Talk:Race and intelligence

Ashkenazi Jews

@NightHeron: Could you please explain why you feel that a section on the Ashkenazi Jews is WP:UNDUE and out of place? Their high average IQ is well-documented, and the argument that this developed due to genetic selection is compelling and highly relevant to the article, and has been covered by numerous reliable sources. Stonkaments (talk) 21:12, 15 February 2021 (UTC)

@Stonkaments:: Let me recall that you've already been informed (please see my comment on your user talk-page this past 6 Oct) about the consensus achieved at the RfC on Race and Intelligence at [1]. The result of that RfC was that the view that some races are genetically superior to other races in intelligence is a fringe POV. We already discussed this once in connection with Heiner Rindermann, and there's no need to re-discuss it. NightHeron (talk) 21:28, 15 February 2021 (UTC)

Thank you, and yes I am aware of the WP:FRINGE determination. But there seems to be some misunderstanding on this page of what exactly WP:FRINGE does and doesn't imply. For example, Generalrelative said in a previous discussion, The issue of the admissibility of sources claiming a genetic link between race and intelligence was settled at the Fringe theories RfC last year. [2] This seems to be a misreading of WP:FRINGE—nowhere does it state that sources containing a fringe view are inadmissible.

Specifically, being a fringe POV doesn't imply that the content is not suitable for the article, as long as it is notable (per independent reliable sources) and presented in proper context, without undue weight. I believe my addition met all of these criteria, as the study concerning Ashkenazi Jews' intelligence was published in a high-quality peer-reviewed journal, has been cited 262 times according to Google Scholar, and covered in numerous independent media outlets.[3][4][5][6]

Indeed, we have an entire section of the article dedicated to "Research into the possible genetic influences on test score differences", so clearly there is some agreement that such content is relevant and notable, despite the WP:FRINGE consensus. Maybe the information on Ashkenazi Jews would better fit in that section? Stonkaments (talk) 18:35, 16 February 2021 (UTC)

(added after Generalrelative's comment below) Concerning your specific point above, the section "Research into the possible genetic influences on test score differences" explains in some detail the methodological fallacies that mainstream scientists have found in the various theories about genetic racial superiority/inferiority. It also explains the difference between genetic theories of individual variation and genetic theories of group differences. The section does not give credence to the latter racialist theories; to do so would violate WP:FRINGE. What you proposed to add states that Azhkenazi Jews were "selected for intelligence" starting in medieval Europe; this suggests a genetic role in the IQ scores, which is a fringe POV. In addition, discussing Jewish people as a group in this article is problematic, because mainstream Jewish organizations dispute the notion that it is a "race" (see [7]), and in fact identify the notion of Jews being a "race" as a key feature of anti-semitism (see [8]). NightHeron (talk) 20:40, 16 February 2021 (UTC)

Could you please point me to where in the policy it says that violates WP:FRINGE? Per WP:FRINGENOT: "The purpose of this project is not to determine truth, but to accurately cover the worldwide view with appropriate weight given to notable viewpoints." This is clearly a notable view, and is being narrowly attributed to the authors of the study (rather than making or "giving credence" to the claims in wikivoice), so it seems entirely appropriate here. Maybe your concerns could be best addressed by further contextualizing the study by adding critiques and counterarguments?

I recognize your concern about discussing Jewish people as a group, but the study (and my addition to the article) discusses specifically Ashkenazi Jews, who were reproductively isolated for roughly a thousand years and have been found to be a clear, distinct genetic subgroup.[9] Stonkaments (talk) 21:19, 16 February 2021 (UTC)

The lead of WP:FRINGE says that a Wikipedia article should not make a fringe theory appear more notable or more widely accepted than it is. Right now the section on "Research into the possible..." adheres to this policy. The text you proposed adding does not. A lot of fringe views, including racial ones, are notable in the sense that a lot of people believe them, they influence how those people behave, and they keep appearing in print and online in various venues. Theories that some races are genetically superior/inferior to others in intelligence are such a viewpoint.

As I understand it (and it's not something I've studied), there are Ashkenazi Jews and Sephardic Jews, differing in what region their ancestors lived in. Neither is a "race", and together they don't form a "race". Discussing Ashkenazi Jews in an article on Race and intelligence is out of place. BTW, it strains credulity to say that Ashkenazi Jews were "reproductively isolated for roughly a thousand years". A huge number of Ashkenazi Jews have ancestry that is partly Slavic for the same reason that many African Americans have substantial caucasian ancestry. The anti-semitic pogroms in the Russian empire included rampant rape as well as murder, just as slaves in the US were often raped by their owners. NightHeron (talk) 22:22, 16 February 2021 (UTC)

The relevant question isn't whether or not Ashkenazi Jews comprise their own "race" (and I agree with you that they don't), but whether reliable sources find that the study of Ashekenazi Jews informs the broader discussion and study of Race and intelligence—in particular regarding a possible genetic influence on differences in groups' intelligence test scores. And the answer to that question is clearly yes.[10][11][12][13] (In the same way that I don't think white British, British Nigerians, British Ghanaians, etc. are distinct races, but their test scores are seen as notable and relevant to the discussion).

I understand your concern for wanting to convey the information in a way that does not make it appear more notable or widely accepted than it is, and I appreciate your patience in helping me understand your point of view. I'll try re-writing it to include a more clear and thorough context including common critiques such as the theory's implausibility, lack of mainstream support, etc. Stonkaments (talk) 00:15, 17 February 2021 (UTC)

I'd suggest that if you write some text on that subject, it would be best first to propose it here on the talk-page, so that editors who watchlist this page might weigh in. Thanks. NightHeron (talk) 01:02, 17 February 2021 (UTC)

NightHeron is correct here. Being FRINGE means that we handle sources in a specific way. See, e.g. Bigfoot. The text that you tried to add, on the other hand, presented FRINGE material as though it were factual / accepted science. The distinction is really not that hard to understand. Generalrelative (talk) 19:12, 16 February 2021 (UTC)

I disagree that it presented the material as accepted science. I was careful to explicitly attribute the claims to the authors of the study: Cochran, Hardy and Harpending (2006) argue that these IQ differences arose due to "the unique demography and sociology of Ashkenazim" in medieval Europe, which selected for intelligence. I'm willing to collaborate to improve the wording if you think it is unclear, but first we need to agree on whether the material deserves mention in the article at all.

And I believe the comparison to Bigfoot is wrong for a few different reasons:

1. One important barometer for determining the notability and level of acceptance of fringe ideas related to science, history or other academic pursuits is the presence or absence of peer-reviewed research on the subject.[14] The study on Ashkenazi Jews' intelligence was published in a high-quality peer-reviewed journal, which differentiates it from outlandish Bigfoot-esque claims.
2. Per WP:FRINGE/PS, on the spectrum of fringe theories, the view that racial differences in intelligence may have a genetic component is a credible alternative theoretical formulation, rather than "obviously bogus" pseudoscience or a hoax like Bigfoot. Stonkaments (talk) 19:42, 16 February 2021 (UTC)

The crucial point here is that this article is not an appropriate platform for alternative theoretical formulations. We do have an article called History of the race and intelligence controversy which is much more capacious in terms of junk science, but I'm not sure that the stuff you're seeking to add belongs there either. In any case that would be a separate question.

The paper you cited by Cochran, Hardy and Harpending simply "elaborates the hypothesis". It doesn't even count as primary evidence (like a controlled study), let alone the kind of reliable secondary source that would cause us to reevaluate what belongs in an article about mainstream scientific understanding.

I'm not sure if you're aware of all the previous on-Wiki debate that has gone down surrounding "Ashkenazi Jewish intelligence". If not I would suggest that you check out the talk page archives of the deleted article of that name. You might also see Wikipedia:Fringe_theories/Noticeboard/Archive_75#Ashkenazi_Jewish_intelligence and Wikipedia:Articles_for_deletion/Ashkenazi_Jewish_intelligence_(2nd_nomination). TLDR: the topic itself may be notable but much of the speculation and primary research surrounding it are utterly unscientific. In other words, much more like Bigfoot than one might initially suspect. Generalrelative (talk) 01:32, 17 February 2021 (UTC)

Adoption studies page

Probably makes sense to mention that the Minnesota adoption study is the only one that followed up the subjects to adolescence. Typically the effects of shared environment (what most would call family environment/upbringing) disappear or at least diminish dramatically when adolescence is reached while the power of genetics to explain a larger share of the variation of the IQ increases.

