Big Fat Fake

The Atkins diet controversy and the sorry state of science journalism.


It was exactly what millions of obese Americans wanted to hear: Diet guru Robert Atkins has been right all along; conversely, the "medical establishment" that has routinely criticized him has been entirely wrong. Unlimited-calorie, high-fat meals are the key to low-fat bodies. So claimed award-winning science writer Gary Taubes in an 8,000-word New York Times Magazine blockbuster that appeared last July, "What If It's All Been a Big Fat Lie?"

The magazine's cover was even juicier than the title: It featured a slab of steak topped with butter and asked, "What If Fat Doesn't Make You Fat?" In fact, Taubes declared in his article, the consumption of too little fat could explain the explosion in obesity.

Atkins quickly wrote an editorial for his Web site claiming the article "validated" his work. Gushingly favorable follow-up stories appeared on NBC's Dateline, CBS' 48 Hours, and ABC's 20/20. Dr. Atkins' New Diet Revolution, with 11 million copies already in print, shot up from No. 5 to the top spot on the New York Times paperback bestseller list for "Advice, How-To, and Miscellaneous" books. It went from No. 178 to No. 5 in Amazon's rankings. Taubes himself landed a book contract from publisher Alfred A. Knopf for a big fat $700,000.

But there were serious problems with this revolutionary argument about one of our nation's most serious health problems. For example, Taubes omitted any reference to hundreds of refereed scientific studies published during the last three decades that contradicted his position. Researchers from whom he could not pull even a single useful quote supportive of his thesis were banished from the piece, while many of those whom Taubes did end up quoting now complain that he twisted their words.

"I was greatly offended by how Gary Taubes tricked us all into coming across as supporters of the Atkins diet," says one such source, Stanford University cardiologist John Farquhar. "I think he's a dangerous man. I'm sorry I ever talked to him."

Upon closer examination, Taubes' "What If It's All Been a Big Fat Lie?" turns into a big fat mess. The misguided hoopla over the New York Times Magazine article and the Atkins Diet is a short study in the sorry state of scientific and medical reporting, not to mention a diet industry that routinely panders to people's worst impulses.

The Fat Shall Set Ye Free?

In Dr. Atkins' New Diet Revolution, Robert Atkins claims that by simply minimizing your carbohydrate intake you can quickly lose massive amounts of weight, even while pigging out daily on fatback, pork rinds, and lard. He also claims his diet will relieve "fatigue, irritability, depression, trouble concentrating, headaches, insomnia, dizziness, joint and muscle aches, heartburn, colitis, premenstrual syndrome, and water retention and bloating."

Claims like those should make anyone suspicious, even those who have barely scraped through high school biology. Gary Taubes has gone well beyond that level. He's a contributing correspondent to America's preeminent scientific journal, Science. He has won the National Association of Science Writers' Science in Society Journalism Award three times—the maximum allowed. Only one other writer has ever achieved that status.

Nonetheless, at the very outset of his piece (viewable in its entirety at Taubes set forth the proposition that Atkins was crucified by the "American medical establishment," which claimed his diet was ineffective and possibly dangerous and in so doing encouraged the "rampaging epidemic of obesity in America."

There is a nugget of truth in Taubes' criticisms of establishment dietary fat advice. Well-meaning but misguided health officials and health reporters, joined by opportunistic anti-fat diet book gurus, have convinced much of the public that the major culprit—perhaps the only culprit—in obesity is dietary fat. Avoid fat, we were told, and you won't get fat. Given license to eat as many calories as we wanted from the other nutrient groups, many of us have done exactly that. This goes far to explain why almost one-third of us are obese and almost two-thirds of us are overweight. But even here Taubes is no pioneer; the damage caused by fat-free fanaticism was pointed out long before. (See, for example, my own 1997 book, The Fat of the Land.)

Moreover, the Atkins-Taubes thesis of "fat won't make you fat" encourages obesity in a similar way: It offers carte blanche for consuming limitless calories, only this time swapping carbohydrates for fat. Taubes made that swap while presenting a far less scientific case than is presented in an Atkins infomercial.

