The Unz Review • An Alternative Media Selection$
A Collection of Interesting, Important, and Controversial Perspectives Largely Excluded from the American Mainstream Media
Email This Page to Someone

 Remember My Information

Topics Filter?
Academia Africans AI American Media Antiracism Artificial Intelligence Black Crime Blacks Brain Scans Britain Coronavirus Culture/Society Diet Disease Economics Feminism Flynn Effect Gender General Intelligence Genetics Genomics Heredity History I.Q. Genomics Ideology Immigration Intelligence IQ Iq And Wealth Political Correctness Psychometrics Public Schools Race Race And Iq Race/IQ Race/Ethnicity Racial Reality Racism Science Statistics Vaccines Academy Awards Adoption Affirmative Action Afghanistan Africa African Genetics Airlines Ancient Genetics Animal IQ Antisocial Behavior Arthur Jensen BBC Behavior Genetics Belgium Bill Gates Blogging Bmi Boeing Boris Johnson Brain Size Brain Structure Brighter Brains Capitalism Celebrity Censorship Chanda Chisala Charles Murray Child Abuse China Christianity Christmas Cohorts Conspiracy Theories Correlation Crispr Davide Piffer Deep Sleep Development Diversity Dogs Donald Trump Dysgenic Education EU Exercise Eyes Face Shape Faces Finance Floyd Riots 2020 Foreign Policy Free Market Game Theory Gays/Lesbians Gender Equality Genetic Diversity Genetic Engineering Genetics Of Height Geography Google Government Spending Group Intelligence GWAS Hbd Health Height Heritability High Iq Fertility Historical Genetics Human Biodiversity Human Genetics Income Indian IQ Inequality Jeremy Corbyn Lead Poisoning LGBT Longevity Love And Marriage Mental Health Mental Illness Mental Traits Microaggressions Multiculturalism Muslims Nassim Nicholas Taleb Nature Vs. Nurture Netherlands Nigeria Nutrition Personality Physical Anthropology Pollution Prince Harry Psychology Race Riots Racial Intelligence Rationality Richard Lynn Robert Plomin Schizophrenia Scrabble Sex Differences Sleep Smart Fraction South Africa Stephen Jay Gould Taxes Team Performance Terrorism Traffic Fatalities Uruguay Vaccination Wealth Wealth Inequality Weight Weight Loss Welfare State Woodley Effect Working Memory World War II
Nothing found
Print Archives3 Items • Total Print Archives • Readable Only
Nothing found
 TeasersJames Thompson Blogview

Bookmark Toggle AllToCAdd to LibraryRemove from Library • B
Show CommentNext New CommentNext New ReplyRead More
ReplyAgree/Disagree/Etc. More... This Commenter This Thread Hide Thread Display All Comments
These buttons register your public Agreement, Disagreement, Thanks, LOL, or Troll with the selected comment. They are ONLY available to recent, frequent commenters who have saved their Name+Email using the 'Remember My Information' checkbox, and may also ONLY be used three times during any eight hour period.
Ignore Commenter Follow Commenter
We have to be resigned to living in a world where social outcomes are substantially determined at birth

I do not wish to accuse my readers of being economists, sociologists or anthropologists, but I am willing to bet that some of you think that the way your parents brought you up, and the schools and community you were raised in, had a big influence on your later achievements in life.

A reasonable belief, but probably a mistaken one.

In fact, it is likely that all that matters is who your parents were, by which I mean your blood parents. Furthermore, conceiving you was the big step, and the rest was due to your being kept alive, and little more.

Here is a discussion paper, written for a conference-attending professional audience, which gives a technical account of the preliminary results of a large study still in progress. I will concentrate on some of the main points, and will leave discussion of some other matters (like assortative mating) to another later post.

For Whom the Bell Curve Tolls: A Lineage of 400,000 English Individuals 1750-2020 shows Genetics Determines most Social Outcomes

Gregory Clark, University of California, Davis and LSE (March 1, 2021)

Economics, Sociology, and Anthropology are dominated by the belief that social outcomes depend mainly on parental investment and community socialization. Using a lineage of 402,000 English people 1750-2020 we test whether such mechanisms better predict outcomes than a simple additive genetics model. The genetics model predicts better in all cases except for the transmission of wealth. The high persistence of status over multiple generations, however, would require in a genetic mechanism strong genetic assortative in mating. This has been until recently believed impossible. There is however, also strong evidence consistent with just such sorting, all the way from 1837 to 2020. Thus the outcomes here are actually the product of an interesting genetics-culture combination.

Greg Clark says:

It is widely believed that while social status – measured as occupational status, income, health, or wealth – is correlated between parents and children, this correlation is driven by parental investments in children, or by cultural transmission. This belief has profound influence on peoples’ perception of the fairness of social rewards, of the need for government intervention in the lives of disadvantaged children, and of the social value of education. In this paper I test whether culture/human capital or genetics offers a better explanation of the inheritance of social attributes, using a lineage of 402,000 English individuals 1750-2020. To do so we have to specify both a general model of cultural/human capital inheritance, and one of genetic inheritance. There is already a well established model of additive genetic inheritance, formulated by Fisher in 1918. This I test against the data below. Specifying a model of cultural/human capital transmission as an alternative is more difficult. The ways culture/human capital has been hypothesized to operate are many and varied.

So, Clark offers us a straight fight between a simple genetic formula and the more amorphous, all-encompassing but vague cultural explanations.

The genetic formula was proposed by Fisher, but since putting a formula in the text cuts readership in half I will eschew it, and instead describe it in plain English: most complex human traits are influenced genetically by the additive effect of many locations in the DNA where there are variants in the base pairs (none, one, or two positive variants), where each location itself has a very small effect on the trait in question. So, you just add up all those small effects to get a total score for the trait in question, which is the additive inheritance. That’s it.

For example, Galton noticed that parent’s height was passed on to their children, though the precise mechanism was not known. The long run intergenerational correlation should be close to 0.5.

Now we would calculate height accurately with polygenic risk scores, but that is not essential. With the additive model you don’t have to worry about fancy stuff like dominant or recessive genes, or interactions between different genes at different locations. This is a very simple and clear model of the intergenerational transmission of social status. In this model, you don’t even have to worry about the environment. Genetics is all you need in order to predict your achievements in life.

