Can mutation load explain the rise of leftism?

A potential proxy for mutational load is positively related to certain leftist beliefs, and public opinion has shifted both between and within cohorts.

Written by Noah Carl and Bo Winegard.

In a recent article in Aporia, Joseph Bronski and Matthew Archer argued that the rise of leftism since 1960 is partly explained by mutational pressure – the accumulation of harmful mutations in the gene pool. Elsewhere, Bronski has gone as far as suggesting that “Western political change is solely due to evolutionary pressure” – by which he means a combination of mutational pressure (favouring leftism) and selection pressure (favouring rightism).¹

This is certainly an interesting and provocative thesis. If true, it would upend our understanding of changes in the political landscape over the last sixty years. But is it true? We are not convinced. In the remainder of this article, we will lay out reasons why. Note that Arctotherium has already published a three-part critique of the mutational load hypothesis which we would recommend reading.

Mutational load and leftism

The mutational load hypothesis makes two key assumptions:

1. Mutational pressure has increased over the last sixty years (due to a combination of rising parental age and declining infant mortality).

2. Mutational load causes leftist political beliefs.

The first assumption seems highly plausible. The second, however, demands closer scrutiny. As evidence in favour of the assumption, Bronski points out that leftism is correlated with paternal age and mental illness – two potential proxies for mutational load. A major problem with his theory is that there’s another potential proxy for mutational load, namely IQ, which is positively correlated with certain leftist beliefs.

What makes IQ a potential proxy for mutational load? Well, it is positively associated with indicators of physiological integrity such as grip strength and longevity. It is negatively associated with indicators of developmental instability such as various measures of fluctuating asymmetry. And individual differences in IQ are partly explained by rare, presumably deleterious, mutations.

Bronski conceptualises leftism in terms of one’s views on three issues: women, homosexuals and racial minorities. He hypothesises that individuals with more “leftist” (i.e., less socially conservative) views on these issues have higher mutational load. Since IQ is an obvious potential proxy for mutational load, his theory predicts that such views should be negatively correlated with IQ. The problem is that studies consistently find the exact opposite.

A 2015 meta-analysis analysed 67 studies and found an average correlation between IQ and social conservatism of r = –.20. The relationship between IQ and political beliefs does vary depending on the issue: measures of fiscal conservatism are often associated with higher IQ. Yet measures of traditional attitudes to women, homosexuals and racial minorities are invariably associated with lower IQ. They are even associated with DNA-based predictors, suggesting that IQ actually causes such attitudes.

To take just one example: in the US General Social Survey, whites who score 0–1 out of 10 on the Wordsum vocabulary test are 33 percentage points more likely to say homosexuality is “always wrong” than whites who score 9–10 out of 10.² Which group is likely to have higher mutational load: the one scoring 0–1 or the one scoring 9–10? Surely, the former. Perhaps mutational load causes social conservatism.

You might say that Wordsum is an imperfect measure of intelligence which is in any case biased in favour of leftists, who tend to be verbally tilted. But even studies that extracted a latent g-factor from multiple subtests have reached the same result. For example, Ian Deary and colleagues observed that general intelligence at age 10 was positively associated with “liberal, non-traditional social attitudes” at age 30. (This was also true of both non-verbal subtests they analysed.)

Chart from ‘Bright Children Become Enlightened Adults’.

The negative association between IQ and social conservatism casts serious doubt on the claim that mutational load causes leftist political beliefs – especially if we define leftism as non-traditional attitudes to women, homosexuals and racial minorities.

Cohort or period?

It is clear that most Western countries, not least the US, have become more “leftist” (i.e., less socially conservative) over the last sixty years. The question is why? We can distinguish two broad classes of explanation: period effects and cohort effects.

Period effects involve multiple cohorts within the population simultaneously shifting their views. For example, suppose we have three groups: young, middle-aged and old. A period effect would be where all three groups become, say, 10% of a standard deviation more leftist – and as a result the overall population becomes 10% of a standard deviation more leftist. In the case of period effects, individuals actually change their views during the course of their lives.

