Figure 16-3: Years of schooling, 1870–2010
Source: Our World in Data, Roser & Ortiz-Ospina 2016a, based on data from Lee & Lee 2016. Data are for the population aged 15–64.
Though we see little or no global convergence in the length of formal schooling, an ongoing revolution in the dissemination of knowledge makes the gap less relevant. Most of the world’s knowledge is now online rather than locked in libraries (much of it free), and massive open online courses (MOOCs) and other forms of distance learning are becoming available to anyone with a smartphone.
Other disparities in education are shrinking as well. In the United States, measures of school readiness among low-income, Hispanic, and African American children increased substantially between 1998 and 2010, possibly because free preschool programs are more widely available, and because poor families today have more books, computers, and Internet access and the parents spend more time interacting with their children.24
Even more consequentially, the ultimate form of sex discrimination—keeping girls out of school—is in decline. The change is consequential not just because women make up half the population, so educating them doubles the size of the skill pool, but because the hand that rocks the cradle rules the world. When girls are educated, they are healthier, have fewer and healthier children, and are more productive—and so are their countries.25 It took the West centuries to figure out that educating the whole population, not just the half with testicles, was a good idea: the line for England in figure 16-4 shows that Englishwomen did not become as literate as Englishmen until 1885. The world as a whole caught on even later but quickly made up for lost time, going from teaching only two-thirds as many girls as boys to read in 1975 to teaching them in equal numbers in 2014. The United Nations has announced that the world has met the 2015 Millennium Development Goal of achieving gender parity in primary, secondary, and tertiary education.26
Figure 16-4: Female literacy, 1750–2014
Sources: England (all adults): Clark 2007, p. 179. World, Pakistan, & Afghanistan (ages 15–24): HumanProgress, http://www.humanprogress.org/f1/2101, based on data from UNESCO Institute for Statistics, summarized in World Bank 2016f. Data for the world are averaged over slightly different sets of countries in different years.
The other two lines tell their own story. The country with the worst gender ratio for literacy is Afghanistan. Not only is Afghanistan near the bottom in almost every measure of human development (including its overall literacy rate, which in 2011 stood at an abysmal .52), but from 1996 to 2001 it was under the control of the Taliban, the Islamic fundamentalist movement that, among other atrocities, forbade girls and women from attending school. The Taliban has continued to intimidate girls from getting an education in the regions of Afghanistan and neighboring Pakistan it controls. Starting in 2009 the twelve-year-old Malala Yousafzai, whose family ran a chain of schools in the Swat district of Pakistan, publicly spoke out for girls’ right to an education. On a day that will live in infamy, October 9, 2012, a Taliban gunman boarded her school bus and shot her in the head. She survived to become the youngest winner of the Nobel Peace Prize and one of the world’s most admired women. Yet even in these benighted parts of the world, progress can be seen.27 In the past three decades the literacy gender ratio has doubled in Afghanistan and increased by half in Pakistan, whose ratio now matches that for the world in 1980 and for England in 1850. Nothing is certain, but the global tide of activism, economic development, common sense, and common decency are likely to push the ratio to its natural ceiling.
* * *
Could the world be getting not just more literate and knowledgeable but actually smarter? Might people be increasingly adept at learning new skills, grasping abstract ideas, and solving unforeseen problems? Amazingly, the answer is yes. Intelligence Quotient (IQ) scores have been rising for more than a century, in every part of the world, at a rate of about three IQ points (a fifth of a standard deviation) per decade. When the philosopher James Flynn first brought this phenomenon to psychologists’ attention in 1984, many thought it must have been a mistake or trick.28 For one thing, we know that intelligence is highly heritable, and the world has not engaged in a massive eugenics project in which smarter people have had more babies generation after generation.29 Nor have people been marrying outside their clan and tribe (thus avoiding inbreeding and increasing hybrid vigor) in great enough numbers for a long enough time to explain the rise.30 Also, it beggars belief to think that an average person of 1910, if he or she had entered a time machine and materialized today, would be borderline retarded by our standards, while if Joe and Jane Average made the reverse journey, they would outsmart 98 percent of the befrocked and bewhiskered Edwardians who greeted them as they emerged. Yet surprising as it is, the Flynn effect is no longer in doubt, and it has recently been confirmed in a meta-analysis of 271 samples from thirty-one countries with four million people.31 Figure 16-5 plots the “secular rise in IQ scores,” as psychologists call it (secular in the sense of long-term rather than irreligious).
Figure 16-5: IQ gains, 1909–2013
Source: Pietschnig & Voracek 2015, supplemental online material. The lines display changes in IQ measured by different tests starting at different times and cannot be compared with one another.
