Paleo asks a fair question, but many hunter-gatherers ate more carbs than it allows

Key takeaways

• A 2015 review of four small trials with 159 people found paleo diets trimmed waists by about 2.4 centimeters, roughly an inch, and lowered triglycerides more than guideline-based diets over six months or less¹⁰.
• In a two-year Swedish trial of 70 women, the paleo group lost more body fat at six months, but by two years the gap was no longer statistically significant⁹.
• Among Tanzania’s Hadza, hunter-gatherers whose diet researchers have tracked month by month, carbohydrates supply a median of about 61% of calories, from foods such as honey, berries, tubers and baobab⁴.
• Starch grains on 40,000-year-old Neanderthal teeth show they cooked and ate grains such as barley, along with legumes and dates³.
• Researchers reported that women on the paleo diet in that Swedish trial developed mild iodine deficiency, so cutting dairy and iodized salt calls for a plan¹¹.

Melvin Konner once lived among foragers in Botswana whose diet, by anthropologists’ estimates, was about 70% plant foods². In 1985, he and S. Boyd Eaton published a paper in the New England Journal of Medicine on what Stone Age people ate and what that might mean for modern disease¹.

That paper helped launch the paleo idea. Three decades later, in a 2016 Wall Street Journal essay titled “Confessions of a Paleo Diet Pioneer,” Konner wrote that the popular version of paleo, which treats carbohydrates as a condiment, didn’t persuade him².

That twist says a lot. Paleo asks a fair question: what did humans evolve eating? The best answers don’t match many paleo rulebooks.

This guide covers what anthropologists have learned about ancient diets, what paleo trials found, what paleo gets right and what its rules cost. A table near the end sorts the parts worth keeping.

Paleo began as a theory about mismatch

Eaton and Konner’s argument is often called the discordance, or mismatch, hypothesis. Human genes adapted to life as hunter-gatherers, modern life changed too fast for genes to keep up, and that gap helps drive chronic diseases such as heart disease and diabetes¹.

Their estimates shifted as data improved. In 1985 they put hunter-gatherer fat intake at about 20% of calories¹. By 2010 they had revised it to 20% to 35%, raised their protein estimate to 25% to 30%, and put fiber at 70 grams a day or more¹.

Their ideal food pyramid would rest on fruits and vegetables, with meat, fish and low-fat dairy on the next level and whole grains above that, since even whole grains were rare in forager diets¹. Notice the dairy: the founders’ own pyramid kept it. They called the hunter-gatherer model a provisional guide and said much more research was needed¹.

Popular paleo plans turned those ideas into stricter rules. A typical version, like the one tested in Swedish trials, centers on lean meat, fish, eggs, vegetables, root vegetables, fruit and nuts⁷. Grains are out because they came with farming⁵. The trademarked Paleo Diet also bans dairy, legumes such as beans and lentils, added sugar and most starchy tubers⁶.

That trademarked plan sets specific targets, too: 19% to 35% of calories from protein, 22% to 40% from carbohydrates and 28% to 47% from fat⁶. Cordain and colleagues had earlier estimated ancestral diets from an anthropological database of world cultures⁶.

Real Stone Age menus were all over the map

Here is where the record gets interesting. Across hunter-gatherer and other small-scale societies, meat supplies anywhere from about 5% to 90% of daily energy needs, according to a 2021 review by Duke University anthropologist Herman Pontzer and colleague Brian Wood⁴⁵. Warm-climate foragers tend to land near a 50-50 split of meat and plants, while cold climates and fishing push meat higher⁴.

The Hadza. Researchers have tracked the diet of the Hadza of northern Tanzania month by month. Over the long run, meat supplies about 38% of their energy needs⁴. Carbohydrates make up the biggest share, a median of about 61% of calories, with protein near 21% and fat near 18%. Across different observation periods, the carb share has run anywhere from 21% to 71%⁴.

The menu swings with the seasons, Pontzer told Knowable Magazine: February is honey season, July belongs to berries and September to meat. Honey alone can reach about 20% of the group’s calories⁵. Studies of Hadza adults find low cholesterol and almost no high blood pressure, even among older people⁴.

The Hadza are a present-day people, not a museum exhibit, and their health reflects movement as well as food. Hadza men walk about 14 kilometers, close to 9 miles, a day on average⁴. Yet their total daily calorie burn is similar to that of far less active people in industrial countries⁴.