When we come to social outcomes the idea here will be that people inherit a set of abilities that determine, whatever their parents’ circumstances, their ultimate outcome in terms of occupational status, education, health or longevity. For wealth, where there is an actual transfer between generations, we would not expect the Fisher rules to hold.

Now a tiny bit of jargon. Each parent transmits their genes, plus a random element. You get exactly 50% from each parent (except the sex genes) but the 50% you get from your father (or mother) are slightly random samples of his genes. Your brother gets a different sample. That is why siblings are similar, but uniquely slightly different. So, what you see in each person is the (phenotype) which is their ancestry (genotype) plus the slightly random assortment they got from their parents. On average, mothers will contribute as much as fathers, so if the genetic theory is correct, mothers will contribute as much to the status of their children as do fathers. (Probably not what the cultural view would predict).

It seem very likely that people choose each other by carefully getting to know their partners (the phenotype), and in that way they will also pick out their partner’s underlying family qualities (the genotype).

The simple genotypic model allows you to work out the degree of relatedness of all your relatives, as shown below. As you get further away from yourself the correlations will go down in a lawful way caused by the genetic totals.


This little chart is very interesting, in that it allows the testing of genetic links in various ways. Your correlation with your cousin is the same as your correlation with your great-grandparent whom you probably haven’t ever met. So, on the Fisher equation, finding out about the achievements of a dead great-grandparent should tell you as much about your eventual social status as your still-alive cousin, even though the latter might have helped you with a job offer. On the cultural model, a cousin will have more potential influence on you than a dead great-grandparent.

We can see if the simple genotype model explains the obtained data.

Clark says, about assortative mating, or marital selection (marriage partners choosing who they marry) which he labels “m”:

There is no intrinsic reason that people should match in marriage based on their social abilities. They could match purely on physical characteristics, or on personality traits unrelated to social and economic outcomes. They do match, in some societies, on whatever cousins are available of the appropriate age and gender. Interestingly, though, if matching is just to a random cousin then in equilibrium in such a society “m” will be quite low at around 0.23, whereas in England the evidence for “m”, as mentioned, is in the order of 0.6-0.8.

• Category: Economics, Science • Tags: Inequality 

For decades, homosexuality has been known to be associated with psychological disorder. In the past, the interpretation was that social ostracism caused stress, and that in turn led to psychological distress. If that was true, the massive changes in the acceptability of homosexuality should have reduced the pressures of social rejection, and led to an improvement in psychological well-being. So, how are results turning out now?

A review studied epidemiological studies to look at the mental health of the non-heterosexuals.

Sexual orientation and symptoms of common mental disorder or low wellbeing: combined meta-analysis of 12 UK population health surveys
Joanna Semlyen, Michael King, Justin Varney & Gareth Hagger-Johnson
BMC Psychiatry volume 16, Article number: 67 (2016)

Sexual orientation and symptoms of common mental disorder or low wellbeing: combined meta-analysis of 12 UK population health surveys – PMC (

Around 1–2 % of the United Kingdom’s adult population identify as lesbian, gay or bisexual (LGB) and 5 % as non-heterosexual, although because sexual orientation comprises identity, behaviour and attraction, the chosen definition used can lead to variability in these estimates. We know that sexual minority populations experience poorer physical heath and engage in riskier health behaviours such as smoking and hazardous drinking. These inequalities may emerge in adolescence and early adulthood, then persist throughout the life-course.

Symptoms of poor mental health (e.g. anxiety, depression) and low wellbeing (e.g. not having ‘positive mental health’) are common in the adult population but there is established evidence that adults who identify as lesbian, gay or bisexual are at higher risk of experiencing these symptoms than adults who identify as heterosexual. A systematic review of the prevalence of mental disorder, substance abuse, suicidality and self-harm in LGB people showed that these populations experience a greater incidence of depression, anxiety, suicidality and substance misuse than heterosexuals. Meta-analysis following this review found that LGB people were around twice as likely to have attempted suicide in their lifetime and have around 1.5 times higher prevalence of depression and anxiety disorders in the preceding 12 months. Associations between minority sexual orientation and poorer mental health have persisted over time with recent studies showing the same effects as older studies. Such disparities are thought to emerge early in adolescence and persist into adulthood.
Of the 94,818 participants in the analytic sample (those with available data on sexual orientation identity, mental health and covariates), 97.2 % as heterosexual, 1.1 % identified as lesbian/gay, 0.9 % as bisexual and 0.8 % as ‘other’ (Table 1). People meeting the threshold of common mental disorder or low wellbeing were significantly different across all study variables (using bivariate t-test or chi-square tests): they were younger, comprised more females, and had lower levels of educational attainment, more current smokers, more longstanding illness/disability and fewer married/co-habiting participants than those below the threshold (Table 2). Significantly higher proportions of those who identified as lesbian/gay, bisexual and ‘other’ were found among those who met the mental disorder threshold.
Our results are consistent with evidence internationally that non-heterosexual adults are at increased risk of mental health symptoms compared to heterosexuals, but provide important new insights by suggesting that younger and older non-heterosexual adults are particularly vulnerable (compared to those at mid-life).

To my mind the “minimally adjusted” figures are most reliable, since the “additionally adjusted” list of factors includes ones that might be a consequence of sexual orientation.

Bisexual men 2.66
Gay men 2.25
Bisexual women 2.23
Other women 1.69
Other men 1.52
Lesbian women 1.38

Non hetero Men (total) 2.14
Non hetero Women (total) 1.76

Leaving aside the sex of the person, the relative odds are worst for Bisexuals 2.37, Lesbian/Gay 1.82, Other 1.59. This is a significant and real increase in the risk of psychological distress. Why is this the case?

The authors suggest that stigmatisation, in various forms and operating from adolescence is the most likely cause. They do not raise any other hypotheses.

One possible interpretation is that bisexuals are the most confused about their identity, which suggests that a sense of clear identity about sexual orientation is partly protective in mental health terms, although non-heterosexuals are more distressed. It was sometimes asserted that bisexuals had “the best of both worlds” in that they had the double the number of prospective sexual partners, but even if that is the case, the result is that this is a very disturbed group.

More generally, it could be that, even given social acceptance, something about the non-heterosexual mode has an intrinsically unsettling effect. A psychological interpretation is that one source of meaning, the strengthening of a personal commitment to children and the resultant future perspective/slow life history is denied to many non-heterosexuals. (This could be tested by comparing them with childless heterosexuals).

