Peter Attia MD167 - Gary Taubes: Bad science and challenging the conventional wisdom of obesity
CHAPTERS
- 0:02 – 8:27
From Physics Major to Journalism: Origins of Gary's Skepticism
Peter asks about Gary's background in physics at Harvard, aerospace at Stanford, and his move into journalism. Gary describes how the Shroud of Turin story at Discover taught him that scientists and journalists are not so different, and raised the question of what a negative result means.
- •Gary started in astrophysics, switched to applied physics, then did a Stanford master's before turning to journalism
- •He links his skepticism partly to growing up competing with an older brother
- •His Shroud of Turin story had him phoning a Los Alamos researcher at home to challenge the interpretation
- •A negative result can mean the effect is absent or that the equipment is inadequate, a debate he sees again in obesity research
- 8:27 – 18:55
Boxing, Risk Perception and the Playboy Golden Gloves Piece
The conversation turns to smoking and then to boxing. Both men discuss how young people fail to perceive danger, Gary describes his best friend's death in the ring, and he tells the story of boxing in the Golden Gloves for a Playboy article.
- •Gary says he started smoking in 1978 and took twenty years to quit
- •Peter describes never feeling boxing was dangerous until a severe concussion at 20
- •Gary's college friend died after his first amateur fight
- •Gary suggests older people become more risk averse as they have less life to lose
- •He lost his second Golden Gloves fight by knockout and never boxed again
- 18:55 – 29:53
Learning to Write and What Great Science Writers Do
Peter asks how Gary developed his writing craft. Gary credits harsh editing, the reporter-and-writer system at Time Inc., and rewriting, then discusses writers he admires and what makes science writing good.
- •A fiction editor's "puerile" critique and relentless criticism shaped Gary's writing
- •Gary cites Calvin Trillin's idea of the first draft as the "vomit out" followed by repeated rewriting
- •He declines to name the best living science writer, praising several peers
- •Good science writing combines rigorous reporting, strong prose, and a story that grows larger than its subject
- •He praises The Ghost Map as an example
- 29:53 – 43:24
Rubbia, CERN and the Path to Nobel Dreams
Gary explains how his profile of Carlo Rubbia as Scientist of the Year led him to embed at CERN to write a book. He and Peter then explain how particle colliders work, the signal-to-noise problem, and why physicists demand five sigma.
- •Rubbia's group discovered the W and Z bosons, and Gary profiled him for Discover
- •Gary left Discover to live at CERN for the nine months of the experiment
- •Colliders look for rare events against billions of collisions, so detector flaws must be understood extremely well
- •Unknown unknowns are why physicists require five sigma
- •Independent detectors are needed for replication
- 43:24 – 54:11
Theorists vs. Experimentalists and the Rise of Pathological Science
Peter asks Gary to explain how physics separates theory from experiment, and Gary contrasts that with biology and medicine. He then describes how Rubbia's public commitment to a premature result produced groupthink and what Feynman's principle of not fooling yourself requires.
- •Physics splits theorists and experimentalists, while biology and medicine merge both roles in one person
- •Experimentalists must know how their equipment can fool them
- •Pathological science involves committing publicly to a result on premature evidence
- •The right approach is to try to prove yourself wrong before publishing
- •Once leaders commit, they listen to supporters and dismiss skeptics, a pattern Gary says appears in COVID, climate and nutrition
- 54:11 – 1:05:40
Rubbia's Reaction, the Book's Reception and Two Kinds of Bad Science
Gary describes Rubbia as a pathological liar who never grasped that Gary saw things differently, and the threatened lawsuit after the book came out. He distinguishes fraud from fooling yourself by neglecting the background, and wrestles with Rubbia's mixed legacy.
