The Church of Kale

There is no such thing as a clean food. There is no dirty one, either. Nutrition science does not recognize either category, and no clinical textbook uses the phrase. Yet at some point in the last two decades, millions of people began treating their plates as though a cookie could leave a residue on the conscience, as though a green smoothie could scrub something away. The language of hygiene crept into the language of eating, and almost nobody noticed the substitution.

Ask ten people what clean eating means and you will get ten answers. To one, it means no refined sugar. To another, no gluten, no dairy, nothing that comes in a box. To a third, it means organic, local, unprocessed, whatever that word turns out to mean. The vagueness is not a bug. It is the entire point. A phrase that can mean anything can be sold to everyone, and a rule with no fixed edges can never quite be satisfied. That is a useful property if you are building a movement. It is a dangerous one if you are trying to feed yourself.

The striking thing about clean eating is how much moral weight it carries for a term with so little scientific content. Eating well should feel like maintenance. Instead, for a great many people, it has come to feel like penance. Understanding why requires going back further than the green smoothie, further than the fitness magazine that first printed the phrase. It requires going back to a man who believed that bland food could save your soul.

The sanitarium that started it

Before kale, there were cornflakes. And before cornflakes, there was John Harvey Kellogg, a physician and Seventh-day Adventist who ran the Battle Creek Sanitarium in Michigan from the 1870s onward. Kellogg was not a quack in the modern sense. He was a trained surgeon, a prolific inventor, and for a time one of the most famous health authorities in America. He was also a man who believed, with total conviction, that digestion and salvation were the same project.¹

At Battle Creek, patients ate a diet designed to be as inoffensive to the body as possible: no meat, no spices, no coffee, no alcohol, nothing that might stimulate the appetite or, worse, the passions. Kellogg regarded pleasure at the table as a kind of leakage, an energy drain that pulled the body toward corruption. Bland food was not a compromise. It was the goal. The cornflake itself was invented partly in service of this philosophy, a food so plain it could not possibly inflame anything.²

What matters here is not the specific menu but the underlying logic. Kellogg fused physiology with morality until the two were indistinguishable. To eat correctly was to be good. To crave the wrong things was to be weak, unclean, in need of correction. He wrapped this in the confident language of science, which gave it authority, but the engine underneath was religious. Purity of the gut stood in for purity of the person.

That idea did not die with Kellogg. It went dormant. A century later it woke up wearing activewear.

From fitness magazines to hashtags

The modern phrase surfaced in American fitness and bodybuilding culture in the early 2000s. In that context it had a fairly narrow and defensible meaning: eat whole foods, fresh vegetables, lean protein, complex carbohydrates, things close to their natural state, and go easy on the packaged, the sugary, the fried. For an athlete tracking macronutrients, clean eating was a shorthand, not a moral system. It described a shopping list.

Then the internet found it. Around 2010, food blogging and photo-sharing platforms turned a dietary preference into an aesthetic and, eventually, an identity. The vague idea acquired a visual grammar: mason jars, overnight oats, avocado on sourdough, kale massaged with lemon, everything shot in cool morning light. By 2014, the hashtag #eatclean had gathered tens of millions of posts, and clean eating had transformed from a bodybuilder’s rule of thumb into a lifestyle you could perform.³

This is where the trouble began. A shopping list cannot make you virtuous, but a lifestyle can promise to. As clean eating migrated onto social platforms, it absorbed the moral charge that Kellogg had built into it a century earlier, and the platforms amplified that charge because virtue photographs beautifully. A bowl of quinoa and pomegranate seeds is not just food in that frame. It is evidence. It says something about the person who arranged it.

The word doing the heaviest lifting in all of this was processed, which quietly became the villain of the story. Anything processed was suspect. Anything natural was blessed. It sounded like a clear line. It was not.

What processing actually means

Processing is not a modern industrial sin. It is one of the oldest things humans do to food. Cooking is processing. So is freezing, drying, canning, fermenting, milling, and chopping. When you freeze a bag of peas, you are processing them, and doing so preserves nutrients that would otherwise degrade. When you dice garlic on your own cutting board, that is processing too. Bread is processed. Cheese is processed. Yogurt, olive oil, and coffee are all processed. Almost nothing a human eats arrives entirely untouched, and much of what we prize as wholesome is the product of deliberate transformation.

So the clean-versus-processed framing collapses the moment you examine it. The real question was never whether food had been changed by human hands. Nearly all of it has. The useful question is how much it has been changed, and toward what end. That distinction required a better map than the moral one the wellness world was using, and in 2009 a Brazilian researcher provided one.

