Gary Taubes and his flawed logic — 35 Comments

  1. Japan: Using the WHO criteria Japan has the lowest rate of obesity among the OECD member countries at 3.2%. However, as Asian populations are particularly susceptible to the health risks of excess adipose tissue the Japanese have redefined obesity as any BMI greater than 25. Using this cut off value the prevalence of obesity in Japan would be 20%, a three fold increase from 1962 to 2002.

  2. We still must compare apples-with-apples in the same chart between countries though.doug

  3. Hi Doug, A few years ago I had the opportunity to look into this when I was in Japan. You can do it yourself and probably should if you’re going to talk about this. From the the National Nutrition Survey in Japan showed the trend of sugar intake. YEAR INTAKE 1946 0.5 g/day 1950 7.2 1955 15.8 1960 12.3 1965 17.9 1970 19.7 1975 14.6 1980 12.0 1985 11.2 1990 10.6 1995 9.9 1999 9.5 Now high fructose corn syrup — a Japanese invention — may have replaced the sugar, but still, these numbers are miniscule compared to the 180 or so grams per day (150 pounds per year) of “caloric sweeteners” that is the average in the U.S. Since you’re in Japan maybe you can look this up for us and see whether these numbers I had were correct and get back to us. That would be a great service you could do. Merely pointing out that Japanese sweets are, well, sweet, doesn’t say much of relevance at all. Most of U.S. consumption is not actually in sweets, but in the sugar and HFCS that permeates the food supply in general and, of course, in sodas and fruit juices. Keep up the good work.SF

  4. Thanks for your added info.My point about Japanese sweets being sweet was just part of it. The other part was that Japanese people actually eat these things all the time and it so it is silly of Taubes to say Japanese people have no sweet tooth. It’s just something made up on his part.In addition to sweets, there is sugar in just about everything. So it permeates the overall food supply as well. During the times I did low-carb it was very hard to find foods that didn’t have sugar added. Even things you wouldn’t normally think of as being “sweet” such as prepared salads have sugar as a high-order ingredient.Here is an interesting article on the history and use of sugar in Japan from Kikkoman corporation, the well-known soy sauce maker: didn’t realize that sugar has been used in Japan since the mid 8th century!In agreement with my perception that sugar is typically used throughout its cuisine is this statement from the article:”Used not only in beverages and confectionery, sugar is an ingredient in many other recipes as well. Japanese cuisine has earned a reputation for its many sweet dishes; for example, sugar is an important flavoring in sukiyaki and is often used in preparing simmered vegetables.”The Kikkoman data disagrees with the data in your table.Your table’s latest year in 1999 and Kikkoman’s data is from 2001, but that’s only 2 years apart so it’s probably close. Kikkoman says the per-capita sugar consumption is 18 kg per year. Your 1999 value of 9.5 g/day multiplies out to just 3.5 kg/year. So there is a considerable discrepancy there. Perhaps we need to investigate these numbers some more.The article also points out the other kinds of sweeteners used in Japanese cuisine, such as wasanbonto and mirin and others.I think this, combined with the fact that Japanese undeniably eat lots of high-glycemic white rice pretty much lays waste to Taubes’ statements about Japanese diet in that interview.My complaint about Taubes is that he is ignoring data which doesn’t fit his theory, which is not what an educated person reporting on science should do. What a scientist (or anybody interested in getting at the truth) would do is adjust or discard a theory shown incorrect by contradictory data – not ignore the data or try to make up facts which aren’t true.If he wants to modify his theory and say that refined carbs, such as white rice, per se are not the cause of obesity and it’s fine to eat a diet with lots of white rice if you cut your sugar intake back to just 18 kg per year that would be a whole different discussion. But it’s up to him to change his claim. Right now it seems he is just making things up.Thanks,doug

  5. Hi Doug, Brief point: even the 18 kilograms per year per capita that Kikoman claims for sugar is only 40 pounds (rounding upward) per capita per year. That still supports Taubes’ point more than your own. Doesn’t it. Americans eating some 150 pounds per year of caloric sweeteners. Japanese 40 pounds. Couldn’t that explain it? It seems to me your logic isn’t all that great either. Again, since you’re in Japan and presumably speak Japanese, which I don’t, why don’t you do a little more research before you go throwing stones.Sidney

