Alfred Binet, a French psychologist working in Paris at the turn of the twentieth century, created the first practical test of intelligence in 1905. His purpose was narrow and humane: the French government had asked him to find a reliable way to identify schoolchildren who were struggling academically so they could receive extra support. But within a decade of Binet’s death in 1911, his invention had been repurposed, rescaled, and exported in ways he would scarcely have recognized, turning a diagnostic tool for individual children into a mass sorting instrument that shaped immigration law, military policy, and generations of educational practice.
The Problem Binet Was Trying to Solve
In 1904, France’s Ministry of Public Instruction appointed a commission to address a practical question: how could schools reliably distinguish children who genuinely had intellectual disabilities from those who were simply unmotivated or poorly taught? Until that point, such judgments were left entirely to teachers and doctors, whose assessments varied wildly. Binet, along with his colleague Théodore Simon, set out to create an objective, standardized method. The resulting Binet-Simon Intelligence Scale, published in 1905, was designed to measure intellectual function as a whole rather than any single narrow ability.1Springer Link / The ECPH Encyclopedia of Psychology. Binet-Simon Intelligence Scale
Binet’s approach was deliberately practical. He assembled a series of short tasks arranged by difficulty and grouped by age level. A typical five-year-old might be asked to copy a simple drawing or count four coins. An older child might be asked to define abstract words, spot absurdities in a story, or arrange sentences in logical order. A child’s performance was then summarized as a “mental age,” the age level at which most of the child’s correct answers clustered. A seven-year-old who passed the tasks typical of a nine-year-old had a mental age of nine. The concept was elegant and intuitive for educators. It was also, as Binet himself insisted, limited. He viewed the scale as a snapshot of a child’s current functioning, not a permanent verdict on their capacity.
A Multifaceted View of Intelligence
Binet was deeply skeptical of the idea that intelligence could be boiled down to a single number. He saw the mind as a collection of distinct capacities, including memory, attention, reasoning, and imagination, and he believed that any meaningful assessment had to probe many of them. This conviction shaped the structure of his test and the way he interpreted scores. He argued repeatedly that mental ability was not fixed at birth; it could be cultivated, and education could raise it. He coined the term “mental orthopedics” for exercises aimed at strengthening weak cognitive skills, much the way physical therapy strengthens weak muscles.
This philosophy persisted in France well after Binet’s death. French psychologists such as Henri Piéron and Henri Laugier continued to develop intelligence assessments in the 1920s and 1930s that reflected Binet’s original concept of multifaceted intelligence requiring lengthy and varied testing.2Wiley Online Library. After Binet: French intelligence testing, 1900–1950 While Americans and the British were increasingly drawn to theories and practices that reduced intelligence to a single score, most French researchers maintained a subtler, more complex understanding of what their tests could and could not measure.
The Atlantic Crossing and the Birth of “IQ”
Binet’s test traveled to the United States largely through the work of Lewis Terman, a psychologist at Stanford University. In 1916, Terman published a heavily revised version of the Binet-Simon scale, renormed on American children and expanded with new items. He called it the Stanford Revision of the Binet-Simon Scale, soon shortened to the Stanford-Binet. Terman also popularized a numerical device that had been suggested by the German psychologist William Stern: the intelligence quotient, or IQ. The original formula divided a child’s mental age by their chronological age and multiplied by 100. A ten-year-old performing at the level of a typical twelve-year-old scored 120. The simplicity of the number made it irresistible.
The shift was not just cosmetic. Binet had used mental age as a descriptive label, a rough guide for educators. Terman and many of his American contemporaries treated IQ as a measurement of an innate, largely hereditary trait. Where Binet had cautioned against treating a test score as a permanent label, Terman argued that IQ was essentially fixed and that differences between individuals reflected genuine biological endowment. This philosophical reversal happened quickly and with enormous institutional momentum behind it. Within a generation, the test Binet had built to help struggling children was being used to rank entire populations.
The Mental Age Paradox
Binet’s foundational insight was that children’s increasing ability to solve harder problems as they age could generate a single scale of intelligence. That insight turned out to be more complicated than it looked. Recent theoretical work has argued that the developmental changes captured by mental age and the individual differences captured by IQ are actually related to different underlying mechanisms, not just two ways of expressing the same thing.3Europe PMC. Binet’s Error: Developmental Change and Individual Differences in Intelligence Are Related to Different Mechanisms In plain terms, what makes an eight-year-old smarter than a five-year-old (brain maturation, experience, education) is not the same thing that makes one eight-year-old score higher than another (genetic variation, environmental differences within the same age group). Binet’s scale conflated these two dimensions, and that conflation has rippled through a century of intelligence testing.
