Over the course of the twentieth century, average scores on standardized intelligence tests rose steadily across the industrialized world, in some nations by nearly thirty points. Because the tests are periodically re-normed so that the mean holds at one hundred, this climb went largely unnoticed until researchers compared cohorts against a fixed standard. The finding, once documented, posed an awkward question. If the scores are to be believed, the average person of a century ago would today be classed as intellectually deficient - a conclusion no one is prepared to accept. Yet if the scores are not to be believed, the most widely used instrument in psychology would seem to be measuring something other than what it claims. Two responses have dominated. The first treats the rise as real: better nutrition, smaller families, longer schooling, and healthier childhoods have, on this view, produced genuinely more capable minds. The second treats it as an artifact: nothing about human intelligence has changed, but successive generations have grown more familiar with the peculiar demands of test-taking - with abstract symbols, timed problems, and the convention of treating hypothetical questions as if they were sincere. Each response captures something and misses something. The gains, tellingly, were not uniform. Scores on tests of learned content - vocabulary, arithmetic, general information - barely moved. The dramatic increases appeared on precisely those subtests meant to be culture-free: pattern completion, analogies, classification by abstract category. This distribution embarrasses both camps. It is hard to credit that only the most abstract faculties improved while ordinary knowledge stagnated; it is equally hard to dismiss as mere test-wiseness a change concentrated in the tasks psychologists had deemed the purest measures of reasoning. A more persuasive account locates the change not in raw capacity but in habits of mind. Modern life, saturated with formal schooling and technical work, rewards the disposition to reason about the abstract and hypothetical - to sort the world into logical categories rather than concrete functions. Asked what a dog and a rabbit have in common, an earlier respondent might note that dogs hunt rabbits; a modern one answers, without hesitation, that both are mammals. Neither reply is stupid. But only the second is the kind the tests reward, and only the second reflects a mode of thought that a century of scientific culture has made second nature. If this is right, the rise measures neither an increase in some fixed quantity called intelligence nor a mere corruption of the instrument. It measures the spread of a particular cognitive style - powerful, genuinely useful, and historically specific. That conclusion should temper two temptations at once: the triumphalism that reads rising scores as proof of a smarter species, and the skepticism that dismisses the tests as measuring nothing real. What they measure is real; it is simply narrower, and more contingent on culture, than the word intelligence invites us to assume.