David Heron was named the first Francis Galton Fellow in National Eugenics in 1907, plying his trade in the Eugenics Laboratory of the University of London. He wanted to measure whether external factors had an impact on intelligence, and in 1910 published on “The Influence of Defective Physique and Unfavourable Home Environment on the Intelligence of School Children.” He found what he expected: environment, he said, had little effect on intelligence. Not surprisingly, some critics disagreed with his analysis, but not always for reasons we might think today.
Heron’s most potent critic was George Udney Yule, a mathematician interested in applying statistical methods to social sciences (and author of An Introduction to the Theory of Statistics, a standard textbook published first in 1911 and revised through many subsequent editions). Yule’s main objection was that Heron’s research relied on teachers' assessments of intelligence in order to evaluate and categorize students, and that these assessments were vague and flawed.
Heron had asked teachers to classify their students as "brilliant," "above the average," average," "below the average," or "very dull and backward"; he also drew on another study in which teachers assigned their students to the categories “excellent,” “good,” “moderate,” or “dull.” Yule, on the other hand, endorsed measurements gathered between 1888 and 1894 by Dr. Francis Warner, who claimed to have assessed some 50,000 “dull” children for various physical characteristics, including the form of the external ear, quality of hand balance, and degree of “grinning or over-smiling.” In Yule’s estimation, Warner’s warehouse of anthropometric data provided a physical shape for “dullness” – it didn’t rely simply on the assessments of teachers, but instead was shaped by a hefty load of apparently related information.
Heron responded in “Mental Defect, Mal-Nutrition, and the Teacher's Appreciation of Intelligence,” a short booklet published in 1911. First of all, he wrote, many of Warner’s measurements were entirely unrelated to intelligence – at any rate, no evidence clearly linked ears to intelligence. Secondly, he took aim at Warner’s methodology, questioning just how precise it could be: after all, in order to examine all of the people he claimed, he must have galloped along at a pace of well under 20 minutes per individual, if working each day from 1888 to 1894. Then, he pointed out, Warner does not bother to provide measurements from all the “normal” subjects he also claimed to have examined in his survey. And, finally, wrote an exasperated Heron, Warner’s subjects were categorized – by teachers, no less! – as simply “under the average,” or left entirely uncategorized. Warner’s data, he claimed, was worthless.
Lying at the heart of this dispute was not simply a disagreement over how to collect data, but more fundamentally over how to identify intelligence. Researchers were assessing intelligence more and more often at the turn of the century, mainly because of fears that the allegedly “feeble-minded” would procreate wildly, thus weakening the national breeding stock (Heron’s pamphlet was part of a series published by the Francis Galton Eugenics Laboratory called “Questions of the Day and of the Fray,” so the researchers clearly imagined a social dimension to their work – this was no abstract ivory tower stuff). Subjective assessments of intelligence, including the “idiot/imbecile” diagnoses used by 19th century physicians, had worked just fine for most of the nineteenth century. But clearly something more precise, more nuanced, was required to confront the twentieth century’s eugenic challenges.