Notes on “Why We Get Fat and What to Do About It” by Gary Taubes

This is my diary of notes taken while reading “Why We Get Fat and What to Do About It” by Gary Taubes, who was kind enough to provide me with a copy after we had an email exchange.

Please note I am reviewing this from the perspective of a person who is obese and who has been unsuccessful losing weight after multiple attempts at very low-carb (less than 20 grams of carbohydrate per day) dieting. I have, however, lost 100+ lb multiple times by ordinary calorie restriction (unfortunately regaining after each loss). 

My skepticism about many passages is unavoidably colored by my own personal experience, which seems to often disagree with Taubes’ conclusions. So I admit my impressions are colored by what I know to be true in my case.

1. The book begins with an author’s note from Taubes: “My one request is that you think critically while you’re reading. I want you to keep asking yourself as you read whether what I’m saying really makes sense.” 

This is what I did, as mentioned above. So please take my comments and criticim in that vein. It is not my goal in life to attack Gary Taubes. But I do very much want to get to the bottom of things and find out what the true facts really are.

2. On page 10, I’m reading about how when insulin is elevated we accumulate fat in our fat tissue and when insulin levels fall we liberate fat from the fat tissue and burn it for fuel and how our insulin levels are effectively determined by the carbohydrates we eat. That the more carbohydrates we eat, and the easier they are to digest and the sweeter they are, the more insulin we will ultimately secrete resulting in greater amounts of fat retained in our fat cells. That “carbohydrate is driving insulin is driving fat.”

I’m thinking, while reading this, three things:

1. I’ve read this many times before, because in Atkins’ books he explains the same thing.

2. I wonder why, then, cutting back to 20 carbs per day doesn’t help me lose weight, yet eating more than 20 carbs per day while limiting myself to 1800 calories I do lose weight.

3. I wonder why Atkins claims calories still count, even given this mechanism, and why Taubes says they don’t. But I realize I’m getting ahead of myself here because Taubes has just implied that calories don’t count so far without specifically saying so.

But my “skeptical hackles” have been raised already, because my own repeated experiences with low-carb dieting contradict the idea we’re moving forward with so far. At the beginning of his book Taubes says scientists have become “trapped in a paradigm” of believing that calories-in/calories-out is true despite evidence to the contrary. Knowing what I read from Atkins so far (and I admit I’m still just at the beginning of Taubes’ book) I am wondering if he is trapped in a low-carb paradigm, despite evidence that calories-in/calories-out works. Anyway, going on with the book…

3. P.24 – The discussion of the Pima really leaves us with no objective explanation. For example, presumably the diet of the Pima men and women were the same so why was obesity observed in greater amounts by the women? Taubes points with suprise at the fact that women were more obese even though they performed more labor, yet didn’t put forth any explanation for why the Pima men were less obese. And while sugar consumption increased with government rations, Taubes admits it was at a level far lower than in today’s America. In separate correspondence, Taubes makes the point to me that the lower consumption of sugar in Japan relative to modern-day America can explain why the obesity rate in Japan is 1/10th that of the U.S. So why, similarly, didn’t a lower rate of sugar consumption among the Pima compared to modern day America lead to less obesity among the Pima? The whole discussion seems very anecdotal, inconclusive and contains contradictions.

4. P.31 – I’m again struck by a possible contradiction in the discussion of overweight women and their thin and stunted children. The fact that the children were growing was not mentioned. Nor was the possibility that the children were simply more active, running around and playing games with their friends or engaging in sports, as children do, rich or poor. I don’t know that this explains it. But it bothers me that those possibilities are not even considered. It’s as though no other explanations are possible other than the one Taubes wants to lead us to.

5. P.39 – Taubes says, “And if undereating isn’t a treatment or a cure, this certainly suggests that overeating is not a cause.” Sounds nice. But really don’t make sense. For example, if undereating isn’t a treatment or cure because it’s unsustainable, what about eating just right instead – aiming for the amount of calories a person your sex, age and level of activity requires to maintain a normal weight? It seems that possibility is just dismissed without any serious consideration. The entire question of “rebounds” is a serious issue. But that’s not a reason to say eating too many calories and lack of activity isn’t a cause of obesity. There’s just not a connection from A to B there.

6. P.39 – In the notes, Taubes reports that WHI investigators reported that the low-fat diet failed to prevent heart disease, cancer or anything else. We do, however, know there is contradictory evidence as well – that extremely low fat diets (Ornish, Esselstyn) have been shown to prevent and even reverse heart disease. I’m hoping Taubes discusses this later in the book.