Carlos Monteiro, an epidemiologist and nutrition researcher at the University of Sao Paulo, proposed a classification system called NOVA. Instead of sorting foods by their nutrient content or their supposed purity, NOVA sorts them by the nature and extent of the industrial processing they undergo. At one end sit unprocessed and minimally processed foods: fruits, vegetables, eggs, milk, grains, meat. In the middle sit processed culinary ingredients like oil, butter, and sugar, and then processed foods such as canned vegetables, cheese, and fresh bread. At the far end sits the category that actually earns concern: ultra-processed foods.⁴

Ultra-processed foods are not simply cooked or preserved versions of whole ingredients. They are industrial formulations built largely from substances extracted or synthesized from foods: refined starches, protein isolates, hydrogenated oils, along with additives like emulsifiers, sweeteners, colors, and flavor compounds you would never find in a home kitchen. Think of packaged snacks engineered to be irresistible, sweetened breakfast cereals, reconstituted meat products, sodas, and most of what fills the center aisles of a supermarket. These are the products designed less to be food than to be consumed, in volume, with minimal friction.

This was the signal buried inside the clean-eating noise. There really is something meaningful about the far end of the processing spectrum. But you needed evidence to prove it, not vibes. And in 2019, someone finally ran the experiment properly.

The experiment that found a real effect

Kevin Hall is a physicist turned metabolism researcher at the United States National Institutes of Health, and he is known for a rigor that most nutrition science lacks. Most dietary studies rely on people reporting what they ate, which is notoriously unreliable, or on comparing populations whose lifestyles differ in a thousand uncontrolled ways. Hall wanted to isolate a single variable, and to do that he brought his subjects into a metabolic ward and controlled almost everything.⁵

In a 2019 study published in Cell Metabolism, twenty adults lived in Hall’s research facility for a month. For two weeks they ate a diet built from minimally processed whole foods. For the other two weeks they ate a diet built from ultra-processed foods. Crucially, the two diets were matched, meal for meal, for calories offered, sugar, fat, sodium, and fiber. On paper, in terms of the nutrients that clean-eating discourse obsesses over, the two menus were nearly identical. Participants were told to eat as much or as little as they wanted at each meal.⁶

The result was stark. On the ultra-processed diet, people ate about 500 additional calories per day, and they gained weight. On the whole-food diet, they ate less and lost weight. Same nutrients on the label. Wildly different effects on the body. The food itself, its texture, its density, the speed at which it could be eaten, was driving people to overeat, independent of the macronutrients it contained.⁶

This is an important finding, and it deserves to be stated carefully. It suggests that ultra-processing is not merely a proxy for eating too much sugar or fat. Something about the physical and industrial nature of these foods, perhaps how quickly they can be consumed, perhaps how they interact with the body’s fullness signals, changes eating behavior itself. The processing category, defined properly by NOVA rather than vaguely by wellness bloggers, appears to matter.

So there was a kernel of truth inside the movement. Eating mostly whole foods and going lighter on ultra-processed products is a defensible, evidence-supported idea. But notice what that evidence does not say. It does not divide food into clean and dirty. It does not attach virtue to a carrot or sin to a cracker. It describes a spectrum of industrial transformation and its measurable effects on appetite. The science pointed at something real. Clean was still the wrong word for it, and the wrong word did real damage.

When the pursuit of pure turns into illness

In 1997, an American physician named Steven Bratman noticed a pattern among his patients that did not fit any existing diagnosis. These were people obsessed not with eating too much or too little, but with eating correctly. Their concern with the quality and purity of their food had grown so consuming that it was making them unwell. Their diets shrank steadily as more and more foods failed to clear the bar of purity, until, in some cases, almost nothing was clean enough to eat.⁷

Bratman gave the condition a name: orthorexia nervosa, from the Greek orthos, meaning correct or right. Unlike anorexia, which centers on quantity and body weight, orthorexia centers on quality and moral righteousness. The sufferer is not trying to be thin so much as trying to be pure. It is not yet a formally recognized diagnosis in the major psychiatric manuals, and researchers still debate how to define and measure it. But clinicians report seeing it more often, and it maps with uncomfortable precision onto the culture that clean eating built.⁷

Estimates of how common orthorexic symptoms are vary widely, in part because the measurement tools are still being refined, and some early studies used instruments that likely overcounted. But research consistently finds elevated rates among people heavily invested in wellness, fitness, and dietary self-monitoring, with some studies of such groups reporting symptoms in a substantial minority. Broadly, the more intensely a person pursues dietary purity, the higher the risk that the pursuit tips into anxiety, guilt, and restriction.⁸

The mechanism is not mysterious once you see the moral framing at its center. If food can be clean or dirty, then eating becomes a series of tests you can pass or fail. A meal that falls on the wrong side of the line is not just less healthy. It is a small transgression, a stain, a failure of will. The glowing wellness aesthetic that made clean eating so shareable also made it a perfect engine for shame. And shame, unlike a nutrient, compounds.