- •Rubbia tried to sue, but his executive committee refused, though many had critiqued drafts for Gary
- •Gary says fooling yourself by ignoring the background is different from fraud
- •Peter points out that the paradox of Rubbia's real discovery alongside sloppy method is unnerving
- •Gary says the Nobel-winning papers were probably wrong too, though the particles had to exist
- •Some reviewers argued physicists' character flaws were not a proper subject
- 1:05:40 – 1:16:42
Cold Fusion and the Making of Bad Science
Gary explains how he came to write about Pons and Fleischmann's cold fusion claim. He covers the competition that drove it, how replication failed, and why he treated it as a case study in how scientists fool themselves.
- •Gary assumed the claim was wrong as soon as he saw a competing group in the first article
- •Competition compromised the goal of establishing knowledge beyond reasonable doubt
- •Cold fusion was cheap to attempt, creating a "bad science generator"
- •Within a few months most labs concluded cold fusion did not exist
- •Gary wanted the book to be required reading for graduate students
- 1:16:42 – 1:24:46
Langmuir's Pathological Science and the Georgia Tech Neutron Detectors
Peter and Gary discuss Langmuir's description of pathological science. Gary illustrates it with the Georgia Tech chemists whose neutron detector responded to humidity, then covers calorimetry errors and why Pons and Fleischmann never conceded.
- •Langmuir defined pathological science as the science of things that aren't so, with effects near the detection limit
- •Georgia Tech's neutron counts were an artifact of humidity because they used the wrong control
- •Non-specialists doing calorimetry saw false excess heat
- •Gary says it takes superhuman intellect to admit a staked belief is wrong
- •He explains his remark about never changing his mind and his skepticism about cold fusion
- 1:24:46 – 1:47:22
Epidemiology, Confounding and the Conflict of Public Health Science
Gary explains how his friends in physics steered him toward power-line cancer studies and a critique of epidemiology. He and Peter explore why association is not causation, how confounders like socioeconomic status arise, and the tension between honest uncertainty and getting people to act.
- •Physicist friends pointed Gary to the power-line cancer debate, which led to his Atlantic and Science pieces
- •Peter notes that scientific rigor is only about 400 years old, and Gary adds that Bacon laid it out in 1620
- •Confounders such as socioeconomic status and participation bias are hard to test without experiments
- •Epidemiology accepts a 95% confidence level where physicists demand far more
- •Gary cites Feynman's call to bend over backwards about limitations and describes the public health conflict this creates
- 1:47:22 – 2:05:05
Salt, Dietary Fat and the Road to the New York Times Magazine Article
Gary tells how a rent-money assignment on the DASH diet led him into the salt-blood pressure controversy and then dietary fat. He explains how the Atkins trials he found while researching his New York Times Magazine article challenged two core hypotheses.
- •The DASH assignment revealed an unexpected controversy over salt
- •Gary spent nine months and about 85 interviews on salt and a year on fat for Science
- •His New York Times Magazine article asked what caused the obesity epidemic and weighed low-fat advice against high-fructose corn syrup
- •Five unpublished trials found Atkins dieters lost more weight and had better heart risk factors
- •Gary tried Atkins himself, lost 25 pounds, and the article led to the Good Calories, Bad Calories book deal
- 2:05:05 – 2:13:34
Good Calories, Bad Calories and the Neglected Science of Fat Metabolism
Peter asks how Gary's thinking developed while writing Good Calories, Bad Calories. Gary lays out the book's structure and argues that obesity research ignored decades of work on fat tissue, insulin and fuel partitioning in favor of an energy-balance view.
- •The book first deconstructs the dietary fat hypothesis, then presents a carbohydrate-centric alternative and addresses obesity
- •Gary says whatever makes us fat also drives diseases such as type 2 diabetes
- •Pre-war European researchers saw obesity as a hormonal regulatory disorder
- •Fat tissue was little studied until the 1930s, and fat metabolism research was later left out of obesity discussions
- •1960s obesity research was dominated by psychologists and psychiatrists trying to get fat people to eat less
- 2:13:34 – 2:32:38
History of Obesity Research: Newburgh, Brobeck and Hypothalamic Lesions
Gary traces how Newburgh's 1930s experiment and Brobeck's hypothalamic lesion work cemented the idea that overeating causes obesity, while Bauer, Ranson and Hetherington offered an alternative reading. Peter highlights how the same observation supports opposite interpretations and Gary describes his proposed mechanism through insulin.