This does not mean the original test was useless. It worked well for its intended purpose: spotting children who were falling behind their age group. But it does mean that treating the resulting score as a window into some stable, general cognitive capacity was always an overinterpretation, and one that Binet himself warned against.
From Classrooms to Trenches
The first truly large-scale application of intelligence testing came during World War I. When the United States entered the war in 1917, the Army needed a fast way to sort nearly two million recruits into roles. A team of psychologists, including Terman and Robert Yerkes, developed two group-administered tests. The Army Alpha was a written test for literate recruits. The Army Beta was a nonverbal, pictorial test for recruits who could not read English, including recent immigrants and those with limited schooling.
The Army Beta in particular has been a flashpoint for criticism ever since. Stephen Jay Gould, in his influential book The Mismeasure of Man, argued that it was poorly designed, inappropriately administered, and unable to measure intelligence. A detailed historical examination and pre-registered replication of the test found that Gould mischaracterized it in several ways, concluding instead that the Army Beta was well-designed by the standards of its era and that all evidence indicates it measured intelligence then and can, to some extent, still do so today.4Europe PMC. Stephen Jay Gould’s Analysis of the Army Beta Test in The Mismeasure of Man: Distortions and Misconceptions Regarding a Pioneering Mental Test The debate matters because the Army testing program became the template for mass cognitive assessment in education, employment, and policy.
IQ as a Political Weapon
The Army test data did more than sort soldiers. When the results were analyzed after the war, they revealed stark differences in average scores across racial and ethnic groups and between native-born Americans and immigrants. Eugenicists seized on these findings. Low scores among immigrant groups, particularly those from southern and eastern Europe, were presented as evidence of genetic intellectual inferiority. These claims were used to lobby the U.S. Congress for restrictive immigration legislation.5eLS. Intelligence Tests and Immigration to the United States, 1900–1940
The Immigration Act of 1924, which imposed severe national-origin quotas that dramatically reduced immigration from the countries whose citizens had scored lowest on the Army tests, owed a significant intellectual debt to this line of argument. The fact that the score differences almost certainly reflected language barriers, cultural unfamiliarity with the test format, and differences in educational opportunity rather than innate ability was largely brushed aside. Binet, who had died more than a decade earlier, would have been horrified. He had explicitly warned that his test should never be used to label people as inherently limited or to rank ethnic groups, and he would have viewed the eugenic application of his work as a fundamental betrayal of its purpose.
The General Intelligence Debate
Running parallel to the story of test development is a theoretical question: is there such a thing as general intelligence? The British psychologist Charles Spearman noticed in the early 1900s that people who scored well on one kind of cognitive task tended to score well on others. He proposed a single underlying factor, which he called “g,” that accounted for the positive correlations among different test scores. Brain-imaging research has lent some support to this idea, finding that tasks with high g-involvement activate overlapping regions in the lateral frontal cortex.6PubMed. A neural basis for general intelligence
Binet’s position was more cautious. He acknowledged that many cognitive abilities tended to travel together, but he resisted collapsing them into a single number. The tension between the “g is real and measurable” camp and the “intelligence is too complex for one score” camp has never fully resolved. Modern intelligence tests attempt a compromise: they yield both an overall composite score and a profile of subscores in distinct cognitive domains. The composite score satisfies those who want a summary. The subscores honor Binet’s intuition that a single number conceals more than it reveals.
Scores Keep Rising, and Nobody Fully Knows Why
One of the most striking discoveries in the history of intelligence testing is the Flynn effect, named after the New Zealand researcher James Flynn. IQ scores have been rising steadily across populations over time, a trend documented across more than a century of data. A meta-analysis covering 285 studies found that scores rose by roughly two to three points per decade, and comparisons involving modern Stanford-Binet and Wechsler tests specifically showed gains of about three IQ points per decade.7PubMed Central. The Flynn effect: a meta-analysis A separate global meta-analysis drawing on 271 independent samples from 31 countries and nearly four million participants confirmed that these gains extend worldwide, though they vary across countries and types of cognitive ability.8PubMed. One Century of Global IQ Gains: A Formal Meta-Analysis of the Flynn Effect (1909-2013)
The causes remain debated. Better nutrition, more years of schooling, smaller family sizes, greater exposure to abstract thinking through technology and media, and reduced childhood disease have all been proposed. No single explanation accounts for the full pattern. The gains tend to be larger on tests of abstract reasoning than on tests of vocabulary or general knowledge, which is puzzling if you assume better education is the main driver.
