Modern WisdomHarvard Professor: “I Tried Every Diet. This Is By Far The Worst.” - Daniel Lieberman
CHAPTERS
- 0:00 – 2:55
From confused reader to self-experimenter: why Lieberman tried so many diets
Daniel Lieberman explains that his diet journey began while writing a book on the evolution of diet and deciding to test major diets firsthand. He describes how contradictory findings in top medical journals pushed him to integrate evolutionary biology with biomedical and anthropological evidence.
- •Tried multiple diets as “participant observation” while writing his book
- •Nutrition research often appears self-contradictory (fat, salt, etc.)
- •Goal: synthesize evolutionary, biomedical, cultural, and experiential data
- •Evolution helps answer “why” questions behind diet and physiology
- 2:55 – 5:57
Diet as identity: why nutrition debates become tribal and faith-based
Chris and Daniel discuss how modern diets resemble religions, complete with dogma and tribalism. Lieberman argues that in a globalized world, diet has become a key source of identity and meaning, especially because outcomes take decades and require “leaps of faith.”
- •Diet choices are increasingly used to signal identity and belonging
- •Long time horizons make diet a belief system more than a testable claim for individuals
- •People define themselves by what they exclude as much as what they eat
- •Online communities can function like “nutritional cults”
- 5:57 – 11:30
Plant-based vs meat-heavy: what the evidence suggests—and why it’s messy
Lieberman says large-scale evidence tends to favor plant-forward diets, while emphasizing that “plant-based” can still be unhealthy if it’s ultra-processed. He outlines dose-response cancer associations with meat and potential mechanisms (e.g., TMAO, Neu5Gc), while stressing trade-offs and individual variability (including saturated fat response).
- •Meta-analyses often show better outcomes for plant-forward patterns
- •Ultra-processed foods can be plant-based yet harmful
- •Red meat: epidemiological dose-response associations; mechanistic pathways discussed (TMAO, Neu5Gc)
- •Saturated fat responses vary widely by individual genetics/physiology
- 11:30 – 14:05
Why carnivore keeps growing: simplicity, masculinity, and “better than junk” effects
Chris asks why carnivore is booming despite skeptical long-term data. Lieberman suggests cultural signaling (masculinity), persuasive media, and the reality that almost any structured diet can feel better than a baseline ultra-processed diet—without necessarily being best long-term.
- •Carnivore’s appeal can be partly tribal/cultural (masculinity narratives)
- •Anecdotes can be misleading when baseline diet was very poor
- •Short-term improvements don’t prove long-term superiority
- •Diet comparison is hard because replacing one food means adding another
- 14:05 – 19:47
Proteinmaxxing reality check: essential nutrient, limited upside to extremes
They unpack why protein is having a cultural moment and clarify what’s true. Lieberman emphasizes protein’s necessity and satiety, but notes the body can’t store excess protein and that benefits plateau—plus protein doesn’t require meat (legumes/grains/soy combinations can suffice).
- •Protein is essential; some amino acids must come from diet
- •Higher protein can aid satiety and supports aging, pregnancy, training
- •Excess protein is converted to energy/fat; extreme benefits are overstated
- •Protein ≠ meat: beans, legumes, grains, soy, quinoa can contribute substantially
- 19:47 – 24:00
Mediterranean diet, explained: the most evidence-rich “pattern,” not a magic food list
Lieberman argues the Mediterranean diet ‘wins’ largely because it’s been studied more and performs well across study types. He explains epidemiology vs randomized trials and highlights PREDIMED as a landmark study, while noting the pattern’s success likely comes from a constellation of nutrient-dense foods and fewer refined/added ingredients.
- •Mediterranean diet has unusually strong support across study designs
- •Epidemiology shows associations; RCTs and mechanistic studies add context
- •PREDIMED: large Spanish trial halted early due to strong benefit signals
- •Core features: whole grains, olive oil/unsaturated fats, plants, fish; low added sugar/refined carbs/red meat
- 24:00 – 42:21
Can we trust nutrition science? Healthy-user bias, study limits, and the “fox vs hedgehog” mindset
Chris challenges whether epidemiology is too confounded to be useful. Lieberman argues no single method is sufficient and introduces the hedgehog/fox analogy: single-factor diet ideologies vs nuanced synthesis across evidence types, trade-offs, and individual differences.
- •Healthy-user bias and covariation (diet + exercise + smoking) complicate conclusions
- •Epidemiology can’t prove causation but remains valuable when triangulated
- •Hedgehog diets rely on one big idea; fox thinking integrates complexity and trade-offs
- •Modern problem: we no longer need to get food—we need to choose among 40,000 options
- 42:21 – 53:58
Low-carb vs low-fat: efficacy vs effectiveness and the carbohydrate–insulin model
Lieberman explains that both low-carb and low-fat diets can work for weight loss, with low-carb often showing higher efficacy but lower real-world adherence. He gives a lay explanation of insulin’s role and stresses that “carbs” are not one thing—fiber-rich, minimally refined carbs behave very differently from sugar/refined starches.
- •Weight loss: many non-junk diets work; sustainability is the bottleneck
- •Efficacy (ideal adherence) vs effectiveness (real world adherence)
- •Carb–insulin model: carbs raise glucose → insulin promotes storage; low-carb lowers insulin
- •Not all carbs spike glucose equally; fiber and processing matter; activity changes insulin sensitivity
- 53:58 – 1:06:12
Calorie restriction & intermittent fasting: metabolic switching, but limited longevity proof
They examine claims that calorie restriction reliably extends lifespan. Lieberman argues evidence in humans (and even primates) is not definitive, explains why rodent results may not generalize, and frames fasting as a way to cycle between growth (mTOR) and repair (AMPK) rather than living in constant deficit.
