Pseudoscience, Edward Dutton, and Christianity

I have in my head a framework for a multi-post series on rationality and Christianity, building in part on insights from a couple of books I am reading, including Inventing the Individual by Larry Siedentop (2020) (I commented on his refreshing historical approach in my last post) and Christianity and History by Herbert Butterfield (1949). But my time is short and I get distracted. But this is at least tangentially related.

Earlier this week a book recommendation came up on Amazon for a book with the provocative title Witches, Feminism, and the Fall of the West, by Edward Dutton. The cover was a picture of a woman being burned at the stake with flames rising all around her. My wife saw this recommendation come up and jokingly expressed some concern about what I must have been reading that would prompt such a suggestion. She read the overview on Amazon out of curiosity. From the summary, we agreed with Dutton that there has been a decline in our culture, but we were highly skeptical of his proposed mechanism of "dysgenic breeding," and especially turned off when his argument seemed to culminate in advocacy for the oppression of women by men. I assured my wife I wasn't planning to buy this book.

A couple days ago I was reading articles in VoegelinView online magazine, which I often enjoy. I was surprised to find an article by associate editor Richard Cocks promoting Dutton's dysgenic breeding hypothesis. I sent him an email expressing my concerns, he wrote me back with some additional thoughts, and as it turned out I ended up downloading the book after all, to better evaluate his sources. Dr. Cocks also posted an essay on his blog riffing on a comment in my email. [update 7-19-22: also posted today at VoegelinView] For items that more directly address the topic of his essay, skip down to section V.

I.

To be fair, let me start with a quote from Dutton's book that I agree with. 

Put simply, postmodernism is the idea that there is no such thing as objective “truth” and that all “truth” is ultimately reducible to power dynamics. What we are told is the “truth” is, in fact, a means of maintaining the powerful in their position of power. ... [P]ostmodernism ultimately strips us of truth itself. Indeed, it strips us of reason. It takes us into a fantasy world in which everything is “subjective,” meaning that anyone can assert anything, and it is “true” for them, and you are being willfully imperialistic (and thus evil) if you question or do not accept their “truth.”

Mainstream scholars today are afraid to challenge the prevailing postmodern narrative for fear of being cancelled or losing status among their peers. This means scholarship that agrees with the narrative is not put to sufficient scrutiny, the scientific method is short-circuited, and the result is a rise in left-wing pseudoscience. On the other hand, scientific exploration of hypotheses (and even entire fields) that don't support the leftist narrative has become taboo for the same reason. This does two things: 1) It destroys trust in academic expertise among conservatives and the right. 2) It creates a vacuum in the problematic fields that gets filled by right-wing scholars who thrive on controversy. Because mainstream scholars refuse to engage these fields and also because they have lost the trust of conservatives, the right-wing controversialist narratives don't get the scrutiny that they deserve, the scientific method is short-circuited, and the result is a rise in right-wing pseudoscience.

One example of a whole field that has become taboo is intelligence, including heritable intelligence and group differences in intelligence. Mainstream scholars won't touch this field with a ten foot pole, for fear of discovering truths that might get them cancelled. Enter Edward Dutton stage right. Due to the polarization of our times, his reputation could only increase among his target audience if he is "cancelled," and he thrives on the controversy.

II.

This post is about pseudoscience. Let me take a moment to share a few of my favorite examples of modern pseudoscience. Pseudoscience typically occurs when an unfalsifiable hypothesis or statement is taken as scientific truth. Now, every rational person must base their reasoning on a worldview built of unfalsifiable assumptions - whether it is a religious worldview, reductive materialism, Marxist humanism, or whatever. The pseudoscience error comes in treating one's unfalsifiable assumptions as scientific truth. Of course there are homeopathic remedies, much of pre-replication crisis psychology and social science, and all kinds of politically convenient takes on COVID and vaccines on all sides, oh my goodness.  But the three examples that I want to talk about are different in that they become foundational building blocks for how people see the world. 

