Gerald Crabtree, a prominent geneticist at Stanford University, made a startling claim. If an average citizen from Athens around 1000 BC were suddenly transported to the present day, he or she would stand out as one of the brightest and most emotionally stable people in the room – sharper memory, clearer thinking, a broader grasp of ideas. Modern humans, Crabtree argued, have been slowly losing intellectual and emotional fitness for thousands of years. The culprit? Genetic mutations that natural selection no longer weeds out as ruthlessly as it once did.
His two-part essay in the journal Trends in Genetics, titled “Our fragile intellect”, was both provocative and meticulously reasoned. Crabtree didn’t claim we’re all turning into dimwits overnight. The decline, he calculated, is gradual – perhaps a few per cent per generation – but inexorable. And it began when our ancestors swapped the high-stakes life of hunter-gatherers for the relative safety of farms, cities and, later, modern medicine.
The genetic arithmetic behind the warning
To understand Crabtree’s case, start with the numbers. The human genome mutates. Each new generation inherits roughly 50–100 de novo mutations – changes not present in either parent. Most land in non-functional “junk” DNA and do nothing. But a small fraction hit the 20,000 or so protein-coding genes that matter.
Crabtree’s key insight was how many of those genes matter for intelligence and emotional stability. Drawing on studies of X-linked intellectual disability (conditions that appear in boys because they have only one X chromosome), he estimated that mutations in any one of 2,000–5,000 genes can subtly impair cognition or mood. These aren’t a handful of “smart genes”; they form a long, fragile chain. Break any link and performance dips.
In a pre-agricultural world, anyone whose mutations made them too slow to spot danger, too impulsive to cooperate, or too forgetful to plan ahead was less likely to survive and reproduce. Natural selection kept the mutations in check. Once civilisation arrived, the pressure eased. A Wall Street executive who miscalculates can still thrive; a hunter-gatherer who did the same probably didn’t. Mutations began to accumulate.
Crabtree’s back-of-the-envelope calculation was sobering. With 2,000–5,000 relevant genes and a mutation rate that delivers roughly one new hit in these genes every 20–50 generations, each of us today carries at least two or three mildly deleterious changes that our distant ancestors would not have tolerated. Over 120 generations (roughly 3,000 years), that adds up. “We are almost certainly losing these abilities,” he concluded.
But are we really?
The idea triggered immediate pushback – and rightly so. Other geneticists pointed out that Crabtree’s model assumes selection has completely vanished. It hasn’t. Even in wealthy countries, at least a third of conceptions end in miscarriage, often because of severe genetic errors. People with serious intellectual impairments still face reproductive disadvantages. And beneficial mutations, though rarer, can spread.
Kevin Mitchell, a neurogeneticist then at Trinity College Dublin, published a direct rebuttal in the same journal. He argued that Crabtree had overstated the fragility. Intelligence is polygenic – influenced by thousands of variants of tiny effect – and the genome has built-in redundancy and repair mechanisms. “Genetic entropy” (the notion that genomes inevitably degrade) is a real concern in small, inbred populations, Mitchell noted, but humans are numerous, mobile and still under selection.
More recent experiments bolster the sceptics. In 2024, evolutionary geneticist Peter Keightley at the University of Edinburgh ran a 21-generation mouse experiment that mimicked relaxed natural selection. The fitness decline was tiny – less than 0.4 % per generation, and probably even smaller when extrapolated to humans, where some selection still operates.
What about IQ scores?
On the surface, the data look alarming. The famous “Flynn effect” – the steady rise in IQ scores throughout the 20th century – has reversed in several rich countries since the mid-1970s. Studies in Norway, Denmark, the UK and elsewhere have recorded drops of several points per generation. Some commentators seized on these as proof of genetic decline.
But most researchers attribute the reversal to environmental factors: changing education patterns, screen time, nutrition shifts, even the way IQ tests are administered. Twin and family studies show that the recent decline tracks within families, pointing away from broad genetic changes in the population and toward shared environments. Genomic data from large biobanks also show no clear signal of accumulating intelligence-damaging mutations at the scale Crabtree predicted.
Michael Lynch, another leading geneticist, had made similar warnings in 2010 about relaxed selection leading to a 1–5 % fitness drop per generation. Yet even he acknowledged the effect would take centuries to become noticeable – and that cultural evolution (medicine, education, technology) can more than compensate.
So where does this leave us? Crabtree’s core insight – that our intellectual abilities rest on an unusually large number of genes and are therefore vulnerable to mutation – remains scientifically respectable. The human mutation rate is relatively high, especially from older fathers. And modern medicine does allow mildly harmful variants to persist that would once have been eliminated.
Yet the evidence for a measurable, ongoing decline in innate intelligence is weak. The genome is not a Swiss watch that inevitably winds down; it is a dynamic, buffered system shaped by millions of years of evolution. Selection still acts, beneficial mutations still arise, and our species’ greatest strength has always been cultural, not purely genetic.
Crabtree himself was careful not to sound apocalyptic. He was raising a flag about long-term genetic hygiene, not predicting a new Dark Age. In an era of CRISPR, embryo screening and growing awareness of paternal-age effects, we have tools our ancestors could only dream of.
The real story, then, is not that we are doomed to get dumber. It is that human cognition is more genetically complex – and therefore more evolutionarily interesting – than we once thought. Whether that fragility becomes a problem or merely a fascinating footnote depends on how wisely we wield our growing power over our own genomes.
References
- Crabtree, G.R. (2013) Our fragile intellect. Part I. Trends in Genetics 29, 1–3.
- Crabtree, G.R. (2013) Our fragile intellect. Part II. Trends in Genetics 29, 3–5.
- Mitchell, K.J. (2013) Genetic entropy and the human intellect. Trends in Genetics 29, 59–60.
- Crabtree, G.R. (2013) Our fragile intellect: response to Dr Mitchell. Trends in Genetics 29, 60–62.
- Lynch, M. (2010 Rate, molecular spectrum, and consequences of human mutation. Proceedings of the National Academy of Sciences 107, 961–968.
- Chebib, J. et al. (2024) An estimate of fitness reduction from mutation accumulation in a mammal allows assessment of the consequences of relaxed selection. PLOS Biology 22, e3002795.