Eleven tropical groups. In 2023, Harvard’s Daniel Lieberman and colleagues compared 11 tropical hunter-gatherer groups with the trademarked Paleo Diet’s ranges⁶. Few fit. The Hadza and several others ate much more carbohydrate, some groups ate much more fat, and a few ate less protein⁶. The authors concluded that no single mix of protein, fat and carbohydrate represents ancestral diets⁶.

In fairness to paleo fans, the same team found that the most careful studies, which weighed food eaten both in and out of camp, recorded more animal food⁶. Less careful studies may have undercounted meat.

The record has blind spots in both directions. Early observers focused on men’s hunting, and stone tools and butchered bones survive far better than plant remains, Pontzer notes⁵.

Neanderthals. Starch shows up even there. Amanda Henry, Alison Brooks and Dolores Piperno examined hardened plaque on 40,000-year-old Neanderthal teeth from caves in Iraq and Belgium³. They found starch grains from date palms, legumes and grains such as barley, and changes in the grains showed the food had been cooked³.

Evolution kept going, and grocery food isn’t wild food

The mismatch idea has a weak spot: it assumes human bodies stopped adapting long ago. Konner now says that assumption was wrong². In 1985, he wrote, scientists believed few genetic changes had happened in the roughly 10,000 years since people were all hunting and gathering. Researchers have since found examples such as Inuit genes for processing the fatty acids in Arctic fish².

A second weak spot: you can’t buy Stone Age food. Pontzer and Wood point out that farming changed the food itself. Crops were selected over generations for more calories, less fiber or both, and domesticated foods as a group carry less fiber and protein than wild ones⁴. So a paleo plate from the supermarket won’t match an ancestral one nutritionally⁴.

Still, the mismatch idea gets one big thing right. No forager ever ate factory-made food, Pontzer points out, and he counts it among the big forces behind obesity. The trouble, in his view, is twofold: manufacturers blend sugar, salt and fat in ways nature never does, and they strip out the fiber and protein that keep hunger in check⁵. Konner, too, now puts his emphasis on what not to eat, including refined carbohydrates and excess salt, and on getting more exercise².

For natural-health readers, that’s the core of paleo worth keeping: whole foods, home cooking, plenty of plants, a healthy suspicion of factory-made food and a body that moves every day. Our article on ultra-processed food weighs the broader evidence.

Small trials found real short-term gains

Researchers at Lund University in Sweden ran some of the first paleo trials. They were small, and they were randomized, meaning chance decided who ate which diet.

• Heart patients, 2007. Twenty-nine people with heart disease and either high blood sugar or type 2 diabetes ate a paleo diet or a Mediterranean-like diet with whole grains and low-fat dairy for 12 weeks⁷. Blood sugar after a glucose drink fell 26% on paleo and 7% on the comparison diet, and waists shrank 5.6 versus 2.9 centimeters⁷.
• Type 2 diabetes, 2009. Thirteen adults ate a paleo diet and a standard diabetes diet for three months each⁸. On paleo, HbA1c, a measure of average blood sugar, was 0.4 points lower, weight 3 kilograms (about 6.6 pounds) lower and diastolic blood pressure, the bottom number, 4 points lower⁸. They also ate fewer calories on paleo⁸.
• Women with obesity, two years. Seventy postmenopausal women ate as much as they wanted of either a paleo diet or a diet following Nordic nutrition guidelines⁹. At six months the paleo group had lost 6.5 kilograms of fat (about 14 pounds) versus 2.6 kilograms (about 6 pounds)⁹. At two years, the figures were 4.6 versus 2.9 kilograms, no longer a significant difference, though triglycerides stayed lower on paleo⁹.

In 2015, a review in the American Journal of Clinical Nutrition pooled four trials with 159 people¹⁰. Over six months or less, paleo diets trimmed waists by about 2.4 centimeters and lowered triglycerides, a blood fat, more than diets based on national guidelines¹⁰. Blood pressure dipped slightly, but not by enough to rule out chance, and HDL cholesterol and fasting blood sugar didn’t change clearly¹⁰. The reviewers rated the evidence moderate in quality and called for more research¹⁰.

Read those results with their limits in mind. The trials were small and short. They measured risk markers, not heart attacks or deaths. And in at least one trial, people ate fewer calories on paleo, which could explain some of the gains⁸.