  6. Sidney.Thanks for your post.Taubes made two claims in the interview:1. That Japanese “weren’t eating white rice, they weren’t eating highly refined wheat.”2. That Japanese “eat virtually no sugar.”Both of Taubes’ claims are blatantly false. How does the fact that Japanese eat only 40 lb of sugar a year in any way support either of the two assertions Taubes made?And shouldn’t the issue with the white rice alone cause an honest scientific researcher to back off or modify his claims about carbohydrates?The facts are simply contrary to what Taubes is strongly and concretely asserting.So, I’m sorry, but I don’t see the equivalence. If Taubes wants to say that it’s ok to eat multiple servings of white rice per day as long as you keep your sugar intake below 40 lb. a year that would be a different discussion. But he’s not saying that, is he? He’s saying that the carbs are bad and mis-stating the reality about what Japanese eat to people who question him about it because it doesn’t fit into his theory. That’s not an honest scientific discussion.doug

  7. Here is some more interesting data on per-capita global sugar consumption from the World Health Organization (, which also corresponds with Kikkoman’s data:The 2005 U.S. per capita sugar consumption was 31.3 kg.The 2005 Japan per capita sugar consumption was 18.8 kg.So the U.S. consumption is just 1.66 times that of Japan. Is that difference alone enough to state that Japanese “eat virtually no sugar” compared to the U.S.? I don’t think so.Also, note that Canada’s per capita consumption in 2005 was 44.2 kg, which is 1.41 times that of the U.S. Yet the obesity rate in Canada ( is just 14.3% compared to 30.6% for the U.S. How does the fact that Canadian’s eat more sugar jive with the data which shows the obesity rate in the U.S. is more than twice that of Canada?And what about Switzerland? The per-capita sugar consumption is a whopping 70.3 kg, which is 2.2 times that of the U.S. But the obesity rate in Switzerland is just 7.7%, which is 1/4th that of the U.S.And there is still the whole white rice issue which needs to be explained.As I wrote earlier, one of my favorite Murphy’s laws is, “For every complicated problem there is a simple, easy-to-understand, wrong answer.” Taubes is saying it’s just carbohydrates and blithely ignoring any data which doesn’t fit his theory. This should upset anybody who is trying to figure out the truth.doug

  8. Hi Doug, This will be my last post then, as my sister used to say when we were five, I’ll let you get the last word in and we can both get back to reality. The data you site are for sugar alone. If you look at the asterick on the website it says very clearly the numbers r “are for centrifugal sugars (cane and beet sugars) only.” The problem is high fructose corn syrup is another form of sugar and those numbers have to be factored in as well. According to wikipedia, high fructose corn syrup consumption in Japan is a quarter of all sweetener consumption and in the U.S. it is half. means that the U.S. numbers you gave above must double and the Japanese numbers must be increased by a third. For the numbers you gave, that would put U.S. consumption at around 64 kg and Japanese consumption at 24 kg, or about 40 percent. According to Taubes’s book. the 24 kilograms would be about what we were eating in the U.S. in the middle of the 19th century. I agree with you, though, if the Swiss eat 70 kilograms a year of sugar it would seem to refute Taubes notions, but that number could be incorrect. The WHO says the Swiss were eating 47 kilograms per capita in 2000, 52 kilograms in 2002 and then 70 kg per capita per year three years after that. That’s a fifty percent increase in five years — from 47 kg to 70 kg. There’s either something wrong with the numbers or maybe the Swiss are an obesity epidemic waiting to happen.With fondest wishes,Sidney

