Is Fat Killing You, or Is Sugar?

What we do and don’t know about dietary science.
Nutritional science is too complex to furnish easy answers about what to eat.Illustration by Ben Wiseman

In the early nineteen-sixties, when cholesterol was declared an enemy of health, my parents quickly enlisted in the war on fat. Onion rolls slathered with butter, herring in thick cream sauce, brisket of beef with a side of stuffed derma, and other staples of our family cuisine disappeared from our table. Margarine dethroned butter, vinegar replaced cream sauce, poached fish substituted for brisket. I recall experiencing something like withdrawal, daydreaming about past feasts as my stomach grumbled. My father’s blood-cholesterol level—not to mention that of his siblings and friends—became a regular topic of conversation at the dinner table. Yet, despite the restrictive diet, his number scarcely budged, and a few years later, in his mid-fifties, he had a heart attack and died.

The dangers of fat haunted me after his death. When, in my forties, my cholesterol level rose to 242—200 is considered the upper limit of what’s healthy—I embarked on a regimen that restricted fatty foods (and also cut down on carbohydrates). Six months later, having shed ten pounds, I rechecked my level. It was unchanged; genes have a way of signalling their power. But as soon as my doctor put me on just a tiny dose of a statin medication my cholesterol plummeted more than eighty points.

In recent decades, fat has been making a comeback. Researchers have questioned whether dietary fat is necessarily dangerous, and have shown that not all fats are created equal. People now look for ways of boosting the “good cholesterol” in their blood and extol the benefits of Mediterranean diets, with their emphasis on olive oil and fatty nuts. In some quarters, blame for obesity and heart disease has shifted from fat to carbohydrates. The Atkins diet and, more recently, the paleo diet have popularized the idea that you can get slim eating high-protein, high-cholesterol foods.

Still, I remained wary of the delicacies of my childhood. Surely it was wiser simply to avoid fats altogether? I wavered, though, in 2013, when The New England Journal of Medicine published an article endorsing the salubrious effects of Mediterranean eating habits. The article detailed the results of a study, the most rigorously scientific one yet conducted on the issue, which showed that following a Mediterranean diet rich in either olive oil or nuts could reduce the risk of heart attack, stroke, or death from cardiovascular causes by thirty per cent. I was elated until my wife, an endocrinologist who is an expert on metabolism, pointed out that the headline number of thirty per cent emerged from the complex statistical way that the study’s results were projected over time. If you looked at what happened to the people in the study, the picture was less encouraging: 3.8 per cent of the people consuming olive oil and 3.4 per cent of the people eating nuts suffered cardiovascular misfortune, compared with 4.4 per cent of the group on a regular diet. The true difference in outcome between the two diets was, at best, one per cent.

It’s one of many cautionary tales about assessing dietary data. Everyone wants to be healthy, and most of us like eating, so we’re easily swayed by any new finding, no matter how dubious. Publishers know this all too well and continually ply us with diet and health books of varying degrees of respectability and uplift. The most prominent on the current menu are Sylvia Tara’s “The Secret Life of Fat” (Norton) and “The Case Against Sugar,” by Gary Taubes (Knopf). Both present a range of cutting-edge dietary research, both say that fat is unfairly maligned, and both inadvertently end up revealing that the science behind their claims is complex and its findings hard to translate into usable advice.

Sylvia Tara is a freelance writer who holds a doctorate in biochemistry and an M.B.A.; she has worked at McKinsey and on the management side of various biotech companies. Drawing on insights from both science and consulting, she has produced a book that is part physiology and part marketing pitch. Tara wants us to view lipids positively. Once we stop treating fat “like a vicious enemy,” she argues, it “could become beloved once again.”

But Tara’s attitude to fat is more ambiguous than this statement suggests. She claims to be obsessed with her figure, measuring her worth by how well she fits into skinny jeans. In her telling, the spur to her investigations comes from her envy of a friend who stays svelte despite gorging on beer and burritos, drinking sugary lattes, and never exercising. Tara, who writes that she gains weight easily, is interested in the question of why some people eat like hogs and stay thin, while others expand no matter how abstemious they try to be.