One more detail from the two-year trial: the women didn’t eat as much protein as the paleo plan called for⁹. The high-protein target proved hard to keep.

What paleo’s rules can cost you

Every food you remove takes nutrients with it. Paleo’s exclusions are big ones.

Iodine. Swedish researchers measured iodine in the women from the two-year trial, in a study published in the European Journal of Clinical Nutrition and described by the trade outlet NutraIngredients¹¹. Iodine in the paleo group’s urine fell but held steady on the Nordic diet, and the authors concluded that long-term paleo eating was tied to mild iodine deficiency¹¹. They suggested iodine supplements be considered for people on paleo¹¹. Our article on nutritional adequacy lists iodine sources and explains how to check your intake, especially in pregnancy.

Calcium. Without dairy, calcium has to come from other foods. Our vitamins and minerals guide lists sources, and a clinician can tell you whether you need a supplement.

Starch and fiber. In a four-week Australian trial of 39 healthy women at Edith Cowan University, the paleo diet lowered intake of resistant starch, a kind of starch that resists digestion and feeds gut bacteria¹². Blood levels of TMAO, a compound the gut and liver make from animal protein that has been linked to heart disease, didn’t change significantly in that short trial¹².

In a separate study, the same team compared 44 long-term paleo followers with 47 people on a typical Australian diet¹³. Paleo followers may have had about twice as much TMAO, and eating few whole grains tracked with higher levels, The Atlanta Journal-Constitution reported¹³. That study compared people who chose their diets, so it can’t prove cause.

Cost. We couldn’t find an independent, recent U.S. study that priced a paleo diet. What’s clear is the trade: the rules remove beans, lentils, oats, rice and milk, and they shift spending toward meat, fish, eggs, nuts and fresh produce. Our guide to the government’s priced grocery plan shows where the cheaper staples fit.

Carnivore diets are a survey, not a test

Carnivore eating, nearly all meat and other animal foods, pushes paleo’s logic further. Its best-known study is a 2021 survey of 2,029 adults by Lennerz and colleagues¹⁴.

The researchers recruited people on social media in 2020 who had eaten a carnivore diet for at least six months¹⁴. Most did it for health, and 95% said their overall health had improved¹⁴. By their own reports, median body mass index fell from 27.2 to 24.3¹⁴. Among those who shared lab results, LDL cholesterol was markedly high, at 172 mg/dL¹⁴.

A survey like this can’t show that the diet caused the reported benefits. There was no comparison group, the results were self-reported, and anyone who felt worse and quit before six months couldn’t take part. The authors themselves wrote that long-term effects need more study¹⁴.

The ancestral record doesn’t support an all-meat pattern as typical, either. Meat’s share ranged widely, and most groups studied ate plenty of plants⁴⁶.

Keep the good parts: a paleo rule-by-rule guide

Paleo rule	Evidence rating	What to keep	What to reconsider
Eat whole, home-cooked foods	Promising but early	Meals built on meat, fish, eggs, vegetables, fruit and nuts; four small trials showed short-term gains10	Long-term results are unknown910
Cut added sugar and processed snacks	Promising but early	Pontzer sees processed food as a big driver of obesity5; the trademarked plan bans added sugar6	Packaged “paleo” snacks are still processed food
No grains	Not enough evidence	Cutting refined flour and sweets	Neanderthals cooked grains3; the long-term follower study tied missing whole grains to higher TMAO13
No legumes	Not enough evidence	Nothing in particular	Legumes show up in Neanderthal plaque3 and in other Paleolithic evidence6
No dairy	Not enough evidence	Skipping it if you’re allergic or it upsets your stomach	Iodine fell in the two-year trial11; plan calcium
More meat and protein	Not enough evidence	Fish, eggs and poultry in reasonable amounts	Carnivore eaters reported high LDL14; TMAO findings are mixed1213
Eat low-carb, as ancestors did	Evidence says no	Nothing, as a claim about history	Hadza median about 61% carbohydrate4; few forager groups fit paleo ranges6
Skip iodized salt along with dairy	Evidence says no, or unsafe	Unprocessed food that’s lower in salt	Mild iodine deficiency in a two-year trial11

Questions readers ask

The bottom line

Paleo got something important right: modern processed food is a poor match for the human body, and whole foods cooked at home are a better one. Short trials back that up with smaller waists and lower triglycerides.

