Is the World Becoming Smarter or Dumber?

A comment by Greg Cochran is going viral:

There must have been some selection for IQ – without it, our brains would have disintegrated. But that selection can’t have been very strong, or intelligence would have gone up like a rocket. Today it’s going down at a rate of something like three points a century – think what would have happened if it had changed that rapidly, either up or down, over the last couple of millennia.


In the past few weeks there has been considerable interest in dysgenics. Many in the NRx-community subscribe to the dumbing-down or ‘idiocracy’ hypothesis that global intelligence is declining. Note: a distinction must be made between biological intelligence and intelligence as measured by an IQ test. Due to re-norming, the average of the latter will always remain at ‘100’ even if biological IQ falls or rises. The debate concerns the former.

There are three approaches to testing the the dumbing-down hypothesis: phenotypic (skull measurements), genomic (genetic markers linked to IQ), and empiric (IQ tests and other data).


A handful of studies have shown a decrease in polygenetic markers linked to IQ. A study of Icelanders by Kong et al 2017 showed a .3 point decline/decade of a polygenic measure of IQ:

However, this seems gradual enough that technology should be able to reverse it.

One scientist argues the world is becoming dumber due to the accumulation of various genetic mutations:

Human intelligence may have actually peaked before our ancient predecessors ever left Africa, Gerald Crabtree writes in two new journal articles. Genetic mutations during the past several millennia are causing a decline in overall human intellectual and emotional fitness, he says. Evolutionary pressure no longer favors intellect, so the problem is getting exponentially worse. He is careful to say that this is taking quite a long time, so it’s not like your grandparents are paragons of brilliance while your children will be cavemen rivaling Hartman’s SNL character. But he does posit that an ancient Athenian, plucked from 1000 BC, would be “among the brightest and most intellectually alive of our colleagues and companions.”

His central thesis is that each generation produces deleterious mutations, so down the line of human history, our intelligence is ever more impaired compared to that of our predecessors.

The major problem with this thesis is trying to ascertain the intelligence of ancient humans. Intelligence tests obviously didn’t exist back then, so all we have are ‘proxies’ for intelligence (such as written works, architecture, skull volume, etc.), as well as possible genomic markers.


The ‘Encephalization quotient’ seems to be a good proxy for intelligence, in which humans rank the highest.

Studies have also shown a positive correlation between head size and IQ. Studies that show Americans’ heads are getting bigger, could imply Americans are also becoming smarter.

Specifically, the researchers found skull size in white men has grown by 200 cubic centimeters, which is about the volume of a tennis ball. Skull height, from the base to the top of men’s heads, has increased by 8 millimeters—so about 0.3 inches. Among white women, skull size has grown by 180 cubic centimeters and height has increased by 7 millimeters.

Overall, skull height has grown 6.8 percent since the late 1800s. That’s a bigger percentage than the growth of body height, which has increased 5.6 percent. And comparatively, femur length has only increased about 2 percent. The researchers said that skull shape in Europe has also changed, but not as drastically as in the U.S.

However head size is an imperfect measure of intelligence. Although earlier humans may have had larger brains, they may not have been smarter.

Other evidence suggests brain shrinkage, going as far back as the Stone Age: If Modern Humans Are So Smart, Why Are Our Brains Shrinking?

Over the past 20,000 years, the average volume of the human male brain has decreased from 1,500 cubic centimeters to 1,350 cc, losing a chunk the size of a tennis ball. The female brain has shrunk by about the same proportion. “I’d call that major downsizing in an evolutionary eyeblink,” he says. “This happened in China, Europe, Africa—everywhere we look.” If our brain keeps dwindling at that rate over the next 20,000 years, it will start to approach the size of that found in Homo erectus, a relative that lived half a million years ago and had a brain volume of only 1,100 cc. Possibly owing to said shrinkage, it takes me a while to catch on. “Are you saying we’re getting dumber?” I ask.


The Flynn Effect suggests biological intelligence is rising. But other data suggests the Flynn Effect has stalled or is even reversing.

