions between psychiatric disorders and such, and states that genetic correlations between traits may be inflated by assortative mating (i.e. people high in trait X marry people high in trait Y). Genetic correlation "is defined as the proportion of the heritability that is shared between two traits divided by the square root of the product of the heritability for each trait". What is meant here, concretely? Ancestry is not really a "heritable trait", is it? And race is just a category, plus a bad proxy for ancestry, as far as HBD is concerned.
11. HBD is not generally accepted in academia. – this is just an appeal to authority, plus misleading. It's a single highly technical paper by some Kevin Bird, "Department of Horticulture Michigan State University", can it be considered an authoritative source on what academia thinks? And from the abstract, it attacks a very strong form of HBD reasoning, using data that cannot plausibly be conclusive: Evidence for selection was evaluated using an excess variance test. Education associated variants were further evaluated for signals of selection by testing for excess genetic differentiation (Fst). Does it strike you as plausible that we know enough about "education associated variants" to impute effects of prehistoric selection on intelligence? This ought to mean that the science of genetics of intelligence is vastly more mature than people think, than you suggest, too, and that within-group intelligence heritability is understood really well! Why hasn't this made the news yet? (And how does this address obvious low-tech HBD arguments, such as admixture studies and adoption studies?)
12. Many public HBD figures have been found guilty of fraud. Cyril Burt would literally forge results, while Lynn would take the average of two neighboring countries' IQ in order to derive "data" from a country's unknown national IQ. – That's an isolated demand for rigor. What field doesn't commit fraud? Were public anti-HBD figures never found guilty of fraud? Is the fraud rate different enough to affect our priors? And your link does not show that Burt's forgery was positively proven, but it admits that figures of heritability arrived at by independent researchers do not differ from Burt's, so why should we care? Assuming that a layman could track it down from here, I'll allow myself to quote Richard Haier (The Neuroscience of Intelligence, Cambridge University press, 2017): Subsequent twin studies done by different investigators around the world with large samples arrive at an average value for the correlation of intelligence scores among identical twins raised apart of .75 (Plomin & Petrill, 1997). Burt’s value was .77. For comparison, based on 19 studies ranging in sample sizes between 26 and 1,300 identical twin pairs, the average value for identical twins raised together is about .86 (see Loehlin & Nichols, 1976, table 4.10, p. 39)... Thus, the .771 “fraud” ends with recognition of overwhelming data from independent researchers that are fully consistent with Burt’s analyses, flawed as they may have been. Any single study, or any one researcher, can be flawed, but the basic conclusion that genes play an important role in intelligence is consistently supported by data from numerous studies of twins, adoptees, and adopted twins. This is an excellent example of looking at the weight of evidence (recall my three laws from the Preface: no story is simple; no one study is definitive; it takes many years to sort out conflicting and inconsistent findings and establish a weight of evidence). ... The weight of evidence summarized in this chapter leaves no reasonable doubt. Only extreme ideologues are still in denial. As for Lynn's country data, well, the same logic applies. Do we have any more trustworthy data? Does it refute Lynn's? Then why not just refer to it instead? Please don't say that it's not very interesting and nobody has bothered to collect proper measurements, IQ and race (or rather, ethnicity) is literally the most painful question in modern science, and it's evident from such dramas that a great many researchers are emotionally invested in proving the relationship wrong.
Ultimately, exposure to this sort of content has done for me what it has done for this person:
[…] Well, OK, I could believe that; visible traits consistent over entire populations like skin color might differ systematically because of sexual selection or something, but why not leave IQ following the exact same bell curve in each population? There was no specific thing here that made me start to wonder, more a gradual undermining (Gould’s work like The Mismeasure of Man being completely dishonest is one example - with enemies like that…) as I continued to read studies and wonder why Asian model minorities did so well, and a lack of really convincing counter-evidence like one would expect the last two decades to have produced - given the politics involved - if the idea were false. And one can always ask oneself: suppose that intelligence was meaningful, and did have a large genetic component, and the likely genetic ranking East Asians > Caucasian > Africans; in what way would the world, or the last millennium (eg the growth of the Asian tigers vs Africa, or the different experiences of discriminated-against minorities in the USA), look different than it does now? But having said that, and admiring things like Plantinga’s free will defense, and the subtle logical issues in formulating it and the lack of any really concrete evidence for or against Jesus’s existence, do I take the basic question of God seriously? No. The theists’ rearguard attempts and ever more ingenious explanations and indirect pathways of reasons and touted miracles fundamentally do not add up to an existing whole. The universe does not look anything like a omni-benevolent/powerful/scient god was involved, a great deal of determined effort has failed to provide any convincing proof, there not being a god is consistent with all the observed processes and animal kingdom and natural events and material world we see, and so on. The persistence of the debate reflects more what motivated cognition can accomplish and the weakness of existing epistemology and debate.[…]
I want to make it perfectly clear: those question marks in my point-by-point do not actually indicate uncertainty. They could as well have been references to papers. The field really is about as advanced as Bird's study suggests – only in the direction he disapproves of. But this isn't the place for it, surely people can go to some edgier venue and ask for receipts. The point I'm trying to make is: you say "Okay, if there's anyone here who actually believes in HBD, here's a couple reasons why you shouldn't." On an ignorant but moderately skeptical person your little list can, and likely will, have an effect that's the opposite of what you intend to achieve. To "who actually believes in HBD", it's utterly unconvincing. If I may be so blunt, it's almost as sad as quasi-scientific gotchas of flat earthers.