The first is the Many Worlds or Multiverse Theorem. It is the unfalsifiable hypothesis that [insert techno-babble about quantum mechanics] causes there to be an infinite number of universes, in which every permutation of values for the important physical constants are sampled. As far as I can tell, this theory arose as a response to observations that the various physical constants seem to be fine-tuned to allow life to exist. Also, due to the nature of infinity, anything improbable on Earth (e.g. complex molecules like DNA arising from a primordial chemical soup; sufficient genes arising simultaneously to enable a replication-capable cell, rapid increase of biological complexity via pure evolutionary chance) suddenly becomes probable somewhere in the Multiverse even without a God to drive the process. The only problem is that it is pure unfalsifiable conjecture.

The second is the idea that we are living in a simulation created on a whim or as a science project by some advanced alien civilization. This cosmology is surprisingly popular among techno-futurist rationalists worried about artificial intelligence. (OK, maybe I read more blogs that overlap that circle than you do.)

And the third is the field of evolutionary psychology (EP). EP assumes that just as our biology evolved from primitive forms of life, our culture and psychology evolved from a primitive animalistic state to more complex and functional society by natural selection / group selection. Higher fitness individual traits and higher fitness social relationships and beliefs result in greater group survival, and less successful traits and social structures were lost by negative selection. This makes a certain amount of sense. It is great as a way to explain the development of society without the need for a God to drive the process. The only problem is that EP is nothing more than unfalsifiable conjectures, an exercise in storytelling. The benefit for the scholar is that with sufficient creativity EP can support any narrative, though it helps for believability to have data points around which to weave the narrative. For example, Joseph Henrich's anthropology (which I briefly addressed here) relies heavily on EP but he spends most of his time doing fieldwork to generate data. In my opinion his EP narratives are dismissible, but the data provides value. However, around any data set there will remain enough flexibility to craft a narrative that conforms to the biases of any worldview. Enter Edward Dutton stage right. About 10 years ago he decided that EP was the field for him.

III.

The first page of the Introduction in Witches, Feminism, and the Fall of the West starts off with an evolutionary psychology argument about how group selection had created an amazing society in Europe until...

The Industrial Revolution, with all its marvels, changed everything. In 1800, half of all those born died as children; two centuries later, almost none did. More and more people who would not have survived in the old times walked among us. They were mistakes made flesh. They no longer uplifted the established rules; they endlessly criticized and undermined them, like the witches of yore. Unwell in body and mind, they were, at best, selfish and impulsive; at worst, they promoted depression and despair.

And he doubled down on this, repeating this paragraph word for word in his final chapter. This is very troubling dehumanizing language. And of course there is no way to know whether the children who tragically passed away in times past would have been rule-breakers, critics, unhealthy and mentally unstable, selfish and impulsive, or whether they would have promoted depression and despair. That seems like a lot to just assert. But lets put that rhetorical flourish aside for a second and try to get to what Dutton is actually contending. His argument at the most basic is the following:

1. Much of intelligence, wellness, and psychology is heritable.
2. There is less evolutionary selection pressure today than in the past, so lower fitness phenotypes are allowed to survive. This is a process he calls "dysgenics."
3. Dysgenics causes lower fitness mutations to stay in the gene pool, increasing the "mutation load" in contemporary westerners compared with before the Industrial Revolution.
4. This has resulted in an overall reduction in intelligence and health today versus before the Industrial Revolution.
5. The "mutants" (a term Dutton uses frequently) are more likely to be feminists (who are like the witches of old times), fighting against the patriarchal order which was necessary for the stability of society.
6. Hints that maybe our culture would be better if men were still allowed to control and suppress women today. (I'm not even kidding - this is where he goes with his logic.)

Because there is no way to measure selection pressure or mutation load (as Dutton himself concedes), the critical aspect in all of this is Point 4. If we see intelligence in the west declining, it wouldn't prove the mechanism in Points 2-3, but the next step would be to evaluate whether this hypothesis has merit. But if we can't demonstrate that intelligence is in decline, then we can reject the dysgenics hypothesis as unsupported.

