Interview with Gary Taubes: Obesity & Calories [2012]

Around 2010-2011 my “Italian & Management Studies” course at University College London started boring me to the brink of death. Thankfully, I regained some child-like curiosity, pushing me to ask questions about our natural world. I also gained a newfound appreciation for just how good a companion the scientific method would be in this exploration.

At the same time, I started learning a lot about cannabis and soon thereafter nutrition grabbed my attention. Good Calories Bad Calories was the scientific ‘tour-de-force’ which laid the foundation for my voracious appetite concerning all things health, science (& more). I am currently in a part-time distance-learning MSc in Molecular Biology at Staffordshire University (UK).

 I emailed Gary Taubes asking him for an interview while making clear that I was essentially a fan, interested in pursuing a career in science/nutrition/something vague. He gladly accepted. I was and still am very grateful for the kindness he showed.


Caveat #1: my level of understanding about probiotics back then was essentially 0. Now it is significantly more informed. Yes – ‘the more you know, the less you know’ thing applies here: answers bring up more questions. Probiotics may be likened to the ‘other side’ of antibiotics, naturally granting them a fundamental role to play in our understanding of human health. A fascinating area of science!

 Caveat #2: This is the ‘full sentences’ transcript. I can provide the unedited one upon request. The write-up is mine alone and as far as I know Gary Taubes has not yet read it.


YouTube isn’t happy with me uploading the .MOV audio interview file so here it is provided as DropBox link:



 NuSIGary Taubes

Interview with Gary Taubes: Obesity & Calories [Transcript 11/06/2012]


Gary Taubes: […] Obesity is caused merely by taking in more calories than we expend, that’s the conventional wisdom. The argument that I’ve been making is that obesity is a hormonal regulatory defect just like any other growth disorder and the obvious trigger for it is the carbohydrate content of the diet working through the hormone insulin.

Raphael Sirtoli: Ok, so we can establish that first of all. What do you see as the major challenges to the food industry over the next 10 years?

GT: A large part of the food industry is selling processed carbohydrates and sugars, the bulk of which comes from supermarkets here with grains and cereals and products with sugars laced into them. So the question is what about fruit juices and sodas then?

R: We would effectively have to change the whole base [of food products] for the majority of the supermarket food stores.

GT: Let me rephrase this. I was thinking about this recently because it is not like we have stopped selling cigarettes because we have accepted that cigarettes cause lung cancer and probably heart disease and emphysema and a host of other ailments. We have just put warnings on them. It is hard to imagine warnings on cans of Coca-Cola and Pepsi and on bottles or cartons of apple juice. I’m not so optimistic that the medical community will ever get around to accepting that these things are kind of inherently deleterious.

R: It’s one thing to get these things accepted as empty calories, but another to get them to be accepted as something actually poisonous (with regards to fructose & sucrose).

GT: Exactly. It is something that literally causes chronic disease, so I do hope we do win that battle and it does eventually get accepted. However this challenges the existence of many food companies. These are not people who see themselves as peddling poison. They are people like you and me who basically think that they are doing good things with their lives and paying for their children’s college education. So I don’t know, I really don’t know. I hope the next 10 years will play out because the message is that you can think of them as a type of poison and you could argue that there is a dose beneath which they are harmless – which is probably true – but we don’t know what that dose is.

R: It would be hard to address that.

GT: Yes, and I would argue there is probably a dose at which cigarette smoking is effectively harmless, but the medical community disagrees on that too.

R: How do scientists typically investigate whether certain foods are good for us or not? Is there a ‘standard model’ of approach when studying the health aspects of food? Or does it mostly depend on what you’re looking for?