Who gets to eat clean

Here is the part the sunlit kitchens rarely acknowledge. Clean eating was never only about health. It was also about virtue, control, and who could afford to signal them.

Fresh organic produce costs more than its conventional equivalent, which costs more than the ultra-processed products engineered to be cheap and shelf-stable. Time to shop for whole ingredients, prepare them, and cook from scratch is itself a resource, one distributed as unequally as money. The person working two jobs and feeding a family on a thin budget is not choosing dirty food out of moral weakness. They are navigating constraints that the language of purity conveniently ignores.

This is where the clean framing becomes not just unscientific but quietly cruel. To call some foods clean is to imply that other foods, and by extension the people who eat them, are dirty. It converts a question of access and economics into a question of character. It lets the person with the mason jars and the morning light feel not merely healthier but better, in the moral sense, than the person eating what they can afford. Kellogg would have recognized the move exactly. Purity of the plate standing in for purity of the person, with the added modern feature that purity is now for sale.

Your body, for what it is worth, does not participate in any of this. It does not read food as moral. It reads food as chemistry: as energy, as building blocks, as signals to a nervous system that evolved long before anyone thought to photograph a smoothie. A carrot is not virtuous. A cracker is not a sin. Both are collections of molecules that your digestive system will process according to laws that take no interest in your Instagram feed.

Nourishment without fear

Strip away the moral vocabulary and a modest, well-supported picture remains. The diets that hold up best in research are not built around a magic ingredient or a forbidden list. They share, more than anything, one unglamorous quality: variety. Mostly whole and minimally processed foods, plenty of plants, a range of nutrients, and, yes, room for the processed and the pleasurable, because a diet you cannot enjoy is a diet you will not keep. The evidence from Kevin Hall’s ward supports going lighter on ultra-processed products. It does not support treating a single cracker as a moral catastrophe.⁶

The genuine insight of the last two decades of nutrition research is quieter than any hashtag. It is not that some foods cleanse and others contaminate. It is that industrial ultra-processing appears to alter how much we eat, and that eating a varied diet of largely recognizable foods is a sensible response. That finding does not require virtue. It does not require shame. It does not require a mason jar.

The great trick of clean eating was to take a reasonable idea, eat more whole foods, and wrap it in a moral language borrowed from a Victorian sanitarium, then sell that language back to us as science. The idea was never the problem. The morality was. The goal of eating well was never purity. It was nourishment, and if possible, nourishment without fear.

So the next time a meal makes you feel like you have failed at something, it is worth remembering where that feeling came from, and how little it has to do with your body. Food has no morals. Only the people looking at it do.

Sources

1. Wilson, Brian C., Dr. John Harvey Kellogg and the Religion of Biologic Living, Indiana University Press, 2014. — https://iupress.org/9780253014474/dr-john-harvey-kellogg-and-the-religion-of-biologic-living/
2. Markel, Howard, The Kelloggs: The Battling Brothers of Battle Creek, Pantheon Books, 2017. — https://www.penguinrandomhouse.com/books/223424/the-kelloggs-by-howard-markel/
3. Nevin, Suzanne M. and Vartanian, Lenny R., The stigma of clean dieting and orthorexia nervosa, Journal of Eating Disorders, 2017. — https://jeatdisord.biomedcentral.com/articles/10.1186/s40337-017-0168-9
4. Monteiro, Carlos A. et al., Ultra-processed foods: what they are and how to identify them, Public Health Nutrition, 2019. — https://www.cambridge.org/core/journals/public-health-nutrition/article/ultraprocessed-foods-what-they-are-and-how-to-identify-them/E6D744D714B1FF09D5BCA3E74D53A185
5. National Institutes of Health, NIH study: Ultra-processed diets cause excess calorie intake and weight gain, NIH News Releases, 2019. — https://www.nih.gov/news-events/news-releases/nih-study-finds-heavily-processed-foods-cause-overeating-weight-gain
6. Hall, Kevin D. et al., Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain, Cell Metabolism, 2019. — https://www.cell.com/cell-metabolism/fulltext/S1550-4131(19)30248-7
7. Bratman, Steven and Knight, David, Health Food Junkies: Orthorexia Nervosa, Broadway Books, 2000. — https://www.orthorexia.com/
8. Dunn, Thomas M. and Bratman, Steven, On orthorexia nervosa: A review of the literature and proposed diagnostic criteria, Eating Behaviors, 2016. — https://www.sciencedirect.com/science/article/abs/pii/S1471015315300362

Related reading

• The Quiet Sweetening of the Western Diet
• The Clock Diet and Its Discontents

More from the Plate edition →