Another hypothesis is that whatever the primary cause which leads to non-procreative sexuality, it is also a cause of mental distress, either by some direct route, or as a strong consequence. In this interpretation, even open support of non-hetero orientation by the public is of little help, because the set of choices, orientations and behaviours is inherently damaging to well-being. How might that happen? Does it lead directly or indirectly to a more vivid emotional sensibility, so that emotions become a part of identity?

Usually, women are more emotionally vulnerable than men, and twice as likely to be emotionally upset. This is not the pattern here. Both are more disturbed than normal, but non-hetero men are more feminine in their vulnerability, with high levels of distress; and non-hetero women, while still being distressed, are relatively more masculine, and a bit more resilient than the men. It seems that there has been an inversion of the usual sexual differences in psychological vulnerability.

Another interpretation is that being non-hetero is inherently unsettling. How could one test this? There has been a massive change in the public status of homosexuals, which ought to have had a very big effect. Results as shown in the above paper should be very unlikely if the main cause of distress is lack of public acceptance. The fact that the pattern is international also makes it seem that social acceptance is not the key variable, since this still varies in different countries.

The cause of non-hetero preferences is not determined. The usual interpretation is that these preferences would not of themselves cause distress. However, perhaps the cause of the preferences also brings about vulnerabilities, as a result of increased emotionality.

• Category: Science • Tags: Gays/Lesbians, LGBT, Mental Illness 

One conception of race is that it is skin deep, and is no more than a matter of skin pigmentation. By implication, such a categorisation is superficial, trivial, and unlikely to be an explanation of any presumed racial differences in behaviour. There may be effects due to people making unwarranted assumptions based on skin colour, but that says more about them than anything else.

According to the skin-pigmentation theory, an Xray should see right through that, to the reality of the bones underneath. This would reveal, the theory says, that people are alike under the skin. Perhaps so, but is it true of their bones?

[Submitted on 21 Jul 2021]
Reading Race: AI Recognises Patient’s Racial Identity In Medical Images
Imon Banerjee, Ananth Reddy Bhimireddy, John L. Burns, Leo Anthony Celi, Li-Ching Chen, Ramon Correa, Natalie Dullerud, Marzyeh Ghassemi, Shih-Cheng Huang, Po-Chih Kuo, Matthew P Lungren, Lyle Palmer, Brandon J Price, Saptarshi Purkayastha, Ayis Pyrros, Luke Oakden-Rayner, Chima Okechukwu, Laleh Seyyed-Kalantari, Hari Trivedi, Ryan Wang, Zachary Zaiman, Haoran Zhang, Judy W Gichoya

Background: In medical imaging, prior studies have demonstrated disparate AI performance by race, yet there is no known correlation for race on medical imaging that would be obvious to the human expert interpreting the images.

Methods: Using private and public datasets we evaluate: A) performance quantification of deep learning models to detect race from medical images, including the ability of these models to generalize to external environments and across multiple imaging modalities, B) assessment of possible confounding anatomic and phenotype population features, such as disease distribution and body habitus as predictors of race, and C) investigation into the underlying mechanism by which AI models can recognize race.

Findings: Standard deep learning models can be trained to predict race from medical images with high performance across multiple imaging modalities. Our findings hold under external validation conditions, as well as when models are optimized to perform clinically motivated tasks. We demonstrate this detection is not due to trivial proxies or imaging-related surrogate covariates for race, such as underlying disease distribution. Finally, we show that performance persists over all anatomical regions and frequency spectrum of the images suggesting that mitigation efforts will be challenging and demand further study.

Interpretation: We emphasize that model ability to predict self-reported race is itself not the issue of importance. However, our findings that AI can trivially predict self-reported race — even from corrupted, cropped, and noised medical images — in a setting where clinical experts cannot, creates an enormous risk for all model deployments in medical imaging: if an AI model secretly used its knowledge of self-reported race to misclassify all Black patients, radiologists would not be able to tell using the same data the model has access to.

This is an astounding paper. It appears to reveal that, using artificial intelligence, deep learning methods lead to the detection of race in X ray images, even if all possible giveaway signals are stripped out of the image. That is extraordinary. The next point of interest is that the authors have made it very clear that they are alarmed that this is possible, and warn that it might lead to evil consequences.

They assert that race is not based on ancestry, but is a social construction and based on self-report.

So, to take the authors at their word, this paper is a warning, as well as the publication of a finding. The underlying results show that you cannot get rid of race in an Xray, even though radiographers are reportedly unable to detect race in such images, and do not know how artificial intelligence achieves such good results.

I read these results some months ago, and now come back to them, seeing this as an important finding, and a vivid example of what authors must do in academia when they report unwelcome results.

First, let us look at what the authors say to deflect critics. Here is their statement:

We emphasize that model ability to predict self-reported race is itself not the issue of importance. However, our findings that AI can trivially predict self-reported race — even from corrupted, cropped, and noised medical images — in a setting where clinical experts cannot, creates an enormous risk for all model deployments in medical imaging: if an AI model secretly used its knowledge of self-reported race to misclassify all Black patients, radiologists would not be able to tell using the same data the model has access to.

Bias and discrimination in Artificial Intelligence (AI) systems has been heavily studied in the domains of language modelling1, criminal justice 2, automated speech recognition 3 and various healthcare application domains including dermatology 4,5, mortality risk prediction 6andhealthcare utilization prediction algorithms7 among others.

While AI models have also been shown to produce racial disparities in the medical imaging domain 9,10 , there are no known, reliable medical imaging biomarker correlates for racial identity. In other words, while it is possible to observe indications of racial identity in photographs and videos, clinical experts cannot easily identify patient race from medical images

Race and racial identity can be difficult attributes to quantify and study in healthcare research 15, and are often incorrectly conflated with biological concepts such as genetic ancestry 16. In this work, we define racial identity as a social, political, and legal construct that relates to the interaction between external perceptions (i.e. “how do others see me?”) and self-identification, and specifically make use of the self-reported race of patients in all of our experiments

After all those virtuous statements, they hope they are in the clear. They are against the evil discrimination shown by artificial intelligence. Only then can they turn to the very bad news.