There are also signs that the effect may be slowing or reversing in some countries. Data from Austria showed continued IQ gains across all test domains between 2005 and 2018, with effect sizes ranging from small to large depending on the domain, but also hinted at a weakening in how strongly different cognitive abilities correlate with one another.9PubMed Central. Increasing IQ Test Scores and Decreasing g: The Flynn Effect and Decreasing Positive Manifold Strengths in Austria (2005-2018) Similar patterns of stagnation or reversal have been reported in Scandinavian countries. Whether this represents a ceiling on environmental improvement, a genuine shift in cognitive profiles, or a measurement artifact is an open question that researchers are still working through.
The Flynn effect has a practical consequence that matters for anyone who has ever taken an IQ test: norms go stale. A person tested using a ten-year-old version of the Stanford-Binet would appear to score higher than if they took the newly normed version, simply because the average performance of the norming sample has risen. Test publishers periodically re-norm their instruments for exactly this reason, and clinicians are trained to account for norm obsolescence when interpreting scores. If you scored 115 on a test normed fifteen years ago, your score on a freshly normed version of the same test would probably be a few points lower, not because you got less intelligent, but because the yardstick shifted.
The Stanford-Binet Today
Despite the controversies that have swirled around intelligence testing for more than a century, the Stanford-Binet remains in clinical use. The current edition, the Stanford-Binet Fifth Edition (SB5), assesses five main cognitive domains and is used in educational and clinical settings to evaluate both verbal and nonverbal abilities.10PubMed. Assessment of verbal and nonverbal cognitive abilities in children with stuttering using 5th edition Stanford Binet It is used alongside other major intelligence tests, particularly the Wechsler scales, in contexts ranging from learning disability diagnosis to gifted education screening to neuropsychological assessment after brain injury.
The SB5 bears only a family resemblance to Binet’s 1905 original. It has been renormed, restructured, and expanded repeatedly. The underlying philosophy, though, retains traces of Binet’s original vision: it produces a profile of abilities rather than only a single number, and it is intended as a clinical tool for understanding an individual person, not as a label to be carried for life. Whether practitioners actually use it that way is another matter. In educational and legal settings, the composite IQ score still carries outsized weight, and the rich subscale information often gets ignored in favor of the easier-to-digest headline number.
What Gets Lost When a Test Becomes a Number
The recurring tension in the story of IQ testing is between specificity and simplification. Binet designed a nuanced, task-by-task assessment that painted a detailed picture of what a particular child could and could not do. Terman simplified it into a single quotient. The Army psychologists turned it into a mass sorting instrument. Eugenicists turned the mass data into a political argument about which human populations were superior. At each step, information was stripped away and the remaining number was asked to carry more meaning than it could support.
This is not to say that IQ scores are meaningless. They predict academic performance reasonably well, and at the extremes they flag genuine developmental concerns. But the number’s power to predict does not mean it captures everything important about cognition, and it certainly does not mean it reflects a fixed biological endowment immune to environment, education, and opportunity. Binet understood this. He saw his test as a starting point for intervention, not as a sorting hat. “Some recent thinkers,” he wrote, “seem to have given their moral support to these deplorable verdicts by asserting that an individual’s intelligence is a fixed quantity, one which cannot be augmented. We must protest and react against this brutal pessimism.” That protest, made more than a hundred years ago, still has not fully been heard.
Why Some Psychologists Have Moved Beyond the IQ Framework
The field of intelligence research is broader today than at any point in its history, and not everyone working in it is comfortable with the IQ paradigm Binet inadvertently launched. Howard Gardner’s theory of multiple intelligences, Robert Sternberg’s triarchic theory, and various models emphasizing emotional intelligence or practical intelligence all represent attempts to capture aspects of human cognitive life that traditional IQ tests miss. None of these alternatives has displaced the IQ test in clinical or educational practice, largely because they are harder to measure reliably and have weaker track records of predicting real-world outcomes. But they reflect a legitimate dissatisfaction with the narrowness of what a two-hour paper-and-pencil test can assess.
Neuroscience has added another layer. Brain-imaging studies have found that the general factor “g” corresponds to identifiable patterns of brain activity, particularly in the lateral prefrontal cortex. This gives g a certain biological respectability, but it does not settle the debate about whether a single number is the best way to describe a person’s cognitive profile. A brain region’s involvement in many tasks does not mean all tasks reduce to one ability; it may just mean that particular region acts as a bottleneck or coordination hub for diverse processes. The honest state of the science is that intelligence is real, measurable to some degree, influenced by both genes and environment, and more multifaceted than any single test score can convey. That is roughly what Binet believed in 1905, and the field has spent a century working its way back to something close to his original position.