The book is a useful primer on the biology of fat. Fat comes in different forms, categorized by color. White fat, the type that we seek to lose when overweight, stores energy. Brown fat, normally found in the neck, back, and around the heart, is filled with tiny structures called mitochondria, and serves as a furnace to burn energy for body heat. A third type, beige fat, was identified some five years ago; during exercise, it receives messages from our muscles to morph into brown fat. Moreover, fat should not be characterized simply as inert blubber. It is the vehicle by which our cells receive certain essential nutrients, like Vitamins A, D, E, and K. The myelin sheaths around our nerves are eighty per cent lipids, “which means fat is actually required to think,” Tara writes. Studies by Jeffrey Friedman, at the Rockefeller University, have shown that the hormone leptin travels from fat cells to the hypothalamus, a part of the brain which is involved in regulating appetite. “Friedman’s discovery redefined fat,” Tara writes. “It was a verifiable endocrine organ with wide influence to our bodies. Through leptin, fat could talk. It could tell the brain to stop eating.”

All this will be illuminating for many readers, but Tara is a less reliable guide when she uncritically embraces various new theories about the causes and effects of obesity. She trumpets the findings of a Turkish physician, Gökhan Hotamisligil, whose work suggests that a molecule known as TNF-alpha, which has potent inflammatory properties, may be the link between obesity and Type 2 diabetes—a condition arising when the body becomes resistant to insulin, a hormone that we need in order to process sugar. (Though there’s a clear correlation between diabetes and obesity, no one has yet discovered a causal link.) Hotamisligil’s experiments showed that not only is TNF-alpha produced by fat; it also can cause resistance to insulin. “This discovery was big news,” Tara writes. However, she fails to specify that the finding was in rodents, and that subsequent studies in humans, including some by Hotamisligil, did not show the same results.

Tara also speculates that viruses may cause obesity. The research she draws on here is obscure and unconvincing. It concerns a virus called Ad-36, which infects fowl and can make chickens fat. In the studies Tara cites, more overweight people appeared to have antibodies to Ad-36—suggesting that they had been infected in the past—than slim people did. There are many reasons to be skeptical: there’s no evidence that fowl can pass Ad-36 to humans, and there are many viruses that could easily be mistaken for Ad-36.

As with many books on diet, “The Secret Life of Fat” alternates exposition with prescription. But the idea that understanding lipids at a molecular level will help you stay trim seems far-fetched. It’s telling that Tara’s final triumph—managing to fit back into her skinny jeans—has little to do with her sophisticated understanding of fat. Rather, she follows the advice of Mark Sisson, a “fitness educator” who fasts eighteen hours a day, and who, at sixty-two, she writes, is “muscular and fit and looks every bit like the Malibu surfer he is.” Tara lost weight by restricting her daily intake to at most a thousand calories and by intermittent total fasts.

This is hardly a healthy note to end on, yet elsewhere Tara seems to take aim at our destructive cultural fixation on body image. Fat was prized in the past, she notes, with big bellies signalling access to plentiful food and, thus, prosperity. The Buddha’s belly “is a major part of his brand,” she writes. (Such consultant-speak seems odd in the context of religion.) The porcine aristocrats one sees in eighteenth-century portraits are frequently shown near tables overflowing with delicacies. The women’s bodies depicted in canvasses by Peter Paul Rubens have long since made “Rubenesque” a euphemism for plus-size. And, if one goes far enough back, the huge bellies and buttocks of the Paleolithic “Steatopygian Venus” figures that have been found across much of Europe suggest that fat can connote fertility and desirability.

Tara digs up examples of Americans celebrating fat as late as the latter half of the nineteenth century. Ladies’ Home Journal gave tips on gaining weight, as did an 1878 book titled “How to Be Plump.” Still, the nineteenth century in general was more notable for a growing concern with being slim, as has been shown by previous writers, such as Gina Kolata, whose “Rethinking Thin” (2007) itself draws heavily on Hillel Schwartz’s remarkable history “Never Satisfied” (1986). Lord Byron, who struggled with his weight, swore by vinegar; at other times, he ingested just a single raisin a day, supplemented by a glass of brandy. Women in the nineteenth century stuffed themselves into near-suffocating corsets to achieve an hourglass figure with an unnaturally tiny waist. Weight-loss regimens included consuming soap, chalk, pickles, digitalis, camphor tea, grapefruit (which was thought to contain fat-dissolving enzymes), potassium acetate (a diuretic), and ipecac (which induces vomiting). People tried sweating their fat away in rubber suits, or squeezing it away in a pressurized reducing machine.