GT: That is a complicated question as well. Our Food and Drug Administration has mechanisms in place for determining whether or not a product is what they call generally recognized as safe. There are obligations that food companies have to fulfil if they have to add additives to foods so that the additives are considered safe enough to use. If you took naturally occurring foods like half a dozen apples and turned them into a glass of apple juice for instance, I don’t know if there is any precedent for establishing whether or not you have now turned that into something that in the long term will cause chronic disease. So how would you test it? Say you had a hypothesis that said that juice was toxic and so now you wanted to test that hypothesis. We are talking about diseases that develop over decades. The question is, what do you compare the juices to? If you just remove juice – say you take 20,000 people and randomize 10,000 of them to a juice-free diet – you are probably going to lower their caloric intake because you are going to remove the calories from the juice as well as the sugars in the juice that represent those calories. So, then you decide, what to replace those calories with and if I should give the other 10,000 people beverages that are sweetened only with glucose? In which case they might still drink less of them because they don’t stimulate their taste buds and their physiology the way beverages sweetened with sugar do. It is surprisingly difficult and complex.

I can tell you one thing we have done. We have actually started a not-for-profit organization, ‘We’ being my colleague Peter Attia and I. It is called the Nutrition Science Initiative and we now have backing from a very well endowed foundation from Houston in Texas. Our argument is that fundamentally the science of nutrition has been substandard over the years. Nutritionists and disease researchers have relied on very poorly controlled experiments or observational studies or animal models to come to their conclusions. None of these are then sufficiently rigorous or up to the task to determine what we really need to know about a healthy diet. So we are going to fund very well controlled experimental trials with humans and come up with innovative ways to do these studies. We can keep human subjects on the diet for 6 months or a year if necessary and will be able to establish if they did not eat or did eat and what was given to them. All of this would be a pipe dream if not for this Houston foundation, the ‘Lauren & John Arnold Foundation’ that is basically making it possible for us to do something we never imagined we could do before. We’re launching July 1st 2012.

R: That’s good news!

GT: Yes, that’s very exciting. We had our first meeting in Bethesda Maryland with members of our scientific advisory board and researchers from the NIH and elsewhere who want to do the first experiment that we want to fund.

The next couple of years are going to be learning experiences on how excruciatingly difficult this is – even when you have the money – but we are going to do it, and it is pretty fascinating.

R: There is a definite need for that. There is too much contrary evidence to the calories-in calories-out hypothesis to just have it be accepted as such.

GT: It is interesting. I already had a conversation with a physician researcher at Yale who very much believes the conventional wisdom. He effectively told me nothing we funded would make a difference in what he ultimately believes. This is because what he ultimately believes is determined by observational studies like the nurses’ health study.

The argument is we could do a 6 months study that could tell you that weight regulation is not determined by caloric intake and expenditure but by the nutrient content of the diet. This is the first study we want to do. It is the kind of study that should have been done rigorously in the 1960s but was not. Even if the counter argument will be, ‘well look, that only tells us what happens over 6 months, it does not tell us what happens over a lifetime’, well we just cannot do studies over a lifetime, so…

 R: It is a self-defeating argument: ‘we cannot do it because it is too difficult [& thus never try to]’. It completely ignores the fact that other studies encounter these barriers as well.

GT: In the mean while we use these observational studies which I believe are worthless and cannot be interpreted correctly. You know in the ‘Step 1’ of science they generate hypotheses, but that is all they do. However, [the over-interpretation of observational studies] is still rampant in medical research. They are everywhere in nutrition and chronic disease these days. One of our goals at the Nutrition Science Initiative (NuSI) is also to get people to understand how misleading these observational studies are and why they are not a substitute for hard experiment. If you don’t have the rigorous experimental trials you don’t know anything and it’s that simple.

R: We’ll definitely take note of that. Is a healthy diet the same for all populations or do certain populations require different diets? Like when considering geographic populations for instance.

GT:  That is an interesting question. I think the argument that I’m making in my book is that sugar and refined grains are sort of the fundamental problems, the primary nutritional evils in our diet. I would argue that that is true for every population. Although there might be populations that are more adapted to particular carbohydrate rich food. For instance, maybe South East Asians can tolerate rice better than other populations because the have been eating it for thousands of years. Although when white rice was introduced in the 20th century to populations it did cause Berri Berri I think. We will have to check this.

 R: Yes, it was a lack of Vitamin B12 I think.