In this study, we investigate a large number of publicly and privately available large-scale medical imaging datasets and find that self-reported race is trivially predictable by AI models trained with medical image pixel data alone as model inputs. We use standard deep learning methods for each of the image analysis experiments, training a variety of common models appropriate to the tasks. First, we show that AI models are able to predict race across multiple imaging modalities, various datasets, and diverse clinical tasks. The high level of performance persists during the external validation of these models across a range of academic centers and patient populations in the United States, as well as when models are optimised to perform clinically motivated tasks. We also perform ablations that demonstrate this detection is not due to trivial proxies, such as body habitus, age, tissue density or other potential imaging confounders for race such as the underlying disease distribution in the population. Finally, we show that the features learned appear to involve all regions of the image and frequency spectrum, suggesting that mitigation efforts will be challenging.

Essentially, a good data-crunching program can work out the race of people in X rays, and nothing can be done about it. This is a horrible result which cannot be “mitigated”. The authors do not for a moment consider the possibility that race is biologically real.


Headlines have to grab attention, and the two headlines in the Sunday Times certainly did that. Usually considered a mildly conservative Sunday paper, with a circulation of 648,000 it is twice as popular as the next rival, the Sunday Telegraph. A Sunday paper is often the one that families are most likely to read and discuss together.

Sajid Javid orders racial bias review after Covid deaths

Medical devices ‘made for white people’ may have driven higher minority fatality rates



‘Racist’ oxygen device may explain why Covid hit minorities so hard


So, that is two “racist” implications to begin your family Sunday.

Notice that the paper did not say that there was to be an investigation into the possible causes of Covid fatalities, which would be the sensible way to approach the topic, since all possible causes could be mentioned and discussed.

For example, the death toll in the first wave of Covid hit over-weight people particularly hard. Racial differences in body mass index, and overall high national body mass index may have been an important cause, but there are many others.

My own view is that we do not yet know what caused the racial differences in fatalities. In the latter phase of the pandemic, differential uptake of vaccinations will have been a new factor.

Now, back to the main story. Racism has been given pride of place, to the detriment of other factors. Perhaps Sunday newspaper editors and writers find that irritating their readers actually boosts readership, but this is a distinctly odd way to approach an unresolved issue.


The Health Minister has said:

When I walk to my office, there’s a board showing everyone who’s held this role for over a century, and being the first name on that list from an ethnic minority is a responsibility I take very seriously.

I’m determined to take a fresh perspective to this position, and do whatever it takes so that in this country, your health and your experience of health and care isn’t dictated by where you live or where you come from.

Because although we’ve come together as a nation to fight this virus, the pandemic has shown that in many areas we’re far apart. At the height of the Covid peak last winter, black, Asian and other minority ethnic groups made up 28 per cent of critical-care admissions in England — about double their representation in the population as a whole. So one of my first visits in this role was to Blackpool, one of the parts of this country where life expectancy is in decline. I spoke about the Office for Health Improvement and Disparities, an organisation that was launched last month and has so much potential to tackle these injustices.

Notice the phrase “these injustices”. Different outcomes may be due to differences in life styles and differences in genetic susceptibilities. At the moment, we are not sure, but the condemnation has already been issued by the Health Minister. In his judgment, someone has perpetrated an injustice on ethnic minorities.

Odd that a national priority is being favoured on the basis of the genetic background of a Health Minister. Imagine if the next one says: “being White British, I think that poor whites should be a priority, because after all, they are the majority of the users of the health service, and also, they have short lifespans”.

Also, it is a bit odd to get into this issue without talking about lifespan data, this being the keystone of most health inequality debates.

Here is what the Office for National Statistics says about lifespan and racial background:

  • In the period 2011 to 2014 in England and Wales, both males and females in the White and Mixed ethnic groups had lower life expectancy at birth than all other ethnic groups, while the Black African group had statistically significant higher life expectancy than most groups.
  • Statistically significant higher age-standardised mortality rates from cancer were present among males and females of the White ethnic group compared with Black and Asian ethnic groups.
  • Statistically significant higher age-standardised mortality rates from circulatory (heart and related) diseases were present among Indian, Bangladeshi and Mixed males and Pakistani, Indian and Mixed females compared with the White group.
  • Cancers and circulatory diseases account for 61% of male and 53% of female deaths in the study and are therefore an important influence on the life expectancy differences seen between ethnic groups.
  • These results reveal important patterns in life expectancy and mortality by ethnic group which are complex but nevertheless consistent with most previous studies; further research is required to investigate the reasons for the differences, with potential explanations including past migration patterns, socioeconomic composition of the groups, health-related behaviours, and clinical and biological factors.

So, whites and part-whites (and some other mixed groups) live shorter lives, Black Africans longer ones. (For the avoidance of doubt, I am not accusing Black Africans of racism towards white people, but in these extraordinary times I feel I need to make that clear).

Sajid Javid is of Pakistani origins, so his children would have lifespans which are higher than those of White British. Here is the relevant Table, ranked by male lifespan.

This is a very interesting, and somewhat unexpected, set of results. As usual, there are potential complicating factors, including the younger ages and better health of some recent immigrant groups. Infant mortality, cardiovascular disease (CVD) and diabetes are higher among Black and South Asian ethnic groups. An analysis of data from the Clinical Practice Research Datalink (Lawson et al 2020) showed that people in the South Asian ethnic group (including Bangladeshi, Indian and Pakistani ethnic groups) had higher ischaemic heart disease, hypertension and diabetes prevalence than those in the White ethnic group. Conversely, those in the Black ethnic group had lower ischemic heart disease than those in the White ethnic group.

In fact, different racial groups have different patterns of health problems. If lifespans are seen through the lens of racism, then whites and part-whites have a grievance against all more recent arrivals.

It would have been useful to include these findings in the article, but we are reading a newspaper, after all.

The Health Minister continues:

For example, research has shown that oximeters, which monitor oxygen levels and are used to see whether treatment is needed for Covid-19, are less accurate on people with darker skin. One of the founding principles of our NHS is equality, and the possibility that a bias — even an inadvertent one — could lead to a poorer health outcome is totally unacceptable.


Sajid Javid is working with his American counterpart, Xavier Becerra, on introducing new international standards to ensure that medical devices have been tested on all races before they are allowed to be sold.


Is it ever possible to work out whether media are biased? Bias may be in the eye of the beholder, and perhaps we are all too prone to seeing bias whenever our preferences are challenged.

One way is to study the stated political preferences of journalists, and to compare them with national political preferences as shown by election results. Will journalists be to the left, to the right, or bang in the middle of the national political dimension?

