Gary Taubes has written two books explaining why people get fat and why a low-carb diet is the solution to preventing and treating obesity. He didn’t like what I had to say about his books on this blog back in 2011. I was not the only one to criticize. Another reviewer accused him of “abandon[ing] journalistic and scientific integrity in place of observational data, straw men and logical fallacy.” He says he agrees with Taubes’ premises but that his “arguments made me cringe,” and he goes into considerable detail to explain why. His analysis is worth reading.

photo by Fj.toloza992 [CC-BY-SA-3.0], via Wikimedia Commons]

photo by Fj.toloza992 [CC-BY-SA-3.0], via Wikimedia Commons]

Rather than engaging in the Comments section, Taubes complained to me in a somewhat offensive personal e-mail, saying I had failed to understand what he wrote. Recently he e-mailed me again, condescendingly suggesting that I might understand his arguments better if I read an article he wrote last year for the British Medical Journal. I read it, and confirmed that I had understood perfectly well the first time around and that it was Gary Taubes who didn’t understand my criticisms. I pointed out some omissions and inconsistencies, but my major criticisms boiled down to two:

  1. The clinical evidence isn’t yet sufficient to convincingly prove his thesis. (He himself admitted this.)
  2. He strongly recommended that everyone adopt a low-carb diet, essentially insisting that we act on insufficient evidence. And this was after he had devoted whole chapters of his books to demonizing the low-fat diet advocates for doing exactly that: acting on insufficient evidence.

His basic thesis is that conventional wisdom is wrong when it says we get fat because we overeat, and that counseling patients on the basis of the energy balance hypothesis has been remarkably ineffective. He thinks we get fat not because we eat too much but because of carbohydrates.

He says:

What is wrong with the mechanism that normally adjusts appetite to caloric output? What part of this mechanism is primarily disturbed? Any regulatory defect that drove people to gain weight…would induce them to take in more calories than they expend. Positive caloric balance would be, then, a result rather than a cause of the condition.

I agree completely. I think we are talking at cross-purposes and getting hung up on semantics. It is obvious to me that “something” (likely more than one “something”) causes us to take in more calories than we expend. And then that positive energy balance is what causes us to gain weight.

Taubes says the real question is not why we eat too much, but why we store too much fat. The endocrinological hypothesis attempts to explain the root cause of obesity on the basis of insulin and/or the concept of lipophilia. Insulin is fattening, and since serum insulin levels are driven by the carbohydrate content of the diet, it seems only logical that a low carb diet would tend to reduce weight. Lipophilia (“love of fat”) explains that fat deposition is not uniform throughout the body and that must have a basis in biology; and that people who are constitutionally predisposed to fatten must have adipose tissue that is more lipophilic than that of lean individuals. The lipophilic tissue gloms onto calories even in the presence of under-nutrition and deposits them as fat, which deprives other organs and cells of energy and leads to hunger or lethargy. Withdrawing fat from the circulation after meals results in a delayed sense of satiety and a yearning for carbohydrates. This is supported by the conventional wisdom that carbohydrates are fattening.

But Taubes said in his book that simply restricting carbohydrates

…leads to weight loss and particularly fat loss, independent of the calories we consume from dietary fat and protein. We know that the laws of physics have nothing to do with it.

That is simply wrong. The only grain of truth is that a low-carb diet may result in decreased hunger pangs so that total calorie intake drops. The laws of physics tell us that if you ate 7000 calories of protein and fat during a period of time when you only expended 3500 calories, you would gain a pound, even if you ate no carbohydrates at all. There is no getting around the physics. If you expend more calories than you ingest, you will lose weight. No diet has ever been shown to produce weight loss without a reduction in calories. The problem is that reducing calorie intake is fiendishly hard to accomplish for many reasons, both physiological and psychological. No one has ever denied that.

Low carb diets came into fashion in the 1970s; and Taubes says they worked “remarkably well” but were criticized because they were rich in saturated fats, which the medical community believed caused heart disease. Sure, lots of people lost weight by controlling the quantity and quality of carbohydrates in their diet, but the evidence does not support the claim that these diets worked “remarkably well.” In comparative studies, people achieved sustained weight loss just as well on diets with different macronutrient compositions. Taubes concludes that there is little long-term compliance with any diet, which seems to me to undermine his recommendations.

He stresses that knowing the underlying cause of obesity is a different question from advising someone to follow a Mediterranean, low-fat, or low-carb diet. Yes it is! Which is exactly my point. There is no evidence that any diet works unless it results in lower total calorie intake. Does it really matter how that lowered calorie intake is achieved? Some people prefer one diet to another; if they are able to lose weight on a diet that includes carbs, why insist that they change their habits?

Taubes agrees with me that diet studies are inherently problematic (poor compliance, poor self-reporting of intake, etc.) and that rigorous studies are required to test hypotheses. He has cofounded a not-for-profit organization called the Nutrition Science Initiative to fund and facilitate rigorous well-controlled experimental trials carried out by independent, skeptical researchers. He says we must “refuse to accept substandard science as sufficient to establish reliable knowledge, let alone for public health guidelines.” I couldn’t agree more, and my criticism of Taubes is based on his track record of accepting substandard science and basing recommendations for public health on that rather than on rigorous science. Where we part company is that I am content to withhold judgment until the hypothesis is properly tested, while Taubes is so convinced that his hypothesis is right that he has urged everyone to change their eating habits before the test results are in. In his books, he says that obesity is such a serious problem that it is urgent that we institute his diet recommendations now, without waiting for the evidence. Which is exactly what he criticized the low-fat diet campaign for doing.