  9. Hi, Sidney.OK. I will take the last word, which is something I have never been able to accomplish with my own sister. :)I was talking about sugar because I was replying to Taubes and he talked about sugar. The effects of fructose corn syrup or honey or whatever were not part of Taubes’ answer and I was just commenting on what Taubes said. He only mentioned sugar when answering the interviewer’s question. At the end of the day, though, we are still talking about Japanese eating lots of refined carbs – something Taubes has not yet explained away.Taubes was asked about the apparent contradiction where, in countries like Japan, the obesity rate is extremely low but people eat lots of carbs.It was a call-in show and according to the transcript the caller asked, “How does he explain the difference in the Asian cuisine where they eat a lot of rice, and they’re all thin all the time. You don’t see fat Asians. How do you explain that?”So that was the question, to which Taubes begins his reply with, “Yeah, that’s a typical question. I mean issue I’ve been getting that question. And I’ve asked them myself, the experts over the years.”So it’s not something he was suddenly stumped with or replying to off the cuff. It’s been something he’s supposedly been thinking about for years.Yet he then goes ahead and claims the reasons are (1) that Japanese are not eating white rice and (2) Japanese do not have sweet tooth’s and eat virtually no sugar.Both of his statements are completely false. It raises questions about his method of scientific inquiry at the least, and also his honesty I think. Instead of modifying his theory to fit the actual data, he’s ignoring or changing the data!I notice you didn’t mention the rice part yet at all. That’s a major major part of daily Japanese carbohydrate consumption. The caller’s question was answered by Taubes with false information. Shouldn’t that cause anybody to wonder whether there is a correct answer which also fits with Taubes’ theories?doug

  10. Doug, Historically, I guess the Portuguese were the early birds to bring the castile in that has a lot of sugar. When I came here in 1969 the western style bakery was practically non-existent. There was Anderson’s that developed into a chain. We also had the craze buying your own electric bread maker which quickly fizzled. Now,when we look for a bakery, we don’t have to go far at all. Luminé, the chain, has a number in one store. In other words they are all over the place after 40 years. I wouldn’t vouch for anything Taubes declares. All I am saying is that at the initial invasion by Macdonald’s, the lay of the land, changed rapidly, not only caused by other outsiders but the locals almost immediately started imitating the invaders with there own similar brands of quickie food fixing stalls. I doubt if any surveys during this period could have an accurate base, the change was so rapid. Now more than ever, today, I see, especially more Japanese women than men on the streets of Tokyo, looking more and more like the women in Wichita, Kansas, at least from a rear view, pushing carts in the chain grocery stores–lumbering dump trucks ahead at each. Once again Lumine in Ogikubo, 1st. floor. Walk down the row of glass cases. Tons of icing, cakes, chocolate, candies, with Starbucks at the far end with scones and doughnuts, all competing on the same floor. The second floor is pasta heaven. And the 5th floor, if he or she eats only there, will guarantee an overweight, undernourished human being. The changing Japanese diet , within 10 years, will close the percentage gap for Japan and the US regarding obesity. This is one person’s observation–not based on any lengthy, “thorough” research. I am not impressed with any so-called research until the source has been verified and the connections between the industries being researched and the creators of research papers. Too much of the research is financed by the industries being researched. So there now too. dgf

  11. Love that you are addressing this. Spent 3 months last fall in Japan. All they do is eat carbs – sugar, noodles, rice, ice cream….and fat. And in generous amounts. I was never able to finish a bowl of ramen, but watched 75 year old grannies polish them off every time. I’m also tired of people characterizing the Japanese diet falsely. People act like they do nothing but eat edamame and grilled fish. It’s just nuts. I brought this up at a Primal forum, and people got annoyed…and fell back on the calories in, calories out explanation. sighI just started low carb and I like it….but I sure wish someone in the science world WOULD address why the Japanese are all thin. I don’t know what “dgf” is talking about. I didn’t see fat people. (okay – I saw 2 – both young women, both obese) And I lived with a Japanese family in 1972 in rural Japan who owned one of MANY bakeries in town. Even without bakeries, wagashi? Is that not a sweet? oyatsu? Has that not always meant some form of high carb food???I’ll be in Tokyo again in May for 2 months. Maybe we’ll meet, Doug….over some nice low carb yakitori! yokattara…. 🙂