This is also true of the Minnesota adoption study - during childhood it seemed that the shared environment had an impact, but when reaching adulthood the effect disappeared. So the studies of children cited as somehow evidence contrary to what the Minnesota study shows is misleading. However, I'm not accusing the editors of purposefully misleading - it was obvious from early discussions that many of the editors were not aware of how adoption studies worked and indeed thought they studied twins reared apart.

https://www.cambridge.org/core/journals/twin-research-and-human-genetics/article/wilson-effect-the-increase-in-heritability-of-iq-with-age/FF406CC4CF286D78AF72C9E7EF9B5E3F — Preceding unsigned comment added by 2A00:23C7:EE82:7701:84C5:2D7D:23DE:3559 (talk) 13:05, 1 March 2021 (UTC)

The article does mention the Minnesota follow-up study at age 17. As is clear from the article on the Minnesota study (see [15]), multiple interpretations are possible. The authors of the study themselves took an "agnostic" view about hereditarian explanations. Several later commentators, noting all the confounding variables that were not controlled for, have made it clear that the study conveys little of scientific value. NightHeron (talk) 15:39, 1 March 2021 (UTC)

Yes, but it doesn't mention that it's the only study that does that.

It shows similar results to all other adoption studies - that the effect of shared environment on IQ disappears entirely or is reduced dramatically as the adoptees reach adolescence. Its design and results are little different to all other adoption studies. — Preceding unsigned comment added by 2A00:23C7:EE82:7701:15CE:3B4A:8B47:E2B8 (talk) 00:46, 3 March 2021 (UTC)

Well, if you're right that the Minnesota study is the best of the adoption studies, and if the Minnesota study was inconclusive and had methodological flaws (confounding variables), then all that means is that the other adoption studies were even worse. So it's not clear what's accomplished by pointing out that the other adoption studies didn't bother to do a follow-up. NightHeron (talk) 03:04, 3 March 2021 (UTC)

It's worth mentioning that the impact of shared environment disappears or is dramatically reduced by adolescence. You can see the link I linked earlier to see this is what all adoption show. Twin studies also show this. Whether one is raised by high IQ adoptive parents makes little to no difference to ones IQ when one reaches adulthood. The biological parents IQ predicts ones IQ very well, whether the biological parents raised them or not. This is the conclusions from both adoption and twin studies, that are much more rigorous methodologically than most research cited in this Wikipedia article.

Wikipedia reports reliable sources, preferably academic studies. You have failed to cite any credible bases for your assertions. When you find such, come back and produce them. Until you do, no-one will take seriously anything you have to say, so you are wasting your breath. --John Maynard Friedman (talk) 16:15, 3 March 2021 (UTC)

Agreed. There also seems to be a misunderstanding here about the distinction between individual and group-level differences. The article cited by the IP is already discussed in Heritability of IQ, as is the fact that Although IQ differences between individuals have been shown to have a large hereditary component, it does not follow that mean group-level disparities (between-group differences) in IQ have a genetic basis. To successfully argue for a change in this article, the IP would need to provide a reliable secondary source that explicitly comments on the relationship between race and intelligence, which the article they've cited above does not. Generalrelative (talk) 16:34, 3 March 2021 (UTC)

Why is the study I linked not reliable? Since all adoption studies show that the effect of shared environment disappears or is dramatically reduced as adoptees reach adolescence, why should the Wikipedia article presume this is not the case when it comes to interracial adoption, especially since the only (imperfect) study of interracial adoption also indicates this?

What is the sort of evidence John Maynard Friedman you mean? — Preceding unsigned comment added by 2A00:23C7:EE82:7701:88D7:EE00:9A6F:9050 (talk) 18:09, 3 March 2021 (UTC)

Please read our policy no original research. Your question "why should the Wikipedia article presume this is not the case..." may seem rhetorical to you, but in fact there are a ton of reasons why heritability at the individual level does not translate to group-level differences, and a whole host of confounding factors which emerge at the group level that are not necessarily apparent to armchair observers. The way we handle complex scientific matters like this is through our policies and guidelines, for example no original research and reliable sources. If your interpretation of the science were correct (i.e. that twin studies provide evidence for genetic differences in intelligence between racial groups), it would be possible to cite a reliable secondary source that states this explicitly. Generalrelative (talk) 18:25, 3 March 2021 (UTC)

I'm not doing any original research. I'm just pointing out a well established finding from adoption studies that shows that the studies cited in the article that are supposed to show findings contrary to the one from Minnesota Transracial Adoption study do not show such findings. Comparing adoption studies of children that show shared environment having an impact with a study of adults showing shared environment does not have an impact is misleading, as the impact of shared environment is reduced dramatically or disappears when children reach adulthood according to all adoption studies. Right now the studies are portrayed as giving results different from the Minnesota one, which is not the case - they show same results and are in no way contradicting each other or providing findings that contradict that of the Minnesota adoption study.

It is original research to indicate that the adoption studies cited as showing results different from Minnesota ones are doing so. All studies (Minnesota one and the studies cited as showing findings contrary to that of Minnesota one) show shared environment during childhood. — Preceding unsigned comment added by 2A00:23C7:EE82:7701:88D7:EE00:9A6F:9050 (talk) 19:00, 3 March 2021 (UTC)

Source used in the article contradict the article

For example, Mackintosh "IQ and Human Intelligence, Second Edition" cited in the article (163) states about whether black-white IQ gap is genetic in origin: "it would probably be even more reasonable to aknowledge that the evidence is simply not sufficient to provide a definitive answer one way or the other - and possibly never will." (page 344).

Should we remove Mackintosh as an unreliable source, since it contradicts the idea that there is a scientific consensus that the gap is not genetic in origin? Or mention that the scientific consensus has not established evidence that has given a definitive answer one way or the other? Seems to be contradictory to define the same source as reliable in some instances while not seeing it as reliable in others. Especially since the claim that scientific consensus exists that it's entirely environmental is not backed up by any sources. — Preceding unsigned comment added by 2A00:23C7:EE82:7701:B58F:CF5A:2948:DE82 (talk) 21:01, 3 March 2021 (UTC)

Nope. Once again, please familiarize yourself with our policies and guidelines, for example no original research and reliable sources. It is perfectly normal for a source to be considered reliable for a certain set of claims (about which the author has special expertise) but not all claims (i.e. when they stray into speculation, as with the quote you've referenced above). Generalrelative (talk) 21:23, 3 March 2021 (UTC)

Currently there is no source cited for the claim that there is a scientific consensus that differences are 100% environmental. How is that allowed? How is the quote I cited anymore speculative or out his expertise than the claims he makes cited in the Wikipedia article? How can we establish objectively which claims are such that his expertise is credible and which are not? And how is quoting this person original research? Familiarising yourself with our policies and guidelines might also be good for you. Making claims that the scientific consensus is that the differences are 100% environmental without any sources is definitely not in line with the guidelines. — Preceding unsigned comment added by 2A00:23C7:EE82:7701:B58F:CF5A:2948:DE82 (talk) 21:41, 3 March 2021 (UTC)

I mean, words like "probably" and "possibly" are a good sign that the author is straying into speculation. And it's because Mackintosh was a good scholar that he provided these signposts to make such an evaluation easy. In other cases we need to exercise our collective judgment. As to your other (falsely predicated) questions, I won't repeat what has been stated repeatedly above. Feel free to search the archives for past discussions. I will remark that it is odd for someone to present themselves as being already familiar with our guidelines and policies while apparently unaware of how to sign their comments. In any case, unless I see evidence that you have actually become more familiar with these policies and guidelines –– and are thus able to present a cogent argument –– I will no longer be responding to your posts. This should not be taken as tacit consent for your views. Generalrelative (talk) 22:12, 3 March 2021 (UTC)

The same words are used when he discusses the Minnesota adoption studies? How do we establish what are the areas he has expertise to make claims about? — Preceding unsigned comment added by 2A00:23C7:EE82:7701:C100:2955:CCAB:E435 (talk) 00:05, 7 March 2021 (UTC)

No source for the claim that the scientific consensus is that the differences are 0% genetic?

The first paragraph states "Today, the scientific consensus is that genetics does not explain differences in IQ test performance between racial groups, and that observed differences are therefore environmental in origin."

All the surveys of intelligence researchers seem to suggest otherwise, unless there are some that I'm not aware, in which case they should be cited as a source. What are the sources for the claim in the first paragraph? Seems like POV pushing.