Regarding the reversal or stalling of the Flynn Effect, a possible explanation is that that earlier gains in IQ were attributed to rapid improvement in living conditions due to industrialization, not genetics, from the earlier post Idiocracy in America? Probably not:

Even if the FLynn effect is tapering off, that doesn’t mean it will reverse. Anther possibility is that early gains in IQ are attributable to environment, and now that essentials such as food, shelter, sanitation, clean water, electricity, and literacy are much more common, the ‘low hanging’ fruit has been picked, putting more precedence on genetic factors, which are much slower to evolve than environmental ones, which is why it may seem like the FLynn effect is reversing.

However, some counter that the Flynn Effect is only tenuously related to g, but rather representative of gains in abilities that are the least g-loaded:

Because children attend school longer now and have become much more familiar with the testing of school-related material, one might expect the greatest gains to occur on such school content-related tests as vocabulary, arithmetic or general information. Just the opposite is the case: abilities such as these have experienced relatively small gains and even occasional decreases over the years. Recent meta-analytic findings indicate that Flynn effects occur for tests assessing both fluid and crystallized abilities. For example, Dutch conscripts gained 21 points during only 30 years, or 7 points per decade, between 1952 and 1982.[7] But this rise in IQ test scores is not wholly explained by an increase in general intelligence. Studies have shown that while test scores have improved over time, the improvement is not fully correlated with latent factors related to intelligence.[14] Rushton has shown that the gains in IQ over time (the Lynn-Flynn effect) are unrelated to g.[15][16] Researchers have shown that the IQ gains described by the Flynn effect are due in part to increasing intelligence, and in part to increases in test-specific skills.[17][18][19]

According to a study of reaction times, perhaps Victorians were smarter than their modern-day counterparts, but the study is beset by a lot of possible methodological flaws and is hardly conclusive.

However, fertility and IQ appear to be inversely related, leading to a possible dysgenic effect over many generations:

…intelligence is negatively correlated with fertility rate, and positively correlated with survival rate of offspring.[1] The combined net effect of these two forces on ultimate population intelligence is not well studied and is unclear. It is theorized that if an inverse correlation of IQ with fertility rate were stronger than the correlation of survival rate, and if heritable factors involved in IQ were consistently expressed in populations with different fertility rates, and if this continued over a significant number of generations, it could lead to a decrease in population IQ scores

In 1982, Daniel Vining sought to address these issues in a large study on the fertility of over 10,000 individuals throughout the United States, who were then aged 25 to 34. The average fertility in his study was correlated at −0.86 with IQ for white women and −0.96 for black women. Vining argued that this indicated a drop in the genotypic average IQ of 1.6 points per generation for the white population, and 2.4 points per generation for the black population.

In a 1988 study, Retherford and Sewell examined the association between the measured intelligence and fertility of over 9,000 high school graduates in Wisconsin in 1957, and confirmed the inverse relationship between IQ and fertility for both sexes, but much more so for females. If children had, on average, the same IQ as their parents, IQ would decline by .81 points per generation. Taking .71 for the additive heritability of IQ as given by Jinks and Fulker,[15] they calculated a dysgenic decline of .57 IQ points per generation.[16]

One study investigating fertility and education carried out in 1991 found that high school dropouts in America had the most children (2.5 on average), with high school graduates having fewer children, and college graduates having the fewest children (1.56 on average).[19]

In addition to having fewer children, smarter women tend to have children later in life, increasing the likelihood of complications both in terms of pregnancy and birth defects.

It’s possible in the next 100 years if birth rates for high-IQ regions fall and populations for low-IQ regions surge, average global biological IQ will fall. But paradoxically, total intelligence may still rise, meaning that the ‘smart faction’ will grow in absolute population size but fall in relative population size, as explained in Idiocracy in America? Probably not. As well as other factors, this is due to assortative mating, in which smarter people choose smart companions.

What matters is that more of these people exist now than existed 10 years ago or 100 years ago. Even if fertility rates among high-IQ people are low, just by having 7+ billion people in the world, by virtue of the normal distribution of IQs you will still get many geniuses.