Indeed, the weight-loss fads of past centuries include precedents for all the main contemporary diets, from low-fat, low-calorie ones to high-fat, low-carbohydrate ones, like the Atkins diet. In 1825, a French lawyer, Jean Anthelme Brillat-Savarin, wrote a famous treatise, “The Physiology of Taste,” in which he contended that true carnivores and herbivores did not get fat; it was only when one ingested grain—read: bread—that the trouble started. Around the same time, an American Presbyterian minister, Sylvester Graham, reasoned that, as gluttony was the greatest sin, abstinence must lead to virtue; he advised eating vegetables and drinking water, eschewing meat, coffee, spices, and alcohol. For a while, students and faculty at Oberlin College were made to follow Graham’s diet; graham crackers were so named in order to appeal to his acolytes. Several years later, Horace Fletcher, known as “the great masticator,” touted very slow chewing as the remedy for obesity; adherents included normally skeptical people like Upton Sinclair and John D. Rockefeller.

A genuine advance, which put nutrition on a solid scientific footing for the first time, was the work of the chemist Wilbur Atwater. In the eighteen-nineties, he began studying how the body converted food to energy, by placing subjects in a closed chamber and measuring the amount of carbon dioxide they produced and oxygen they consumed after eating various foods. Atwater came up with the idea of the food calorie, adapting a measurement previously used for heat energy. In 1917, Herbert Hoover, then the head of the U.S. Food Administration, worked to publicize Atwater’s findings. “I eat as little as I can to get going,” he said. Low-calorie foods and skipping meals became popular. The importance of calories—if energy gained exceeds output, the excess becomes fat—remains one of the few unchallengeable facts in the field of dietary science. Still, further research has shown that calories eaten are only part of what determines weight. Our metabolism reflects an interplay of things like genes, hormones, and the bacteria that populate the gut, so how much energy we absorb from what we eat varies from person to person.

“You can get that out with a little seltzer.”

In the nineteen-fifties, the American Medical Association identified obesity as the country’s No. 1 health problem, and the diet industry exploded. The end of that decade brought the idea of the liquid diet—skimmed milk, supplemented with bananas or other fruit—which, in turn, eventually gave rise to products like Metrecal, Carnation Slender, and SlimFast. Self-help groups modelled on Alcoholics Anonymous began proliferating with the establishment, in 1948, of a movement called TOPS (the acronym stood for “take off pounds sensibly”). Overeaters Anonymous followed, in 1960; Weight Watchers, in 1963; and Jenny Craig, in 1983.

The immediate postwar years also brought the first sustained scientific assault on dietary fat. Ancel Keys, a physiologist at the University of Minnesota, who had spent the war developing nutritionally optimal Army rations and studying the effects of starvation, became interested in the high rates of heart attack among a seemingly well-fed sector of the population—American businessmen. He soon became convinced that the saturated fats found in meat and dairy products were the cause, and thus began the war on fat that galvanized my parents. Keys became, with his wife, Margaret, an advocate for the Mediterranean diet of unsaturated fats. Their books promoting the diet were best-sellers, and Keys, who spent his latter years in Italy, lived to the age of a hundred. (Margaret lived to ninety-seven.)

The author of “The Case Against Sugar,” Gary Taubes, gained prominence as a science writer in 2002, with a cover story in the Times Magazine—“What If It’s All Been a Big Fat Lie?”—which challenged the orthodoxy of restricting dietary fat. Carbohydrates were the real danger, he wrote—not just processed foods containing refined sugars like sucrose and fructose but also easily digestible starches from grains and vegetables. He expanded these arguments in a book, “Good Calories, Bad Calories” (2007), and, in his new book, he goes much further. Though he now allows that people can eat some carbohydrates and still live a “relatively” healthy life, he sees sugar as the devil incarnate, doing harm independent of its known role in causing obesity. Taubes believes that a wide range of seemingly unrelated diseases—“diabetes, heart disease, cancer, stroke, and Alzheimer’s, which account for five of the top ten causes of death in the U.S.”—are in fact linked, and that dietary sugar is the cause of them all, as well as of “other disorders that associate with these illnesses, among them polycystic ovary syndrome (PCOS), rheumatoid arthritis, gout, varicose veins, asthma, and inflammatory bowel disease.” In addition, he aims at showing that the food industry has systematically tried to obstruct scientific research that exposes the dangers of sugar, just as tobacco companies tried to hide the risks of smoking.