  12. Dave has been in Japan even longer than I have, so he has seen some gradual changes in body types over many decades which have occurred as more and more people adopt western eating habits, like fast food.But you are also right – there is just no comparison with the U.S. Just glancing around a crowded street in Japan sure, you can see a few fat people. But the overwhelming majority of people are thin.The stats verify this. The obesity rate in Japan is just 3.2% compared to over 30% (!) in the U.S.When I get off the Narita to Chicago airplane and look around I feel like I’m in the “Land of the Giants.” :)Getting back to Garry Taubes, which is what my original post was about, he hasn’t explained this. Instead he made false claims about what Japanese people eat, claiming they don’t eat white rice and eat practically no sugar. Both claims are not true.He and the low-carb crowd have to explain this.My explanation is simple: Americans eat too much and Japanese, by and large, don’t. American portions are typically huge by comparison. And it’s not just lots of refined carbs.  Americans eat loads of fat too. I don’t see how you can separate that out because it all adds up to lots and lots of calories in. Much more than the body needs during a day.Japanese undeniably eat a high carb diet, but are by and large much thinner than Americans. I think the reason for that is also obvious. They eat smaller portions and overall less calories per day.If you look at this country-by-country average calorie map: can see that the per-capita caloric intake in the U.S. is 3754 calories per day, while in Japan it is 2768 calories per day.That by itself is enough to explain the difference in obesity rates, even without considering relative daily activity. It’s the only answer that actually fits the data. Taubes explanation doesn’t fit the data.doug

  13. But….they are smaller. I don’t mean thinner….I mean smaller. So eating fewer calories makes sense. I just don’t think the calories argument works….

  14. I don’t think there is much difference in height. Maybe there used to be pre-war, but not these days.This reference lists the average adult heights by country: average Japanese male is 1.715 m tall. The average U.S. male is 1.763 m tall. That’s not so much difference – just 1.89 inches.Let’s say the maximum ideal BMI is 24.9. That means to barely not be overweight an average Japanese should weigh no more than 24.9 * 1.715^2 = 73.24 kg and an average American should weigh no more than 24.9 * 1.763^2 = 77.39 kg. The BMR ( for a barely non-overweight 20 year old Japanese would thus be 1,709 calories/day and for the American it would be 1,781 calories/day. In other words, the caloric requirements are not that much different – just 72 calories per day. But Americans eat on the average 986 calories per day more than the average Japanese person! They should be eating just 72 calories more.The calories argument makes more sense than the carbs argument here, doesn’t it? The carbs argument doesn’t work at all for Japan.doug

  15. Very interesting…making the calorie argument more feasible….although I am not sure height alone is enough. I’m 5’2 and even when I only weighed 140, I couldn’t find any clothes in Japan to fit! They are just so tiny. One of my best friends there is a man probably around 5’10”, a soccer-playing athelete….with arms the size of an American 6 year old. Seems about the norm there. So wouldn’t more than height have to come into play?

  16. I don’t know what could account for “tininess” other than height and weight. Japanese people are proportioned just like other humans. :)When you say a 5’10” soccer player had the arms of a an American 6 year old you meant the arms were very thin, right? Not that the arms were foreshortened! If the person’s arms were the length of a 6 year old that would be a deformity and not any more typical of a Japanese adult than it would be of an American adult.Speaking of calories, I think we would all agree that the most famous proponent of very-low-carb eating and ketogenic diets is Atkins. But take a note of this from Atkins’ own site.At it says, and I quote (from Atkins!):BE SENSIBLE, NOT OBSESSIVE, ABOUT PORTIONS. There’s no need to count calories on Atkins, but we do ask that you use a little common sense. You probably could guess that too many calories will slow down your weight loss, but here’s a surprise—too few will slow down your metabolism and slow weight loss. You only need worry about calories if, despite following Atkins to the letter, you cannot lose weight. Then a calorie reality check may be in order. Depending upon your height, age and metabolism, you may need to play with the following calorie ranges to lose weight: Women: 1,500–1,800 calories a day.Men: 1,800–2,200 calories per day.So even Atkins admits calories do count and the appropriate calories ranges are 1500-1800 for women and 1800-2200 for men.I say if you maintain those calorie ranges you can lose weight and ignore the amount of carbs you are eating. It seems to me that people who are successful on Atkins are able to take advantage of the appetite suppressant effect of low-carb eating (which is real) and naturally bring themselves down to within that calorie range and that’s all there is to it. Unfortunately, that doesn’t happen with me – on Atkins, followed to the letter – I still eat too many calories to lose weight.doug