Where paleo goes wrong is in the rules. Hunter-gatherers ate many different diets, often rich in carbohydrates, and evolution didn’t stop when farming began. Keep the whole foods and the daily movement, think twice before dropping grains, beans and dairy, and talk with your clinician first if you take medicines or are pregnant.

Related reading

• The case against ultra-processed food keeps growing, though defining it is harder
• Half of adults overrated a 12-grain bread, so learn to spot real whole grains
• Missing the RDA doesn’t mean you’re deficient, and you can check your diet
• Some healthy habits clash with your medicines, so check these goals first
• Cooked beans are cheap and well studied, and the lectin scare applies to raw ones

Sources

1. Konner M, Eaton SB. “Paleolithic Nutrition: Twenty-Five Years Later.” Nutrition in Clinical Practice, 2010;25(6):594–602 (revisits Eaton SB, Konner M, New England Journal of Medicine, 1985;312:283–289). melvinkonner.com
2. Konner M. “Confessions of a Paleo Diet Pioneer.” The Wall Street Journal, January 20, 2016 (author’s archive copy). melvinkonner.com
3. Smithsonian Science. “Starch grains found on Neandertal teeth debunks theory that dietary deficiencies caused their extinction” (on Henry AG, Brooks AS, Piperno DR, Proceedings of the National Academy of Sciences, 2011). January 2011. smithsonianscience.org
4. Pontzer H, Wood BM. “Effects of Evolution, Ecology, and Economy on Human Diet: Insights from Hunter-Gatherers and Other Small-Scale Societies.” Annual Review of Nutrition, 2021;41:363–385. pubmed.ncbi.nlm.nih.gov
5. Kwon D. “Our ancestors ate a Paleo diet. It had carbs.” Knowable Magazine (Annual Reviews), September 28, 2022. doi.org
6. Lieberman DE, Worthington S, Schell LD, Parkent CM, Devinsky O, Carmody RN. “Comparing Measured Dietary Variation Within and Between Tropical Hunter-Gatherer Groups to the Paleo Diet.” American Journal of Clinical Nutrition, 2023;118(3):549–560. pubmed.ncbi.nlm.nih.gov
7. Lindeberg S, Jönsson T, Granfeldt Y, et al. “A Palaeolithic diet improves glucose tolerance more than a Mediterranean-like diet in individuals with ischaemic heart disease.” Diabetologia, 2007;50(9):1795–1807 (Lund University record). lup.lub.lu.se
8. Jönsson T, Granfeldt Y, Ahrén B, et al. “Beneficial effects of a Paleolithic diet on cardiovascular risk factors in type 2 diabetes: a randomized cross-over pilot study.” Cardiovascular Diabetology, 2009;8:35 (Lund University record). portal.research.lu.se
9. Mellberg C, Sandberg S, Ryberg M, et al. “Long-term effects of a Palaeolithic-type diet in obese postmenopausal women: a 2-year randomized trial.” European Journal of Clinical Nutrition, 2014. link.springer.com
10. Manheimer EW, van Zuuren EJ, Fedorowicz Z, Pijl H. “Paleolithic nutrition for metabolic syndrome: systematic review and meta-analysis.” American Journal of Clinical Nutrition, 2015;102(4):922–932 (abstract record). nutrition-evidence.com
11. NutraIngredients. “Should women on a paleo diet be taking iodine supplements?” September 28, 2017 (on Manousou S, et al., “A Paleolithic-type diet results in iodine deficiency: a 2-year randomized trial in postmenopausal obese women,” European Journal of Clinical Nutrition). nutraingredients.com
12. Genoni A, Lo J, Lyons-Wall P, et al. “A Paleolithic diet lowers resistant starch intake but does not affect serum trimethylamine-N-oxide concentrations in healthy women.” British Journal of Nutrition, 2019;121(3):322–329 (Edith Cowan University repository). ro.ecu.edu.au
13. The Atlanta Journal-Constitution. “Paleo ‘caveman’ diet linked to heart disease biomarker, study finds” (on Genoni A, et al., European Journal of Nutrition, 2019).
14. Lennerz B, Mey J, Henn O, Ludwig D. “Behavioral Characteristics and Self-Reported Health Status among 2029 Adults Consuming a ‘Carnivore Diet.'” 2021 (LSU Pennington Biomedical Research Center repository). repository.lsu.edu