The latter claim is the more persuasive. Taubes, a pugnacious writer who clearly relishes the role of muckraker, digs up a long history of attempts to discredit charges against sugar and to point the finger at fat as the primary dietary cause of disease. In 1943, when sugar, dismissed by the government and medical organizations as “empty calories,” was being rationed as part of the war effort, sugar companies formed a trade association “to set the record straight.” It devised a two-pronged strategy: support scientists who endorsed the notion that sugar was a valuable source of dietary energy without any specific health risks; and then mobilize these experts in a public-relations campaign. A prominent Madison Avenue firm was hired to design a public-health campaign that would “establish with the broadest possible audience—virtually everyone is a consumer—the safety of sugar as a food.” Among the scientists they supported was Ancel Keys, the Mediterranean-diet pioneer; his work influenced the dietary guidelines of the American Heart Association and the American Diabetes Association. Fred Stare, an influential nutritionist at Harvard, received not only research funding but a donation of more than a million dollars, from the General Foods Corporation (whose products included Kool-Aid and Tang), to build a new department. He proclaimed that it was not even “remotely true” that “modern sugar consumption contributes to poor health.” Taubes recounts that Stare appeared on talk shows on more than two hundred radio stations as a “front man to dismiss anti-sugar sentiments publicly.”

The spread of diet crazes and of obesity anxiety in the fifties alerted the sugar industry to the fact that its product was at risk. Diet sodas with artificial sweeteners were gaining market share. The sugar industry responded in two ways: by stressing how important sugar was as an energy source for children (“neither a weight reducing nor fattening food”); and by discrediting artificial sweeteners such as saccharin and cyclamates as health dangers. It funded research at the Wisconsin Alumni Research Foundation and at the Worcester Foundation for Experimental Biology, which managed to find various adverse effects from consumption of cyclamates in rats. The latter achieved this by giving rats an absurd dose—the equivalent, in human terms, of five hundred and thirty cans of Fresca. Nonetheless, the F.D.A. eventually banned cyclamates as a health risk.

Though Taubes’s account of these little-known manipulations is useful, he overreaches in blaming sugar for such a wide range of diseases. In attempting to lump them together, he cherry-picks from a variety of recent research. For instance, some epidemiological surveys have shown that when people move from the developing world to the West they change diet and often become obese, leading to an increased incidence of diseases, including diabetes and cancer. And other diseases, such as Alzheimer’s, appear on Taubes’s list, because researchers have studied whether they are linked to insulin resistance.

Synthesizing these conjectures, Taubes sees insulin resistance as the bedrock disturbance in the body which unleashes a cascade of other hormonal and inflammatory molecules that attack the brain (causing dementia), degrade the heart and the blood vessels (causing heart attack and stroke), disturb the metabolism of uric acid (causing gout), and so on. He then attempts to build his case as a prosecuting attorney by means of a chain of “if/then” statements, such as “If sugar and high-fructose corn syrup are the cause of obesity, diabetes, and insulin resistance, then they’re also the most likely dietary trigger of these other diseases.” He invokes Occam’s razor, a concept adopted by medieval philosophers and theologians, which holds that explanations should rely on the smallest possible number of causes. “If this were a criminal investigation, the detectives assigned to the case would start from the assumption that there was one prime suspect,” Taubes writes.

Occam’s razor is hardly a fundamental law of the universe, however. No credible scientist would ever think of using it to prove or disprove anything. And Taubes neglects findings that contradict his idea that diabetes—and, by extension, sugar—is at the root of all our troubles. A study of the diabetes drug metformin, published two years ago in The Lancet, failed to show any impact on the treatment of pancreatic cancer. A placebo-controlled trial in which insulin was given to dementia patients did not find a meaningful improvement in cognition. Indeed, there is no conclusive evidence that excess dietary sugar per se causes diabetes. Most important, Taubes’s assertion that all these diseases are “closely related” is not scientifically supported. The biological behavior of cancer—DNA mutations, aberrant growth, metastatic spread—is nothing like that of diabetes. Inflammatory-bowel disease, a complex disorder that appears to have a variety of genetic underpinnings, does not seem to be caused by any particular diet or substance, and there is no evidence that restricting sugar ameliorates it. The attempt to characterize Alzheimer’s as “type-III diabetes,” linking it to insulin resistance and inflammation, is likewise speculative.