And this is how all of anti-HBD rhetoric is, in my experience. It crumbles under basic scrutiny, links do not show what they are purported to show, there are simple misunderstandings of what terms mean, there is no coherent epistemology or a single model, there's suppression of inconvenient evidence, there's substitution of evidence with confident op-eds in Vox from people who are supposed to be experts (but whose legitimate work doesn't support their confident claims), there are cascades of internally inconsistent Gish Gallops and other fallacies; worst of all, the reader is assumed to just not be all that bright. It's a collection of purely reactive objections that might come across as persuasive to like-minded people, but are not battle-tested – and indicate general unwillingness to test one's beliefs.
I expect very little payoff from this labor. But it would be nice if EAs were to become a little more reserved on this topic, and at least stopped turning off potential recruits with irrefutable displays of irrationality.
There is no such consensus, though. Your links do not support your very strong claim.
E.g. Vox:
Do most experts think genes make a substantial contribution to the black-white difference in intelligence? There have been several surveys of expert opinion over the years. Perhaps the first was described in a 1988 book by Snyderman and Rothman. The most recent was described in a 2013 blog post about a conference presentation. The survey described in that post has resulted in two publishedarticles, neither of which presents data on opinions regarding the black-white difference. The studies do, however, report that only about 5 percent of people who were invited to participate responded to any one set of items. Given this very low response rate, along with the potential for bias in which scientists were invited in the first place, we doubt that these results are an accurate representation of the field.
Still, in both the Snyderman and Rothman book and in the more recent survey, more than half of respondents selected one of two response categories that included zero (one option was “0 percent of [black-white] differences due to genes” and the other was “0-40 percent of differences due to genes”). Much more important, however, is that respondents were not allowed to endorse what in my view is the only reasonable response: It is not possible to give a meaningful estimate of the percentage.
Education was rated by N = 71 experts as the most important cause of international ability differences. Genes were rated as the second most relevant factor but also had the highest variability in ratings
If anything, the consensus seems to be that genes play some role here.
To anyone who sincerely wonders if there's anything to the "race-iq realism" theory, I ask you to consider this great point point made Ezra Klein to Sam Harris when Harris embraced The Bell Curve: very rarely will you see any serious consideration of the possibility that environmental factors explain >100% of the black-white iq gap. In other words, on a purely genetic basis that black people may be more intelligent than white people is alien to the discourse.
I do not see why this hypothetical is impressive? The best that could be said for it is that it is logically sound and novel. But heritability and norms of reaction impose limits on such explanations. If X percent of a trait's variance can be explained by a factor, then there's only so much you can get by changing the sum of non-X factors. Adult intelligence has roughly 80% heritability (equally within white and black populations; actually this alone invalidates the idea). For 1 d of difference in intelligence to be explained away by the environment, the gap in environmental quality must be 2.24 d. This is implausibly large for intra-national racial differences, contradicted by direct measures of environmental quality and indirect proxies of deprivation (such as stress and self-esteem), made suspect by the fact that there's been a great deal of improvement in race relations and equalization of living standards since the 60s, yet no large narrowing of the IQ gap; and for the case where black people have higher "genotypic IQ", environmental deprivation must be even greater than 2.24 d.
I don't want to be rude, but this appears to be just shoddy overuse of rationalist lingo in the name of shoehorning a myopic and empirically unsupported political agenda into the consequentialist framework.
What observed empirical effects? You link to a very strange post saying, concretely, that
Pardon me, what? Is this your evidence base?
№1-2 might as well be considered arguments for lesser demonization of HBD. There is nothing inherently political about thinking one way or another about sources of cognitive differences; the political valence is imposed on such hypotheses by external forces. If smart people independently arrive at HBD as a morally neutral explanation for generally available observations, then it's not very prudent on part of "the left or center-left" to baselessly label them racists, supporters of genocidal far-right ideologies, insane cranks and such and leave them no choice except break their own minds into an Orwellian mold, learn to live in falsehood, or go rightward. When they say, like Bostrom, that they are motivated by humanitarian impulses, they can be taken at their word.
You, however, seem to conclude that the only problem is insufficient intensity of vilification of HBD, now as a "cause area" unto itself; that these people can be intimidated into not believing what they see, through pure peer pressure and pushing the topic to the fringe instead of rational persuasion.