Before getting into that topic, start with another question: why would anyone be a journalist? The only qualification is to be able to write, which most people can do. Having an opinion helps, but other skills are not essential. If you want to write on any topic, you simply ring people up, interview them, and write up your story. If people read it, you are a journalist.

Seen from a career point of view, writing things is not a very exclusive occupation. The bar to entry is not high. Teaching maths at school is more demanding, as are keeping accounts, servicing washing machines and repairing engines.

Nonetheless, why be a journalist when you can do other things? One commonly stated reason is: To make a difference. By implication, journalists want to change things by exposing them.

Many studies have indicated that there is a left-wing bias in the media. Indeed, lists of the most watched TV news channels and most read newspapers suggest that there are simply more left-wing outlets than right wing ones, so it is more than bias, it is hegemony. A simple explanation is that journalism is a left-wing activity, by and large. Left wing people are trying to change society. Right wing people are trying to make money, not scribbling.

The authors took great care to find objective ways to categorize political parties on the left to right dimension, only to find that these detailed methods correlated at .85 with the Wiki descriptions. However, the independent raters were far less likely to rate political parties as being “far Right” than was the case for Wikipedia descriptions.

They studied 17 countries, mostly European ones. They researched journalist’s actual voting behaviour, or voting intention

It can be seen that, apart from in Slovenia, journalists are to the left of the countries in which they work. This is a massive effect. It holds true even when countries lean to the left, as some European countries do.

All their data can be found in a publicly available data repository.

The main finding is simple: journalists favour left-wing parties, with a correlation of .5 though this is mostly due to their support of centre left parties, not the far left ones.


Compared to the general voting population, journalists prefer parties that are associated with the following ideologies: green parties/environmentalism, feminism, support for the European Union, socialism. Conversely, journalists are less likely than the general voting population to support parties associated with the following ideologies: national conservatism, libertarianism, populism, nationalism and conservatism.

the general population votes about 6.1 times more for national conservative parties as journalists do, whereas journalists vote about 3.0 times more for green parties.

Journalists lean left overall. Another group who write for a living are academics. Which way do they lean?

Langbert (2018) found that the ratio of Democrat to Republican professors was 17.4:1 in History, 43.8:1 in Sociology and 133:1 in Anthropology.

This may lead to a self-confirming amplification effect: journalists are more likely to quote left-leaning academics, who will thus have a higher profile, and will get cited and funded more often, and have more influence on other academics, and thus lead their fields in particular directions. If the left dominate in both media and academia, then the best-known research will be left-inspired.

Is the bias media and academia a bad thing? Yes, and it would have been as bad if a bias to the right had been revealed. The ideal is that both journalists and academics should be even-handed, and give a balanced evaluation of the available evidence. Devoutly to be wished, but rarely achieved. Perhaps people love a fight, and love taking sides.

I doubt that any steps can be taken to ensure balance in journalists. Many people on the right will feel that they have better things to do than go about convincing people. If that is the case, the best that can be hoped for is that journalists nail their colours to the mast so that readers can be warned where their deepest preferences lie. Readers have to pick their way through different news sources, trying to put together the least implausible account. Let a thousand flowers bloom, as Mao said before cutting off their heads.

As for academia, academics have made a profound life choice: they are not direct producers of wealth, but provide a service at a cost. It is natural that, having chosen that life, they are in favour of more funding for research, and willing to countenance more taxation to fund that research. If their research is based on solid methods their personal preferences need not be fatal to truth seeking. But I am left with a feeling that if a few more right-leaning persons might be willing to enter academia, they would provide a small dissenting voice which might illuminate some blind spots.

Currently, journalists in 17 countries lean left, and that distorts the basis on which those democracies work.

• Category: Science • Tags: American Media, Political Correctness 

Books have titles so that readers are tempted to buy them. Such titles are a general indication, and the text will give the further explanations. Neither Plomin nor Harden need be taken literally, but their choice of analogy reveals a general attitude: Plomin sees genetics as being more causal than does Harden. His reference to a blueprint may seem out of place, but genes do unfold in a predictable manner most of the time. How appropriate is the analogy of a lottery?

Here we have to assume that the lottery is an honest one, where every ticket holder has the same chance of winning as any other ticket holder. The chance of a win is winners divided by the number of lottery tickets, and the return on the lottery is the cost of a ticket compared with the probability of getting a particular prize. For example, in the UK there is a lottery (Premium Bonds) in which you keep your stake, but share with others the interest due on your stake. There are £1 million prizes, which tend to attract investor’s attention, and far more common far smaller prizes. How lovely to have a monthly chance of winning a million pounds for a one-pound stake! Yet, however many tickets you buy, the overall return is 1%, so it is not even compensation for inflation. All lotteries are a snare for the gullible.

Is genetics a true lottery? Of course not. Simple examination reveals the analogy to be misleading. Not all outcomes are equi-probable. Intelligence and other characteristics are heritable, and lotteries do not show any effects of inheritance. You pay for a ticket with a fair chance of winning, you do not roll up to claim a prize for privilege. Nor are you rewarded for being a frequent gambler, merely fleeced because of your foolishness. Even your identical twin has no higher chance than you do of winning a real lottery.

Also, people do not mate at random. They make choices. You can only regard your birth as random if you were conceived as a result of a vast masked orgy.

However, some chance is involved in genetic transmission, and that is why the analogy flourishes.

Two bright parents have a higher-than-average chance of getting a bright child, but it is not guaranteed. Birth injuries are one reason, rare mutations another. The other important reason, apparently difficult for some to accept, is that a high correlation is not a perfect correlation. There will be some surprises and disappointments, all as part of the genetic package. Eggs and sperm fuse, and although each parent as a result contributes exactly half of the DNA, the additive genetic inheritance of each child will vary somewhat around the average of the two parents as a result of Mendelian segregation. Kids vary somewhat, though less in families than in the general population.

All parents will have children of similar ability to them because intelligence is 0.8 heritable. Furthermore, instead of a mean absolute difference of 17 IQ points which would obtain if there was zero correlation between parent and child intelligence, their children will have a 12-point mean absolute difference. (I assume a sibling correlation of 0.5 for intlligence, which is the best-supported estimate).

As a rule of thumb, any family will show about two thirds of the variation found between unrelated persons. Children will have different outcomes (and the brighter ones in each family will earn more).

As to regression to the mean, most of the regression will tend to be to their own ancestral family mean. Think of it as a weighted mean of your ancestors, with more recent relatives having greater weight.