We need to understand why some people gain weight easily and others don’t. Taubes doesn’t have an answer for that: his “cause” of obesity is more of a “mechanism” that doesn’t really get at the underlying genetic, environmental, and behavioral causes. While we are waiting to understand that, we still have the practical problem that overweight people need to lose weight now. It is undeniable that if you can find a way to reduce total calorie intake sufficiently, you will lose weight. I maintain that there is more than one way to accomplish that, and that we don’t know enough yet to dictate one method. Most people find it difficult to reduce total calorie intake without restricting carbohydrate intake, particularly simple carbohydrates. But that doesn’t mean everyone must follow a strict low carb diet.

A 2003 systematic review found insufficient evidence to make a recommendation for or against low-carb diets, and found that weight loss on a low-carb diet was principally associated with decreased caloric intake.

The Swedish Council on Health Technology Assessment recently reviewed the dietary treatment of obesity. They said a low-carb, high-fat diet is the most effective for weight loss in the first 6 months and it improves health markers like HDL cholesterol. But they also said long term studies show no statistically significant differences among different diets, probably because compliance decreases over time. And they didn’t deny that other diets can also be effective for weight loss and can also improve health markers. And contrary to popular wisdom, they found that the benefit of adding exercise to diet is marginal or non-existent. Another 2014 meta-analysis found that the short-term benefits of higher protein, lower carb weight loss diets appear to persist to a small degree over a longer period.

My prescription for weight loss is this:

  1. Start keeping a food diary. Write down everything that goes in your mouth and track how many calories you are eating to maintain your present weight.
  2. Whatever number you come up with, cut it down by 500 calories a day; this should result in loss of a pound a week.
  3. If you are not losing a pound a week, cut down by increments until you are.
  4. Exercise will allow you to eat more calories but is not absolutely essential; those who abhor exercise or who are unable to exercise due to physical limitations can still lose weight.
  5. It doesn’t matter if your calorie estimates are inaccurate; reducing calories on the basis of inaccurate estimates will still result in weight loss.
  6. It is the average calorie intake that matters. If you are aiming for 1200 calories a day, you could splurge on a 2400 calorie Thanksgiving dinner or binge on 2400 calories of birthday cake and ice cream and make up for it by cutting down by 100 calories on each of 12 other days.
  7. There is no need to deny yourself anything you crave as long as you control total calories.
  8. Try to include a variety of foods to insure adequate nutrient intake. If you are concerned that you may have cut calories so much that good nutrition is suffering, it’s better to consult a dietitian than to rely on self-prescribed vitamins and supplements.
  9. Try to pick foods that are filling but low in calories (low calorie density foods). Use the diet diary to study your eating habits, identify situations or foods that particularly tempt you to overeat, and look for ways to outwit the temptations. You might want to try tricks like using smaller plates, eating more slowly and consciously savoring every bite, eating in the dining room instead of in front of the TV, etc. One of my patients would walk the dog whenever she felt hungry; when she got back home, the cravings had passed and she and the dog had benefited from the exercise.

I’m not claiming that everyone can do this successfully, but then Taubes admits that there is poor long-term compliance with a low-carb diet, too. He even admits that the diet can cause side effects, which he attributes to (1) eating too much protein and too little fat, (2) attempting strenuous exercise without taking the time to adapt to the diet, and (3) most importantly, to the body’s failure to compensate for the lower insulin levels. He admits that carbohydrate cravings can be difficult to overcome, and that high protein diets can be toxic.


Taubes’ hypothesis about the underlying cause of obesity may be right, and a low-carb diet may be best not only for weight loss but for disease prevention. I look forward to the results of rigorous studies. Meanwhile, I favor moderation: going easy on the carbs, especially simple carbs with low nutrient content, but not following a low-carb diet per se. The recommendations in Taubes’ books go beyond the currently available evidence. In his understandable enthusiasm for his attractive and plausible hypothesis, he has jumped the gun and committed the very sin he criticized in others. Perhaps he has reconsidered: the BMJ article is more about finding the underlying cause of obesity than about pushing a specific diet. Until the evidence is in, I see no compelling reason to prefer Taubes’ low-carb recommendations to the sensible advice of Michael Pollan to “Eat food. Not too much. Mostly plants.”

Posted by Harriet Hall

Harriet Hall, MD also known as The SkepDoc, is a retired family physician who writes about pseudoscience and questionable medical practices. She received her BA and MD from the University of Washington, did her internship in the Air Force (the second female ever to do so),  and was the first female graduate of the Air Force family practice residency at Eglin Air Force Base. During a long career as an Air Force physician, she held various positions from flight surgeon to DBMS (Director of Base Medical Services) and did everything from delivering babies to taking the controls of a B-52. She retired with the rank of Colonel.  In 2008 she published her memoirs, Women Aren't Supposed to Fly.