7. P.40 – Now this might be a problem with the English language in general, when trying to explain a biological or physical process with words. One can make “cute sounding pseudo-logical arguments” which are not logical at all. On this page, where Taubes is talking about the invitation to a feast he says, “You might try to eat less over the course of the day – maybe even skip lunch or breakfast…” and he goes on to say you might do more exercise and walk to the dinner, rather than drive. 

I agree. Yes I would do those things. 

But the next paragraph really took me by surprise when Taubes wrote, “Now let’s think about this for a moment. The instructions that we’re constantly being given to lose weight – eat less and exercise more – are the very same things we’ll do if our purpose is to make ourselves hungry, to build up an appetite, to eat more.” 

This is, I believe, fallacious logic. 

In my case, with the expecation of eating more, I would do the things mentioned – not to build up an appetite, but to burn off calories so eating more would not adversely affect my weight. Surely Taubes realizes that exercising doesn’t necessarily make one hungry. In fact, it can often have the opposite effect. Eating less can make one hungry – but not necessarily for more than you would have otherwise eaten during the day. 

That whole argument just seems entirely out of kilter, and detracts from the point he’s trying to make. In fact, the problem going forward is that this is the beginning of a chain of logical conclusions, one leading to the other. But the first link in the chain is a weak one, therefore the conclusions derived from the starting point are all suspect. These are the kinds of niggling contradictions which always leap out at me. What can one say about the rest of the chapter if it’s all built on a faulty premise? And what can one say about the rest of the book if it is built on a faulty chapter? Well, let’s see how much the rest of the argument and book depends on this strange and fallacious analogy he’s starting off with.

8. P.41 – Taubes says, “I want to begin with the obvservation I made in Chapter 1 that obesity associates with poverty. In the United States, Europe, and other developed nations, the poorer people are, the fatter they’re likely to be.” 

Taubes did not show such an association! He gave some anecdotal stories about some poor populations where women, in particular, were a lot more obese than men. Period. He showed no association. He didn’t even show a correlation. And he completely ignored the richer populations in the U.S. (the vast U.S. middle class) which are also plagued by obesity. In fact, the middle class wasn’t mentioned at all! I feel like screaming at this point, “What on earth is Taubes talking about here?!” How can he make such unsubstantiated leaps in his argument? Agh! Anyway, persisting…

9. P.43 – The quote from the AHA and ACSM joint guidelines on physical activity and health where Taubes quotes the article regarding daily exercise limiting weight gain – “So far, data to to support this hypothesis are not particularly compelling.” – was interesting and thought provoking. 

In my own diets over the years, I have counted my calories expended on excercise as a credit toward how many calories I can eat during the day. I have successfully lost weight this way, but it’s entirely possible the entirety of my exercise was so little it didn’t really make a difference either way. I have never been a big exerciser. 

Yet, on the other hand, you hear about people like Olympic swimmers who train hard for hours each day and pack away 10,000 calories, including lots of carbs. So isn’t there evidence from people on that extreme that large amounts of exercise does prevent getting fat regardless of what you’re eating? 

Anyway, the whole section on exercise and weight loss is interesting, and I don’t have enough personal experience with sustained exercise over long periods of time to contradict anything written. Maybe Taubes points this out later though: if you perform exercise which increases muscle mass, the body’s metabolism goes up and your body requires more calories to maintain weight. So it is a natural way of increasing your personal metabolism. On the other hand, if Taubes doesn’t believe calories matter at all, maybe he won’t mention it. I’ll wait and see and read on!

10. P.45 – This section is a particularly troubling one. Taubes’ conclusion about running and calories-in/calories-out has no foundation based on the research report he summarized. Assume the runners were thin to begin with and they eat a lot each day. As they get older, their metabolism starts slowing down. If they don’t reduce what they eat each day and maintain the same exercise regimine it make perfect sense that they gain weight as they age. Taubes doesn’t even mention here how a person’s metabolism changes as they get older here, nor does he say anything at all about whether the runners changed their eating habits as they got older. Instead he uses this observation about runners, age and weight to obliquely question calories-in/calories-out. But how can he do this without having said anything at all about the runners’ calories-in? 

Note: Later, in chapter 12, he does bring up metabolism changing as we get older, but in terms of “insulin resistance.” There he says, “You don’t get fat because your metabolism slows; your metabolism slows because you’re getting fat.” Whatever. It’s a phenomenon that could explain the runners gaining weight with age consistent with calories-in/calories-out.