It's also in contradiction with articles such as this. — Preceding unsigned comment added by 2A00:23C7:EE82:7701:B053:9A89:8E8F:7F86 (talk) 00:02, 13 March 2021 (UTC)

This was discussed at great length last year, see [16]. The result of that RfC was that the view that some races are genetically superior to other races in intelligence is a fringe POV. There's no need to re-discuss this here. NightHeron (talk) 00:23, 13 March 2021 (UTC)

Clearly a faulty result, the opinion of that Wikipedia admin is contradicted by mainstream academic sources. This issue needs to be examined at a higher level I think. Spork Wielder (talk) 07:44, 13 March 2021 (UTC)

Spork Wielder, It was not the opinion of a single admin, but process that gathered opinions from many editors. This is pretty much the top of the pyramid as far as making content decisions on Wikipedia goes. Overruling it would require running another RFC, with similar levels of attendance, that goes the other way. That's not really likely to happen. MrOllie (talk) 13:19, 13 March 2021 (UTC)

It should also be noted that the result of the RfC was examined at two higher levels (AN and ArbCom), and was sustained. NightHeron (talk) 13:44, 13 March 2021 (UTC)

@NightHeron: Could you point me to the AN and ArbCom review of this RfC? I can only find the ArbCom review back in 2012 [17]. Thanks. Stonkaments (talk) 20:13, 14 March 2021 (UTC)

The AN closure review is here: [18]. I've had difficulty finding the ArbCom discussion, perhaps because no action was taken. A complaint to ArbCom was made by one of the editors who was unhappy with the closure of the RfC, and an IP-editor then joined in to attack me directly. That's the only time I've ever been brought into a discussion on ArbCom, and I'm not knowledgeable about how ArbCom works or why it's hard to find a record of that discussion. There is a reference to the ArbCom discussion and a link to the IP's attack on me here: [19]. Sorry I haven't been able to locate the full ArbCom discussion. NightHeron (talk) 21:25, 14 March 2021 (UTC)

The ArbCom request related to the RFC was filed as a clarification request on the old ArbCom case, such requests are archived on the talk page of the ArbCom case in question. ArbCom chose to do nothing (unsurprising, since reviewing content decisions is outside their mandate.) Reviewing the case did remind me that discretionary sanctions are authorized for this topic area, though. Continued warring against the RFC result could be handed with topic bans on WP:AE. MrOllie (talk) 21:54, 14 March 2021 (UTC)

• Did you read the article on the letter you linked? Most of the people who signed it were not experts in the field they were commenting on. The opinions of non-experts expressing opinions outside their area of expertise doesn't carry much weight. A similar strategy is frequently used to attack the science on climate change and even, in some cases, evolution - throwing together a list of scientists from unrelated fields who know nothing about the topic but who have strong feelings about it regardless; it's easy to find a ton of such names on any culture-war issue, but it certainly doesn't represent any sort of meaningful academic consensus. If you want some recent research rather than culture-war salvos from 30 years ago, see [20][21][22][23], in addition to many sources already in the article. --Aquillion (talk) 14:01, 13 March 2021 (UTC)

Your own source contradicts the claim that there is a scientific consensus that genetics explain 0% of the variation. First of all it states that there is ongoing debate, not a consensus about the topic. Second of all, it states the following for example:

" When the between group variance attributable to trait-associated SNPs is compared to the observed phenotypic between-group variance, over 85% of the between-group variance in IQ is not attributable to additive genetic effects, where at most 4.7-8.7 IQ points could be attributed to such genetic effects"

How can a source that says at most 4.7 - 8.7 IQ points could be attributed to genetic effects be used to make the claim that there is consensus that the effect of genetics is 0? The paper says that there is a possibility that the contribution of environment is 0, but surely this is different from claiming there is a consensus its zero?

Here is another Wikipedia article in direct contradiction with the opening paragraph of this article.

2A00:23C7:EE82:7701:101A:AB08:BD42:6253 (talk) 14:41, 13 March 2021 (UTC)

What you're claiming does not make logical sense. (1) Of course there's still "controversy" about whether some races are genetically inferior to others, just as there is still controversy over whether anthropogenic climate change is just a fabrication of climate scientists, about whether evolution is a fabrication of atheists, about whether aliens landed at Roswell, and so on. Controversies persist because huge numbers of people persist in believing in fringe theories. (2) When researchers write that they have rigorously shown that the genetic effect could not be more than a few percent, directly contradicting what racialist hereditarians say, that does not imply that they believe that it is greater than zero percent. It just means that they cannot prove that it's zero percent. The inability to prove rigorously that a fringe theory is false is not evidence that it's true. We can't prove that Bigfoot doesn't exist. (3) The Wikipedia article you cite describes a book that supports the POV of the racialist hereditarians, but the fact that Wikipedia describes that book and the reactions to it does not imply that Wikipedia supports its POV. NightHeron (talk) 14:59, 13 March 2021 (UTC)

@NightHeron: You're of course absolutely correct here (except that "Mainstream Science on Intelligence" was an open letter published in the Wall Street Journal, not a book). But I'd argue that it's entirely okay at this point to ignore this IP range. Picking one very specific statement about the limits of one specific method and using it to misrepresent the overall argument of the cited piece is really beyond the kind of thing we need to take seriously, even if we assume good faith. See their repeated exercise of similar strategies above. Seems to me that it is quite enough, as you did in your initial reply here, to refer them to last year's RfC. Generalrelative (talk) 15:26, 13 March 2021 (UTC)

@Generalrelative: Sorry -- I didn't notice that the IP referred to two different Wikipedia articles, and my comment related only to the IP's second example.

The IP doesn't seem to understand the difference between something being notable enough to have an article and being mainstream. Of course all sorts of fringe people and books have Wikipedia pages.

You make a good point that, when confronted with illogical POV-pushing, it's best to resist getting drawn into repetitive arguments. NightHeron (talk) 18:11, 13 March 2021 (UTC)

@NightHeron: Aha, my apologies for the overzealous correction. I hadn't noticed the second wikilink. Generalrelative (talk) 18:35, 13 March 2021 (UTC)

Except there is no similar debate in the scientific community about whether creationism is true or whether humans contribute to climate change. I find it unconvincing that the following professional bodies, from which the experts surveyed where randomly chosen, would be filled with "racialist hereditarians". It seems to me that Wikipedia has chosen that its editors are in a better position to know what the research says than the experts from the following institutions:

• American Educational Research Association (120)
• National Council on Measurement in Education (120)
• American Psychological Association:
• Development psychology division (120)
• Educational psychology division (120)
• Evaluation and Measurement division (120)
• School psychology division (120)
• Counseling psychology division (60)
• Industrial and organizational psychology division (60)
• Behavior Genetics Association (60)
• American Sociological Association (education) (60)
• Cognitive Science Society (60)

2A00:23C7:EE82:7701:101A:AB08:BD42:6253 (talk) 15:54, 13 March 2021 (UTC)

I agree with the IP. Any claims that the scientific consensus believes genetics plays no role in intelligence differences are simply not supported by the evidence in the field.

I would like to remind editors that there is a spectrum of fringe theories, and arguments in favor of a genetic component to racial differences in intelligence are a clear Alternative theoretical formulation on the far end of the spectrum, rather than obvious pseudoscience like creationism on the other end. Continued attempts to compare a reasonable alternative theoretical formulation, with a well-documented line of inquiry and support in the mainstream scientific community, to something like creationism or Big Foot, strains the limits on assuming WP:GOODFAITH. Likewise for implying that the science is settled in this area of inquiry, and that the only controversy is due to "racialist hereditarians" akin to creationists or climate deniers. Such framing of the debate, combined with the ongoing refusal to allow any additions to the article that suggest a genetic component may be involved, strike me as exhibiting WP:OWNERSHIP over the article and POV pushing.

Edited to add: When a source says "over 85% of the between-group variance in IQ is not attributable to additive genetic effects", it is a BIG stretch to conclude that the author really believes the effect is 0% but just can't prove it. Notice the author himself concludes the evidence is "consistent with genetic differences contributing little [emphasis added] to the Black–White gap". If he really thought the evidence showed that genetics had no effect, why wouldn't he say that, rather than little?Stonkaments (talk) 22:38, 13 March 2021 (UTC)

1) It seems you have not understood the source you are quoting at all. The author is saying that even using the flawed methodology favored by hereditarians "over 85% of the between-group variance in IQ is not attributable to additive genetic effects". The author is very explicit about this, stating that because "the assumptions of this model violate core principles of modern population genetics (such as no gene-environment interplay, gene-gene interactions, and similar allelic effect across populations), there is little reason to expect the genetic contribution to be this large in reality."