  17. Hi Doug,I agree with you on Taubes. Germans eat lots of potatoes and drink a LOT of beer and have according to nationmaster only have an obesity rate of 12%. Here in Brazil we eat rice, beans, and salad for lunch every day. In the rural areas there is VERY little obesity. In urban areas where people eat in a rush and consume lots of american style fast food obesity is becoming more common. Also in very poor slums where people eat almost exclusively refined carbs and have malnutrition there is obesity too, esp. among females. I have read everything Gary has written on diet and heard all of his online lectures too. The guy had me so convinced that I tried an atkins diet for a while and ended up gaining 5 kilos. Then I read “in defense of food” by Michael Pollan and that was just a real wake up call to forget fad diets and just eat real food in a traditional way.I am not familiar with Japanese culture but I am sure they have many rules and cultural traditions that help them eat less and maintain their weight.

  18. Good luck. You’re going to need it. I’m not leaving any more comments on GT’s blog. I think that 195 is quite sufficient! People love to use the strawmen argument that you, I , CarbSane etc are discrediting Gary Taubes and/or low-carb diets. We just want some proper science, not cherry-picked science.So, how active are Japanese compared to Americans?

  19. “Cherry-picked science” is a perfect way of putting it.I don’t have any data on relative activity. It would be interesting to see what objective information is available.It’s just my personal impression, but I do think more people in Japan at least walk to the train station and take public transportation every day while Americans tend to rely more on cars for everything. But that’s not to say cars aren’t popular in Japan either. So I can’t really quantify it.People who are slim here basically are those who stick with the traditional Japanese diet of mainly white rice, with smaller portions of fish or meat or pork as a side dish, as a relatively small part of the meal. In the U.S. it seems flipped upside down: the fish/meat/port, etc., tends to be the main dish and the rice and veggies become side dishes. Even if Japanese eat fried chicken or tempura, the relative amount is small compared with the main rice dish. If you see a lunch bentou (box lunch) with pieces of fried chicken in it it is a few small cut-up pieces on top of a bed of rice. In the U.S. it’s the opposite. I was shocked when I visited a Japanese restaurant in the U.S. that served tempura and, accommodating U.S. tastes, they had piled a huge plate high with tempura, while in Japan it would just be a few pieces as a side dish. It was like two different worlds.Anyway, all the people who are complaining about Gary Taubes are asking is, “please make all the data fit – not just the data which agrees with Taubes.”doug

  20. White rice is what I call an “intact” grain i.e. it’s not been turned into dust. I think that that makes a lot of difference. Also, rice isn’t as moreish as chips.The Filipinos that work in mums care home tell me that when they come to England, they eat bread & chips instead of rice and put on weight.

  21. If Gary Taubes wants to make that argument it would be interesting to hear him say something like “refined carbs are bad except for white rice” for whatever reason. But he isn’t saying that. Instead he’s claiming Japanese eat brown rice!doug

  22. Hello again,I’m amazed at your patience with some of the commenters on GT’s blog. Apart from Galia L, who leaves intelligent comments, GT’s blog is like a car crash.New posters come hurtling along, don’t bother to read any of the other comments and post a load of uninformed crap that’s been discussed & dealt with already.Oy!

  23. Well, that’s way these discussions go. From years of experience in forums, I’ve learned to try to be patient, avoid getting personal and try to side-step possible “flame wars.” :)doug

  24. My family is from the Philippines and we eat either rice, bread, or noodles with every meal. I’ve been following a traditional filipino natural diet for the past month and the caloric distribution is more like 40% carbs, 20% protein, 40% fat. This is neither low carb, nor low fat. Also, Filipinos drink coke (with real sugar) with every freaking meal, yet still stay skinny. So far, here are my conclusions why American is becoming unhealthy1. Our aversion for fat, especially saturated fat. By eating less fat, we’re consuming more carbs2. Our switch from animal oils to processed seed oils. Lard -> vegetable shortening, Butter -> Margarine, Soybean oil everywhere3. Our switch from sugar to high fructose corn syrup4. All the added chemicals in our food

