(I've also written a shorter summary of Taubes's longer previous book, Good Calories, Bad Calories.)
A brief note, August 23, 2014: While I credit Good Calories, Bad Calories with getting me interested in nutrition and health in the first place and still believe that many mainstream dietary beliefs are poorly supported, I've since become less enthusiastic about some of Taubes's core claims. This change was due in large part to the writings of Stephan Guyenet at his blog Whole Health Source, including his very public spat with Taubes concerning the scientific evidence supporting the "insulin theory of obesity". Seth Yoder's critical review of Nina Teicholz's The Big Fat Surprise (part two here) also highlights many instances where Taubes allegedly cherry-picked data or misrepresented research (which Teicholz subsequently plagiarized). With that said, I am still super-excited about the work that Taubes and Peter Attia are leading at NuSi and am optimistic that the results will be reported honestly, whatever they are.
Original review, March 6, 2011: Even though it's based on almost entirely the same material as his earlier Good Calories, Bad Calories, I found Gary Taubes's Why We Get Fat to be entertaining and informative, largely due to the different route that the author takes to reach what are essentially the same conclusions. GC,BC was structured in three sections: the history of the "fat-cholesterol hypothesis" (that dietary fat increases the risk of obesity and heart disease), then that of the "carbohydrate hypothesis" (that carbohydrates are instead responsible), and finally an examination of the processes responsible for weight regulation. In WWGF, Taubes opens with an attack on the "calories-in, calories-out" view of weight regulation and then explains and argues for an alternate theory.
GC,BC's argument against "calories-in, calories-out" triggered a sizable amount of criticism, with many reviewers claiming that Taubes was denying conservation of energy. Taubes in an interview, on being confronted by a trainer from The Biggest Loser on Larry King Live: "I remember sitting there at the time thinking, 'I did study physics at a pretty good college; I think I understand this.'" He seems determined to head off the accusation this time around with chapters entitled "Thermodynamics for Dummies, Part 1" and "Part 2" that attempt to make his reasoning clear. Taubes is not denying that a person must take in more calories than they expend in order to gain weight; rather, he's stating that the statement is so obvious as to be useless when searching for the causes of obesity. It's no more meaningful than answering the question, "Why is this room getting crowded?" with, "Well, because more people entered [it] than left it." (p.74-5)
Taubes's point here remains the same as in GC,BC: despite what we've been told by organizations like the World Health Organization and National Institutes of Health, long-term weight loss can't be brought about by just eating less or exercising more to bring about a caloric deficit. When the processes regulating energy use and storage aren't functioning correctly, the body compensates for the deficit by increasing appetite or decreasing the rate at which stored energy is burned to bring the system back into equilibrium (and in the process, makes it extremely difficult or impossible to keep up the diet or exercise regimen). Taubes states his case with historical examples (a long list of impoverished populations containing malnourished children with obese mothers, with the implication that obesity can occur even in the presence of a inadequate supply of calories), research (including the NIH's Women's Health Initiative studies in the 1990s, where women on a low-fat diet that lowered their caloric intake lost an average of only two pounds after undereating for eight years and actually ended up with an increased average waist circumference, implying that the lost weight was muscle rather than fat), and with appeals to the reader's personal experiences:
Imagine you're invited to a celebratory dinner. The chef's talent is legendary, and the invitation says that this particular dinner is going to be a feast of monumental proportions. Bring your appetite, you're told — come hungry. How would you do it?
You might try to eat less over the course of the day — maybe even skip lunch, or breakfast and lunch. You might go to the gym for a particularly vigorous workout, or go for a longer run or swim than usual, to work up an appetite. You might even decide to walk to the dinner, rather than drive, for the same reason.
Now let's think about this for a moment. The instructions that we're constantly being given to lose weight — eat less (decrease the calories we take in) and exercise more (increase the calories we expend) — are the very same things we'll do if our purpose is to make ourselves hungry, to build up an appetite, to eat more. Now the existence of an obesity epidemic coincident with half a century of advice to eat less and exercise more begins to look less paradoxical. (p.40, with this example credited to Chris Williams)
Taubes then provides a succinct and (he claims) non-controversial explanation of how fat is regulated in humans. Here's my attempt to summarize the key points:
This leads to a straightforward solution: to lose weight, modify your diet to reduce blood sugar and insulin levels, which will make your body use its existing fat store instead of storing incoming calories immediately. Your insulin level is determined primarily by the quantity and quality of the carbohydrates you consume. Quickly-digestible carbohydrates have the biggest effect: Taubes lists "anything made of refined flour (bread, cereals, and pasta), liquid carbohydrates (beers, fruit juices, and sodas), and starches (potatoes, rice, and corn)." (p.134)
Much of the rest of the book mirrors the first two sections of GC,BC, with Taubes diving into a history of how dietary beliefs have changed over the past two hundred years (albeit, with much less detail than in GC,BC). Interestingly, researcher Ancel Keys doesn't get a mention this time. I found the explanation of the transition from the long-held-until-the-1960s low-carb conventional wisdom to the 1970s-and-onward low-fat movement more satisfying in this book. Taubes describes the science of fat regulation as being omnipresent in physiology, endocrinology, and biochemistry journals at the time but rarely crossing over into medical journals. (p.159) Obesity came to instead be viewed as an eating disorder in the 1960s, with the result that much of the earlier beliefs were ignored.
Taubes describes a health campaign launched by the National Heart, Lung, and Blood Institute in 1984. $115 million was spent on a decade-long clinical trial to test whether eating less saturated fat would reduce heart disease, with no connection found. An additional $150 million was used to test the efficacy of a cholesterol-lowering drug in reducing heart disease, with successful results. Rather than do additional time-consuming and costly tests before making recommendations, and despite the results from the first trial, the researchers made the assumption that a low-fat, cholesterol-lowering diet would be likely to produce the same results as seen with the drug (which itself depends on the assumption that the drug was effective because it reduced cholesterol). (p.182-3)
It's worth noting that Taubes additionally denies low-fat diets' abilities to improve cholesterol levels; he describes the negative effects of replacing saturated fat with carbohydrates: increased triglycerides and decreased HDL ("good cholesterol"). (p.187-8) Taubes acknowledges that saturated fat increases the level of LDL ("bad cholesterol") but states that carbohydrates make LDL particles smaller and denser, increasing the risk of atherosclerosis. (p.193)
Taubes opens the final chapter, entitled "Following Through", with the sentence, "This is not a diet book, because it's not a diet we're discussing." Nevertheless, Why We Get Fat is aimed at a different audience than Good Calories, Bad Calories, and acknowledging as much, this chapter contains advice for readers looking to put the book's conclusions into practice. It is, in fact, followed by an example diet based on guidelines from the Lifestyle Medicine Clinic at the Duke University Medical Center, which is run by Eric Westman, the author of the latest Atkins book. I'd summarize the diet as:
This diet obviously won't work for vegetarians or vegans (although the book's key point, that easy-to-digest carbohydrates should be avoided, can embraced by anyone). A diet as meat-heavy as this may even be tough for many other people to stick to — looking around my corner health-concious grocery store, I'd estimate that at least 90% or so of the food on the shelves and in the coolers is off-limits. It's frustrating that the difficulty springs not from this being a difficult diet to stomach (which is a matter of personal opinion, of course), but rather from the changes to the dietary landscape brought about over the last half-century; I'd have no trouble finding a large variety of acceptable food for a low-fat, high-carbohydrate diet. Fortunately, this is something that books like Why We Get Fat hope to change.