2) After pointing out a whole host of questionable methodological assumptions underlying this "best case estimate for genetic contributions to group differences in cognitive performance", the author concludes that "the mean-expected-difference provided here [4.7-8.7 IQ points] is likely an overestimate and should be thought of as the maximum mean difference attributable to genetic variation due to genetic drift. It is also important to note that the direction of the mean difference could favor Africans or Europeans with equal likelihood." This may be too subtle an argument for someone trying to trawl the article for shreds of doubt as to the state of the scientific consensus, but for a scientist this is actually a pretty devastating take-down.

3) You ask If he really thought the evidence showed that genetics had no effect, why wouldn't he say that, rather than little? If you had read carefully you would have seen that the author actually does remark that his results are consistent with the possibility that "the genetic contribution to the IQ gap is zero". Because this is a scientific paper, (rather than, say, an editorial) one wouldn't expect him to state his opinion on the matter, no matter how strongly held, so the absence of such a statement shows nothing. What the author does emphasize is that his study "demonstrates that patterns of genetic differences between African and European populations in the 1000 Genomes Project dataset is consistent with neutral evolution and insignificant genetic contribution to the Black-white IQ gap. In other words, the patterns observed in this study can be explained without appealing to the core tenets of the hereditarian hypothesis." That really is scientist-speak for the hereditarian hypothesis is like belief in the existence of Bigfoot: not based on evidence.

4) The article already does discuss alternative theoretical formulations, as well as the reasons why the scientific mainstream rejects them. Presenting them as though they were valid would violate WP:FRINGE per the findings of the RfC. Generalrelative (talk) 23:37, 13 March 2021 (UTC)

Thank you for the clarification, I admit I did only a cursory scan of the source based on the discussion above. Nevertheless, my point stands that it is flawed WP:OR to conclude that a best-case estimate of 15% genetic contribution "just means that they cannot prove that it's zero percent". This source is very clear that it is a take-down of the "hereditarian hypothesis", which he defines as "a significant portion of differences in cognitive performance between Black and White populations are caused by genetic differences due to natural selection". As the IP correctly pointed out, refuting this "hereditarian hypothesis" is different from the question of whether there is any genetic component.

Anyway, that was a minor side point. My larger concern is what I see as WP:OWNERSHIP and POV pushing by relegating the genetic component argument to the realm of pseudoscience like creationism and Big Foot. Per IP's second source, in response to the question, "Which of the following best characterizes your opinion of the heritability of black-white differences in IQ?", only 15% of experts who responded voted that the gap was "due entirely to environmental variation", vs. 45% who voted that it was a "product of genetic and environmental variation" (and 1% who said it was entirely due to genetic variation). Granted that book was written nearly 40 years ago, but shouldn't that give us pause? Can you honestly say that the article accurately portrays this level of dissent? Even the "History of the controversy" section gives the impression that the argument for a genetic component was always a fringe view held only by racists to justify slavery and other atrocities. Stonkaments (talk) 02:35, 14 March 2021 (UTC)

Thanks for being amenable to clarification. But I was responding to you, not to the IP's 0% argument, which seems like a red herring to me. The actual language that appears in our article is: "the scientific consensus is that genetics does not explain differences in IQ test performance between racial groups", and this is very well sourced. Whether that means literally 0% or merely an "insignificant" contribution isn't necessary for us to parse because the language ("genetics does not explain") comes straight out of Nature.

As to my own beliefs, one of the recent reports cited by Aquillion above stated them quite well: "[T]he claims that genetics defines racial groups and makes them different, that IQ and cultural differences among racial groups are caused by genes, and that racial inequalities within and between nations are the inevitable outcome of long evolutionary processes are neither new nor supported by science (either old or new)." [24] I don't know enough about that IQ Controversy book to speculate on its merits –– but even if it did accurately capture the state of professional understanding 40 years ago that would hardly be relevant to this article. We have a separate article on the History of the race and intelligence controversy for just such information.

In any case, let's all please try to respect the way Wikipedia works, which includes respecting (and not continually relitigating) the results of high-profile RfCs like [25]. There is really no room for ambiguity in the result, and no reason to accuse others of violating policy for working to keep this and related articles in accord with that result. Generalrelative (talk) 03:37, 14 March 2021 (UTC)

Do you have a link to the Nature article? I know it's been referenced before but I don't have it in front of me. I think it's important to look at the context of the article before concluding that there's no need to distinguish between no contribution and a >0% but insignificant contribution. I'll also note that Nature included the qualifier "an emerging consensus" that the Wikipedia article fails to reflect. The Wikipedia article also adds "and that observed differences are therefore environmental in origin", which I don't believe is reflected in the Nature article. Would it be better to simply quote Nature verbatim? And the Wikipedia article also states later on, "The current scientific consensus is that there is no evidence for a genetic component behind IQ differences between racial groups."--which is problematic because of the additional assertion of "no evidence", which I don't think is supported by the Nature article nor the RfC.

Surely the history of the debate is relevant to the article, hence why we have the "History of the controversy" section at all. Maybe it should rewritten as a WP:Summary style article, but as is it is very biased and misleading IMO.

I disagree that there is no ambiguity here. The fringe determination doesn't remove the responsibility to do the work and evaluate the evidence to determine to what extent there is a consensus that there is no genetic component, how widely held and what is the evidence in favor of the minority view that there is a small genetic component, etc. Can we agree that it's wrong to characterize a proposed genetic component as obvious pseudoscience akin to creationism? Stonkaments (talk) 05:04, 14 March 2021 (UTC)

I'll reply to your questions in turn but will not be continuing to debate these issues ad infinitum. My silence should not be confused with tacit agreement. I am simply tired of debating the same points again and again.

1) The Nature editorial from which this language comes is [26]. The quote is "There is an emerging consensus about racial and gender equality in genetic determinants of intelligence; most researchers, including ourselves, agree that genes do not explain between-group differences."

2) With regard to the statement "observed differences are therefore environmental in origin", this is WP:BLUE. There are only two options here (i.e. genetics or environment), and it is not WP:OR to make this clear to the reader. Further, in the body of the article we have four WP:RSs supporting the statement "Growing evidence indicates that environmental factors, not genetic ones, explain the racial IQ gap." The Nature editorial we rely on for framing our "genetics does not explain" language is far from our only source here; it's just explicit in a way that research articles typically cannot be.

3) Would it be better to simply quote Nature verbatim? No, this sentence in the lead summarizes a wide variety of WP:RSs. Again, we simply follow the phrasing of the Nature editorial because editorials are written in explicit, everyday language, and editorials in Nature are the pinnacle of the genre, at least as far as science is concerned.

4) With regard to the statement "The current scientific consensus is that there is no evidence for a genetic component behind IQ differences between racial groups", this is backed up by six WP:RSs. See especially Hunt and Mackintosh for a discussion of the lack of evidence. Both are reliable WP:SECONDARY sources, and neither of them can be accused of being overly harsh toward the hereditarians, yet both acknowledge the total lack of evidence to support the idea of a genetic link between race and intelligence.

5) We include only the most relevant information in the schematic History section in this article, saving all the nitty-gritty for History of the race and intelligence controversy. If the IQ Controversy book is not discussed at length in recent secondary or tertiary sources, it doesn't rise to the level of inclusion here. Your opinions about what is "biased" and "misleading" are not going to persuade anyone who doesn't already agree with your POV.

6) The RfC found: "There is consensus that the theory that a genetic link exists between race and intelligence is enough of a minority viewpoint in the scientific consensus that it falls under Wikipedia's definition of a fringe theory". That is indeed unambiguous. We do not promote WP:FRINGE on Wikipedia. We do report on its existence where appropriate, as we have done in this article (and more extensively in the History article). Pretending that there is a scientific rationale for believing that some genetic link exists between race and intelligence would clearly violate this finding, and no amount of hair-splitting will alter that.

7) I'm not sure about creationism, but I stand by my Bigfoot analogy. I think it captures well the persistence of belief in a thing about which there is absolutely no evidence but which some people want desperately to believe in anyway. The only difference is that believing in Bigfoot is harmless. Generalrelative (talk) 07:04, 14 March 2021 (UTC)

1) That Nature article is not an editorial. Holding it up as the definitive statement of the scientific consensus is wrong. It was the opinion of two researchers on one side of the debate; it was not written or endorsed by the editorial board. They also note that "plenty of scholars remain unpersuaded", and describe the hostile environment for any research that challenges the "politically correct" consensus. Here was the actual editorial in that issue[27].