Ask Stanford endocrinologist Gerald Reaven. He's best known for calling attention to "Syndrome X," a cluster of conditions that may indicate a predisposition to diabetes, hypertension, and heart disease. Among Reaven's recommendations for lowering the risk of that syndrome is to reduce consumption of highly refined carbohydrates such as those present in soft drinks and table sugar. But that's where the overlap with Atkins ends.

"I thought [Taubes'] article was outrageous," Reaven says. "I saw my name in it and all that was quoted to me was not wrong. But in the context it looked like I was buying the rest of that crap." He adds, "I tried to be helpful and a good citizen, and I ended up being embarrassed as hell. He sort of set me up." When I first contacted Reaven, he was so angry he wouldn't even let me interview him.

But his position on Atkins was all over the Internet in interviews posted long before Taubes talked to him. Do "low-carb diets like The Zone [by Barry Sears] and Atkins work?" one asked. Answer: "One can lose weight on a low-calorie diet if it is primarily composed of fat calories or carbohydrate calories or protein calories. It makes no difference!"

The very person with whom Taubes chose to end his article, Stanford's John Farquhar, was as livid as Reaven. Taubes said that Farquhar had sent Taubes "an e-mail message asking the not-entirely-rhetorical question, 'Can we get the low-fat proponents to apologize?'" On this powerful note, the article ended.

But it's Taubes whom Farquhar wants to apologize. "I was greatly offended by how Gary Taubes tricked us all into coming across as supporters of the Atkins diet," he wrote in an e-mail he broadcast to reporters and to colleagues who were stunned that Farquhar might actually hold the beliefs Taubes attributed to him. "We are against the Atkins Diet," he wrote, speaking for himself and Reaven. "I told him [Taubes] there is the minor degree of merit" to the idea that "people are getting fatter because too much emphasis is being placed on just cutting fats," Farquhar told me. But "once I gave him that opening—bingo—he was off and running, even though I said about six times that this is not the cause of the obesity epidemic."

Diets and Data

Taubes proved as adept at clipping data as at clipping quotes. Thus he claimed that one of the "reasons to suggest that the low-fat-is-good-health hypothesis has now effectively failed the test of time" is "that the percentage of fat in the American diet has been decreasing for two decades." (Emphasis added.)

That's true, but irrelevant. The amount of fat consumed has been steadily climbing, as has consumption of all calories. Individual caloric consumption jumped from 3,300 calories per day in 1970?79 to 3,900 in 1997, an 18 percent increase. Per-person consumption of fat grams increased from 149 to 156, a 4.5 percent increase. "We're eating just too darned much of everything," says Farquhar.

Taubes also shoved aside decades of published, controlled, randomized clinical trials comparing nutrient intake and weight loss. His apparent justification in the article was that the "research literature [is] so vast that it's possible to find at least some published research to support virtually any theory." But that's sheer nihilism. Good science is cautious and skeptical, not permanently open-ended. That's why terms like weight of the evidence are used. And the evidence against Atkins-like low-carbohydrate diets is crushing.

In April 2002, for example, the Journal of the American Dietetic Association (JADA) published a review of "all studies identified" that looked at diet nutrient composition and weight loss. It found over 200, with "no studies of the health and nutrition effects of popular diets in the published literature" excluded. In some, subjects were put on "ad libitum" diets, meaning they were allowed to eat as much as they wanted as long as they consumed fat, protein, and carbohydrates in the directed proportions. In others, subjects were put on controlled-calorie diets that also had directed nutrient proportions. The conclusion: Those who ate the least fat carried the least fat.

An alternative method of comparing diets is a meta-analysis, which means not looking at the sum of the whole but actually combining the data. One such meta-analysis, covering 16 ad libitum studies and almost 2,000 people, appeared in the International Journal of Obesity and Related Metabolic Disorders in December 2000. The conclusion: Those on low-fat diets had "a greater reduction in energy intake" and a "greater weight loss than control groups."

