How Our Ancestors Ate Calories

Nearly nothing had low nutrient density. You can put yourself in the ancestral environment by controlling what you allow on your plate and in your mouth.

A Calorie Is a Calorie: A Useful Fiction

America has achieved a strange biological success. We made calories cheap, portable, shelf-stable, intensely flavored, and available everywhere. Then we built public nutrition around the one number that made that achievement look neutral.

The result should disturb anyone who still thinks the problem is simple arithmetic. By NHANES-based estimates, more than nine in ten Americans fall below the Estimated Average Requirement for vitamin D from foods and beverages. About half fall below the EAR for magnesium. Potassium intake remains below recommended levels across the population, and fiber intake remains so low that one NHANES 2013–2018 analysis found only 7.4% of U.S. adults met the fiber target of 14 grams per 1,000 kilocalories. These are intake estimates, not blood-level diagnoses. The caveat matters. But the direction does not disappear when the caveat is stated.

The same population lives in one of the most calorie-abundant food environments ever constructed.

Hold those two facts together. The standard story begins to crack. This is not a morality play about weak consumers failing to obey correct advice. It is not even, at root, a story about Americans failing to count accurately. It is a story about what the counting system was built to see — and what it was never built to see.

Read on to see how an energy unit became the blind spot at the center of American nutrition affects our perception of “food”.

The calorie was not wrong. It was too narrow.

The calorie is an energy unit. That is its virtue and its limitation.

Historically, it entered nutrition through the attempt to quantify food as potential energy. W.O. Atwater introduced the Calorie to American audiences in 1887 as a practical unit for food energy, and the modern food-label Calorie is equivalent to the kilocalorie: the energy needed to raise one kilogram of water by one degree Celsius. That history matters because the popular critique often becomes sloppy. Atwater was not stupid. The system he helped build was not simply “burn food and pretend humans are furnaces.” Atwater’s work tried to estimate metabolizable energy by accounting for losses.

That is precisely why the deeper error is harder to see.

The calorie does a narrow job. It estimates energy. It does not measure nourishment. It does not measure satiety. It does not measure absorption rate. It does not measure food structure. It does not measure whether the energy arrives with potassium, magnesium, vitamin E, essential fatty acids, intact plant cell walls, water, fiber, polyphenols, or any of the other non-caloric features that make food biologically different from fuel.

“A calorie is a calorie” is true in the limited thermodynamic sense. The first law of thermodynamics has not been repealed. But the slogan smuggles in a second claim: that two foods with the same energy value are nutritionally and metabolically interchangeable. That claim is not physics. It is a category error.

Once you see the calorie as an energy instrument rather than a nourishment instrument, the modern paradox stops being paradoxical. Food has at least two independent properties. Energy density is calories per gram. Nutrient density is nourishment per calorie. Industrial food can maximize the first while stripping the second.

Oil is almost pure energy. Spinach is not. A snack made from refined starch, refined oil, sugar, salt, flavor chemistry, and texture engineering can be extravagant on the first axis and nearly barren on the second. Count only calories and the food becomes a number. Read both axes and it becomes visible again.

The ancestral issue is not nostalgia. It is calibration.

The point is not to romanticize a single Paleolithic menu. There was no single ancestral diet. Human beings ate differently across latitude, season, ecology, and culture. Eaton and Konner made that clear in 1985, and Cordain and colleagues later emphasized that the modern Western diet differs from ancestral dietary patterns across multiple features at once: glycemic load, fatty acid composition, micronutrient density, sodium-to-potassium ratio, and fiber content.

That is the relevant point. Not “eat like a caveman.” The relevant point is that the biological systems regulating appetite, digestion, glucose handling, mineral balance, and satiety evolved in food environments where energy usually arrived inside structure.

Our ancestors packed fat away during times of plenty in biologic anticipation of future periods of food austerity. They also at whole foods.

Plants came with cell walls, water, fiber, minerals, and phytochemicals. Animal foods came with protein, fats, minerals, and tissue structure. Food required chewing. It occupied volume. It carried sensory signals that bore some relationship to its biological content. The package was not perfect. Evolution does not engineer perfection. But energy and nourishment were far less separable than they are now.

Industrial refining broke that relationship.

Refine grain and you keep the fast starch while discarding much of the bran, germ, fiber, minerals, B vitamins, and lipids. Press oil and you concentrate energy while leaving behind most of the matrix. Isolate sugar and you extract sweetness from biological context. Reassemble starch, oil, sugar, salt, emulsifiers, flavors, colors, and stabilizers, and the result is not merely “food with calories.” It is a manufactured delivery system for energy stripped from many of the constraints that once governed intake.

This is why “overfed and undernourished” is not a contradiction. It is the predictable result of an inverted food supply meeting a measurement system that notices energy first and structure last.

The cleanest test broke the slogan - nearly.

The standard story says weight is an arithmetic problem: calories in, calories out. It is not false. It is incomplete. The human question is not whether energy balance matters. It is what kind of food environment drives intake before arithmetic begins.