(As an aside, since there is no way to measure "mutation load," the argument in Point 5 that feminists have higher mutation load is an unfalsifiable assertion. Similarly, the idea that patriarchy was needed for the stability of society is an unfalsifiable EP story. I'll address this idea briefly at the end of this post.)

IV.

Dutton cites 3 journal articles in support of the central claim that intelligence is in decline since the Industrial Revolution. I will go through each one and evaluate the evidence. Again, if the evidence for a decline in intelligence is weak, then we have no obligation to accept Dutton's EP narrative. Here is where I get to say that I am not a statistician, but I have a PhD in chemical engineering and I do  data analysis as a material scientist every day in my research job. 

1. Reaction times

The first is a paper called "Were the Victorians cleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time", by Woodley, et al., in the journal Intelligence. The authors are looking for evidence to support the dysgenics theory. They start off with the admission that "The presence of a dysgenic effect on intelligence has proven difficult to detect via direct measurement" because IQ scores have tended to increase through much of the 20th century. It may be true that the Flynn effect is due to education or culture or other differences that don't imply a hereditary increase in cognitive ability. But still, it must be said that IQ increases do not support the dysgenics theories. Back to the paper, they use a test of reaction times - the time it takes test subjects to respond to a sensory stimulus - as a measure of intelligence. The authors state that reaction times are 58% hereditable and modestly correlated with IQ (correlation coefficient of -.31). There are two datasets from the late 1800s and a much larger number from the late 20th century. Here is the key plot (I have no idea why the axes are labelled "Effect Size" and "Independent Variable" rather than something more meaningful):

While the regression coefficient is statistically significant, there are very large error bars on the slope coefficient (0.07 - 1.3), and the trend is entirely driven by the two low points in the late 1800s. There is no obvious trend from 1940-2002, so either all of the increase happened between 1890 and 1940, or else we just don't have enough data to know the true mean value for the late 19th century, given how much variability there is between studies in the later time periods. I would say we just don't have enough data to draw strong conclusions, and the data we do have is very weak evidence for a cognitive decline. 

2. Color acuity

The second paper explores a measure of color acuity - the ability of test subjects to differentiate between different hues - in the quest to find evidence for the hypothesized dysgenic theory. The paper is called "Showing their true colours: Possible secular declines and a Jensen effect on colour acuity — More evidence for the weaker variant of Spearman's Other Hypothesis," also by Woodley, et al., published in Personality and Individual Differences. If you'll excuse me for jumping straight to the figure, this is what they find:

They have four data points, plenty of variability, and I just don't see how you can conclude that there is a trend in this data. (And it looks to me like the dashed line is in the wrong place - I don't know what is going on there.) Note that the four larger points are one dataset, and the four smaller points are a second dataset. The author concedes "Four observations may not seem like much on which to base a secular trend, however it is important to note that these four normative studies represent the virtual totality of the data that can be compared." That's understandable, but it doesn't strengthen the conclusions, and I am not impressed.

3. Genetic markers for educational attainment in Iceland

The third paper looks at genetic markers for educational attainment in Iceland in a quest to find the hypothesized data that would support the idea of a cognitive decline. The paper is called "Selection against variants in the genome associated with educational attainment" by Kong et al. (No Woodley on this one), published in PNAS. The authors trace a polygenic score that accounts for 3.74% of the variance in actual educational attainment from 1910 to 1990. They find a small decrease of 0.14 standard deviations over the 80 years. Here is the figure (the polygenic score is measured in standard units, which they have scaled such that one standard deviation is 1.0):

(Error bars are the standard error of the mean, not the standard deviation of the distribution.) Of the three papers, this is for me the most solid indication of a decline. But it is a very small decline in a poor proxy of educational attainment, which itself is a poor proxy for cognitive ability. And the data is only for Iceland, which Dutton wants us to think represents all of the West. So it is a small decline 4 steps removed from actual cognitive ability in the West. Another question I have to ask myself is, why didn't they look at a polygenic score for IQ? Very good polygenic scores for IQ exist, and these Icelandic genomes are available for analysis. I don't know for sure, but my guess is that they did look but just didn't find what they wanted to see. How many datasets are there (like IQ) that would oppose the idea of a cognitive decline were kept in the desk drawer while they looked for evidence that better suited their biases?