2) It is indeed WP:OR. It is making the false assumption that the statement implies genes do not explain ANY between-group differences, when it could just as easily mean genes do not explain ALL between-group differences, genes do not explain any significant amount of between-group differences, or any number of other possibilities.

3) But we explicitly aren't following the source on a few important details. In addition to #2 above, the lead also fails to mention the authors describe an emerging consensus, and that a significant number of scholars disagree.

4) Thanks, I do plan to do a deeper dive on these sources (and others) concerning the claim that there is no evidence for a genetic component.

5) I disagree that this section adequately and neutrally summarizes the most relevant historical information. I think the fact that (assuming the source is correct) as recently as the 1980s the scientific consensus held that genetics played a role is highly relevant context for the article. I'll add that section to my to-do list.

6) "Pretending that there is a scientific rationale for believing that some genetic link exists between race and intelligence would clearly violate this finding"--this is simply not true, and is a misinterpretation of fringe, which says: "Alternative theoretical formulations generally tweak things on the frontiers of science, or deal with strong, puzzling evidence—which is difficult to explain away—in an effort to create a model that better explains reality." That completely contradicts your claim that fringe necessarily implies that there is no scientific rationale for the argument.

7) According to your source (and others provided by IP), the belief in a genetic component to the racial IQ gap has significant minority support in the mainstream academic community; this is simply not the case for Bigfoot or creationism. Additionally, the Nature commentary describes an environment that is overly hostile to research that dissents from the dominant view, which can "lead to a one-party science that squelches divergent views". I would suggest you reflect on that dynamic, and the motives for supporters of the dominant view to so fiercely attack any dissenting research, before casting aspersions on the motives of others. Stonkaments (talk) 15:58, 14 March 2021 (UTC)

This really needs administrative review because Generalrelative cherry picks and misrepresents a handful of sources as "consensus" in order to censor the view he doesn't like. I'm not sure someone who would claim "someone wrote it in Nature" makes something an uncontested fact should be editing Wikipedia. Spork Wielder (talk) 17:42, 14 March 2021 (UTC)

(ec) When an SPA comes in to make ridiculous accusations against a careful editor, in violation of WP:NPA, that tells us that this discussion has reached a dead end. As repeatedly mentioned above, Wikipedia editors arrived at a consensus in an RfC last year on Race and intelligence. If opponents of this consensus wish to continue bludgeoning, they should not be surprised if other editors decline to respond. NightHeron (talk) 18:45, 14 March 2021 (UTC)

Your stonewalling based on the faulty RfC is what needs to be reviewed. Spork Wielder (talk) 07:28, 15 March 2021 (UTC)

I should clarify something about point #4 above. The sources for the statement "no evidence for a genetic component" support what that part of the article used to say, which is "no direct evidence". This wording was changed to the current one based on the outcome of the RFC, without changing what sources were cited. It was stated explicitly that the wording was changed "without having to cite a source for the change in wording." The quoted comment was referring to the lead section, but the same wording was subsequently copied to the body of the article, again without a change in sourcing.

See also the summary given by Literaturegeek here. At the time when this wording was changed last year, multiple editors were pointing out that the sources did not support the change, but the outcome of the RFC was understood to require the change regardless of whether it was an accurate summary of what sources said.

The original phrasing "no direct evidence" was a verbatim quote from the Kaplan source, while the Nisbett et al. paper says "no new direct evidence". The Mackintosh source says (p. 358): "In spite of claims to the contrary, there is remarkably little evidence that the difference is genetic in origin." This is somewhat close to what the article cites him to say, but "remarkably little" does not mean the same thing as "none". This is not just a matter of phrasing: when Mackintosh discusses the hereditarian argument based on brain volume (p. 339), he does not reject that argument as entirely invalid, but instead argues that it could only account for a very small portion of the difference in average test scores. So, Mackintosh's phrasing "little evidence" is in fact a more accurate summary of his overall position than "no evidence".

The new wording is an especially severe distortion of what the Hunt source says. Here is a quote from that source (pp. 434-435):

"Rushton and Jensen (and Lynn) are correct in saying that the 100% environmental hypothesis cannot be maintained. Nisbett's extreme statement [that the differences are 100% environmental] has virtually no chance of being true. However, the 100% environmental hypothesis is something of a stalking horse. Many researchers who are primarily in environmental differences associated with racial and ethnic differences in intelligence would not be at all perturbed by an ironclad demonstration that, say, 3% of the gap is due to genetic differences. The real debate is over the identity and size of genetic and environmental influences on group differences in intelligence, not the existence of either one."

When someone tried to remove the same "no evidence" wording cited to the same sources (Hunt, Mackintosh, Nisbett and Kaplan) from a different article, Generalrelative explained in this edit summary that the new wording is required by the RFC. So it isn't necessary for him to re-explain why consensus requires the new wording. But let's please not pretend this change is something other than what it was widely acknowledged to be at the time: it is based on the consensus of Wikipedia editors in the RFC, not based on the sources. Gardenofaleph (talk) 18:35, 14 March 2021 (UTC)

Thank you for that additional context, that is very helpful. It seems clear to me when we have wording that is directly contradicted by the sources being defended by appealing to the RfC, something has gone very wrong in either the RfC itself, or the interpretation and implementation of that RfC (or both). Do you think we should revisit the RfC, or would it be more productive to examine how we can rewrite the text to stay consistent with the RfC while also aligning with the sources? Stonkaments (talk) 19:11, 14 March 2021 (UTC)

No "direct" evidence for a genetic component

Breaking this out in a new thread, as the previous discussion has stalled. I want to revisit this claim in the article: The current scientific consensus is that there is no evidence for a genetic component behind IQ differences between racial groups.

Gardenofaleph made a very strong argument that this sentence, as written, is wrong and is not supported by the sources cited[28]. Others have made similar arguments previously, as shown here[29].

For context, the original phrasing was "no direct evidence for a genetic component", but the word "direct" was removed as alleged MOS:WEASEL wording[30]. The fringe determination in the RfC[31] was also cited in support of removing the word "direct"[32], and the most recent revert said "not good wording, implies there is indirect evidence".[33]

As far as I can tell, no argument has been forwarded that no evidence for a genetic component is supported by the cited sources. To review, here is what the sources say:

• Hunt: "Nisbett's extreme statement [that the differences are 100% environmental] has virtually no chance of being true...The real debate is over the identity and size of genetic and environmental influences on group differences in intelligence, not the existence of either one."
• Mackintosh: "In spite of claims to the contrary, there is remarkably little evidence that the difference is genetic in origin."
• Kaplan: "no direct evidence" (original phrasing was a verbatim quote from the Kaplan source)
• Nisbett et al: "no new direct evidence"

Furthermore, later in the Wikipedia article another quote from Hunt says, in part, "The argument for genetic differences has been carried forward largely by circumstantial evidence."

So we have all of these highly reliable sources referencing some degree of indirect or circumstantial evidence for a genetic component, including Hunt which is arguably the most reliable tertiary source available. It seems clear the current claim of "no evidence" is false and unsupported by the sources; therefore I propose reverting to the original wording of "no direct evidence", or a similar alternative. I don't find the argument that this would somehow violate the fringe RfC to be convincing. And if it truly is the case where we acknowledge the wording is wrong but determine that it would contravene the RfC to fix it, I would remind editors of WP:IGNORE: If a rule prevents you from improving or maintaining Wikipedia, ignore it. Stonkaments (talk) 19:44, 15 March 2021 (UTC)

Maybe I could be convinced we need a modifying word somewhere in that sentence, but I don't think 'evidence' is the right thing to hang it on, because implying a category of indirect or circumstantial evidence leaves open how much (or how little) of it there really is. How about 'The current scientific consensus is that there is no evidence for a significant genetic component behind IQ differences between racial groups.'? - MrOllie (talk) 19:54, 15 March 2021 (UTC)

That is a fair point. I would support that change. Stonkaments (talk) 19:57, 15 March 2021 (UTC)