Note that each line plots the change in IQ scores in a continent relative to the average score in the earliest year for which data are available, which is arbitrarily set to 0 because the tests and periods for the different continents are not directly commensurable. We cannot read the graph as we did the previous ones and infer, for example, that the IQ of Africa in 2007 is equivalent to the IQ of Australia and New Zealand in 1970. Not surprisingly, the rise in IQ scores obeys Stein’s Law: Things that can’t go on forever don’t. The Flynn effect is now petering out in some of the countries in which it has been going on the longest.32
Though it’s not easy to pinpoint the causes of the rise in IQ scores, it’s no paradox that a heritable trait can be boosted by changes in the environment. That’s what happened with height, a trait that also is highly heritable and has increased over the decades, and for some of the same reasons: better nutrition and less disease. Brains are greedy organs, consuming about a fifth of the body’s energy, and they are made of fats and proteins that are demanding for the body to produce. Fighting off infections is metabolically expensive, and the immune system of a sick child may commandeer resources that would otherwise go to brain development. Also helping with brain development is a cleaner environment, with lower levels of lead and other toxins. Food, health, and environmental quality are among the perquisites of a richer society, and not surprisingly, the Flynn effect is correlated with increases in GDP per capita.33
But nutrition and health can explain only a part of the Flynn effect.34 For one thing, their benefits should be concentrated in pulling up the lower half of the bell curve of IQ scores, populated by the duller people who had been held back by poor food and health. (After all, past a certain point, additional food makes people fatter, not smarter.) Indeed, in some times and places the Flynn effect is concentrated in the lower half, bringing the duller closer to the average. But in other times and places the entire curve crept rightward: the smart got smarter too, even though they started out healthy and well-fed. Second, improvements in health and nutrition should affect children most of all, and then the adults they grow into. But the Flynn effect is stronger for adults than for children, suggesting that experiences on the way to adulthood, not just biological constitution in early childhood, have pushed IQ scores higher. (The most obvious of these experiences is education.) Also, while IQ has risen over the decades, and nutrition, health, and height have risen over the decades, their various ascents and plateaus don’t track each other particularly closely.
But the main reason that health and nutrition aren’t enough to explain the IQ rise is that what has risen over time is not overall brainpower. The Flynn effect is not an i
ncrease in g, the general intelligence factor that underlies every subtype of intelligence (verbal, spatial, mathematical, memory, and so on) and is the aspect of intelligence most directly affected by the genes.35 While overall IQ has risen, and scores on each intelligence subtest have risen, some subtest scores have risen more rapidly than others in a pattern different from the pattern linked to the genes. That’s another reason the Flynn effect does not cast doubt on the high heritability of IQ.
So which kinds of intellectual performance have been pushed upward by the better environments of recent decades? Surprisingly, the steepest gains have not been found in the concrete skills that are directly taught in school, such as general knowledge, arithmetic, and vocabulary. They have been found in the abstract, fluid kinds of intelligence, the ones tapped by similarity questions (“What do an hour and a year have in common?”), analogies (“BIRD is to EGG as TREE is to what?”), and visual matrices (where the test-taker has to choose a complex geometric figure that fits into a rule-governed sequence). What has increased the most, then, is an analytic mindset: putting concepts into abstract categories (an hour and a year are “units of time”), mentally dissecting objects into their parts and relationships rather than absorbing them as wholes, and placing oneself in a hypothetical world defined by certain rules and exploring its logical implications while setting aside everyday experience (“Suppose that in Country X everything is made of plastic. Are the ovens made of plastic?”).36 An analytic mindset is inculcated by formal schooling, even if a teacher never singles it out in a lesson, as long as the curriculum requires understanding and reasoning rather than rote memorization (and that has been the trend in education since the early decades of the 20th century).37 Outside the schoolhouse, analytic thinking is encouraged by a culture that trades in visual symbols (subway maps, digital displays), analytic tools (spreadsheets, stock reports), and academic concepts that trickle down into common parlance (supply and demand, on average, human rights, win-win, correlation versus causation, false positive).
Does the Flynn effect matter in the real world? Almost certainly. A high IQ is not just a number that you can brag about in a bar or that gets you into Mensa; it is a tailwind in life.38 People with high scores on intelligence tests get better jobs, perform better in their jobs, enjoy better health and longer lives, are less likely to get into trouble with the law, and have a greater number of noteworthy accomplishments like starting companies, earning patents, and creating respected works of art—all holding socioeconomic status constant. (The myth, still popular among leftist intellectuals, that IQ doesn’t exist or cannot be reliably measured was refuted decades ago.) We don’t know whether these bonuses come from g alone or also from the Flynn component of intelligence, but the answer is probably both. Flynn has speculated, and I agree, that abstract reasoning can even hone the moral sense. The cognitive act of extricating oneself from the particulars of one’s life and pondering “There but for fortune go I” or “What would the world be like if everyone did this?” can be a gateway to compassion and ethics.39
Since intelligence brings good things, and intelligence has been increasing, can we see a dividend from increasing intelligence in improvements to the world? Some skeptics (including, at the outset, Flynn himself) doubted whether the 20th century really produced more brilliant ideas than the ages of Hume, Goethe, and Darwin.40 Then again, the geniuses of the past had the advantage of exploring virgin territory. Once someone discovers the analytic-synthetic distinction or the theory of natural selection, no one can ever discover it again. Today the intellectual landscape is well trodden, and it’s harder for a solitary genius to tower above the crowd of hypereducated and networked thinkers who are mapping every nook and cranny. Still, there have been some signs of a smarter populace, such as the fact that the world’s top-ranked chess and bridge players have been getting younger. And no one can second-guess the warp speed of advances in science and technology of the past half-century.