If everyone always regressed to the population mean, no differences would ever be found as a result of selective marriages. No family would have a tradition of scholarship nor a particular capacity for banking, or sporting prowess. No bloodline of race horses would matter very much.

To the contrary, endogamous marriage traditions follow the breeder’s equation (which applies to all species), and so long as the characteristic is heritable and confers a slight advantage, it will spread in the descendants, giving them propensities which make them differ from other groups. This is a response to selection, and the more severe the selection over the generations, the greater will be the effect.

Some will hold their breath and dive deeply for pearls, others will calculate their business deals carefully so as to buy those pearls and keep them in their bank vaults.

India has been doing this since at least 300 BC and the differentiation of jatis shows up in genetic studies. The prevalence of rare diseases in India is a testament to the longevity of this practice in the subcontinent. Razib Kahn has explained this in detail.

Thousands of years later the upper stratum of Indian society has a noticeable overrepresentation of ancestry from the Sintashta horse lords of the Bronze Age, an aristocracy that has somehow maintained itself for 4,000 years. This is true in every corner of the subcontinent with Brahmins in particular, and genetics indicates that many of the Brahmin groups have had a coherency that dates back thousands of years. They monopolized elite positions in a primitive agro-pastoral Iron-Age society, and they remain overrepresented in elite positions in a world where India is slouching towards its destiny as an advanced technological society. This is a miracle of continuity and stability which the British could never have invented. Some things about India are truly eternal.

So, is having children a gamble? Not really. It is not a true gamble in the sense that everyone has equi-probable outcomes. For example, if every newborn was sequestered and then put up as a lottery prize for the general public, then that would be a real gamble. Horrible, but a gamble. (Sometimes I fear that only a policy as severe as this will ever satisfy social engineers).

In the real world, your children will be very much like you and your partner, with abilities very much like yours. Your children will vary around the parental mean with somewhat smaller variation than that observed in the general population. They are children, not clones, so somewhat different but very like you, and more like each other than the general population.

So, not a guarantee, but a strong likelihood.

• Category: Science • Tags: Genetics, Heredity 
Effects of adoption on intelligence: 42% heredity, 8% environment?

I don’t do policy, but how about this one? In addition to all public policies aimed at getting rid of the achievement gaps between different groups, why not take an intensive approach? Continue with every program which is already under way, but add this one.

Get every child who is under-performing to live permanently from an early age with an adoptive family. This will give the children a deep immersion in all the things that good families can give their own children. We will carefully select adoptive parents who really want to adopt, many of which will have no children of their own, but very much want to bring up children.

This is an extreme experiment, far more profound than having just a few hours of enriching experience at a nursery. It will affect every waking moment, every spontaneous incident and remark, each instant of togetherness. Every shared experience can be used to teach, train and pass on the insights of wealth and status. How much will this boost IQ? To give this experiment every chance of showing its effects, we will wait till the children have grown up, and are 30 years of age. By now they should be well-established into their upward trajectories as a consequence of their enriched upbringing.

We do not actually have to do this experiment, many researchers have followed up adoptees. Here is the most recent study.

Genetic and environmental contributions to IQ in adoptive and biological families with 30-year-old offspring. Emily A. Willoughby, Matt McGue, William G. Iacono, James J. Lee. University of Minnesota Twin Cities, Department of Psychology, Minneapolis.

Intelligence 88 (2021) 101579

The authors say:

While adoption studies have provided key insights into the influence of the familial environment on IQ scores of adolescents and children, few have followed adopted offspring long past the time spent living in the family home. To improve confidence about the extent to which shared environment exerts enduring effects on IQ, we estimated genetic and environmental effects on adulthood IQ in a unique sample of 486 biological and adoptive families. These families, tested previously on measures of IQ when offspring averaged age 15, were assessed a second time nearly two decades later (M offspring age = 32 years). We estimated the proportions of the variance in IQ attributable to environmentally mediated effects of parental IQs, sibling-specific shared environment, and gene-environment covariance to be 0.01(0.00-0.02], 0.04 [0.00-0.15], and 0.03 [0.00-0.07] respectively; these components jointly accounted for 8% of the IQ variance in adulthood. The heritability was estimated to be 0.42 [0.21-0.64]. Together, these findings provide further evidence for the pre-dominance of genetic influences on adult intelligence over any other systematic source of variation.

While adoption studies have provided key insights into the influence of the familial environment on IQ scores of adolescents and children, few have followed adopted offspring long past the time spent living in the family home.

So, bright parents create bright environments (add 1%), brothers and sisters in that family contribute to that environment (add 4%), and the interaction adds another 3%, thus 8% in all. However, if we look at the confidence limits (in brackets above) the low-ball figures could conceivably be nothing for the environment and 21% for genetics. The high-ball estimates would be 24% for environment, 64% for genetics. An even larger sample of adopted children would reduce those confidence limits.

At the Minnesota Center for Twin and Family Research (MCTFR), the Sibling Interaction and Behavior Study (SIBS) has followed a sample of adoptive and biological Minnesota families for nearly two decades. Initial IQ assessments were conducted when offspring were approximately 15 years of age, and this paper reports on new assessments taken at approximately 30 years of age. At age 15 children will still be under family influence, by 30 they will have established their own lives.

A recent development which transforms adoption research is to include entirely gene based polygenic risk scores. These are the new kids on the block, which provide the first approximation to a pure predictive genetic score. At the moment they are not able to capture all the effects which we know are due to heredity, as shown in twin research, but they are steadily increasing in power.

Our polygenic scores for educational attainment provide what is, at moment of writing, the largest R 2 estimates for any cognitive phenotype (0.113 for Total IQ; 0.154 for Verbal IQ). The particularly high predictive validity for PGS EA for verbal IQ is perhaps to be expected given that verbal IQ correlates more strongly than other IQ subscales with educational attainment, particularly for parents. These scores also enable a unique test for the so-called “placement effect,” wherein adoptees (typically twins reared apart) are thought by some skeptics to resemble their adoptive parents prior to placement, thus biasing biometrical estimates. By demonstrating a total lack of evidence (p =0.514) for a correlation between parents and adoptive offspring in polygenic scores, we provide support for the validity of at least some adoption studies in establishing causal inference.

That is to say, the idea that adopted kids are placed with adopting parents of apparently similar ability is not supported by these findings. (Critics said that the finding that identical twins reared apart correlated in intelligence was not due to their genetics, but to them having been placed in similarly bright families).