№3 is honestly horrifying in terms of epistemic integrity. You seem to be dismissive of truth as a terminal value, so let's put it like this: a person who sees nothing wrong with such pseudoreasoning – and, given the score, that's normal on EA forum –can delude oneself into excusing arbitrary atrocities; or less dangerously, draining resources into arbitrarily ineffective causes just to feel good about oneself.
We don't have a good theory, in part, because there's no meaningful way to lump together "far-right ideas" over "the past few centuries", or indeed seriously analyze anything prior to the 20th century through these lens. Do you mean Jacobites or Bourbons by far-right? Why not address la Terreur as an archetypal case of the idea of egalitarianism causing mass death and suffering in the characteristic manner of an infohazard? Should this make us suspicious of egalitarian ideation in general?
Here's a honest thought: the notion of "memetics" or "infohazards" is an infohazard in its own right. It's bad philosophy, and it offers zero explanatory power over traditional terms like "undeservedly popular idea", "misleading idea" or "dangerous idea" but it gives the false impression of such adjectives having been substantiated. It's just a way of whitewashing a classical illiberal and, indeed, totalitarian belief that some ideas must be kept away from the plebeians because they are akin to a plague. In illiberal societies those are "democracy" and "independent thought"; we have a consensus that theories justifying restriction of access to those are vacuous and evil, but those theories at least had some substance, unlike equivocation here about suffering caused by "far right" and, by an entirely frivolous extension, HBD.
In sum, analogizing ideas and their bearers to infectious agents invading and spreading within the body politic is a staple of far-right sociology that exploits deep-seated reactions of disease-associated disgust, fear and distrust of outsiders, and that's all there is to "memetics" in such colloquial use. Perhaps you could do without resorting to such tools for thought.
Perhaps there is, and it's called "law" and "democracy", and you need to argue in a principled way for your cost-benefit analysis that concludes that extant legal and political checks against far-right threats are insufficient, and concludes with embracing of some of the worst totalitarian legacies to ostracize an apparent scientific truth.
But can you be trusted to actually think that, given what you say about utility of public admission of opinions in question? For an external observer, it's a coin toss. And the same for the entirety of your reasoning. As an aside, I'd be terrified of a person who can willfully come to believe – or go through the motions of believing – what he or she believes to be morally prudent but epistemically wrong. Who knows what else can get embedded in one's mind in this manner.
Well, consider that, as it tends to happen in debates, people on the other side may be as perfectly sure about you being misguided and promoting harmful beliefs as you are about them; and that your proud obliviousness with regard to their rationale doesn't do your attempt at persuasion any more good than your unwillingness to debate the object level does.
Consider, further, that your entire model of this problem space really could be wrong and founded on entirely dishonest indoctrination, both about the scholarly object level and about social dynamics and relative effects of different beliefs.
Finally, consider that some people just have a very strong aversion to the idea that a third party can have the moral and intellectual authority to tell them which thoughts are infohazards. If nothing else, that could help you understand how this can happen.
Personally – because I do, in fact, believe that you are profoundly wrong, that even historically these views did not contribute to much harm (despite much misinformation concocted by partisans: policies we know to be harmful are attributable to different systems of views); that, in general, any thesis about systematic relation in the pattern {views I don't like}=>{atrocities} is highly suspect and should be scrutinized (e.g. with theists who attribute Stalin's brutality to atheism, or derive all of morality from their particular religion); and that my views offer a reliable way to reduce the amount of suffering humans are subjected to, in many ways from optimizing allocation of funds to unlocking advances in medical and educational research to mitigating slander and gaslighting heaped upon hundreds of millions of innocent people.
Crucially, because I believe that, all that medium-term cost-benefit analysis aside, the process of maintaining views you assume are beneficial constitutes an X-risk (actually a family of different X-risks, in Bostrom's own classification), by comprehensively corrupting the institution of science and many other institutions. In other words: I think there is no plausible scenario where we achieve substantially more human flourishing in a hundred years – or ever – while deluding ourselves about the blank slate; that it's you who is infecting others with the "Basilisk" thought virus. And that, say, arguments about the terrible history of some tens of thousands of people whom Americans have tortured under the banner of eugenics – after abusing and murdering millions of people whilst being first ignorant, then in denial about natural selection – miss the point entirely, both the point of effective altruism and of rational debate.
This is an impossible standard and you probably know it. Risks of a given strategy must be assessed in the context of the full universe of its alternatives; else the party that gets to cherrypick which risks are worth bringing up can insist on arbitrary measures. By the way, I could provide nontrivial evidence that your views have contributed to making a great number of people more intolerant and more violent, and have caused thousands of excess deaths over the last three years; but, unlike your wholly hypothetical fearmongering, it's likely to get me banned.
Indeed, I could ask in the same spirit: what makes people upvote you? If your logic of cherrypicking risks and demonizing comparative debate is sound, then why don't they just disregard GiveWell and donate all of their savings to the first local pet shelter that gets to pester them with heart-rending imagery of suffering puppies? Maybe they like puppies to suffer?! This is not just manipulation: rising above such manipulation is the whole conceit of this movement, yet you commit it freely and to popular applause.