Most dramatically, an increase in one kind of abstract intelligence is visible all over the world: mastery of digital technology. Cyberspace is the ultimate abstract realm, in which goals are achieved not by pushing matter around in space but by manipulating intangible symbols and patterns. When people were first confronted with digital interfaces in the 1970s, like videocassette recorders and ticket machines in new subway systems, they were baffled. It was a running joke of the 1980s that most VCRs eternally flashed “12:00” at owners who couldn’t figure out how to set the time. But Generation X and the Millennials have famously thrived in the digital realm. (In one cartoon of the new millennium, a father says to his young boy, “Son, your mother and I have bought software to control what you see on the Internet. Um . . . Could you install it for us?”) The developing world has thrived in that realm as well, often leapfrogging the West in its adoption of smartphones and of applications for them such as mobile banking, education, and real-time market updates.41
Could the Flynn effect help explain the other rises in well-being we have seen in these chapters? An analysis by the economist R. W. Hafer suggests it could. Holding all the usual confounding variables constant—education, GDP, government spending, even a country’s religious makeup and its history of colonization—he found that a country’s average IQ predicted its subsequent growth in GDP per capita, together with growth in noneconomic measures of well-being like longevity and leisure time. An 11-point increase in IQ, he estimated, would accelerate a country’s growth rate enough to double well-being in just nineteen years rather than twenty-seven. Policies that hurry the Flynn effect along, namely investments in health, nutrition, and education, could make a country richer, better governed, and happier down the road.42
* * *
What’s good for humanity is not always good for social science, and it may be impossible to unsnarl the bundle of correlations among all the ways that life has improved and trace the causal arrows with certainty. But let’s stop fretting for a moment about how hard it is to disentangle the strands and instead take note of their common direction. The very fact that so many dimensions of well-being are correlated across countries and decades suggests there may be a coherent phenomenon lurking beneath them—what statisticians call a general factor, a principal component, or a hidden, latent, or intervening variable.43 We even have a name for that factor: progress.
No one has calculated this vector of progress underlying all the dimensions of human flourishing, but the United Nations Development Programme, inspired by the economists Mahbub ul Haq and Amartya Sen, offers a Human Development Index that is a composite of three of the major ones: life expectancy, GDP per capita, and education (being healthy, wealthy, and wise).44 With this chapter we have now examined all of these goods, and it’s an appropriate point to step back and take in the history of quantifiable human progress before we turn to its more qualitative aspects in the next two chapters.
Two economists have developed their own versions of a human development index that can be estimated retroactively into the 19th century, each of which aggregates measures of longevity, income, and education in different ways. Leandro Prados de la Escosura’s Historical Index of Human Development, which goes back to 1870, averages the three measures with a geometric rather than an arithmetic mean (so that an extreme value on one measure cannot swamp the other two), and transforms the longevity and education measures to compensate for diminishing returns at their high end. Auke Rijpma of the “How Was Life?” project (whose data have appeared in a number of graphs in this book) developed a Well-Being Composite that goes back to 1820; together with the big three, it throws in measures of height (a proxy for health), democracy, homicide, income inequality, and biodiversity. (The latter two are the only ones that don’t systematically improve over the past two centuries.) The grades for the world on these two report cards are shown in figure 16-6.
Figure 16-6: Global well-being, 1820–2015
Sources: Historical Index of Human Development: Prados
de la Escosura 2015, 0–1 scale, available at Our World in Data, Roser 2016h. Well-Being Composite: Rijpma 2014, p. 259, standard deviation scale over country-decades.
To behold this graph is to apprehend human progress at a glance. And packed into the lines are two vital subplots. One is that although the world remains highly unequal, every region has been improving, and the worst-off parts of the world today are better off than the best-off parts not long ago.45 (If we divide the world into the West and the Rest, we find that the Rest in 2007 had reached the level of the West in 1950.) The other is that while almost every indicator of human well-being correlates with wealth, the lines don’t just reflect a wealthier world: longevity, health, and knowledge have increased even in many of the times and places where wealth has not.46 The fact that all aspects of human flourishing tend to improve over the long run even when they are not in perfect sync vindicates the idea that there is such a thing as progress.