On the very important question as to whether adoption has a long-term impact on the ability of adopted children, the answer appears to be that no such effect can be found.

By examining parent-offspring resemblance in a sample of offspring that are among the oldest of any adoption study of IQ to date, we have effectively tested for the presence of parenting effects that would have persisted for more than a decade after the conclusion of the typical rearing period. No such persistence is found to occur in our unique sample.

This is pretty clear. It is unlikely that adoption has long-term effects on adopted children’s intelligence. This should give pause to all those proposing interventions less intensive than the full time parenting provided by adoption and expecting great results.

• Category: Science • Tags: Adoption, Heredity, Indian IQ 

The Rand Corporation had a look at the factors which led to effect war fighting, and found that ability was a key factor. Thanks to commentator Mac Tonight for the link.

Determinants of Productivity for Military Personnel: A Review of Findings on the Contribution of Experience, Training, and Aptitude to Military Performance
Jennifer Kavanagh
Prepared for the Office of the Secretary of Defense. Approved for public release; distribution unlimited, 1981.

The report reviewed previous studies where experience and training had been evaluated while service personnel carried out various tasks. The historical trends they identified were towards a more cognitively demanding, technically sophisticated type of warfare, which will have made their conclusions even more relevant to today’s warfare. The report illustrates a general principle that holds up in all settings: success is more likely in any enterprise if you recruit bright people.

We know from previous work on training, summarized by Linda Gottfredson, that brighter recruits complete training faster, better, and then go on to apply that training more quickly into new situations. Brighter recruits are faster learners and better appliers. For that reason, training is not quite what is seems. It is not a uniform causal variable. It can be completed faster by brighter recruits, and applied faster and more widely. For that reason, although training is required, it does not follow that more training will boost lower ability recruits to the levels of brighter persons, not in this lifetime anyway. Nonetheless, training has an effect, and skimping on it reduces performance.

All recruits take the Armed Forces Qualification Test, and this can be used to sort soldiers into five categories, from the brightest downwards: I, II, IIIA, IIIB, IV. The first three grades are above average, the last two below average.

In the carrier landing exercise, for example, individuals were scored on a seven-point scale, ranging from dangerous to excellent. The effect of a career decrease in training hours of 10 percent led to a 10 percent increase in the number of unsatisfactory landings, from 14 percent to 24 percent of the total, and a 5 percent decrease in the number of excellent landings, to 28 percent of flights.

Winkler, Fernandez, and Polich (1992)looked at the relationship between AFQT and the performance of three-person teams on communications tasks, including making a system operational and troubleshooting the system to identify faults. They find a significant relationship between the group’s average AFQT score and its performance on both activities. On the first task, they find that if the average group AFQT is lowered from the midpoint of category IIIA to the midpoint of category IIIB, the probability that the group will successfully operate the system falls from 63 percent to 47 percent. Similar results are found for the troubleshooting task; the probability that a group would identify three or more faults falls drastically as average AFQT score fell. Another important observation is that the effect of AFQT is additive, meaning that each additional high-scoring team member increases the overall performance of the team. This is particularly important in the military context, given the number of group-centered tasks the armed forces are required to complete.

On page 27 the author turns to mental ability measures. Once again, the Armed Forces Qualification Test sorts soldiers into five categories, from the brightest downwards: I, II, IIIA, IIIB, IV.

AFQT and experience appear to be fundamentally different measures of quality. While AFQT measures an individual’s innate ability, experience considers personnel performance and skill level as developed and manifested over time. This relationship is an important one from the perspective of our discussion because AFQT as a proxy for personnel quality can be used to guide military recruitment and requirement determinations and can aid in the development of a more effective and cost-efficient military structure.

Tank crews do better when the drivers and gunners are brighter:

For example, they find that an increase in AFQT score from category IV to category IIIA leads to an improvement of 20.3 percentage points in performance. A similar increase in AFQT for the gunner in the same exercise will lead to a performance increase of 34 percentage points. These results are consistent with the arguments that AFQT score is an effective indicator of personnel quality and that having a force made up of personnel with higher AFQT scores contributes to more effective and accurate team performance.

A study by Winkler, Fernandez, and Polich (1992) offers additional support and evidence for this finding. The authors examine the relationship between AFQT score and the performance of two communication activities. The sample included 84 groups from active-duty signal battalions and 240 teams recently graduated from the Signal Center’s advanced individual training (AIT) course. In the first task, the three-person teams were asked to make a communication system operational. In the second, the teams were expected to identify and repair a number of faults in the communication system.

The model predicts that for active-duty units with an average AFQT at the midpoint of category IIIA, there is a 63 percent chance that the unit will successfully operate the system in the allowed time. However, if the average AFQT is lowered to the midpoint of category IIIB, the probability of successful completion falls to 47 percent

The authors also note that the addition of another high- scoring member to the team improved the probability of success by about 8 percent. This suggests that the effect of AFQT on group performance is additive. This finding is significant for an assessment of the optimal force mix because it implies that AFQT continues to make a difference in team performance even when considering the contribution of a second or third team member.

Orvis, Childress, and Polich (1992)used controlled trials to assess how AFQT score was related to various aspects of air defense and Patriot air defense system operation. The study included several types of air defense situations: point defense, asset defense, missile conservation, area defense, and a mixed defense scenario (Table 4.3).

Service members were also tested on their tactical kills/success in air-to-air combat and their overall battlefield survival.

• Category: Science • Tags: Afghanistan, IQ 

Estimates vary, and fall short of basic data quality. The highest estimate I can find for Afghanistan is IQ 83, and it is just that: an estimate. Say IQ 86 for Iran and Iraq, IQ 83 for Pakistan and IQ 86 for Turkey and we need not quibble about individual points, but the general range in the neighbourhood is clear.

Let us take the US Army recruitment requirement of IQ 93 to see what that means for training an army in Afghanistan. On that basis, recruiters would have to reject 75% of the Afghan population, (compared to rejecting 25% of the US population). The high rate of rejection means that the top 25% of the Afghan population may have some easier and safer occupations than to take up arms, which makes it difficult, but not impossible to get the talent you need for modern warfare.