11. P.46 – Taubes says, “Is it true that we can increase our expenditure of calories, burn an extra 150 calories a day, say, or go from being sedentary to active or from active to very active, without changing our diet – without eating more – and without maybe decreasing the amount of energy we expend in the hours between our bouts of exercise? The simple answer, again, is no.” 

But this also contradicts my own experience, and the experience of countless others who exercise. Exercise often makes you feel so good that instead of feeling hungry you actually feel satisfied. It’s a very common phenomenon. I wonder why Taubes is overlooking that.

(Pause to reflect. I know that low-carb eating doesn’t work for weight loss. I’ve tried it multiple times myself. So I am beginning to wonder what the point in going on here is. But I am persisting to the end since I promised. But whatever argument Taubes is leading up to seems already based on unproven premises.)

12. P.54 – Taubes does, in fact, bring up the muscle-increasing-metabolism argument I raised in (9) but dismisses it as an inconsequential increase in metabolism. If what he says is true, that is interesting.

13. P.59 – No matter how many times I read Taubes’ famous “20 calories per day” argument it frankly never makes sense to me. He himself, in his previous chapter, makes the case that the more you exercise the more you eat, etc. So why does maintaining weight over a quarter-century require mainting caloric intake to within 20 calories a day? And he didn’t mention change in metabolism over time either. What about the idea that when metabolism slows down with age (some say 1% per year after the age of 25) some people get less hungry and thus maintain their weight while others don’t lose their appetite and thus get fat? Also he assumes the weight gain is linear throughout 25 years without providing any reason to make such an assumption. This whole argument doesn’t really convince. As mentioned above, later, in chapter 12, he does bring up metabolism changing as we get older, but in terms of “insulin resistance.” There he says, “You don’t get fat because your metabolism slows; your metabolism slows because you’re getting fat.” Whatever the cause and effect, this still needs to be taken account in the 20 calories per day argument.

14. P.50 – In talking about animals, he says, “Yet species that begin their adult lives lean remain lean with little apparent effort.” Has he never seen fat cats and dogs? Animals in the wild spend almost all their time and energy looking for food. This sounds like an argument in favor of calories-in vs calories-out rather than against it.

15. Chapter 5 – Here, Taubes talks about the probable genetic factors which might lead one person (or type of farm animal) to be fat or not. All that is well and good. People are different. Metabolisms are different. That all makes sense. What doesn’t make sense is how he then concludes calories-in vs calories-out doesn’t make sense. I don’t see any statement or picture of lean vs obese twins or any of the unusual disorders he mentions at all making a case against calories-in vs calories-out. To me, all those examples say is that different people have different metabolisms, and therefore different calorie requirments. And that different people tend to have their fat distributed differently on their bodies. How that disputes calories-in vs calories-out is lept to as an assumption without any basis. It’s as though I pointed to the sky and said, “Look – there is a wispy white cloud and over there is a dark, dense cloud. Therefore calories-in vs calories-out makes no sense.” Merely showing a bunch of photos and then leaping to a conclusion without any basis makes no sense either.

16. P.78 – Taubes write, “To a great extent, if not entirely, the energy we expend from day to day and week to week will determine how much we consume, while the energy we consume and make available to our cells will determine how much we expend. The two are that intimately lined.” 

The main problem with this argument, regardless of how true or untrue this may be, is that Taubes repeatedly, throughout his book, claims that if you eat less you get lethargic and if you move more you build up an appetite and here he is implying if you eat more you want to expend more… 

Anybody who is obese knows none of this makes sense. How many obese people know they are obese because they are sedentary and still eat too much? Raise your hands! And how many of us have also experienced a sense of satisfaction from exercise that helps us resist eating more? All these factors are left out of Taubes’ argument. In fact, his argument taken to the logical conclusion would be that nobody gets obese because we all naturally move enough to expend the energy we take in, and we know that is not true. I know he is leading up to carbohydrates, but it’s not true if you eat almost no carbohydrates but too many calories you will lose weight. It’s just not true. I’ve tried it. I’m afraid every chapter I’m reading falls back on my personal experience with this.

17. Chapter 8 – Here Taubes is talking about the dangers of the calories-in/calories-out theory. He writes, in part, “It has done incalculable harm. Not only is this thinking at least partly responsible for the ever-growing numbers of obese and overweight in the world – while directing attention away from the real reasons we get fat – but it has served to reinforce the perception that those who are fat have no one to blame but themselves. That eating less invariably fails as a cure for obesity is rarely perceived as the single most important reason to make us question our assumptions…” and “Rather, it is taken as still more evidence that the overweight and obese are incapable of following a diet and eating in moderation.” 