Cohort effects, by contrast, involve the gradual replacement of cohorts with one set of views by cohorts with a slightly different set of views. Go back to the three groups. A cohort effect would be where the middle-aged turn out to be 10% of a standard deviation more leftist than the old, and the young turn out to be 20% of a standard deviation more leftist – and as a result the overall population becomes 10% of a standard deviation more leftist. In the case of cohort effects, individuals’ views are fixed once they are formed.

It should be obvious that any explanation for the rise in leftism that invokes genetic changes must posit cohort effects: changes that accumulate generation by generation cannot cause multiple cohorts to simultaneously shift their beliefs. The problem for Bronski’s theory is that studies attribute a sizeable part of the rise in leftism to period effects.

A comprehensive review of the literature is beyond the scope of this article, but suffice it to mention a few recent papers.

David Ekstam examines change in attitudes to homosexuality and concludes that “both period and cohort effects have contributed to the increase in tolerance over the past three decades”. Jean Twenge and colleagues examine increased support for gay marriage between 1988 and 2004–2018 and conclude that it was “primarily due to time period”. Likewise, Ashley Kranjac and Robert Wagmiller examine change in attitudes to homosexuality and conclude that “both intracohort and intercohort change played positive and significant roles in the liberalization of attitudes”.³

Elizabeth Curry examines change in attitudes to homosexuality in Britain. She finds that while “there is evidence of the rise in support … being driven by generational replacement”, there is “also considerable evidence of a societal effect”. This can be seen in the chart below, which plots acceptance of homosexuality by birth cohort. The lines for more recent cohorts being higher suggests a cohort effect; the lines sloping upward suggests a period effect.

Chart from ‘A liberalisation in attitudes?’

Technically, the pattern in the chart above could be explained by a combination of cohort and age effects, but that would mean people become more tolerant of homosexuality as they got older. This is contrary to analyses of panel data, which find that people rarely get more liberal as they age (after taking time period into account).⁴

Studies from East Asia also find large period effects on attitudes to same-sex relationships. Rei Naka’s analyses of Japanese data “highlight the role of intracohort change in liberalizing attitudes toward homosexuality”. And in Zhiyong Lin and Jaein Lee’s models of attitudes toward homosexuality in South Korea, “the period effects dominated”.

The evidence reviewed above is not inconsistent with a weak version of Bronski’s thesis whereby the rise of leftism is partly explained by genetic changes, but it does contradict the strong version whereby the rise of leftism is “solely” explained by such changes.

Conclusion

It is far from clear that mutational load causes leftist political beliefs. Such beliefs may be associated with mental illness and paternal age, but they are also associated with IQ. In particular, individuals with lower IQs (who presumably have higher mutational loads) tend to hold more traditional attitudes toward women, homosexuals and racial minorities.

What’s more, studies tracking the liberalisation of attitudes over time find that period effects are at least as important as cohort effects. Period effects cannot be due to genetic changes because they involve multiple cohorts simultaneously shifting their beliefs.

Mutational pressure may have made a small contribution to the rise of “leftism” (i.e., the decline of social conservatism) over the last sixty years. But it seems very unlikely to be a major contributor.

Noah Carl is an Editor at Aporia. Bo Winegard is Executive Editor at Aporia.

---

Consider supporting Aporia with a paid subscription:

To chat with fellow Aporia readers and attend meet-ups, join our Telegram. You can also follow us on Twitter.

1. Selection pressure favours rightism because rightists have higher fertility than leftists.

2. Using only data from 2010 onwards. The difference is even larger when controlling for age and sex.

3. Michael Hout tracks numerous attitudinal variables in the General Social Survey and concludes that “both the climate (cohort) and the weather (period) were decidedly liberal over the long run from 1972 to 2018”.

4. Panel data refers to observations on the same individuals at different time points.