"Aren't all these studies highly relevant to the issue of whether an Atkins-like diet works, and don't they indicate that it does not?" I ask Dr. Louis Aronne, director of the Comprehensive Weight Control Program at New York's Weill Cornell Medical Center. "I agree completely," he says. "You're absolutely right."

This wasn't the first time Taubes had published a lengthy article on fat while leaving out this vital information. He also did so in one of his award-winning pieces, a precursor to the "Big Fat Lie" article called "The Soft Science of Dietary Fat" that appeared in Science in March 2001. In a subsequent letter to the journal, three obesity research co-authors, including James Hill, director of the University of Colorado Center for Human Nutrition in Denver, noted, "What Taubes does not mention are the meta-analyses of intervention studies comparing ad libitum intakes of higher fat diets with low-fat diets that clearly show reduced caloric intake and weight loss on the low-fat diet." Taubes responded to the letter but again refused to address these studies.

Why? "They're not worth mentioning," he told me in a telephone interview. They weren't done correctly. None of them? None. The one meta-analysis Taubes thinks was properly conducted appeared in 2002 in The Cochrane Library. Yet it, too, found no advantage to low-carbohydrate diets, merely that "fat-restricted diets are no better than calorie restricted" ones.

Where, I ask Taubes, did all these researchers go wrong? The problem is inherent to an intervention study, he says. "When you counsel people you change their behavior." But doesn't that apply to all the groups in a study? Yes, he grants. "But the idea is to make the intervention effect equal for everyone, whichever diet they happen to be on," he says. "If the interventions aren't the same, then you just don't know how to interpret the results." That may be true, but it's also irrelevant. There's no reason to think persons on either low-fat or high-fat diets got more or less intervention in these myriad studies. Indeed, in some of them virtually all the intervention emphasis was on exercise, with little nutrition counseling one way or the other.

Finally, the comprehensive JADA review published last April also looked at persons who weren't in intervention studies at all but rather were part of the U.S. Department of Agriculture's Continuing Survey of Food Intake by Individuals. An updated report on the survey appeared last June in the Journal of the American College of Nutrition. Both survey reports came to the same conclusion as the intervention studies.

Dr. Aronne is quick to point out that this wealth of data supporting lower-fat diets "is not an endorsement for eating unlimited amounts of nonfat muffins and soda simply because they're fat-free." All carbohydrate sources are not equal. For example, fiber appears to play a powerful role in weight control, but there is no more fiber in a soda than there is in a steak. That said, a high-fat diet does carry an inherent metabolic disadvantage in that fat has nine calories per gram, while carbohydrates and protein each have four.

Abstract Weight Loss

Having circumvented this mass of peer-reviewed literature readily open to public scrutiny in libraries and often online, Taubes instead tried to make his case with a mere five studies. All five were (and are) available only in abstract form. That is, they are summaries of about 300 words each that have been presented at various obesity conferences. "The results of all five of these studies are remarkably consistent," Taubes averred. "Subjects on some form of the Atkins diet…lost twice the weight as the subjects on the low-fat, low-calorie diets."

One of the five studies, conducted at the Durham Veterans Administration Medical Center in North Carolina, was funded by the Atkins Center. Those researchers repeatedly have publicized their interpretation of their findings and unsurprisingly have conferred their full blessings on the diet.

They did so most recently in late November, garnering tremendous favorable media attention. (See "Hold the Lard" at The authors of the other four studies, however, have been reticent about releasing their data, in part because pre-publicity in the lay press makes it more difficult to get published in medical journals. But when I interviewed researchers for two of the other studies, they all insisted Taubes grossly mischaracterized their findings.

"The Atkins diet produces weight loss, as does the grapefruit diet, the rotation diet, and every other fad diet out there," says one of the researchers, Colorado's James Hill. "I haven't seen any data anywhere saying Atkins is better than these other diets for weight loss. Taubes is trying to fly in the face of the scientific evidence." Referring to the book deal, he says, "Taubes sold out."