Kevin Hall’s 2019 inpatient randomized crossover trial at NIH remains the cleanest demonstration. Twenty weight-stable adults lived in a metabolic ward. Each participant received an ultra-processed diet and an unprocessed diet for two weeks each, in randomized order. The diets were matched for presented calories, energy density, macronutrients, sugar, sodium, and fiber. Participants ate ad libitum. On the ultra-processed diet, they consumed about 508 additional kilocalories per day and gained weight. On the unprocessed diet, they lost weight.

That result does not abolish thermodynamics. It identifies where the calorie slogan fails.

The instrument said the diets were matched in the variables nutrition usually foregrounds. The organism did not behave as though they were equivalent. Form mattered. Processing mattered. The matrix mattered. Eating rate, palatability, texture, protein dilution, and digestive kinetics all remained live mechanisms.

This is where the press release version usually fails in the opposite direction. The Hall trial was small, short, and single-center. It established a powerful internal-validity result; it did not isolate one universal mechanism. The trial cannot tell us whether the decisive driver was eating rate, food texture, protein dilution, reward architecture, altered satiety signaling, or some interaction among them. Ultra-processed foods tend to bundle those features together because that is what industrial formulation does.

The honest conclusion is therefore stronger than a simple slogan. It is not “processing magically causes obesity.” It is this: when diets matched on conventional nutrient variables still produce radically different spontaneous energy intake, the conventional variables are insufficient.

These results do not mean that you will not gain weight if you eat too many calories of nutrient-rich foods. Calories still matter. But one signal tells the body to stop eating - and it’s not calories.

Protein leverage is plausible. It is not a verdict.

One live mechanism deserves special attention because it fits the pattern without reducing the whole problem to sugar, fat, or willpower.

Simpson and Raubenheimer’s protein-leverage hypothesis proposes that organisms defend an absolute protein target. When dietary protein is diluted by non-protein energy, the organism may overconsume fat and carbohydrate to reach that protein target. The model has experimental and epidemiological support, and Martínez Steele and colleagues found U.S. intake patterns consistent with the idea that ultra-processed foods dilute dietary protein while total energy intake rises.

If the implications of their findings are real, it would mean that our bodies gauge our meal based on the amount of protein they have consumed, regardless of what comes with the protein.

That does not make protein leverage the master key. It makes it one serious mechanism among several.

This distinction matters. Nutrition repeatedly destroys itself by replacing one monocausal story with another. Fat was the villain. Then sugar. Then carbs. Then seed oils. Then ultra-processing. Each story captures part of the terrain and then becomes propaganda when it pretends to be the map.

The better view is structural. Ultra-processed foods can combine protein dilution, low fiber, rapid eating rate, high reward value, soft texture, high energy availability, long shelf life, low cost, aggressive marketing, and displacement of intact foods. The organism experiences the bundle. The label reports fragments.

So ask yourself: How many of the calories on my plate beyond the protein are nutrient-poor and not worth the extra pound later?

You cannot fortify your way back to a matrix.

If the diagnosis is stripped nourishment, the institutional fix seems obvious: add the missing nutrients back. That is the logic of enrichment and fortification.

It is not useless. Iodized salt prevents goiter. Folic-acid fortification reduced neural-tube defects. Vitamin D fortification can matter in populations with low sun exposure. The problem is not fortification as a targeted public-health tool. The problem is treating fortification as proof that disassembly has been repaired.

Whole food is not a pile of chemicals. It is a structured biological object.

Fiber, water, intact cell walls, particle size, viscosity, and food texture alter gastric emptying, nutrient bioaccessibility, glycemic response, satiety signaling, and what reaches the microbiome. A review of soluble dietary fibers notes that food composition and structure affect digestion and nutrient absorption, and that plant cell-wall properties influence nutrient bioaccessibility, gastric emptying, transit time, and the extent of digestion and absorption.

That is the part a calorie cannot see.

A whole apple, apple juice, and an apple-flavored fortified bar are not three interchangeable carriers of apple-associated nutrients. They are different biological inputs. They differ in structure, chewing requirement, fiber architecture, absorption kinetics, and satiety. They may share label words. They do not share physiological meaning.

This is also where NOVA became useful, even with its boundaries debated. Monteiro and colleagues define ultra-processed foods not merely by “processing,” since nearly all food is processed in some sense, but by industrial formulations built from fractionated substances, cosmetic additives, and techniques designed to produce convenient, durable, highly palatable products that displace minimally processed foods.

Louzada and colleagues showed the population-level version in Brazil: as the ultra-processed share of the diet rose, the micronutrient profile of the overall diet worsened. That is displacement in nutritional form.

You cannot sprinkle a destroyed matrix back onto food.

The official language has begun to move. The measurement system has not.

The latest U.S. dietary-guidelines language now leans toward whole, nutrient-dense foods and reduced highly processed foods. That shift matters. It concedes that food form and processing cannot be treated as irrelevant decoration around macronutrients and calories.

But the concession does not solve the problem.

The label still begins by foregrounding calories. The marketplace still rewards shelf stability, texture engineering, flavor amplification, and low-cost energy. Nutrition software still trains users to experience meals as arithmetic. Public advice still struggles to distinguish a food from a formulation when both can be made to display similar macro panels.

