Enlightened Self-Interest.

Ashley Hodgson unpacks what economists actually mean by rationality — a word so misread it generates more heat than light in almost every policy debate.

▶ Play video

Video: Ashley Hodgson / YouTube

A word doing too much work.

The word "rational" does different things in different rooms. In everyday speech, rational means something close to cool-headed, deliberate, unswayed by emotion. In economics, it means something closer to goal-pursuing — and the goals can include emotion, instinct, tribal loyalty, and impulsive action. That gap between the two definitions is the source of a staggering amount of confusion in public discourse about markets, policy, and human nature.

Ashley Hodgson's nine-minute video is an exercise in vocabulary repair. She takes the three-word definition — "pursuing enlightened self-interest" — and unpacks each term until the definition earns its place in a serious latticework. Pursuing is not deliberating; it includes gut-driven behaviour. Enlightened is not omniscient; it means informed by local, lived experience. Self-interest is not selfishness; it means vantage-point-bound — not self-seeking, but situated.

These are not word games. They change what you look for when you're trying to understand why a person or system behaves as it does.

Old models, sharper edges.

Hayek's knowledge problem — the insight that local knowledge is irreducibly distributed and cannot be centrally aggregated — maps cleanly onto the "enlightened" component of the definition. When Hodgson says that people inside a system have access to knowledge that system designers cannot access, she is restating the knowledge problem in behavioural terms. A rational actor in the Hodgsonian sense is not all-knowing; she is drawing on a specific, local body of experience that the designer cannot replicate. That is the case for humility in system design.

Korzybski's map is not the territory gets sharpened here too. Language about rationality, Hodgson argues, is itself a map that gets used as a weapon or a shield as often as it gets used as an instrument. The word "rational" can be deployed to dismiss a competitor's argument, to foreclose debate, or to signal in-group membership. The map is unreliable not because the territory changed, but because the mapmaker had a motive.

Emergence — the principle that system-level behaviour can't be deduced from individual-level intentions alone — is quietly endorsed by the whole account. A system's outcome is the product of many vantage-point-bound, instinct-driven actors, each with limited visibility. Predicting that outcome from any one vantage point is a category error.

Models that bend under scrutiny.

The selfish-altruistic binary — a common framing in both economics pedagogy and popular moral philosophy — doesn't survive Hodgson's analysis intact. She introduces the mother-at-the-school-play example: advocating for your child's position in a play is simultaneously selfish (from the perspective of the other mother's child) and altruistic (from the perspective of your child). It is neither purely one nor the other. The binary was always an oversimplification, and a particular costly one in policy debates where it gets used to assign moral valence to economic positions.

The homo economicus assumption — that economic actors are perfectly rational, fully informed, and self-maximising — gets quietly retired without ceremony. The Hodgsonian replacement is not "people are irrational"; it is "people are pursuing goals from within a specific, limited vantage point, using information available at that vantage point, often through gut rather than deliberation." That is a fundamentally different claim, and it calls for fundamentally different system designs.

The assumption that deliberation is the signature of rationality gets challenged too. Hodgson's image of a lion pursuing a gazelle — purposeful, instinct-driven, non-deliberative — is her cleanest illustration: that is not irrational behaviour. It is goal-pursuing behaviour that happens not to run through the prefrontal cortex. The rationality/irrationality binary was always partly a story about which cognitive mode we were willing to dignify.

New entries for the latticework.

The most generative addition is vantage-point self-interest: the principle that "self" in "self-interest" denotes a situated perspective inside a system, not a fixed ego boundary. A nation acting "self-interestedly" in foreign policy is expressing a vantage point. A firm acting "selfishly" is expressing a vantage point. A mother advocating for her child is expressing a vantage point. The concept doesn't privilege any of these; it just says: every actor is operating from somewhere, and that somewhere limits and shapes what they can perceive and pursue.

The second new model is pursuit vs. deliberation: the distinction between goal-directed behaviour that runs through conscious deliberation and goal-directed behaviour that runs through trained instinct or gut reaction. The economic definition of rationality covers both. Most real human behaviour is the second kind. Any model that only accounts for deliberate choice is modelling a small slice of the action. Designers and policymakers who optimise for the deliberative slice are building systems for the exception, not the rule.

The third addition is language-as-instrument vs. language-as-weapon. Hodgson's closing observation — that the same word can be used to understand or to foreclose understanding — generalises to any technical vocabulary in any discipline. The question to ask when terminology is introduced in a policy debate or a meeting is not "what does this word mean?" but "what is this word being used to do?" Those are often very different questions.

When to reach for which.

After the definition.

Language can be used as an instrument to better understand the world. It can be used as a weapon. It can be used as a shield to stop you from understanding something.Ashley Hodgson

The word "rational" is not going away. Neither is the confusion it generates. What Hodgson offers is not a better definition so much as a better question: when you hear the word, ask what the speaker is trying to do with it. Instrument, weapon, or shield. The answer tells you more about the conversation than the definition ever will.