Hill's co-researcher, Gary Foster of the University of Pennsylvania, says "the probable explanation for the greater weight loss in the groups on the Atkins regimen" is that it "gives people a framework to eat fewer calories, since most of the choices in this culture are carbohydrate driven….You're left eating a lot of fat, and you get tired of that. Over time people eat fewer calories." That would make the Atkins plan nothing more than a low-calorie diet in disguise.

Another of the abstracts came from the University of Cincinnati. The Atkins-like group "did have twice as much weight loss, and to completely lose that point would be unfair," says one of the co-authors, Randy Seeley of the university's Obesity Research Center. But his explanation is similar to Foster's, if more colorful. "If you're only allowed to shop in two aisles of the grocery store, does it matter which two they are?" he asks.

All the researchers I interviewed also insisted the studies weren't long enough to be conclusive, with none lasting more than a year. And the kicker is that all five were intervention studies, conducted using the same methodology that Taubes cites to dismiss the mountain of published material that undercuts his position.

Seeley and co-researcher David D'Alessio were also upset that Taubes made use of their material at all and not just because it hurt their chances of publication. "One of the things I object to most in the Taubes article is the idea that we're going to carry out this scientific debate in the lay press with data that's unavailable for scientists to review," says Seeley. "I believe in the peer-review process." Indeed, one "danger of trying to conduct this out in the lay press," he says, is that "you have a guy like Taubes going through it and just picking up the pieces that support his opinion."

3,000 MIAs

Taubes also ignored the approximately 3,000 members of a database called the National Weight Control Registry. For 10 years, the registry has tracked people who have lost at least 30 pounds and kept it off for at least a year. The average member has maintained a loss of about 60 pounds for about five years.

Co-administered by Hill in Denver and Rena Wing of the University of Pittsburgh, the registry is aimed at finding out what works and what doesn't. According to its members, what doesn't work is a high-fat diet. On average, they consume only 23 percent of calories from fat. "Almost nobody's on a low-carbohydrate diet," Hill says. Another important lesson that may be drawn from the registry is that the importance of any type of diet in weight control may be overemphasized. Ninety-one percent of the subjects said they regularly exercised.

While relying on self-interpretation of unpublished abstracts is valid methodology to Taubes, he insists the registry is so unscientific as to be worthless. One problem, he told me, is that it represents only a tiny fraction of all those who have succeeded at weight loss. Further, the sample is entirely self-selected rather than randomized. "Its method of recruiting could bias the selection toward those who use low-fat diets," he says.

Yet the registry data have been considered valid and important enough to have been written up in such peer-reviewed medical publications as the American Journal of Clinical Nutrition, the Journal of the American Dietetic Association, the International Journal of Obesity, Health Psychology, and Obesity Research.

"You can't get around" the problem of self-selection, says Suzanne Phelan, a co-investigator of the registry at Brown University Medical School. "But why would non-Atkins people select in and Atkins ones stay out?" Atkins dieters, she notes, "seem to be very dedicated." (Other researchers have described them as having an almost religious fervor.) Originally, recruits were selected "based on a random-digit dialing procedure," she says. But that proved onerous, and "as media such as USA Today and CNN began talking about the registry, we just let them take over" the recruiting process. "There's no reason to think that people who see those media are more likely to have a certain diet," Phelan says.

Oh, and there's another place where joining the registry has been promoted: the Atkins Web site. So we're left wondering why successful low-fat eaters would be especially likely to select into the registry or why the purchasers of over 11 million Atkins diet books consistently opt out.

Feeling Full

For all its 8,000 words, there were few actual data in the Taubes piece. It was rather like reading a treatise explaining how the Chicago Cubs may well be the best team in baseball history without being informed they haven't won a pennant since 1945. Instead readers were regaled with explanations of physiological mechanisms—the basis for which, Taubes wrote, is "Endocrinology 101"—that might explain how dieters shed pounds and inches. Endocrinology 101 is a term popularized by Dr. David Ludwig, who runs the pediatric obesity clinic at Children's Hospital Boston.