@MrOllie: Thank you for suggesting that wording. I'd also support that change provided that some words are added clarifying that there is also no scientific evidence which racial group would be the beneficiary of any very small genetic difference in intelligence. This was stated in the source quoted above by Generalrelative: "It is also important to note that the direction of the mean difference could favor Africans or Europeans with equal likelihood." Many people would read a statement that there is no "significant" genetic contribution to the black/white IQ gap to mean that there might be a slight amount of genetic inferiority of blacks compared to whites. As the source says, it is equally likely that whites would turn out to be inferior to blacks. Since a genetic component in group IQ differences is a matter of speculation, one can equally speculate that, if whites had been treated over the last 350 years as badly as blacks have been and if blacks had enjoyed the privileges that whites did, then the black-over-white IQ gap would be more than 15 points. In other words, the wording should clearly indicate that there is no scientific evidence of any race being genetically superior or inferior in intelligence to any other race. NightHeron (talk) 20:39, 15 March 2021 (UTC)

A possible sentence to add: "Nor is there any scientific evidence as to whether people of African descent or European descent would be favored by any very small genetic difference in intelligence that might exist." NightHeron (talk) 21:09, 15 March 2021 (UTC)

References

The framing of this debate as "genetic inferiority" is tiring, inflammatory, and arguably disruptive; speculation about a black-over-white IQ gap of greater than 15 points is unfounded and similarly unproductive. As for the claim that "the direction of the mean difference could favor Africans or Europeans with equal likelihood", one primary source should not be relied upon to the exclusion of the many more reliable secondary and tertiary sources that do not support such a claim. Stonkaments (talk) 21:18, 15 March 2021 (UTC)

@Stonkaments: Please stop your personal attacks against me, which violate WP:NPA. I am not being disruptive or inflammatory. In the context of discussing a black-white IQ gap, it is highly misleading to suggest that there might be a very small genetic explanation, since that suggests that a between-group difference, if there is one, would necessarily favor whites. In other words, it reinforces implicit biases against black people. It's not "inflammatory" to acknowledge this reality.

Why are you taking offense and reacting with such hostility to the notion that a racial difference in intelligence could favor blacks? NightHeron (talk) 21:48, 15 March 2021 (UTC)

@MrOllie: @NightHeron: For what it's worth, I oppose these proposed changes in wording, though I recognize that they come from a genuinely collaborative place. As I stated above, pretending that there is a scientific rationale for believing that some genetic link exists between race and intelligence would clearly violate the RfC consensus, as well as the cited sources. The OP is welcome to try to defend their interpretation of WP:IGNORE before WP:AE, but I suspect that the verdict is more likely to be WP:NOTHERE. Generalrelative (talk) 22:09, 15 March 2021 (UTC)

@NightHeron: I agree wholeheartedly that we need to take special care to avoid any misleading or ambiguous claims in this article. I believe the best way to do that is to be thorough and meticulous in presenting well-sourced information in a clear and neutral way. My specific objection to adding the additional sentence "Nor is there any scientific evidence as to whether people of African descent or European descent would be favored..." is with regards to sourcing and WP:UNDUE—that claim is based solely on one primary source, while we have a number of other more reliable source that do not support that claim. Adding that would make the article worse, not better. I feel that the wording MrOllie proposed is careful and neutral as-is.

@Generalrelative: Can you explain how the proposed change in wording to "no evidence for a significant genetic component" would violate the cited sources? As I mentioned in the OP, I have never seen that argument made, and the excerpts provided by Gardenofaleph indicate the change would in fact align more closely with the sources. Stonkaments (talk) 22:28, 15 March 2021 (UTC)

(ec)@Generalrelative: I think you're right. Stonkaments' responding to my attempt at a new wording by personally attacking me -- like the SPA's ridiculous attacks on you earlier -- shows that their only purpose here is to try to overturn the consensus and promote racialist hereditarianism. Their repeated violations of WP:NPA show that it was naive on my part to hope that this could end amicably.

As you have pointed out, we're under no obligation to relitigate the RfC in response to the refusal of some people to accept consensus. All of the claims they are making have been made and refuted many times before. NightHeron (talk) 22:43, 15 March 2021 (UTC)

The most important concerns on this article ought to be WP:Verifiability, and the prohibition against misrepresenting sources. Editorial consensus shouldn't be able to overrule both a basic Wikipedia policy, and one of the administrative remedies affecting this article.

MrOllie's proposed change is an improvement over the current wording, but the only source that it summarizes accurately is Mackintosh. The wording "no evidence for a significant genetic component" seems to be making a statement about the possible size of a genetic contribution, and saying that it must be small if it exists. Mackintosh discusses evidence for a very small genetic contribution in his comments about brain volume, and later concludes that there's "remarkably little evidence" for a genetic contribution. But the other sources say that that the evidence for a genetic contribution is indirect or circumstantial, without taking a position about its possible size. Hunt is very critical of all arguments that the genetic contribution is a specific size (including the argument that it's a size of zero), and describes these arguments (p. 436) as "overly precise statements on either the pro-genetic or pro-environmental side". So I think MrOllie's proposal is not consistent with what most of these sources say, and the original wording "no direct evidence" is the most accurate summary of the sources.

If MrOllie's proposal is the only one that can gain consensus, though, don't count me as opposing it. It's only consistent with one of the four sources, but I guess that's better than contradicting all four of them Gardenofaleph (talk) 00:27, 16 March 2021 (UTC)

Speculation is not science. Scientists cannot prove that the genetic contribution is zero. Nor can they prove that it's nonzero. Nor can they prove that, if it's small but nonzero, then it is positive. Nor that it's negative. (Positive means that if it weren't for environment blacks would still score below whites on average; negative means that they'd score higher.) This is the point that's made in the source that Aquillion, Generalrelative, and I have cited. It's a recent reliable source, published in the journal of the American Association of Physical Anthropologists. Nothing in recent reliable sources contradicts that.

Please keep in mind that the RfC is settled. Wikipedia works by consensus, which means that attempting to continually relitigate settled questions is viewed as unconstructive. NightHeron (talk) 01:03, 16 March 2021 (UTC)

To put the same point another way: A measurement is never a single value, it is always a confidence interval. If the genetic effect is zero, what we measure will always be "somewhere between minus epsilon1 and plus epsilon2", with epsilon1 and epsilon2 being positive numbers. All that happens is that the epsilons will get smaller over time. And the people who believe in a non-zero effect will always be able to say "you did not refute the non-zero assumption". This is logically related to Popper's "possibility of refutation" criterion for science: following Popper, the statement "the value is zero" is science because it is refutable by getting an interval which does not contain zero, while the statement "the value is not zero" is not science because it is not refutable. The burden of proof lies with the non-zero crowd, and until they get an interval without zero in it, science says the effect is zero. The same logic applies to lots of other pseudosciences too. The effect of homeopathy is also measured to be "zero plus/minus epsilon". --Hob Gadling (talk) 10:30, 16 March 2021 (UTC)

Now do the environmental effect to see how meaningless this is. Spork Wielder (talk) 12:01, 16 March 2021 (UTC)

Are you saying that there is no evidence that Blacks are poorer than whites and that they have poorer access to education in the US? That must mean that someone here must have mastered the feat of editing a parallel universe's Wikipedia. --Hob Gadling (talk) 12:27, 16 March 2021 (UTC)

None of the studies show there is a consensus that genetics play no role. The study you cited says that the evidence the study uses is compatible with the claim, just as it's compatible with genetics explaining up to 15%. The scientific evidence is compatible with genetics playing a role, which is also what all the (imperfect) surveys of experts suggests is their opinion. I think the wording should be rearticulated as something akin to "the scientific consensus is that the existing evidence is compatible with environmental factors explaining the differences". Any sources cited don't (and as Nighheron said, cannot) rule out genetics playing a role.

There seems to be a misunderstanding where some editors think that "is compatible with environmental explanations" is thought to mean "is incompatible with genetic explanations".

