Best-selling science writer Gary Taubes should have expected serious pushback when he recently declared to a room full of Tufts University nutritionists that Americans became fat from fruit — not burgers. Yet, he tells me, he was a bit surprised by the negative reaction:

“I think I may have insulted much of the faculty by giving the lecture from the standpoint that I knew the truth and they don’t.’’The “truth’’ as Taubes asserts in his latest book, “Why We Get Fat,’’ is that weight gain isn’t about eating more calories than we burn; rather it’s about basing our diets on starches, cereal, sugar — and yes, berries — rather than on steak, chicken wings, and fried eggs.

Can anyone say Atkins? Taubes, who caused a national stir with a 2002 New York Times Magazine piece questioning the vilification of dietary fat, admits that’s the plan he is advocating. He lost weight on Atkins more than a decade ago and has been off and on it ever since. “I eat scrambled eggs and bacon and lose weight effortlessly,’’ he tells me.

It’s all about hunger, he says. Eating carbohydrates drives up the hormone insulin — which rapidly clears glucose (the component of all carbs) from the blood and shuttles it into fat cells. Blood sugar levels drop and we get hungry again, making it easy to overeat.

“It’s hard to get people to overfeed on fat,’’ Taubes contends, whereas studies have shown that people on high-carbohydrate diets can easily overeat because they don’t feel full for very long.

Exercise presents the same hunger problem, he says. Sure, we burn off calories when we sweat, but then our bodies compensate by making us hungrier to replace all that expended energy.

Taubes insists, though, that the calories in, calories out theory of weight gain and weight loss is just plain incorrect.

Of course, most nutritionists — which Taubes is not — would say that he’s the one who’s wrong.

“The problem is people’s inability to know how many calories they burn and eat,’’ said Dr. George Blackburn, associate director of the Division of Nutrition at Harvard Medical School, in a previous interview with the Globe. “If you put a person in a metabolic chamber, where you know exactly what they eat and what they burn, the calories in, calories out idea is always reconfirmed.’’

Taubes is keenly aware that he’s controversial. “I get accused of being too rigid,’’ he says. “I think the medical community is scared to embrace a diet high in saturated fat.’’ Or one that rejects high-fiber whole grains and fruits bursting with antioxidants.

Since Taubes prefers to write about science rather than dietary advice, the eating plan in his book, developed by Duke University Medical Center, is buried as a seven-page appendix. It calls for eating red meat, chicken, fish, and eggs, as well as salad greens and nonstarch vegetables such as cucumbers.