According to Taubes, Endocrinology 101 "requires an understanding of how carbohydrates affect insulin and blood sugar and in turn fat metabolism and appetite." In brief, it says there are aspects of a high- or low-carbohydrate diet that affect both how much we want to eat (referred to as "satiety") and how efficiently the body converts the various nutrients into body fat. And the theory says an Atkins-like diet is both more satiating and less efficient in converting calories to fat.

Yet the published literature that Taubes ignored says otherwise. The aforementioned review of over 200 studies in the Journal of the American Dietetic Association expressly nixed the idea that any type of food converts less efficiently to body fat. "None of the popular diet research we reviewed suggests a metabolic advantage with respect to weight loss," it declared.

Nor can Taubes fall back on his five studies, according to Seeley.

"Ultimately our data do not support any of the mechanisms" for why a low-carbohydrate diet might be especially effective in inducing weight loss "that Atkins and proponents of the diet have [suggested]," he told me. Indeed, each explanation that Taubes presents for how an Atkins diet might cause weight loss collapses under the weight of the published research he ignores.

Consider the matter of satiety. How, Taubes wondered, could a low-calorie regimen "suppress hunger, which Atkins insisted was the signature characteristic of the diet." One possibility, he said, was, yes, "Endocrinology 101: that fat and protein make you sated and, lacking carbohydrates and the ensuing swings of blood sugar and insulin, you stay sated."

But is there any empirical support for this? No, according to an April 2002 review of studies in the Journal of the American College of Nutrition that summarized "high and low fat treatments when subjects were allowed to eat ad libitum." It found "energy intake on the low-fat diets ranged from 16 percent to 24 percent less than those on high fat diets."

"We've done masses of studies on fat and satiety," says Barbara Rolls, professor of nutrition at Pennsylvania State University, where she has authored four books and written about 60 medical journal articles on human food intake. She's widely considered the nation's top authority on satiety. Some of her experiments involved ingestion; in others, "We directly infused pure fat and pure carbohydrates both directly into [human] veins and directly into stomachs." Says Rolls, "We found very little difference between fats and carbohydrates."

Rolls does say there is some evidence that high-protein diets may be more satiating, but Atkins isn't really high protein; it's just high fat. According to an analysis in the journal Circulation, Atkins starts off at 36 percent protein from calories and declines to 24 percent in the "maintenance" stage.

What really counts when it comes to satisfying hunger, Rolls says, is "foods that give big portions without a lot of calories. We call these low-energy-density foods." She adds, "The Atkins diet would not be a good way to reduce energy density at all, especially with the restrictions on fruits and vegetables that are really the keys to a low-energy diet." Further, because fat contains more than twice the energy per ounce as either carbohydrates or protein, "high-fat foods are so energy-dense that it's really easy to eat excessive portions."

Rolls says she sent a big pile of her material on satiety to Taubes, but he "just brushed it aside." She says he also interviewed her for over six hours, but every last sentence disappeared into a black hole. Likewise for the interviews Taubes conducted with James Hill and at least five other top obesity researchers from whom he apparently couldn't extract even a single useful line: Dr. F. Xavier Pi-Sunyer of St. Luke's-Roosevelt Hospital in New York; Marion Nestle, chairwoman of the Department of Nutrition and Food Studies at New York University; Dr. Arne Astrup of Denmark; and Dr. Jules Hirsch, whom Taubes interviewed in his office at Rockefeller University in New York. "I just kept telling him, it doesn't matter what kind of calories you eat," says Hirsch.

Taubes is "very selective in what he chooses to include because he's trying to sell a specific line," Rolls says. "He is a good writer; that's the thing that scares me. This is such a good example of how you can pick and choose your facts to present the story you want. But that's not how science should be done. You can't interview everybody and simply ignore the people you don't want to hear." She means that rhetorically, of course.

Gorging on Theory

Stacking theory atop theory, Taubes roared on. Something called "hyperinsulinemia" could also favor the Atkins dieter, he insisted. When carbohydrates are ingested they are broken down in the intestine into glucose and other sugars. Glucose then stimulates cells in the pancreas to secrete insulin to remove that glucose and take it into tissues to be used as fuel or stored. Protein and fat consumption don't have nearly the same impact on insulin production because the whole point of insulin is to maintain the stability of the sugar level.