78.16.177.15 (talk) 01:40, 16 March 2021 (UTC)

@NightHeron: Could you clarify what is the point you're trying to make about speculation vs science and what can be proven? None of the proposed changes to the wording are making any claims as to the level of proof that has been presented, and WP:Verifiability and WP:DUE apply to reliable sources across the board–not only to what science has proven–so I fail to see how those distinctions are relevant here. That said, I could probably get on board with a mention of the uncertainty involved, including Hunt's argument criticizing all estimates of the size of the genetic contribution as "overly preciese", if you think that would be important context. Stonkaments (talk) 01:49, 16 March 2021 (UTC)

Please see the above comment by Hob Gadling, who explained this better than I've been able to. NightHeron (talk) 10:49, 16 March 2021 (UTC)

Thank you. But I have a correction and two additions: "science says the effect is zero" should actually be "science says the effect is either zero or so close to zero that nobody has been able to measure it". Please note that this implies that those who believe it is different from zero do not do so because of solid empirical evidence but because of errors such as overgeneralization, cherry picking, wishful thinking, or for ideological reasons. Here another rule of thumb comes into play: Occam's Razor. If you can explain what we see without the assumption that there is an effect, you should. --Hob Gadling (talk) 11:42, 16 March 2021 (UTC)

The garbled attempts at scientific reasoning have no place here, since it is not the place for us to state our opinions. Nevertheless I can't help but laugh at those saying science proves things, or that an observation must be zero, when the explanation of the observation is the point under discussion, and then getting Occam's Razor exactly backwards by positing mysterious unidentified "environmental variables" to reduce an observed difference to zero. I guess this is why we look at reliable sources, not the intellectual mediocrities and social justice activists that edit Wikipedia. Spork Wielder (talk) 11:57, 16 March 2021 (UTC)

Regarding the environmental variables, see my response above.

Those are things scientists understand and apply implicitly. The purpose of explaining them was to make some of the editors understand the parts they are missing when scientific sources are quoted, but it was not to be expected that all of the targets would grasp the concept. I'll just say WP:CIR and drop it.

Reliable sources have been quoted above. It is your choice not to accept what they are saying. Wikipedia's choice is different from yours. --Hob Gadling (talk) 12:27, 16 March 2021 (UTC)

Hunt says: "Nisbett's extreme statement [that the differences are 100% environmental] has virtually no chance of being true...The real debate is over the identity and size of genetic and environmental influences on group differences in intelligence, not the existence of either one." This is arguably the most reliable tertiary source that we have, and presents a clear and direct claim that is as close to a refutation of your argument about the confidence interval as you'll get. Just because he leaves open the remotest of possibilities of the effect being zero (as you say, science can never prove with 100% certainty), per Occam's Razor, the best-guess estimate is a >0% genetic component. However, that claim is covered by the fringe RfC and I am not seeking to re-litigate or overturn that here.

But the disputed sentence in the article makes an even stronger claim, saying there is no evidence for a genetic component. This is not a claim about the confidence interval, and I have yet to see anyone make an argument that it is an accurate representation of the cited sources (Generalrelative made the claim but did not elaborate an argument[34]). This is a problem, especially because there is a special restriction on this page about misrepresenting sources.[35]

refutation of your argument about the confidence interval Wrong. What I wrote was just a further explanation of what NightHeron said, and it is all very basic science and independent of the specific race-and-intelligence example. Hunt's opinion about what Nisbett said does not invalidate it. --Hob Gadling (talk) 15:01, 16 March 2021 (UTC)

Oh, and yes, "no evidence for a genetic component" is a claim about the confidence interval, namely that it contains the value of zero. If it did not, that would constitute evidence for a genetic component. Can we stop this please? --Hob Gadling (talk) 15:04, 16 March 2021 (UTC)

(ec) Actually Hunt makes exactly the same point as Hob Gadling has above. Scientists cannot prove that the genetic contribution to the IQ gap is zero because, as Hunt helpfully points out, "doing so would require proving the null hypothesis and, as any good statistics instructor will tell you, that is a logical impossibility." (Hunt p.447) Hereditiarians, on the other hand, could indeed have proved that it was non-zero if that were the case, and have consistently failed to do so. This is the meaning of the Hunt quote: "Of course, tomorrow afternoon genetic mechanisms producing racial and ethnic differences in intelligence might be discovered, but there have been a lot of investigations, and tomorrow has not come for quite some time now."

Hunt's opinion about Nisbett's view, on the other hand, is just an opinion. And indeed, it conflicts directly with Mackintosh's opinion that "One could reasonably defend Nisbett's argument that the gap was entirely environmental in origin." (Mackintosh p.344)

One more point before I drop this thread: Hunt and Mackintosh are indeed reliable sources on the state of understanding 10 years ago among psychometiricians because that was their area of professional competence. When they speak about genetics specifically (and today's consensus) however, we should take their views with a grain of salt. If there is anyone who is truly open to persuasion that the hereditarian view is wrong (who is not already convinced by the mountain of evidence already presented), I would suggest reading "Race, genetics and pseudoscience: an explainer" by four prominent geneticists –– Ewan Birney, Jennifer Raff, Adam Rutherford, and Aylwyn Scally: [36] I won't be debating this. I know that it's been attacked by know-nothings on all the usual white-supremacist websites. I am also aware that it's a blog post. While blog posts by established experts are sometimes allowed in articles (per WP:RSSELF), I am not arguing that this source be included in this article. I am legitimately trying to inform those who do not yet know why the overwhelming majority of geneticists believe what they do about this topic. Please consider this my last straw of patience for the current crowd of race/IQ truthers. After this my engagement with specious argumentation will be limited to repeating the results of the RfC and reporting sanctionable behavior if necessary. Generalrelative (talk) 15:11, 16 March 2021 (UTC)

@Hob Gadling: Stop what, exactly? You are ok with the fact that the article's claim of no evidence for a genetic component is contradicted by the cited sources? Would you care to address that? I think that is the whole crux of the matter here. Stonkaments (talk) 15:16, 16 March 2021 (UTC)

It's amazing that these prominent geneticists identified the genes responsible for intelligence and found they were uniformly distributed among races. That's way more advanced the state of research I was aware of. Truly stunning and groundbreaking work that ends the debate. Nobel prize winning stuff. One only wonders why they didn't publish their data and methods, rather than asserting it on a blog. Spork Wielder (talk) 09:00, 19 March 2021 (UTC)

It is, as Generalrelative just explained, contradicted not by "the cited sources" but by an opinion in one of the cited sources. You people do not understand how scientists talk and what they mean when they do, nor can you tell the difference between a statement of fact and an opinion. We all tried to explain it but seem to have failed. I don't think any further explanations will do anything to change the situation. --Hob Gadling (talk) 15:24, 16 March 2021 (UTC)

Can you point to just one example (of either fact or opinion) from any of the cited sources that supports the assertion that there is no evidence of a genetic component? Stonkaments (talk) 15:37, 16 March 2021 (UTC)

Stonkaments, That is what Hunt is communicating, that there probably is some difference (as there would be between any two randomly selected groups), but it is within the observational error range of the methods we have. This is scientist for 'no evidence' the same way that 'works as well as placebo' is medical researcher for 'it doesn't do anything'. MrOllie (talk) 16:18, 16 March 2021 (UTC)

I don't agree with that interpretation at all. It seems to me he's saying that it's abundantly clear (based on circumstantial evidence) that there is some genetic component, it's just a question of how much. Stonkaments (talk) 16:29, 16 March 2021 (UTC)

Hob Galding, have you looked at the Hunt source itself, and not just the excerpts from it that various editors are posting here? Hunt definitely is not making the argument that you think he is. He gives a detailed summary of the arguments made by hereditarians, and then says "In general, I find their arguments not so much wrong as vastly overstated. But overstatement does not mean that there is no point to them." This is the context in which Hunt goes on to say, "Rushton and Jensen (and Lynn) are correct in saying that the 100% environmental hypothesis cannot be maintained." So in context, it is very clear that Hunt is saying that the hereditarians have presented enough evidence to demonstrate that the differences are not 100% environmental, but not enough evidence to support Rushton and Jensen's 80% genetic "default hypothesis" (which Hunt calls an "excessively precise statement").

As I said in my earlier post, I understand that consensus is opposed to changing this part of the article, or any of the other articles that the same wording cited to the same sources has been copied to. But I'd like everyone at least be aware of the reality of what's happening on these articles, which is that this is a case of consensus superseding the Verifiability policy. Gardenofaleph (talk) 23:36, 16 March 2021 (UTC)

I only know the few parts that have been quoted here, by any party, but nothing from those quotes convinces me that "no evidence" is wrong. What Hunt says still just sounds like his opinion.

Is there evidence that homeopathy works? No. Well, actually yes, if you insist on answering the question literally, but it is really crappy evidence that does not count because it is the kind of evidence you would expect to be found even if the effect does not exist: bad studies without control groups, small studies with very little power, irreproducible statistical flukes. So, effectively, there is no evidence. If the effect exists, it is so tiny that nobody could reliably reproduce it. Still, there are people who are convinced that it must exist, and continue looking for it.