If you are recruiting officers, they need to be about 2 standard deviations above the population average. (The intermediate level of sergeants at 1 standard deviation above average “translate” between soldiers and officers). So, Afghan sergeants will be IQ 83+15 = 98 and Afghan officers 83+30 = 113

These levels will not be sufficient to handle high technology weapon systems, nor sufficient to maintain them, nor even sufficient to check that maintenance has been completed properly. Not only that, but even that lower ability officer class will be harder to find. Only 16% will make sergeant level, and only 2% officer level. A government job doing something pleasant in an office will be more attractive to brighter Afghans.

These facts might explain why the Afghan Army did not match the performance of the American Army even when provided with American Army equipment. Little details like not paying troops regularly, nor properly supporting those deployed in distant outposts will not have helped morale.

So, how did the Taliban do so well? Here we have to understand the fundamental design principle of the AK-47. Mikhail Kalashnikov knew that for his weapon to be successful, it must be simple enough to be handled by a simple Russian peasant boy. Kalashnikov was himself a soldier, and knew what conditions in the field were like, and the environments in which his rifle must operate without jamming. He also knew exactly what weapons the enemy would use against Russia, and had no need to re-invent the wheel. He stressed simplicity and reliability. Without being stated, it was designed to be easy to use by even the simplest recruit. It has been stripped of intellectual content. The weapon is popular because it is easy to use, and cheap.

As to transport, no need for tanks. The Taliban have used readily available pick-up trucks. Very serviceable. Helicopters are fine, but complicated.

In the early stages of the Vietnam war, historian Arnold Toynbee argued (in Playboy magazine) that a strong external ally always damages the side which it supports. Why? For the obvious reason that if one party relies on a strong external ally, it can relax, knowing that the heavy lifting will always be done by the ally. While the insurgents take casualties, and become bitter, and harder, and better soldiers, more willing to die for their cause (having invested so much blood) the other side enjoy their wages, and become softer, lazier, and very aware of the advantages of an easy life, as opposed to the nasty, noisy and frankly dangerous chore of defending their liberty.

A final observation: in any country where you marry your cousin, your first duty is to your cousin, not your country.


Does the world need another IQ test? There are many well-validated tests, and also a number of short tests suitable for large scale surveys, many of which take less than 10 minutes, and several useful ones which take less than 5 minutes. However, if you are searching for a good measure of the manifold panoply of human achievement, it might be worth spreading the net even wider, so as to capture every gem of intellectual prowess. In that case, who would sit through such an assessment, given that the gold standard Wechsler tests take more than an hour?

One approach, in which the Madrid team under Roberto Colom have been prominent, is to cast the assessment in the form of a computer game. This makes it accessible to a much wider audience, and increases the number of wider-ranging intelligence test which can be used as part of genetic studies of intelligence. In fact, the “gamification” is rather light, a surface gloss only, but it seems to have been enough.

Now Robert Plomin’s team have added further to their own already published game-based intelligence test, and have interesting new results to report. The author’s names are a good roll-call of the new wave of intelligence researchers.

They point out that genome-wide association studies haven’t yet explained as much variance as can be obtained from twin studies. Critics have called this “missing heritability”, which misses the point. We know from twin studies that intelligence is heritable, and genome-wide association studies are trying to identify the genes responsible for this result. (We know that genetics is powerful in real life, now we need to show it in theory). Part of the problem is that larger studies have put together results from disparate tests, so the team has designed a 40 item intelligence game which produces a reliable (internal consistency = .78, two week retest reliability = .88) measure of g which they have given to 4,751 young adults from their twin study.

This novel g measure, which also yields reliable verbal and nonverbal scores, correlated substantially with standard measures of g collected at previous ages (r ranging from .42 at age 7 to .57 at age 16). Pathfinder showed substantial twin heritability (.57, 95% CIs = .43, .68) and SNP heritability (.37, 95% CIs = .04, .70). A polygenic score computed from GWA studies of five cognitive and educational traits accounted for 12% of the variation in g, the strongest DNA-based prediction of g to date. Widespread use of this engaging new measure will advance research not only in genomics but throughout the biological, medical, and behavioural sciences.

So, they have verbal and nonverbal scores, and can generate a gene-based prediction which accounts for 12% of intelligence variance, a good result by current standards. The test used item response theory to select the most powerful items, and get maximum predictive power from the fewest number of items.

The software is free, and if it gets taken up by further genome wide association studies then the variance accounted for may rise from 12% to 30%, assuming that the heterogeneity of test is a complicating factor which universal use of this test would overcome. The great advantage of a polygenic risk score for intelligence is that you can get a prediction from birth onwards, which overcomes the problem that early tests of intelligence do not get reliable till about age 11, and gain in accuracy thereafter. Early predictions might be a more precise way of evaluating whether early teaching has any effect on later intelligence.

To my eye the missing letter test doesn’t seem worth including. Sure, it is a basic process measure, but a bit out on a limb in factorial terms.

This very big sample of 4,751 25-year-olds shows significant sex differences in favour of men. The authors don’t comment on this, but it fits the emerging pattern of a male intellectual advantage in adulthood.

They say:

Heritability was 57%, shared environmental influence was 8% and multivariate polygenic scores predicted up to 12% of the variance. The latter finding –that 12% of the variance of Pathfinder g can be predicted by DNA –makes this the strongest polygenic score predictor of g reported to date. Although 12% is only one fifth of the twin study estimate of heritability, we hope that adding Pathfinder g in large biobanks will improve the yield of 21meta-analytic GWAS analyses by increasing sample sizes and decreasing heterogeneity of cognitive measures. It should be possible to use the brute force method of increasing sample sizes, especially with less heterogeneity of measures, to close the missing heritability gap from 12% to the SNP heritability of about 30%.

Getting above 30% of the variance will be hard, to reach the 60% heritability revealed by twin studies. That will require whole genome sequencing.

The summary is that the team have created a good new 15-minute IQ test which correlates well with the many longer assessments used over the years on their very large sample of twins. It also has good predictive power. If more widely adopted, and the few bits of explanatory English language translated into other languages, it could be a very useful contribution to large GWAS investigation of the genetic basis of intelligence.

You can get Pathfinder here:

• Category: Science • Tags: IQ 
James Thompson
About James Thompson

James Thompson has lectured in Psychology at the University of London all his working life. His first publication and conference presentation was a critique of Jensen’s 1969 paper, with Arthur Jensen in the audience. He also taught Arthur how to use an English public telephone. Many topics have taken up his attention since then, but mostly he comments on intelligence research.