The Atkins hyperinsulinemia theory, ex-plained Taubes, is that carbohydrates can "cause a spike of blood sugar and a surge of insulin within minutes. The resulting rush of insulin stores the blood sugar away and a few hours later, your blood sugar is lower than it was before you ate." The brain receives a signal that the body needs more food, and the vicious circle repeats itself. Carbohydrates at the top of what's called the "hypoglycemic index" are the most evil of the evil, since they cause blood sugar to rise the fastest. The index ranks potatoes as slightly worse than jelly beans.

For support, Taubes once again fell back on "Endocrinology 101." David Ludwig "notes that when diabetics get too much insulin, their blood sugar drops and they get ravenously hungry," wrote Taubes. "They gain weight because they eat more, and the insulin promotes fat deposition." But according to Seeley, this applies to diabetics injecting massive amounts of insulin into the bloodstream, not to carbohydrate consumers.

"Yes," Seeley says, "if you give people a big wallop of insulin they do eat a lot, but do people under normal circumstances ever get close to that by eating? No. Is it possible that some people are that reactive? Yes. Is it likely that lots of people fall into that category? No."

Taubes presented University of Washington endocrinologist Michael Schwartz, whom he had interviewed, as a proponent of the idea that blood insulin levels as altered by carbohydrates could be a significant contributor to weight gain. But a commentary in the same magazine Taubes writes for, Science, sharply contradicted that position. "Although the concept that insulin triggers weight gain has little scientific merit, it remains a key selling point for advocates of diets that are low in carbohydrate and high in protein and fat," it read. "If hyperinsulinemia has adverse consequences, obesity does not appear to be among them," it concluded. Who wrote that? Michael Schwartz.

Indeed, Schwartz was also the primary author of a study concluding that obese people whose systems secrete insulin at high levels may be protected against further weight gain. "Relatively reduced insulin secretion," he concluded, "is a significant and independent predictor of the tendency to gain weight and adiposity in Pima Indians."

The Pima in Arizona have been the focus of a tremendous amount of research because even by American standards they are incredibly obese and suffer horrific rates of diabetes and heart disease. Comparisons of the Arizona Pima with genetically similar Pima in Mexico find that the Arizonans eat about twice as much fat (although the Mexicans also do far more manual labor) and are almost 60 pounds heavier on average. A National Institutes of Health evaluation of the traditional Pima diet (that is, back when they were thin and healthy) found that it was extremely high in carbohydrates, from 70 percent to 80 percent.

Schwartz says it's not that he believes insulin can't play a role in promoting weight gain, but he rejects "Endocrinology 101" based on what he calls "Scientific Methodology 101." "Before you draw conclusions you need data," he says. "There is no compelling evidence that in normal individuals day-to-day fluctuations of the blood glucose level are an important determinant of how much food is consumed."

The Diet Revolution That Isn't

Two distinct controversies have always swirled around the Atkins diet. First, is it effective for long-term weight loss? Second, could those using it be harming themselves by raising their blood lipids (cholesterol and trigylcerides)? The five unpublished abstracts do seem to indicate that for people who manage to stick to a high-fat Atkins diet, it may not be as harmful as was once generally believed. But this finding is quite preliminary and in any case certainly must depend greatly on which types of fat are consumed.

This is a distinction Taubes decided to lose.

Thus he quoted or invoked the name of the chairman of the Department of Nutrition at the Harvard School of Public Health, Walter Willett, seven times during his piece. Willett protests, however, that "I told Taubes several times that red meat is associated with higher risk of colon and possibly prostate cancer, but he left that out." And don't forget the illustration on the cover of The New York Times Magazine; that wasn't a flounder with heart-healthy flaxseed oil sitting on top of it.