To make me or anyone like me change my mind, strong and honest empirical and consequentialist arguments addressing these points are required. But that's exactly what you say is "much less relevant" than just demanding compliance. Well. I beg to differ.
For my part, I do not particularly hope to persuade you or anyone here, and guidelines say we should strive to limit ourselves to explaining the issue. Honestly it's just interesting at this point, can you contemplate the idea of being wrong, not just about "HBD" but about its consequences, or are you the definition of a mindkilled fanatic who can't take a detached view at his own sermon and see that it's heavy on affirmation, light on evidence?
First, that depends on what you mean by "this stuff"; Bird does not study intelligence nor behavioral genetics for a living, he's a plant geneticist. Skewed though the survey may be, it's probably more representative than a single non-expert.
Second, why do you suppose the non-response rate is so high and so skewed? And might it have something in common with your own refusal to continue our conversation on merits of your list?
I suspect that professionals who prefer not to respond, rather than respond in the negative about genetic contributions to the IQ gap, are driven by contradictory impulses: they believe that the evidence doesn't allow for a confident "100% environmental" response and, being scientists, have problem with outright lying, but they also don't want to give the impression of supporting socially unapproved beliefs or "validating" the very inquiry into this topic. So they'd rather wash their hands of the whole issue, and allow their less squeamish colleagues to give the impression of moderate consensus in favor of genetic contribution.
This logic is only applicable to contrived scenarios where there is no prior knowledge at all – but you need some worldly knowledge to understand what both these hypotheses are about.
Crucially, there is the zero-sum nature of public debate. People deliberately publicizing reasons to not believe some politically laden hypothesis are not random sources of data found via unbiased search: they are expected to cherrypick damning weaknesses. They are also communicating standards of the intellectual tradition that stands by the opposing hypothesis. A rational layman starts with equal uncertainty about truth values of competing hypotheses, but learning that one side makes use of arguments that are blatantly unconvincing on grounds of mundane common sense can be taken as provisional evidence against their thesis even before increasing object-level certainty: poor epistemology is evidence against ability to discover truth, and low-quality cherrypicked arguments point to a comprehensively weak case. Again, consider beliefs generally known to be kooky, and what they bring to bear on the opposition. Their standard of rigor alone is discrediting to what they believe in.
Moreover, I've established that, upon checking, some of your links positively provide evidence in favor of HBD, rather than against – at least by the standard of evidence implicit in the phrasing of the list. Returning again to the Identical Ancestors Point, is presented as an Anti-HBD finding in the first place because it implies a very low prior for genetic divergence of populations, migrations somehow averaging it all out: Human migrations over the last thousand years have been such that literally everyone on Earth is a descendant of literally everyone that lived 7000 years ago whose offspring didn't die out. (Is this the wrong takeaway? What, then, did you mean to say by adding it?) Looking into the actual paper, we see: ...For example, a present-day Norwegian generally owes the majority of his or her ancestry to people living in northern Europe at the IA point, and a very small portion to people living throughout the rest of the world. Furthermore, because DNA is inherited in relatively large segments from ancestors, an individual will receive little or no actual genetic inheritance from the vast majority of the ancestors living at the IA point. Not only does this make the original argument invalid (even in a strong absolute sense – there can be zero common inheritance!) – it directly reinforces the HBD conjecture that long-term (i.e. pre-IAP) divergent local adaptation is relevant to current genetic (and trait) differences.
I agree that this is improper and irritating terminology, because doctrinally, LW asserts its allegiance to empiricism, with all the talk about Bayes-updating on evidence and how rationalists must "win". But in practice this isn't so clear-cut: LW is fascinated with armchair thought experiments (that routinely count as evidence to update on), and all the attention devoted to infohazards, Pascal mugging, one-boxing, AI scenarios etc. suggests that they, as a living tradition, are not resilient to speculation the way pure empiricists – say, regular natural scientists – would be. So, not necessarily a misnomer.
They are internally consistent, but I think the point of ethical theory is to clarify the intuitively knowable essence of moral action for purposes of nontrivial decisionmaking, not to assert what morality is and derive an arbitrary decision rule from there. Utilitarianism is often criticized for things like the repugnant conclusion, yet non-utilitarian ethical theories routinely produce more grating outputs, because they fail to capture the most significant part of intuitive ethics, which is mostly about harm reduction under conditions of resource scarcity. They are less consistent with ethics given to us in lived experience, so to speak.