Empirically, I don’t buy the dumbing-down argument. Just skim the Arxiv high-energy physics section to see how far down the high-IQ rabbit hole goes…it’s obvious there are still a lot of intelligent people, many of whom who are producing research of great complexity and depth. The number of research publications in STEM fields, not just on Arxiv but also everywhere else, has surged in the past few decades:

Look at all the research coming out of Silicon Valley: delivery drones, self-driving electric cars, virtual reality, and apps that can almost mimic the intelligence of a human. The ancients produced a substantial body of classic literature, but their math and physics developments were not commensurate, especially given how much time elapsed, whereas in just the past three decades alone two of the hardest math problems ever conceived were proven: Fermat’s Last Theorem and the Poincaré Conjecture. Euclid, Pythagoras, Archimedes and Thales made discoveries, but the rigorous study of complex analysis and calculus eluded Western Civilization until only as recently as the 1700’s. The stagnation of math during antiquity may have been attributed to the cumbersome system of Roman numerals and the difficulty of sharing knowledge, as the printing press had not been invented.

There is also evidence later generations (Millennials and Gen Z) may be smarter than earlier generations, or at least that certain elements of pop culture have become smarter. The Big Bang Theory, which appropriates ‘nerd culture’, is one of the highest rated TV shows. There is a huge outpouring of intellectualism on sites such as Reddit and 4chan, where thousands of smart millennials everyday engage in impassioned debates about finance, political philosophy, futurology, biology, statistics, politics, and economics, whereas generations ago there seemed to be less interest in such complicated topics by the general population and youth. Then there is the rise of ‘esoteric celebrities’, and how people in STEM have become the equivalent of ‘rock stars’ for millennials. Philosopher David Chalmers had a hugely popular AMA on Reddit a few week ago, the success of which goes contrary to the belief by the fake news media that millennials only care about celebrity gossip. When millions of millennials showed up to the polls, they were rejecting the fake news narrative that Trump couldn’t win.

However, there is a caveat: mass immigration may disrupt assortative mating, and it’s possible the global population will stagnate, and coupled with an inexorable decline in genetic IQ, will over many centuries lead to a dysgenic dystopia. But this process may be slow enough that technology may reverse it…given the rate of research being produced, it seems more like likely than not such technologies will be developed. But the questions is, will policy makers implement them or cave in to ‘political correctness’ and ‘slippery slope’ arguments.

A combination of exploding population growth and environmental factors (nutrition, shelter, drugs, etc.) has increased total global intelligence, leading to this explosion in the 20th and 21st century of technologies and STEM research. But now that ‘low hanging fruit’ of environmental factors and modernity may have picked, and if global population stalls at 10 billion people, and if the fertility rate of >100 genetic-IQ populations remains below replacement, and if life extension technology to keep smart people alive longer fails, and if global mean genetic-IQs fall and technology is unable to reverse it, and if IQ not only falls but falls so much that it hits a critical level whereby humanity cannot save itself even if it wanted to, then something resembling ‘idiocracy’ is possible. But that’s a lot of ‘ifs’. I’m optimistic that a combination of both rising life expectancy and genomic technology will keep the total population of smart people unchanged, rather then the entire world plunging into idiocracy.

Let’s assume a genetic IQ of 160 is the minimum required to make significant breakthroughs (to advance the canon of human knowledge). That is the estimated IQ of Newton and Einstein, two of the greatest scientific geniuses who ever lived. A single high-IQ person by himself is useless…you probably need at least 500 to collaborate on scientific research and to have a functioning civilization.

The world population is expected to level off in the year 2100 at 10 billion. The regions that typically produce breakthroughs are Europe, Japan, China, India, Middle East, and North America. The combined projected population for these regions is projected to be 6 billion. Assuming genetic IQ stays at 100, with a SD=15 and mean of 100, there will be 190,000 geniuses (IQ >160), which is quite a lot of people and could explain why there are so many physics publications on Arxiv.

But if we simulate dysgenics, the dysgenic threshold is a genetic IQ of 82-83; any lower and the number of geniuses falls below 500.

Below 65 and there are none.

So to answer the original question, ‘Is the World Becoming Smarter or Dumber,’ the verdict still seems to be out, although the empirical evidence lends itself to optimism (for now at least). Despite this optimism, policy makers should not be complacent…boosting national IQ has many benefits, such as increased economic growth, higher standards of living, less crime, less entitlement spending, and improved overall welfare.