A calorie-centered label misleads by construction when the food supply has been engineered to maximize energy delivery while minimizing structure. It reports the property industrial food can most easily provide and leaves the organism to discover the missing ones later.

This does not mean calories are irrelevant. It means calories are subordinate. They answer one question: how much energy is present or available under a model. They do not answer the prior question: what kind of biological object delivered that energy?

What follows

Three conclusions follow.

First, calorie literacy is not nutrition literacy. A population can learn to count calories and still fail to understand food.

Second, enrichment is not reconstruction. Fortification can correct selected deficiencies, but it cannot restore fiber architecture, intact matrices, eating rate, or the ecological relationship between foods and the microbiome.

Third, the next nutrition instrument cannot simply be another single number. A “nutrient density score” may help, but if it collapses structure, processing, absorption kinetics, satiety, and displacement into another label metric, it will reproduce the same reductionism under new branding.

The right instrument would ask different questions. How much energy per gram? How much protein per unit energy? How much fiber and water? Which micronutrients arrive per calorie? What was removed? What was added? How intact is the matrix? How rapidly can the food be eaten? How much of the original biological structure remains? Does this food displace a whole food or preserve one?

That may not fit neatly on the front of a package. That is the point. The body is not reading a package. It is reading food.

The calorie is not useless. It is useful in the way a ruler is useful. But no one measures temperature with a ruler and then blames the fever on arithmetic. We built a nutrition system around an energy unit, then acted surprised when it could not see nourishment.

The next question is not whether a calorie is a calorie. It is whether we are ready to stop pretending that heat is the same thing as food.

Sign up for Evolutionary Biology of Human Health - 3 lectures in, 13 to go!

Citations

Hall KD, Ayuketah A, Brychta R, et al. Ultra-processed diets cause excess calorie intake and weight gain: an inpatient randomized controlled trial of ad libitum food intake. Cell Metabolism. 2019;30(1):67-77.e3. DOI: 10.1016/j.cmet.2019.05.008. PMID: 31105044. Note: errata have been published and should be checked before final bibliography lock.

Eaton SB, Konner M. Paleolithic nutrition: a consideration of its nature and current implications. New England Journal of Medicine. 1985;312(5):283-289. DOI: 10.1056/NEJM198501313120505. PMID: 2981409. Note: NEJM lists a correction for this article.

Cordain L, Eaton SB, Sebastian A, et al. Origins and evolution of the Western diet: health implications for the 21st century. American Journal of Clinical Nutrition. 2005;81(2):341-354. DOI: 10.1093/ajcn.81.2.341. PMID: 15699220.

Simpson SJ, Raubenheimer D. Obesity: the protein leverage hypothesis. Obesity Reviews. 2005;6(2):133-142. DOI: 10.1111/j.1467-789X.2005.00178.x. PMID: 15836464.

Martínez Steele E, Raubenheimer D, Simpson SJ, Baraldi LG, Monteiro CA. Ultra-processed foods, protein leverage and energy intake in the USA. Public Health Nutrition. 2018;21(1):114-124. DOI: 10.1017/S1368980017001574. PMID: 29032787.

Monteiro CA, Cannon G, Levy RB, et al. Ultra-processed foods: what they are and how to identify them. Public Health Nutrition. 2019;22(5):936-941. DOI: 10.1017/S1368980018003762. PMID: 30744710.

Louzada MLC, Martins APB, Canella DS, et al. Impact of ultra-processed foods on micronutrient content in the Brazilian diet. Revista de Saúde Pública. 2015;49:45. DOI: 10.1590/S0034-8910.2015049006211. PMID: 26270019.

Giuntini EB, Sardá FAH, de Menezes EW. The effects of soluble dietary fibers on glycemic response: an overview and futures perspectives. Foods. 2022;11(23):3934. DOI: 10.3390/foods11233934. PMID: 36496742.

Hargrove JL. History of the calorie in nutrition. Journal of Nutrition. 2006;136(12):2957-2961. DOI: 10.1093/jn/136.12.2957. PMID: 17116702.

Hargrove JL. Does the history of food energy units suggest a solution to “Calorie confusion”? Nutrition Journal. 2007;6:44. DOI: 10.1186/1475-2891-6-44.

Miketinas DC, Tucker WJ, Patterson M, Douglas C. Usual dietary fiber intake in US adults with diabetes: NHANES 2013-2018. Current Developments in Nutrition. 2021;5(Suppl 2):1061. DOI: 10.1093/cdn/nzab053_054.

National Institutes of Health, Office of Dietary Supplements. Vitamin D Fact Sheet for Health Professionals. Accessed June 17, 2026.

National Institutes of Health, Office of Dietary Supplements. Magnesium Fact Sheet for Health Professionals. Accessed June 17, 2026.

National Institutes of Health, Office of Dietary Supplements. Potassium Fact Sheet for Health Professionals. Accessed June 17, 2026.

U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary Guidelines for Americans, 2025-2030. Accessed June 17, 2026.