- •Newburgh's small starvation experiment was taken as proof that obese people get fat by eating too much
- •Bauer argued it was naive
- •Brobeck's lesioned rats showed hyperphagia, and he excluded the three pair-fed animals that still got fatter
- •Ranson and Hetherington proposed that hunger and inactivity respond to a drive to accumulate fat, like thirst in diabetes insipidus
- •Ranson died and Hetherington left the field, which Gary says left Brobeck's view dominant
- •Gary says lesions cause hyperinsulinemia that traps fuel in fat tissue
- 2:32:38 – 2:38:30
Leptin and the ob/ob and db/db Mice
Peter asks about the ob/ob mice. Gary explains that pair-fed ob/ob animals still get obese, that leptin was interpreted as a satiety hormone, and that the diabetic db/db versus ob/ob difference reflects hyperinsulinemia on different backgrounds.
- •Pair-feeding half the intake of a lean mouse to ob/ob mice still produced obesity
- •Friedman's leptin discovery was interpreted through the satiety assumption
- •Gary argues these animals confirm a fuel-partitioning disorder
- •Both animal types are hyperinsulinemic from weaning, but only the db/db background's pancreas fails
- •Gary says the field never considered the competing hypothesis
- 2:38:30 – 2:44:43
Responding to Findings That Challenge the Carbohydrate-Insulin Model
Peter asks how Gary reconciles studies that did not support the carbohydrate-insulin model, and asks how a biased field can make progress. Gary recounts the Pennington talk and describes the paradigm problem, then Peter offers the failure of the current model as the reason to keep questioning it.
- •NuSI funded one group testing each hypothesis, and each interpreted its results as supporting its own
- •Gary says energy-balance thinking is "not even wrong" and recounts being asked whether he thought they were all idiots
- •He argues researchers inherited an untested paradigm
- •Peter says failure of the current model to prevent obesity is the best reason to keep questioning it
- •Gary mentions arguing with Mark Friedman about whether being right proves you are not a quack
- 2:44:43 – 2:58:18
Fixing the Obesity Epidemic: Palatability, Funding and Doritos vs. Wagyu
Peter raises the view that palatable, cheap and convenient food drives obesity, illustrated by his Cool Ranch Doritos story. Gary counters with Wagyu steak and discusses why authorities don't question the paradigm, how research funding favors normal science, and the idea of red-team and blue-team research.
- •Peter argues palatability, availability and convenience drive obesity
- •Gary contrasts Doritos with a filling Wagyu steak, asking whether macronutrient composition drives hyperphagia
- •Blaming industry is popular, while blaming the scientific community gets little traction
- •Gary says R01-style funding supports normal science and has no mechanism to question paradigms
- •He envisions jury-like teams of unbiased scientists to test competing hypotheses
- 2:58:18 – 3:13:03
Measuring Energy Expenditure, Kevin Hall's Paper and the Progress Made
Peter and Gary discuss the Speakman and Hall paper and the disagreement over calorimetry chambers versus doubly labeled water. They conclude with Gary's view that ketogenic diets have gained wide acceptance, even if the science of obesity remains unresolved.
- •Gary describes how Kevin Hall's biases may have locked in as the two argued
- •Peter explains chamber calorimetry versus doubly labeled water and critiques about DLW in carbohydrate-restricted people
- •Gary notes the two methods gave divergent results and questions using DEXA over short periods
- •Gary hopes the question is answerable and says it matters even with GLP-1 drugs
- •He points to wider acceptance of low-carb diets, including American Diabetes Association recommendations for type 2 diabetes