No, the extent of the purported risk matters. You are just falling back on the unsupported prior about cost-benefit ratio because you have preemptively excluded all factors that may change the qualitative conclusion of "not doing that". To give a specific example: under the assumption that HBD is wrong, we must consider disparate outcomes to be a result of some discrimination and devote resources to alleviate it; but if HBD is actually right, this'd necessarily mean that our costly attempts to help low-performing groups are suboptimal or futile (as in, not effective), and that we will have unfairly attributed blame, harming other groups psychologically and materially. Then there are knock-on effects of harming science: for starters, fears of enabling racists can hold back genomic medicine (and population-specific treatment) by increasing hurdles to data collection and access. We do not have a priori knowledge as to which costs are negligible. On a more meta level, Scott Alexander's parable comes to mind.
IMO it's a weak argument because for all the racism, black Americans still report the highest self-esteem of all racial groups; and theories of stereotype threat are apparently unsupported by high-quality data; so it isn't clear what the odds are that some HBD research or whatever would harm people substantially. But even before that – there are laws against hate speech and discrimination, and they can be strengthened if needed; it seems very suboptimal to focus on not developing neutral knowledge only to deny hateful ideologies rhetorical ammo, instead of dealing with them directly. By the way, cannot racists point to censorshipas sufficient evidence of their correctness, if their intent is to spin available facts to their benefit? Actually, doesn't this enable them to – convincingly – claim that facts are much worse than they are, that the genomic gap in cognitive ability is bigger than the non-zero gap we'd have found (and, I believe, have partially found) with proper research (which is currently prohibited)? And in any case, you have to put racism-driven harms in the context of costs of pretending that HBD is certainly false – that is, under the assumption that we are "just not doing that" and have no clue whether it is or isn't true.
As an aside, I am personally puzzled by the strong conviction of many that HBD becoming common knowledge could lead to normalization of racial discrimination. This is a normative, not scientific question. Societies with Social Darwinist values do not need HBD to embrace and exacerbate the status quo of disparate power. Societies with ethnocentric values opportunistically oppose and exploit ethnic outgroups regardless of relative merit. Mainstream modern value system depends on the premise of human rights, not equality of capability. We do not hold that it is normal to oppress individuals who are known to be below average in some morally neutral trait (except maybe for an expansive definition of "oppress" and clinical issues having to do with lack of legal capacity), we have a strong revulsion to identity-based discrimination, and we understand the unreasonableness of treating individuals on the basis of average values.
Uncharitably, in the case of EA, this concern may have to do with the strain (common to EA and LW) of conspiratorial elitism and distrust in the democratic process, and with the unconscious belief that intelligence does define moral worth. That's ...not a very popular belief. I would deeply hate it if my cognitive betters acted like they have greater moral worth than myself, and therefore, to be fair, I cannot deny equal moral worth to people of lower ability. Most people correctly believe that they aren't brilliant, but they're not so dull as to not arrive at this logic. There are some contingent factors that complicate the picture, but not fundamentally.
Without getting into the weeds of stereotype scholarship, the extent of claimed harms, and the irrational denial of the role of merit in achievement (reasoning in that wiki page doesn't even begin to address what would happen in a "proper" meritocratic society after a few generations, because it is premised on genes not contributing to achievement; this is a typical case of an unexamined anti-HBD prior leading to policy errors)... I'll just say that in my opinion both those issues, insofar as they harm anyone, have to do with beliefs about moral qualities. If "Protestant ethic" is alive and prescribes vilification of people of lower morally neutral ability, then that is a problem in its own right and beyond the scope of this conversation. Luckily, Protestant ethic also encourages treating people on a case by case basis.
Crucially, the search for interventions that actually close the IQ gap. As it stands, we have picked low-hanging fruit like lead exposure, malnutrition, iodine deficiency, parasites and such (in developed nations; I expect EA efforts in Africa to keep delivering on this front), and are left with pursuing dead ends of addressing iniquity like the "food deserts" nonsense, or doubling down on stuff like school spending, that has long ran into diminishing or zero returns and is only popular because to point out its inefficacy means to risk being labeled racist. As Nathan Cofnas argues,
Ironically, Cofnas got in trouble for this. If the suppression starts this far upstream from the object level, how can our priors be trusted?
I sincerely doubt you can prove 1 or 2 (given that your critiques of relevant methodology weren't persuasive), and it looks like assigning any value to 3, on its own, is pure spite that is best left out of effective altruism. Making racists mad is not, in fact, a positive good, fun as it may be.
Have read some. I explicitly say I'm addressing the state of discourse here, more than just your comment. I respond to you in particular when I quote specific passages. Sorry if that was unclear.
Again, this is not HBD Central, and it is sufficient to establish that there is legitimate uncertainty, so we cannot fall back on the comfortable prior that costs of repudiating HBD are negligible.
Well, I believe that misleading people, and even wasting people's time on true but irrelevant, misleadingly phrased, blatantly misinterpreting and patently false claims is a form of rudeness that's extra obnoxious, because it craftily avoids opprobrium one could earn with trivial show of disrespect. It's not fair to act indignant about an unflattering comparison after doing that. Even so, I've made peace with Brandolini's law, and kept addressing those claims on the object level, to substantiate my "very civil" summary and so that "EAs were to become a little more reserved on this topic, and at least stopped turning off potential recruits with irrefutable displays of irrationality." To be honest your reaction isn't wholly unexpected, but I did hope that I've been polite enough to merit some tolerance.