- •Primate calorie restriction results depended heavily on control diet quality
- •Mice/yeast/flies don’t map cleanly to human longevity
- •Mechanism: switching between mTOR (growth) and AMPK (repair) may be beneficial
- •Intermittent fasting can be more tolerable than constant restriction; starvation is the extreme trade-off
- 1:06:12 – 1:07:54
Paleo under the evolutionary microscope: no single ancestral diet, and cooking changes everything
Chris asks whether ‘ancestral eating’ can be reconstructed. Lieberman argues paleo gets some basics right (avoid sugar/refined foods) but is wrong to claim humans didn’t eat grains/legumes; evidence shows wide dietary diversity among hunter-gatherers and even Neanderthals, and humans evolved as extreme omnivores aided by cooking and fermentation.
- •Hunter-gatherer diets vary widely; there’s no single ancestral macro split
- •Evidence for starch/grain/plant consumption appears in archaeology and dental remains
- •Elimination rules (no legumes/grains/dairy) are often overstated or incorrect
- •Naturalistic fallacy: “ancient = optimal/healthy” doesn’t follow
- 1:07:54 – 1:19:38
Antinutrients and raw foodism: why cooking is a cornerstone of human biology
They tackle lectins, phytates, and oxalates, with Lieberman arguing dose and preparation matter—cooking/fermentation largely neutralize issues for most people. He then calls raw food the worst diet he tried, explaining that cooking increases net energy and nutrient availability and likely enabled big human brains and high metabolic demands.
- •Antinutrients can hinder absorption in raw foods; cooking/fermentation mitigates them
- •Lectins in raw beans can be problematic; cooked beans are generally safe
- •Raw food diet is difficult to sustain and can leave people hungry due to lower energy extraction
- •Cooking improves digestion, safety, and nutrient release; key to human evolution
- 1:19:38 – 1:24:58
Did farming make us healthier? More calories, more disease, and the rise of refined grains later
Lieberman describes agriculture as an evolutionary success but an individual health trade-off: more food and babies, but more infectious disease, crowding, animal exposure, and famine vulnerability. He also clarifies that traditional farming diets differed from modern refined-grain diets; industrial milling changed the metabolic impact of cereals.
- •Agriculture increased population but also infectious disease and density-related hardship
- •Living with animals and poor sanitation amplified pathogen load
- •Famine and organized violence became more consequential for settled populations
- •Traditional whole grains/legumes differ substantially from modern refined flour products
- 1:24:58 – 1:28:01
Gluten, allergies, and autoimmunity: hygiene hypothesis and antibiotics as suspects
Chris asks why gluten issues seem to be rising. Lieberman points to broader increases in autoimmune conditions and suggests the hygiene hypothesis and widespread antibiotic exposure as plausible contributors via immune ‘boredom’ and microbiome disruption.
- •Autoimmune and allergy rates (gluten, peanuts, etc.) have climbed
- •Hygiene hypothesis: reduced pathogen/worm exposure may dysregulate immune responses
- •Antibiotics—direct and via food supply—may alter microbiome and gut barriers
- •Evidence is suggestive; mechanisms are still being worked out
- 1:28:01 – 1:43:06
Why we’re gaining weight: ultra-processed food design, appetite biology, and activity’s hidden role
Lieberman attributes modern obesity primarily to ultra-processed, hyper-palatable foods engineered around ‘bliss points’ and texture/novelty tricks. They discuss how fat cells, leptin, and appetite signals make weight regain likely, why exercise helps prevent gain (even if it’s hard for loss), and how labels can mislead about true absorbed calories.
- •Hyper-palatable foods combine sugar + fat (+ salt/texture/novelty) to drive overeating
- •Once fat cells expand and multiply, future fat gain becomes easier; appetite regulation fights loss
- •Exercise helps prevent weight gain and improves insulin sensitivity beyond calorie burn
- •Calories on labels are estimates; processing/blending changes absorbed energy; microbiome adds uncounted calories
- 1:43:06 – 1:51:40
“Healthy” foods and processing pitfalls: meal replacements, solvents, and the seed-oil myth
Chris asks what people eat believing it’s healthy but may backfire. Lieberman flags highly processed meal replacements and industrial processing (e.g., hexane solvents in some products) as areas of uncertainty, then pushes back on the idea that seed oils are uniquely harmful, citing mechanistic and epidemiological evidence that omega-6 intake is not clearly pro-inflammatory and may correlate with better outcomes.
- •Concern: ultra-processed ‘health’ products (meal replacements) contain additives with uncertain long-term effects
- •Solvent processing (e.g., hexane) may leave residues; risk framing depends on assumptions and regulation
- •Seed oils: explains omega-3 vs omega-6 chemistry and the inflammation pathway hypothesis
- •Claims that omega-6 drives inflammation broadly are not supported well by controlled and large cohort data
- 1:51:40 – 2:08:13
Why diets fail, GLP-1 trade-offs, and the simplest durable principles for eating well
They close on why most diets stop working: adherence collapses and biology pushes back with hunger signals. Lieberman sees GLP-1s as powerful but imperfect (muscle loss, discontinuation and rebound fat gain, ‘disevolution’), then offers practical principles—avoid ultra-processed foods and misleading additives, improve labeling, and consider policy tools like taxes to rebalance a predatory food environment.
- •Most diets fail due to sustainability; weight loss triggers hunger and hormonal pushback
- •GLP-1s can produce major weight loss but may reduce muscle and require long-term use for maintenance
- •Regaining after stopping GLP-1s may skew toward fat, worsening body composition
- •Core guidance: avoid refined/ultra-processed foods and “weird” additives (e.g., maltodextrin); improve consumer info and consider systemic interventions