Taubes also told readers that a metabolic process called ketosis, often invoked to show the Atkins diet could be dangerous, was quite harmless, providing reassuring words from National Institutes of Health researcher Richard Veech that "ketosis is a normal physiologic state." Veech told me by e-mail that the quote was correct, but that Taubes "omitted to say that I strongly urged people to not use the Atkins diet without the supervision of a physician because of the likely elevation of blood cholesterol and lipid on a high fat diet." But you don't have an impact if you insist that a fad diet be supervised by a doctor.

There's nothing "revolutionary" about the Atkins diet. A similar diet appeared in an 1863 booklet by a British undertaker named William Banting, who got the idea from a surgeon. It has popped up in various guises ever since, including a 1946 book extolling the virtues of eating whale blubber, and a 1958 book, Eat Fat and Grow Slim, written by a psychiatrist.

Likewise, there has long been convincing evidence that the diet fails to live up to its claims. Taubes wrote that "when the American Medical Association (AMA) released its scathing critique of Atkins' diet in March 1973, it acknowledged that the diet probably worked but expressed little interest in why."

The heavily endnoted document, which appeared in the June 4, 1973, issue of The Journal of the American Medical Association but unfortunately is not available on the Internet even in abstract form, was indeed scathing. But the rest of Taubes' description is false.

"The notion that sedentary persons, without malabsorption or hyperthyroidism, can lose weight on a diet containing 5,000 calories a day is incredible," the article says. Statements such as "No scientific evidence exists to suggest that the low-carbohydrate ketogenic diet has a metabolic advantage over more conventional diets for weight reduction," and "there is no reason to associate a diet rich in carbohydrate with obesity" hardly seem to acknowledge "that the diet probably worked."

Other terms the AMA used to described Atkins' theories included "naïve," "biochemically incorrect," "inaccurate," and "without scientific merit."

It also explained why the diet didn't work, mocking Atkins' basic thesis that fat and protein cannot cause weight gain in the absence of carbohydrate consumption as a "thermodynamic miracle."

Three additional decades of research have merely played "pile on" with the AMA's findings. The explanation for weight loss on Atkins given by Foster and Seeley was right there. "When obese patients reduce their carbohydrate intake drastically, they are apparently unable to make up the ensuing deficit by means of an appreciable increase in protein and diet," said the AMA.

Girth of a Nation

What has changed drastically in the last three decades is the girth of a nation. American obesity is increasing at a terrifying rate.

Since the publication of Taubes' article, numerous doctors, scientists, and health writers have picked apart various pieces of his argument. A fatlash has formed against Taubes, The New York Times Magazine, and Knopf. Originally riding an adulatory wave, Taubes complained bitterly to the weekly New York Observer in November that he was "being attacked by sleazebags."

The New York Times Magazine has printed no clarification or retraction of any kind. Yet not only should the magazine's editors have known there were serious problems with the piece, but Farquhar says he told them so outright, based on what he had gleaned from two fact checkers. He says he told those checkers that if Taubes "tries to make it look like I'm saying that I was supporting the idea that the obesity epidemic was from overloading on carbohydrates that this was so far off the mark that I would have to vomit."

At Knopf, Taubes' acquisitions editor, Scott Segal, has wrapped himself in the flag, telling the Observer, "It's a free country: First Amendment," as if he believes the Constitution requires publishers to hand out $700,000 checks to all authors. Equally bizarre is his effort to distance the book acquisition from the article. They "chose to put a certain picture on the cover and to use a certain approach to the subject in 5,000 [sic] words, but that's not the book," he said. Critics "are reacting to a magazine piece I had nothing to do with." Yet Taubes told me that his article had barely hit the stands when Knopf's offer dropped in his lap, as did an even larger offer from another publisher that he says he rejected because it's the publisher of Atkins' book and it might hurt his credibility.

But obesity and the millions of individual tragedies it has produced are ultimately far more important than this skirmish over a single story. Louis Aronne says, "I think people are getting increasingly confused about what to do. I'm afraid they'll just give up." Randy Seeley says journalism like Taubes' "just makes people confused and frustrated."

Taubes "gave his readers what they wanted to hear," says James Hill. "But what people want to hear is killing them."

Gary Taubes responds to Michael Fumento.