Gary Taubes - Why We Get Fat: The Diet/Weight Relationship, An Alternative Hypothesis
What this video covers
An academic-setting lecture hosted by Cornell’s Science & Media initiative, this talk is Taubes at his most didactic: “Adiposity 101 and the Alternative Hypothesis of Obesity.” The framing is explicitly pedagogical — he is teaching the audience a competing model of obesity from first principles, aimed at listeners with enough scientific literacy to follow the endocrinology.
The lecture’s structure mirrors the book’s: first, the conventional account — obesity as the product of eating too much and exercising too little — and why Taubes considers it explanatorily empty; then the alternative — that the carbohydrates we consume act on the hormones and enzymes regulating fat tissue, driving accumulation independent of willpower or activity levels. He works through the logic step by step, presenting the carbohydrate-insulin mechanism as the more parsimonious account of the observed patterns: who gets fat, on what diets, and under what conditions.
He opens with a provocation for his academic hosts: imagine, he says, that the cause of AIDS had been established in the mid-1980s and mortality kept climbing anyway — “there would be national investigations by the CDC, the NIH, and every university in the country.” The fact that obesity and diabetes have only worsened under the accepted explanation, he argues, suggests the field has the etiology wrong: “this happened under your watch.”
As a university-hosted talk, the Q&A (where present in the full recording) gives a sense of how the academic audience received the argument. The lecture is useful as a compact, structured statement of the Taubes model stripped of book-tour anecdote — the version he wanted scientifically literate skeptics to engage with on its own terms.
Key points
- Taubes presents the “alternative hypothesis”: obesity is driven by carbohydrates’ effects on the hormones and enzymes regulating fat tissue, not by overeating per se.
- He argues the conventional eat-less-move-more account fails to explain observed patterns of who becomes obese.
- The lecture is structured as “Adiposity 101” — a first-principles walkthrough of fat-tissue regulation aimed at a scientifically literate audience.
- Taubes positions the carbohydrate-insulin mechanism as the more parsimonious explanation of population-level obesity patterns.
- His central analogy for reversed causality: a growing child goes from 20 to 45 pounds in positive energy balance — but grows because of growth hormone, not because he ate more; the excess intake is the effect of growth, not its cause. A tumor cell, likewise, grows in positive energy balance, but the growth causes the imbalance, not the reverse.
- He cites a then-recent systematic review — 23 reports across 17 clinical investigations covering 1,140 obese patients — in which low-carbohydrate diets were associated with improved weight, BMI, blood pressure, triglycerides, glucose, insulin, and inflammatory markers, plus higher HDL — while noting the standard caveat that long-term effects are unknown.
- Taubes’s memorable closing formulation: “If you don’t want people to get lung cancer, you don’t tell them to eat less and move more — you tell them to quit smoking,” arguing that refined grains and sugars are the analogous causal agent for obesity.
- Delivered in an academic venue, it represents his case as pitched to skeptics rather than to a popular audience.
- The underlying model remains contested; mainstream obesity researchers dispute both the mechanism and the reading of the evidence.