The temptation to draw facile connections is ever-present in medical research, and the most valuable current work on these conditions is a matter not of grand unified theories but of a multiplicity of very fine-grained observations. Taubes is critical of scientists’ tendency to see disorders as “multifactorial” and “multidimensional”—that is, as arising from a complex interplay of factors. Lung cancer, he argues, is also multifactorial (most smokers don’t get it and many non-smokers do), yet no one disputes that smoking is the primary cause. But cigarette smoke contains carcinogens, molecules that have been shown to directly transform normal cells into malignant ones by disrupting their DNA. There’s no equivalent when it comes to sugar. Taubes surmises a causal link by citing findings that cancer cells need glucose to thrive, and absorb more of it than other cells. But this proves nothing: malignant cells consume in abundance not only carbohydrates like glucose and fructose but other nutrients, like vitamins. To imagine that, just because cancer cells like glucose, elevated levels of it might prompt healthy cells to become cancerous is to take a vast, unsubstantiated leap.

Ultimately, Taubes’s indictment of sugar as the leading culprit in virtually all modern Western maladies doesn’t provide enough evidence for us to convict. That doesn’t mean sugar is without dangers: it certainly plays a role in the development of obesity, to say nothing of dental cavities. But these are lesser charges, and they make for a less dramatic headline.

Taubes’s big claims get our attention, of course, but for people suffering from these diseases they’re not just a harmless rhetorical strategy. A woman I know who recently emerged from chemotherapy treatment for ovarian cancer and is now in remission told me that she was terrified after reading Taubes’s book. She asked if eating chocolate would make her tumor start growing again.

The problem with most diet books, and with popular-science books about diet, is that their impact relies on giving us simple answers, shorn of attendant complexities: it’s all about fat, or carbs, or how many meals you eat (the Warrior diet), or combinations of food groups, or intervalic fasting (the 5:2 diet), or nutritional genomics (sticking to the foods your distant ancestors may have eaten, assuming you even know where your folks were during the Paleolithic era). They hold out the hope that, if you just fix one thing, your whole life will be better.

In laboratories, it’s a different story, and it sometimes seems that the more sophisticated nutritional science becomes the less any single factor predominates, and the less sure we are of anything. Today’s findings regularly overturn yesterday’s promising hypotheses. A trial in 2003, led by researchers at the University of Pennsylvania, compared an Atkins diet, high in fat and low in carbohydrates, with a low-fat, high-carbohydrate, low-calorie one. After a year, there were no significant differences in how much weight the people in each group had lost, or in their levels of blood lipids—including their LDL cholesterol, the primary concern for heart attack and stroke. In a follow-up study in 2010, participants who followed either a low-carbohydrate or a low-fat diet ended up losing about the same amount of weight (seven kilograms) after two years. It was impossible to predict which diet would lead to significant weight loss in any given individual, and, as most dieters well know, sustaining weight loss often fails after initial success.

Other research seems to undermine the whole idea of dieting: extreme regimens pose dangers, such as the risk of damaged kidneys from a buildup of excess uric acid during high-protein diets; and population studies have shown that being a tad overweight may actually be fine. Even studying these issues in the first place can be problematic. Although the study of the Mediterranean diet, for example, reflects randomized controlled experiments, most nutritional studies are observational; they rely on so-called food diaries, in which subjects record what they remember about their daily intake. Such diaries are notoriously inexact. No one likes admitting to having indulged in foods that they know—or think they know—are bad for them.

Science is an accretion of provisional certainties. Current research includes much that is genuinely promising—several groups have identified genes that predispose some people to obesity, and are studying how targeted diets and exercise can attenuate these effects—but the more one pays attention to the latest news from the labs the harder it becomes to separate signal from noise. Amid the constant back-and-forth of various hypotheses, orthodoxies, and fads, it’s more important to pay attention to the gradual advances, such as our understanding of calories and vitamins or the consensus among studies showing that trans fats exacerbate cardiovascular disease. What this means for most of us is that common sense should prevail. Eat and exercise in moderation; maintain a diet consisting of balanced amounts of protein, fat, and carbohydrates; make sure you get plenty of fruit and vegetables. And enjoy an occasional slice of chocolate cake. ♦