The race-and-intelligence situation looks very much the same to me: If the effect exists, it is so tiny that nobody could reliably reproduce it. Still, there are people who are convinced that it must exist, and continue looking for it. Hunt is one of them. That is how it looks to me from the quotes.

There are many studies that find that IQs are determined mainly by genetic factors. People who understand statistics will know that such results are not natural laws. They are properties of data sets. If you only use university students in your research, or only US citizens, or only US citizens with specific properties, your results will be determined by the distribution of the parameters you are looking at within your data set. If the environmental conditions of your subjects do not vary a lot - e.g. if you do not look at any people who live in actual slums - the influence of environmental factors will be lower than if you did. If you want to interpret data from another source, which includes people living in slums, you cannot just extrapolate from the more-homogenous-populations studies.

But not everyone who uses statistics understands this. Many people, even scientists, view statistics as a tedious tool, a couple of recipes you follow without having to understand what exactly you are doing. Even published scientific studies are sometimes full of rookie mistakes. Statistical significance is one of the things used by thousands of scientists who do not understand what it really means - see Replication crisis. I think those IQs-are-determined-by-genetic-factors studies are misleading many such math-averse scientists into taking their results for real effects instead of just properties of data sets. Hunt sounds like one of those. Unless he gives actual results from valid studies with the right scope, "no evidence" seems right. --Hob Gadling (talk) 12:12, 17 March 2021 (UTC)

Hrm... Hunt was the president of the International Society for Intelligence Research, which has historically been at least amicable to the hereditarian view (ie. willing to have its more out-there advocates on the editorial board of its journal). And just reading the section of his book being cited here I see several eyebrow-raising things - he devotes an entire aside to James Watson's controversial comments, where he defends them as factually accurate on every point.^Note _{(Including the bit where Watson said His hope is that everyone is equal, but “people who have had to deal with black employees find that this is not true.” Hunt helpfully notes that this is an accurate statement because "As references in this chapter have shown, in the United States the work performance evaluations of African Americans are, on the average, lower than the evaluations received by white workers. This is true for both objective and subjective evaluations. The difference is much less than the difference in test score averages." I couldn't resist including that as a footnote.)} He also gives massive amounts of focus to Rushton, Jensen, and Lynn. I would also point out that while he presents himself as a neutral observer who refuses to take sides, [37] describes him as a heriditarian, which pretty closely lines up with his career, expressed sympathies, and, of course, stated opinions. Like... obviously scholars exist who support the hereditarian position, but that doesn't make it the mainstream consensus. Hunt doesn't present what he's saying as the mainstream consensus, he just says it's what he believes. And it is fairly notable that even the scholars supporting that perspective tend to cite the same small number of people, which doesn't exactly support the argument that their views represent a broad scientific consensus. --Aquillion (talk) 14:24, 17 March 2021 (UTC)

I guess the peer reviewed literature on the heritability of intelligence is written by people who just don't understand. Luckily we have Wikipedia editor Hob Gadling to show us the truth. Spork Wielder (talk) 15:43, 17 March 2021 (UTC)

Not "the peer reviewed literature", only part of it. --Hob Gadling (talk) 16:06, 17 March 2021 (UTC)

So where does this leave us? To summarize:

• The current wording, no evidence for a genetic component, fails WP:VERIFIABILITY and misrepresents the cited sources, all of which reference some level of indirect or circumstantial evidence for a genetic component. The only argument made that the sources do in fact support the current wording came from (in my opinion) a misinterpretation of Hunt's views tied to a tangential discussion on confidence intervals around the null hypothesis, speculation vs. science, and facts vs. opinions.
• It looks like there is strong opposition to the original wording, no direct evidence of a genetic component, despite being the most faithful representation of the cited sources.
• There has been some qualified support for no evidence of a significant genetic component. While still not an entirely accurate representation of the sources, it would be an improvement.

Can we work towards building consensus on how to incorporate a change to no evidence of a significant genetic component? What additional context or qualifying statements, if any, would be needed to go along with that? Stonkaments (talk) 15:47, 17 March 2021 (UTC)

That summary is obviously biased. The truth of the matter is that the discussion has (1) two editors plus one sarcastic SPA who dislike the conclusion of the RfC on race and intelligence and want to undermine it by changes in wording, and (2) five editors who have been arguing against this. A consensus already exists on Wikipedia, and it does not support introducing weasle-words that suggest that there's scientific evidence that some races are genetically inferior to others. NightHeron (talk) 16:02, 17 March 2021 (UTC)

Can we really vote to ignore core policies? Neat. Spork Wielder (talk) 16:12, 17 March 2021 (UTC)

I moved your response where it belongs: after the contribution it responded to. But consensus is not a vote, and your interpretation of the source is just your interpretation while the consensus has a different one. So, your problem, not ours. --Hob Gadling (talk) 10:27, 18 March 2021 (UTC)

Why should "Hunt's views" be relevant? Science is about results, not about views. --Hob Gadling (talk) 16:06, 17 March 2021 (UTC)

The views in Hunt's Intelligence are relevant because:

1. It is being cited as a source (and his views are being used to defend the current wording, when in fact they argue against it)
2. It is considered a reliable tertiary source, which Wikipedia policy notes "can be helpful in providing broad summaries of topics" and "may be helpful in evaluating due weight"[38]. Stonkaments (talk) 16:20, 17 March 2021 (UTC)

It is a tertiary source for the stuff Hunt gets from other sources and summarizes, but a primary source for Hunt's own view about it. Which part of "Science is about results, not about views" did you not understand? --Hob Gadling (talk) 10:27, 18 March 2021 (UTC)

I have one last proposal that I hope might be amenable to everyone, before I bring this to dispute resolution. Would there be support for simply removing the sentence entirely? The sentence that follows, Growing evidence indicates that environmental factors, not genetic ones, explain the racial IQ gap, seems to provide an adequate summary of that section and the current scientific consensus, without the issues of verifiability and misrepresenting sources that have been brought up here. Stonkaments (talk) 01:22, 18 March 2021 (UTC)

I oppose deleting that sentence, which correctly represents current scientific consensus. I would also urge you to drop the stick rather than taking a matter to dispute resolution that was already litigated in great detail in the RfC and elsewhere, including this talk page. NightHeron (talk) 02:33, 18 March 2021 (UTC)

I also oppose deleting that sentence, and I reject the premise that it misrepresents the sources. - MrOllie (talk) 03:30, 18 March 2021 (UTC)

As pointed out in earlier discussion, there is no source that states the existence of such consensus, and all the surveys of experts suggest this is not the case. Even if these surveys like this are imperfect, the burden of proof is upon those claiming consensus is 100% environmental to provide better sources with more rigorous methodology which they have been unable to do. Indeed there seems to be inability to cite any source even with less rigorous methodology that would suggest such consensus exists. It seems people mistake "compatible with entirely environmental explanation" as "incompatible with any genetic factor". I support changing the sentence to something that mentions that something along the lines of "the differences are compatible with environmental explanations".

2001:14BB:70:C4C5:A041:ECC7:B828:6037 (talk) 08:54, 18 March 2021 (UTC)

You're mixing up two very different statements. No one is saying that the scientific evidence is "incompatible with any genetic factor", which would be the same as saying that science has proved that it's zero. Neither is science incompatible with the existence of Bigfoot.

You, too, please drop the stick. NightHeron (talk) 10:09, 18 March 2021 (UTC)

So then we agree that at least some experts think the evidence is at least compatible with a genetic factor? It's kind of an academic question, because we can all see that some do. Anyway, this is clearly going nowhere and needs to go to dispute resolution. Spork Wielder (talk) 12:22, 18 March 2021 (UTC)

You still do not get it. All experts think that, and they will all continue to think it forever, because there is no possible evidential situation which would be incompatible with it. That is trivially true and does not need to be mentioned in the article.

Yes, this is going nowhere because it already has gone somewhere and is there now. The dispute resolution has already happened and does not need to happen again. That is why, as NightHeron said, you should drop the stick. EOD as far as I am concerned. --Hob Gadling (talk) 13:27, 18 March 2021 (UTC)

All you people are doing is putting words in the mouths of experts. Those who say genetics is a likely factor, are not saying "it's impossible to mathematically prove the genetic effect is exactly zero". That's just something you made up. Spork Wielder (talk) 13:58, 18 March 2021 (UTC)

I have raised this issue here: Wikipedia:No_original_research/Noticeboard#Race_and_IQ:_"no_evidence"_for_genetic_component? Stonkaments (talk) 22:47, 18 March 2021 (UTC)