  25. I definitely agree about the fat…and the added chemicals.Have been reading about gut flora. Americans guts are a mess. The Japanese diet does a lot of things right as far as taking care of gut flora. Miso, rice wine vinegar, fish versus meat, tea versus pop…all help not hurt the gut. Bad flora = all kinds of bad stuff, including obesity.Read this the other day, written from a woman who was part of the Chernobyl cleanup crew: How did working in the dead zone begin to affect your health? What happens during first contact with radiation is that your good flora is depleted and the bad flora starts to flourish. I suddenly wanted to sleep all the time and eat a lot.

  26. I know I’ve gone far afield from the original topic of this post…just thought I’d share this one article:…I just was so fascinated last fall to see an entire nation (Japan) effortlessly maintain low body weight that I haven’t quit searching for answers since then. Nursing sets babies up with great flora. Formula wrecks the gut flora. I assume that Japanese are still predominantly breast-fed?

  27. I know I’ve gone far afield from the original topic of this post…just thought I’d share this one article:…I just was so fascinated last fall to see an entire nation (Japan) effortlessly maintain low body weight that I haven’t quit searching for answers since then. Nursing sets babies up with great flora. Formula wrecks the gut flora. I assume that Japanese are still predominantly breast-fed?

  28. It’s actually irrelevant. What isn’t clear is that Japanese people wouldn’t live EVEN longer and be even thinner for longer if they didn’t eat carbs. Also, Japanese people have a different bone structure and many doctors suggest obesity and all its problems manifest in Japanese people at a BMI of 28 rather than 30 for Europeans.

  29. For most Japanese, they really can’t and shouldn’t be thinner. You’re certainly not suggesting the Japanese give up a diet that keeps them slim and with about the longest lifespans in the world and change to low carbs?What about “if something works, don’t fix it?”doug

  30. Check out Dr. McDougall’s work. He is most certainly against Taubes work and he DOES address the white rice issue. McDougall’s work makes sense to me, and he’s in much better shape than Taubes is too 🙂

  31. Hah. Thanks. Actually, since May I’ve gone vegan + no oil/nuts/seeds (like Ornish and Esselstyn). Since then I’ve lost 48 lb, my HbA1c has dropped to normal, my LDL has gone down to 70 and total cholesterol down to 130. And I feel better than I have in years.

  32. Found this site searching for something about Taubes, found the comments interesting, had no intention of comment a 2.5 year old thread, not until I read the last comment. A total cholesterol of 130 is not a good thing. This is actually something GB talks about in this book, where US Doctors tell people to bring there cholesterol down to avoid heart disease, but in Japan doctor tell them to raise it to avoid stroke. If you have really low cholesterol you may lower your risk of heart disease but your risk of dying from things like stroke, cancer, accidents, and suicide go way up. And when your cholesterol gets too low your risk of heart disease goes just up too (it has to do with the particle size of the LDL unit). Check out this graph from WHO data And notice that the graph only goes down to a total cholesterol of 150! It’s wonderful that you’re feeling great, and even though that graph only shows correlations just understand that you are in a super high risk category with those numbers. Just be hyper vigilante about health issues that may be caused by nutritional deficiencies. A lot of vegans fall into the trap of blaming health issues on “detoxing” or old age or whatever cause it cant possibly be there diet.

  33. Well, my total cholesterol is now 122, LDL is down to 64, my HbA1c dropped from a dangerous 10.7 to a normal 5.5 (without medication), my blood pressure was 110/64 yesterday (also without medication) and I’ve lost 85 lb so far. And my one year June heart catheter checkup was so clear that the doctor halved my medications. I have more energy than I’ve had in years, and go for daily bike rides along the nearby rivers. I think I would need more convincing than that graph that I need to increase my cholesterol! There is just such an abundance of information from Ornish and Esselstyn and research data to the contrary.

  34. Pingback:“Own Your Health” – an informative and entertaining read | Doug Reports

Leave a Reply

Your email address will not be published. Required fields are marked *

HTML tags allowed in your comment: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>