I don’t know where to begin here. 

Taubes knows very well, and has stated so publicly, that the important point is what is true and not true, not whether people like to hear the facts.  We are talking about science here. So whether obesity is or is not the fault of the obese is completely besides the point. If it is, it is. 

Nevertheless, the assumption that Taubes is making is false. Not everybody who believes in calories-in/calories-out blames the obese person. There is also recognition that there are nearly irrestible physiological reasons why people almost inevitably rebound after losing weight by calorie restriction (as I have, multiple times). For example, the number of fat cells in our bodies don’t go down. Fat cells are alive and send signals to the brain that they are being starved. This could be one of the many reasons why people rebound. 

And there very well could be psychological reasons preventing a person from getting out and exercising more. As an obese person, I don’t deny that. 

The point here is just that Taubes argument that calories-in/calories out is “bad” partly because it points to behavior and at least partially blames the victim for not getting his or her eating and exercise under control is not a statement of science. It’s not a reason to dispute calories-in/calories out. In other words, just because the truth may not be what we want to hear doesn’t mean it isn’t true. His argument is out of place in this book, and also out of place from what Taubes says in other places.

18. Chapter 10 on “Lipophilia” is, I admit, an interesting read. It doesn’t lead me to any conclusions, one way or another though, except to say that “all people are different” which is something I think most of us already realize. We all know of people who just naturally stay thin and people who struggle with weight their whole life. It seems to me that is all this chapter is really saying. Which raises another question in my mind. Taubes has said in his blog that weight control via carbohydrates will work for everybody and it isn’t a matter of what works for one person might not necessarily work for another person. I believe his chapter 10 somewhat contradicts that. It also contradicts my own experience since I have never been able to lose any substantial amount of weight even by keeping carbs below 20 grams per day. Yet, it’s an interesting chapter on some historic research.

19. Chapter 11, “A Primer on the Regulation of Fat” is the most interesting chapter so far. It explained a lot to me about fatty acids and triglycerides which were never clear in my mind. Unfortunately there is still the nagging fact that low carb dieting doesn’t help me lose weight, so I’m left with a paradoxical feeling after reading that chapter. It seems to make sense when explained in words. But there must be a logical flaw somewhere because I know from experience it doesn’t work, while low calorie eating (not starvation, but limiting to 1800 calories per day) does work, regardless of the amount of carbs I eat. So how to explain this contradiction? 

I’m still left with the impression I’ve always had that people who are successful on low-carb diets are so because it is true that reducing carbs suppresses appetite (which Taubes later briefly mentions in the book). I’ve noticed that myself. I suspect that for people who lose a lot of weight low-carbing it’s because their appetite is naturally suppressed enough for them to lose. But for others, like me, while we experience appetite suppression, it’s not enough to reduce caloric intake enough to lose substantial amounts of weight.

20. Chapter 12, which discusses insulin resistance, also is an interesting chapter. Atkins talks about this too, but he doesn’t claim that caloric intake is irrelevant. In that sense, Atkins research corresponds better with my personal experience than what Taubes claims.

21. Chapter 16, “A Historical Digression on the Fattening Carbohydrate” is also a particularly fascinating chapter as Taubes goes through a recitation of diet research and recommendations since the middle of the 19th century to about 1960 at which time he says all the conventional wisdom about the fattening effects of carbohydrates was tossed overboard in favor of calories-in/calories-out. 

It is a compelling, interesting, broadly-researched chapter. So I’m again left wondering – why doesn’t it work for me? 

In this chapter he does say that some people are fine with carbohydrates. But on the other hand he doubts the idea that there are any other causes other than reducing carbohydrates which will aid in losing weight. He says that all the other forms of diet, such as a typical low-calorie diet, naturally also result in reduced carbs, and that is the factor. This is not a controlled variable experiment though, and it doesn’t explain the paradox I have personally experienced. 

When I am on a low-calorie diet of 1800 calories/day most of my calories are from carbs. And I lose weight. Yes, I regain after about two years, but I do lose weight for those two years – typically more than 100 lb. Yet if I try eating less than 20 grams of carbs per day – much much much less than the amount of carbs I would eat on a low-calorie diet, I do NOT lose any substantial amount of weight. 