If these three papers constitute the best evidence that Dutton can cite for a cognitive decline in the West, despite teams of researchers (at least 2 teams) searching for the evidence that will support their dysgenic theories, then I have to conclude (noting that all scientific conclusions are temporary and subject to review as more data is generated) that there is no measurable cognitive decline since the Industrial Revolution. Look over the data yourself and see whether you agree with me. But I feel quite confident dismissing the idea of a dysgenic effect. The corollaries to this finding are that there no need to worry about your "mutation load," no feminist mutant witches, and no need to start oppressing women for the good of society. My wife will be relieved.

V.

In the preceeding sections I have mostly addressed Dutton's factual or scientific claims. Here I will address religious and philosophical claims. 

First of all, as a Christian, I take issue with Dutton's characterization of Christianity as a vehicle for ethnocentrism and patriarchy (both of which he sees as a positive for adaptive "fitness"). It was not post-Industrial Revolution witches who crushed the patriarchy, but Jesus Christ and his early Christian followers. Larry Siedentop tells the story, in Inventing the Individual, of the ancient patriarchal religions in which the father was the head of the house and the head of the family religion. Everyone else had their place in a more or less rigid hierarchal structure beneath the patriarch who interacted with divinity on behalf of the family. Christ came and spread the revolutionary new idea that each individual has a personal relationship with God unmediated by any earthly status, race, or gender. Roman elites mocked the church a place for women, slaves, and outcasts. This new egalitarian metaphysics led to the demise of the Hellenistic family "church" led by the patriarch, and a re-ordering of legal theory and societal structures on individual agency and value. What emerged from that new social theory (though greatly simplifying, of course), was Western liberalism with the freedom and prosperity we enjoy today (and which is being threatened by postmodern trends). In my opinion Siedentop's story has a far better foundation in truth than Dutton's story. You may disagree, but I highly recommend the book.

From my reading of Dutton, his focus on group identity and promotion of group conflict over individual moral value and individual agency seems to be just as much a rejection of Western liberalism as Marxism, Critical Theory, and postmodernism. I see illiberalism as a threat to the West, whether it comes from the left or the right.

I have gone over the evidence provided by Dutton and found no scientific reason to believe his dysgenics theory or his views on the benefits of patriarchal oppression of women. I think Eric Voegelin was correct when he said that Christianity inoculates against Hegelianism, Marxism, and neo-Kantian philosophy. I similarly think Christianity inoculates against this sort of right-wing illiberalism. But lets say for the sake of argument that I had read Witches and found its ideas convincing that a stable society requires the oppression of women. Or similarly, lets say I believed Girard's idea that social stability requires a scapegoat to kill (which Cocks discusses in his post). Or on the other hand, lets say that I as a Christian believed the structural oppression in society could be finally eradicated, but only through a violent Marxist revolution. Would I be justified in acting on these "truths" via theory or praxis, if I believe it to be immoral but necessary for societal stability? To quote Herbert Butterfield, “historians do not seem to me to spend their time weeping over the downfall of the Roman Empire.” We would do better to trust in divine providence, remembering that from the ashes of Rome rose the West on a higher moral plane. Christians can trust God to bring about (quoting Butterfield again) “the kind of triumph that may come out of apparent defeat - the kind of good that can be wrested out of evil.” In the end it is only our individual agency and virtue that have any value. No abstract collective goal is worth that sacrifice.