OK but please think of your stated desire is to persuade those who happen to believe in HBD to disbelieve it. Obviously you've failed in my case, but I maintain that flaming out like that is detrimental even as far as fence-sitters are concerned. I believe I've provided sufficient receipts for the purpose of showing how your list is inadequate.
It is really not hard to showboat on this topic, by citing from very clearly argued stuff like this or "authoritative" sources like that review or very technical recent papers or just by gesturing in the general direction of environmentalist rhetoric that is... the way I've described, and evident in, e.g., this condemnation of Cofnas, mired in (what I hope is obvious after my initial comment) logical fallacies and half-truths and raw indignation. Or one can just say that if this guy is challenged not by rational and empirical arguments but by being repeatedly called a pseudoscientist and getting a page full of personal attacks on him to the top of search results for his name (a page he responds to with an even pettier page), then he may get a lot of uncomfortable stuff right.
My point is not to showboat but to argue that people who pursue this anti-HBD rhetorical strategy, including you, are probably not succeeding, and are doing the community no favors.
This looks like retconning of history. EA and rationalism go way back, and the entire premise of EA is that determining what makes more good through "rationalist", or more precisely, consequentialist lens is moral. There is no conflict of principles.
The quality of discussion on the value of tolerating Bostrom's (or anyone else's ) opinions on race&IQ is incredibly low, and the discussion is informed by emotion rather than even trivial consequentialist analysis. The failure to approach this issue analytically is a failure both by Rationalist and by old-school EA standards.
This list is a good example of the sort of arguments that look persuasive to those already opposed to HBD, but can push people on the fence towards accepting it, so it may be net-negative from your perspective. This is what has happened to me, and I'll elaborate on why – so that you may rethink your approach, if nothing else.
Disclaimer: I am a non-Western person with few traits worth mentioning. I identify with the rationalist tradition as established on LW, feel sympathy for the ideal of effective altruism, respect Bostrom despite some disagreements, have donated to GiveWell charities on EA advice, but I have not participated more directly. Seeing the drama, people expressing disappointment and threatening to leave the community, and the volume of meta-discussion, I feel like clarifying a few details that may be hard to notice from within your current culture, and hopefully helping you mend the fracture that is currently getting filled with the race-iq stuff.
All else being equal, people who hang around such communities prefer consistent models (indeed, utilitarianism itself is a radical solution to inconsistencies in other ethical theories). This discourse is suffused with intellectual inconsistency, on many levels of varying contentiousness.
4. Race isn't a valid construct, genetically speaking. It's not well defined". – but aren't we already talking of genomic ancestry? So this is a true but irrelevant objection. Now, people are of course free to believe that conventional self-reported "races", which are, as is often correctly said, social constructs, do not correspond to continental-level ancestry – although noisily in many cases. I think this is pretty absurd on its face, but anyway, Googling tells us "In mothers self-identified as Black and White, the imputed ancestry proportions were 77.6% African and 75.1% European respectively" in a "diverse" NYC sample, and I'd expect less cosmopolitan groups to show higher figures. However unfit race is for purposes of cutting-edge research, in the aggregate data it is robustly aligned with ancestry, which is well-defined.
5. Intelligence is not well defined. There's no single definition of intelligence on which people from different fields can agree. – blatantly misinterpreting. The cited paper states: "...Nevertheless, some definitions are clearly more concise, precise and general than others. Furthermore, it is clear that many of the definitions listed above are strongly related to each other and share many common features" and goes on to propose a unified definition: “Intelligence measures an agent’s ability to achieve goals in a wide range of environments.” Features such as the ability to learn and adapt, or to understand, are implicit in the above definition as these capacities enable an agent to succeed in a wide range of environments. Also it is not clear why we'd even need people from different fields (in this case, psychology and AI research!) to agree on a definition of intelligence to have a useful measurement of human smarts. And this is what has happened with IQ:
6. IQ has a number of flaws. It is by definition Gaussian without having appeared empirically first and the g construct itself has almost certainly no neurological basis and is purely an artifact of factor analysis. – this is just some Gish Gallop. To begin with, I don't see your link supporting your summarization – except the vague "number of flaws". If I may, where have you taken this list from? In any case, anything but God has a number of flaws; your link says that "According to Weiten, "IQ tests are valid measures of the kind of intelligence necessary to do well in academic work."" and "clinical psychologists generally regard IQ scores as having sufficient statistical validity for many clinical purposes". It doesn't seem like there's any scientific objection as to the validity of IQ as a measurement of what's casually called smarts and understood to be smarts in the context of this discussion – even though there are weird attempts to drown this fact in caveats. Why is the part about assumed Gaussian even relevant? What would it mean for g to have a neurological basis, and why would that matter in the discussion of HBD? ...And the part about g being "purely an artifact of factor analysis" is plain false, far as I can tell. It comes from Cosma Shalizi's essay that misstates the reason for the existence of positive manifold. "If I take any group of variables which are positively correlated, there will, as a matter of algebraic necessity, be a single dominant general factor... Since intelligence tests are made to correlate with each other, it follows trivially that there must appear to be a general factor of intelligence." This is just a lie: a great deal of effort has been devoted to making cognitive tests comprehensive and diverse assessments of ability, but positive correlations pop out on their own, even in research informed by Shalizi's assumptions, e.g. "The WJ-R was developed based on the idea that the g factor is a statistical artifact with no psychological relevance. Nevertheless, all of its subtests are intercorrelated and, when factor analyzed, it reveals a general factor that is no less prominent than those of more traditional IQ tests". And "...All 861 correlations are positive. Subtests of each IQ battery correlate positively not only with each other but also with the subtests of the other IQ batteries. This is, of course, something that the developers of the three different batteries could not have planned – and even if they could have, they would not have had any reason to do so, given their different theoretical presuppositions."
