Intelligence Solves Only Half of Life's Problems — and Not the Half That Matters

Standard psychology defines intelligence as a general capacity to reason, plan, and solve problems, which implies smarter people should navigate life better and end up happier. The data refuses to cooperate: meta-analyses and a large UK study find essentially no link between IQ and happiness, and 50 years of General Social Survey data (over 30,000 people) show that higher vocabulary scores correlate with a slight decrease in happiness. This holds even though IQ reliably predicts school and job outcomes, and even accounting for the well-documented biases and effort-sensitivity of the tests themselves.

The author traces the puzzle to Charles Spearman, who in 1904 noticed that performance across school subjects was positively correlated and inferred a single underlying ‘general intelligence.’ That statistical pattern (the ‘positive manifold’) is real and endlessly replicated, but the interpretation is flawed. What look like wildly different tests — math, vocabulary, music, spatial rotation — actually share a hidden family resemblance: they are all well-defined problems. They have stable rules, indisputable answers, finite information, and repeatable solution procedures, which is precisely why they can be scored on a standardized test and used to rank people.

The catch is that most of life’s consequential problems are poorly defined. ‘Should I be a dentist or a dancer?’, ‘How do I get my kid to stop crying?’, and ‘What do I do as my parents age?’ have no answer key and no transferable procedure — and skill at the well-defined kind doesn’t carry over. Because well-defined problems are the ones we can measure and rank, we mistake the people who top those rankings (physicists, mathematicians, chess champions) for the smartest people alive, while the messy, poorly-defined work of actually living well — closer to wisdom than to IQ — goes unmeasured and unrewarded by the test.