Lierre Keith suffers from numerous chronic health problems. Unable to secure a diagnosis for most of them, she decided that the vegan diet she had followed for twenty years was to blame. But she wasn’t content to add a few animal products back to her diet. Instead, she set out to prove that healthy diets require copious amounts of animal foods and that small-scale animal farming is the answer to sustainability. To prove it, she has cobbled together information from websites (yes, she actually cites Wikipedia!) and a few popular pseudoscientific books.

It’s next to impossible to review this book; it is so packed with misinformation and confusion that refuting the claims could be another book itself. This is a long post, and it doesn’t begin to address all of the problems in The Vegetarian Myth.

I read the section on nutrition first. Since it’s my area of expertise, I figured it would give me some idea of the quality of her research and analysis. But quality isn’t at issue here because there is no research or analysis. Keith doesn’t bother with primary sources; she depends almost exclusively on the opinions of her favorite popular authors, which she presents as proof of her theories. For example, when she writes about evolution as it affects dietary needs, and suggests that “the archeological evidence is incontrovertible,” she is actually referencing the book Protein Power, written by two physicians who have no expertise in evolution or anthropology. It’s a neat trick, of course, because we have no idea where the Protein Power authors got their information. By burying all of the actual studies this way, she makes it laborious for readers to check her facts.

I doubt she did this on purpose. And I don’t think she was being sloppy or lazy, either. She just doesn’t understand how complex the research is and she certainly doesn’t know much about basic nutrition. Worse, her conclusions are indebted to the Weston A Price Foundation, a non-credible group that bases its recommendations on the opinions of a dentist who wrote up his observations of indigenous populations in the 1930s.

Keith makes a big point about the fact that humans now eat foods—grains—that our Paleolithic ancestors rarely ate. But she never discusses the fact that dairy, a food she heartily endorses, falls into the same category. In fact, while grains could be gathered, ground and consumed by our ancestors, dairy is 100% dependent on agriculture. The fact that normal human development—throughout most of the world, at least—results in a decreased ability to digest dairy foods, should provide a major clue that humans did not evolve to consume them. None of this gets even a mention in the book.

Instead, we get page after page of contradictions, fabrications, and misinterpretations. Not surprisingly, given the sources she uses, Keith is woefully confused about fats. She believes that saturated fat is needed for absorption of vitamins and minerals, that polyunsaturated fat is “low-fat,” and that we have a dietary need for cholesterol. In fact, we have no dietary need for either saturated fat or cholesterol—there is no RDA for either. The liver makes all the cholesterol our bodies require. And the two essential fatty acids required by humans—both unsaturated—are found in plant foods.

On page 172 she suggests that fat intake has dropped by 25% over the past 15 years. Thirty pages later she says it has fallen by 10%. You might think that this discrepancy would send her to the actual data, in which case she would have found that fat intake has increased over the past 15 years. Among Americans, total fat intake is around 33% of calories and a good one-third of that is saturated fat—so her belief that Americans consume 30% of their calories as polyunsaturated fat is also wrong.

There is a long section on eating disorders with the popular claim that vegetarian diets are a cause. But the experts who have done research in this area point out that girls with anorexia may choose vegetarian diets as a way of masking their eating behavior. There is no evidence—according to these experts—that girls who are vegetarian or vegan are any more likely than anyone else to develop an eating disorder.

Like most anti-vegetarians she is vehemently against soy, insisting that it reduces testosterone levels and therefore male libido (there is no evidence of this) and she speculates that African-American girls reach puberty faster because they are more likely to be enrolled as infants in food assistance programs like WIC and therefore, to be fed soy infant formula. It’s true that African-American babies are less likely to be breastfed, but I couldn’t find any indication that they consume more soy formula. And, recent research has linked animal protein to earlier puberty and cow’s milk to excessive growth in children. There is no evidence that soy is involved in either; in fact, recent preliminary research suggests that soy could slightly delay puberty in girls.

On page 227, she notes that “Mark Messina, a champion of soy, thinks the Japanese eat 8.6 [grams of soyfoods] per day,” or less than a tablespoon. Really? Well, I happen to be married to Mark Messina, so I have a fairly good idea of what he “thinks” about soy intake. But even if I didn’t know him, I could read his 2006 analysis of soy intake data that was published in the peer-reviewed medical journal Nutrition and Cancer. Apparently, Keith didn’t or she would have seen that Asian soy intake is the equivalent of 1 to 1 ½ servings or more per day. Why did she get this so wrong? It’s because she doesn’t understand that there is a difference between soy protein intake and soy food intake. A cup of soymilk contains around 7 grams of soy protein, so the 8.6 to 11 grams of protein that the Japanese typically eat is equal to at least a serving per day.

(There is much more on soy in this book, covering all the usual criticisms, but I’m going to address that in another post; it’s a big topic.)

I’m less able to evaluate her discussion of the environmental consequences of animal farming, although it seems reasonable to assume that she gets as much wrong in this section. Some things did jump out. Notably, she points out that ten acres on Polyface Farm can produce enough food to feed 9 people for a year. But on his blog Say What Michael Pollan, mathematician Adam Merberg performs calculations which suggest that Polyface requires more calories in feed (for the chickens) than it produces in food. The numbers aren’t nearly as egregious as those for factory farming, but they suggest that there is no such thing as truly sustainable meat production. (For more on environmental questions associated with Polyface, I highly recommend Adam’s review of the Omnivore’s Dilemma.)

It’s true that Keith is vehemently against factory farming for both environmental and ethical reasons. But she remains convinced that food production is impossible without animal waste, which ignores the value of leguminous cover crops in fixing soil nitrogen.

But Keith didn’t give up veganism because of concerns about the environment; she gave it up because she didn’t feel well. Much of where she goes wrong is in confusing food cravings with biological needs. When she decides to eat her first bite of tuna fish after 20 years as a vegan, she says “I don’t know how to describe what happened next. […] I could feel every cell in my body—literally every cell—pulsing. And finally, finally being fed. Oh god, I thought: this is what it feels like to be alive.”

This, more than anything, shows that Keith’s conviction about her need for meat has to do with something other than nutrition—because food just does not work like that. Eating a bite of tuna—no matter how deficient you might be in a nutrient that it supplies—does not cause all of your body cells to start pulsing. It wouldn’t cause you to feel too much of anything. (At the very least, you’d have to digest and absorb it first!)

It’s true that some vegans are too skinny and are not healthy. They don’t eat enough fat or enough calories or they refuse to supplement with vitamin B12. Or they make any of the many mistakes that people make with all types of diets. But Keith insists that a vegan diet will damage us all–she is 100% certain of this–and it is simply not true. It’s not supported by nutrition science and it isn’t supported by simple observations of long-term vegans, not to mention vegan-from-birth children.

Interestingly, she never tells us what she ate when she was vegan or what she eats now that she is an omnivore. Except to say that she used to eat “all carbohydrates” (All? No wonder she was sick) and that she now eats mostly animals and their secretions. And while she thinks she understands “moral vegetarians,” she reveals her total disconnect from a vegan ethic with three short sentences in the closing paragraphs of the book, “I have looked my food in the eye. I have raised some of it myself, loved it when it was small and defenseless. I have learned to kill.”

This is ultimately a sad book. Lierre Keith has suffered from multiple health problems all of her life and was desperate to find an answer. She landed on vegetarianism and then spun a tale to support her theory. Her intent seems heartfelt; she sees herself very much as a savior of vegetarians and wants us to learn from her mistakes. And the book has been widely embraced by those who want to believe that meat-eating is healthy and just. The problem is that there is truly nothing in this book that accurately supports that conclusion.