7. Twin studies are flawed in methodology. Twins, even identical twins, simply do not have exactly the same DNA. – again, misinterpreting; there are flaws but the method is not summarily "flawed" just because a section about flaws exists. The first link is a list of objections but in no way does it show or argue that they are decisive, or even apply at all to current methods (there are "responses to critiques" subsections). The second is apparently irrelevant, and was already addressed by another user.
8. Evolution isn't just mutations and natural selection. Not every trait is an adaptation. (a link to Wiki on "Evolution – Evolutionary processes") – ...okay but how does this even support your case? I'm honestly unsure what the idea here is. Taken literally, your summary suggests that evolution can produce maladaptive changes, so we cannot assume that all (or any) populations will be maximally fit (for their environment). This is a pro-eugenicist take, if anything. Whereas the page itself discusses mechanisms of change in allele frequency and does not have any clear impact on the validity of HBD one way or another.
9. Heritability does not imply genetic determinism. Many things are heritable and do not involve genes. These include epigenetic mechanisms, microbiota, or even environmental stress on germinal cells. – irrelevant/false. The link is to "Heritability – Controversies" with some nitpicks of unclear truth value. The second is a general overview of possible issues with heritability estimates. It does not weigh in on HBD and accepts the premise of variable genetic contributions to human intelligence: As a case in point, consider that both genes and environment have the potential to influence intelligence. Heritability could increase if genetic variation increases, causing individuals to show more phenotypic variation, like showing different levels of intelligence. On the other hand, heritability might also increase if the environmental variation decreases, causing individuals to show less phenotypic variation. This says, concretely, that in more equal environments we will observe more true genetic effects on variation in intelligence, so whatever differences in genetic effects on this trait there are between groups, they will become more pronounced. By the way this is terrible for the anti-HBD position because it means that the state of perfect environmental equality – one could say equality of opportunity – will collapse into genetic determinism (modulo random noise). Your own idea seems to be that non-genetic mechanisms of apparent heritability can be interrupted by a positive environmental intervention. What share of "heritable" variance can it explain, at a maximum? Like, concretely, to what extent do you think the racial IQ gap is explained by microbiota, epigenetic mechanisms and environmental stress on germinal cells? Those are all quantifiable and falsifiable claims, but you just gesture at them. At this point, a dedicated layman looks it up and sees that they can explain very little indeed.
10. We don't mate randomly, which is an assumption in many genetics studies. – irrelevant applause lights, "genetics bad". Which studies, and does this matter for HBD? I've watched the video; it discusses interactions between psychiatric disorders and such, and states that genetic correlations between traits may be inflated by assortative mating (i.e. people high in trait X marry people high in trait Y). Genetic correlation "is defined as the proportion of the heritability that is shared between two traits divided by the square root of the product of the heritability for each trait". What is meant here, concretely? Ancestry is not really a "heritable trait", is it? And race is just a category, plus a bad proxy for ancestry, as far as HBD is concerned.
11. HBD is not generally accepted in academia. – this is just an appeal to authority, plus misleading. It's a single highly technical paper by some Kevin Bird, "Department of Horticulture Michigan State University", can it be considered an authoritative source on what academia thinks? And from the abstract, it attacks a very strong form of HBD reasoning, using data that cannot plausibly be conclusive: Evidence for selection was evaluated using an excess variance test. Education associated variants were further evaluated for signals of selection by testing for excess genetic differentiation (Fst). Does it strike you as plausible that we know enough about "education associated variants" to impute effects of prehistoric selection on intelligence? This ought to mean that the science of genetics of intelligence is vastly more mature than people think, than you suggest, too, and that within-group intelligence heritability is understood really well! Why hasn't this made the news yet? (And how does this address obvious low-tech HBD arguments, such as admixture studies and adoption studies?)