So I can’t help but feel that Taubes has it backwards again. It seems the reduced hunger in people who restrict carbs helps them control calories and for some that is enough. But it isn’t enough for everybody. Some of us, regardless of how few carbs we eat, still eat too many calories to lose weight. I can’t understand why Taubes doesn’t acknowledge this other side of the coin as at least a possibility.

22. P.174 – Taubes writes, “If you restrict only carbohydrates, you can always eat more protein and fat if you feel the urge, since they have no effect on fat accumulation.” All I can say is I wish that were the case. Why can’t I lose weight when I try it? The logic leading up to this statement in preceeding chapters sounds so compelling. But it just doesn’t seem to be so. I’m curious also why Taubes doesn’t mention Atkins also denying this. Even the current Atkins site specifically says you can’t eat too many calories and lose weight. They clearly say if you aren’t losing weight on a low-carb diet try restricting your calories too. I am curious about what Taubes thinks about the Atkins group’s suggestion and why they arrived at that conclusion, while Taubes did not.

23. P.186 – This is actually a good argument here, where Taubes is talking about the use of statins and how they reduce LDL and also help reduce heart attacks. Yet diets which reduce LDL do not appear to reduce heart attacks. Taubes writes, “The fact that the drugs known as statins lower LDL cholesterol and prevent heart disease does not necessarily imply that they prevent heart disease because they lower LDL. Consider aspirin: it cures headaches and prevents heart disease, but no one would ever suggest that aspirin prevents heart disease because it cures headaches. It does other things as well, as statins do, and any one of these other things could be the reason why either drug prevents heart attacks.” 

That is proper logic about causality. I wish Taubes would use the same reasoning when describing weight loss on low-calorie diets as being due to reducing carbs in the diet. Again, I am coming back to the point that in my own experience, merely reducing carbs without reducing calories fails to help me lose weight.

24. P.205 – And here, finally, is the catch. Taubes mentions studies begun in 1956 and 1957 which reported that a small proportion of their obese patients failed to lose any significant fat even though they faithfully avoided fattening carbohydrates. He goes on to quote one of the researchers saying, “It is quite simply, and sadly, that a point of no return has been reached.” So what Taubes is saying is that low-carb eating must reduce body fat – unless it doesn’t because it’s too late for the individual. But this clearly  contradicts my own experience where I can lose weight by controlling calories but cannot by controlling carbs. So there is something wrong with the observation here.

The book is an interesting and fascinating read. And I am grateful to the author for a chance to read it and report my views on it. But I’m reminded of one of Murphy’s Laws: for every difficult problem there is a simple, easy to understand, wrong answer.



Notes on “Why We Get Fat and What to Do About It” by Gary Taubes — 9 Comments

  1. Hi, Doug, I would like to read your notes based on Gary Taubes’ book; but I have used Format/Style/Bigger 5 times, the max, and I still have to use a magnifying glass which is too straining on my eyes. So, guess I have to pass on this one. Sorry, Dave

  2. Dave,I’ll send you my notes by separate email with enlarged fonts.doug

  3. I’ve been quite successful over the last year with the Weight Watchers variation of calorie counting. That alone makes me unconvinced by the general thrust of Taubes’ arguments (haven’t read the book yet though, only reviews). I’m old enough to remember a time when obesity was rare: in my high school class of over 600 there was only one really obese kid. Try looking at an old photo of a public event like a baseball game from a few decades ago – how many fat people do you see?People didn’t eat low carb diets back then, and yet fat people were unusual. Low carb diets may be a useful weight loss tool for some people who are already overweight, but IMHO they certainly aren’t the only key to a healthy body

  4. How much protein were you taking during your 20g carb experiment? (% of calories?)Also, how long did you do the 20g carb experiment for?Thanks you.

  5. I varied my protein intake depending on recommendations from the Atkins group. Taubes himself says it doesn’t matter though, and that’s the topic of my discussion here. Some Atkins people said I was eating too much protein so upped my fat intake and lowered my protein intake to, about, 25% of calories I believe. I’d have to look up old FitDay numbers to be sure. The Atkins site is all over the place on this though. They say you can eat too much protein, too much fat and even too many calories. So there really is no consistent advice there.I tried the 20g carb experiment for at least 3 months.Honestly, it just doesn’t work. I just eat too many calories.doug

  6. A followup note here:

    After my blog post, since May of this year, I’ve lost 56 lb so far by going vegan, low fat (e.g. Ornish, Esselstyn) and have never been healthier or had more energy.

  7. Pingback: Doug’s Diet Q&A – everything you didn’t want to hear about diets | Doug-at-Large

Leave a Reply

Your email address will not be published.