12. Many public HBD figures have been found guilty of fraud. Cyril Burt would literally forge results, while Lynn would take the average of two neighboring countries' IQ in order to derive "data" from a country's unknown national IQ. – That's an isolated demand for rigor. What field doesn't commit fraud? Were public anti-HBD figures never found guilty of fraud? Is the fraud rate different enough to affect our priors? And your link does not show that Burt's forgery was positively proven, but it admits that figures of heritability arrived at by independent researchers do not differ from Burt's, so why should we care? Assuming that a layman could track it down from here, I'll allow myself to quote Richard Haier (The Neuroscience of Intelligence, Cambridge University press, 2017): Subsequent twin studies done by different investigators around the world with large samples arrive at an average value for the correlation of intelligence scores among identical twins raised apart of .75 (Plomin & Petrill, 1997). Burt’s value was .77. For comparison, based on 19 studies ranging in sample sizes between 26 and 1,300 identical twin pairs, the average value for identical twins raised together is about .86 (see Loehlin & Nichols, 1976, table 4.10, p. 39)... Thus, the .771 “fraud” ends with recognition of overwhelming data from independent researchers that are fully consistent with Burt’s analyses, flawed as they may have been. Any single study, or any one researcher, can be flawed, but the basic conclusion that genes play an important role in intelligence is consistently supported by data from numerous studies of twins, adoptees, and adopted twins. This is an excellent example of looking at the weight of evidence (recall my three laws from the Preface: no story is simple; no one study is definitive; it takes many years to sort out conflicting and inconsistent findings and establish a weight of evidence). ... The weight of evidence summarized in this chapter leaves no reasonable doubt. Only extreme ideologues are still in denial. As for Lynn's country data, well, the same logic applies. Do we have any more trustworthy data? Does it refute Lynn's? Then why not just refer to it instead? Please don't say that it's not very interesting and nobody has bothered to collect proper measurements, IQ and race (or rather, ethnicity) is literally the most painful question in modern science, and it's evident from such dramas that a great many researchers are emotionally invested in proving the relationship wrong.
Ultimately, exposure to this sort of content has done for me what it has done for this person:
I want to make it perfectly clear: those question marks in my point-by-point do not actually indicate uncertainty. They could as well have been references to papers. The field really is about as advanced as Bird's study suggests – only in the direction he disapproves of. But this isn't the place for it, surely people can go to some edgier venue and ask for receipts. The point I'm trying to make is: you say "Okay, if there's anyone here who actually believes in HBD, here's a couple reasons why you shouldn't." On an ignorant but moderately skeptical person your little list can, and likely will, have an effect that's the opposite of what you intend to achieve. To "who actually believes in HBD", it's utterly unconvincing. If I may be so blunt, it's almost as sad as quasi-scientific gotchas of flat earthers.
And this is how all of anti-HBD rhetoric is, in my experience. It crumbles under basic scrutiny, links do not show what they are purported to show, there are simple misunderstandings of what terms mean, there is no coherent epistemology or a single model, there's suppression of inconvenient evidence, there's substitution of evidence with confident op-eds in Vox from people who are supposed to be experts (but whose legitimate work doesn't support their confident claims), there are cascades of internally inconsistent Gish Gallops and other fallacies; worst of all, the reader is assumed to just not be all that bright. It's a collection of purely reactive objections that might come across as persuasive to like-minded people, but are not battle-tested – and indicate general unwillingness to test one's beliefs.
I expect very little payoff from this labor. But it would be nice if EAs were to become a little more reserved on this topic, and at least stopped turning off potential recruits with irrefutable displays of irrationality.
There is no such consensus, though. Your links do not support your very strong claim.
E.g. Vox:
This does not allow to claim consensus, and the way it's worded is obviously motivated by the desire to downplay the belief of experts in causal role of genetics. We have a newer survey they do not mention, too, Survey of Expert Opinion on Intelligence: Causes of International Differences in Cognitive Ability Tests, Rindermann, 2016:
If anything, the consensus seems to be that genes play some role here.
I do not see why this hypothetical is impressive? The best that could be said for it is that it is logically sound and novel. But heritability and norms of reaction impose limits on such explanations. If X percent of a trait's variance can be explained by a factor, then there's only so much you can get by changing the sum of non-X factors. Adult intelligence has roughly 80% heritability (equally within white and black populations; actually this alone invalidates the idea). For 1 d of difference in intelligence to be explained away by the environment, the gap in environmental quality must be 2.24 d. This is implausibly large for intra-national racial differences, contradicted by direct measures of environmental quality and indirect proxies of deprivation (such as stress and self-esteem), made suspect by the fact that there's been a great deal of improvement in race relations and equalization of living standards since the 60s, yet no large narrowing of the IQ gap; and for the case where black people have higher "genotypic IQ", environmental deprivation must be even greater than 2.24 d.
Comment erased due to formatting error; apologies. The correct version is here.