GT: Yes so that is a possibility. The fundamental argument I’m making is that refined grains and sugars are the problem. You remove those from the diet and every population is going to be healthier. You minimize them in every population and they will be healthier. Are some populations more susceptible to these foods than others? I think that is pretty clear, because some populations like African-Americans in the U.S.A or Hispanics in the U.S.A have higher rates of obesity and diabetes than the Caucasian population which suggests that Caucasians had more time to adapt to these foods. Let’s say centuries or rather millennia instead of just a few centuries. So we can tolerate them a little bit easier. The argument that some people lose weight on say, low-fat diets and others lose weight on a low-carbohydrate diet and that that is a genetic issue or that it maybe depends on whether your insulin resistant is looking at it the wrong way I think. I think low-fat diets are just poorly constructed low-carbohydrate diets and some people can still benefit from them even though the carbohydrate restriction is not as great as it could be. If you look at low-fat diets they are just kind of restricted carbohydrate-diets in disguise.

R: […] because of the proportions.

GT: Yes exactly. Some people are metabolically healthy enough that their bodies will still respond to that level of carbohydrate restriction while others are so metabolically disturbed that they need much more carbohydrate restriction. You do that with a higher-fat diet. Not to say that the person who responds to the low-fat diet would not respond better to the low-carbohydrate diet, they just get enough benefit from a low-fat diet or at least from being told to go on a low-fat diet that that can be enough.

R: Should we then be confident in saying that a few thousand years – like in the case of Asians getting used to eating rice – is enough to see a small adaptation to a new introduction of food source?

GT: There is a lesson in evolution there. Have you ever read the book “Beak of the Finch”?

R: No I haven’t unfortunately (laughs).

GT:  Well it won the Pulitzer Prize here in the U.S.A a while back and I forget the author’s name at the moment. It is a terrific science book about evolution. The lesson from the research that the author was following was that 2 or 3 generations are enough to have a sufficient impact on a population’s ability to adapt to a new feature in the environment. So for instance you can imagine a situation in which sugar and refined grains are introduced into a population and the people are really susceptible – particularly the women – and they get obese and diabetic pretty quickly. This would have profound reproductive disadvantages and so would not take many generations before the most susceptible individuals to these aspects of the diet were weeded out because of the reproductive disadvantage. Maternal mortality during birth along with birth defects and the death of the infants are examples. All these things come at a pretty high frequency with uncontrolled diabetes.

R: The associated risks with diabetes will do the selecting.

GT: You can imagine this kind of situation where the women who are most susceptible to these carbohydrates might have been perceived as more or less attractive as mates. It is hard to say. One of the lines that keeps haunting me – I’m doing a new book now, very slowly but I’m working on it. It is on sugar and high-fructose corn syrup and the history and politics. I found an essay online by a physician writing around the 18th century. I think it might have been 1730s and he is actually defending sugar in the diet against claims that sugar is toxic, back then.

R: Back then!

GT: He is saying sugar isn’t toxic, it is completely harmless. But he then has an aside where he says ‘but of course it makes young girls fat’. Other than that…If that was true that it made young girls fat, those young girls in the 18th century would have been eating about 1/20th the amount of sugar our young girls are eating today. Is it possible that this was accurate? You would like a lot more evidence than one doctor’s word. However, it was interesting that it came in the context of an essay defending sugar and not in an essay attacking it.

R: Yes actually making that axiomatic point about sugar; that it will just do that to you [fatten you up].

GT: Yes, so the question is, ‘how much is necessary to do real harm when it is introduced into a population that is completely new to it?

I read a lot of memoirs of naturalists and explorers who interacted with aboriginal populations around the world. I read reports wherever I could find them; government reports, studies of aboriginal population and so on. There were these reports from Australia about the local Aborigines and they would describe them as almost decimated within a decade or one generation by the introduction of Western foods into their diets. The question is how, ‘how much of it was due to alcohol? Or sugar and flour? Or vegetable oil? Saturated fat? Lard?’ Who knows. These Western diets do, however, cause an enormous amount of harm. You would also have to worry about infectious diseases. However, those tended to be pretty obvious when they hit. The point is that the idea that these things did their harm at relatively low doses compared to those we take today and they did it quickly, is at least out there in the literature. Hopefully we will never have a chance to test that one.

R: Hopefully not! This leads me on to my next question. How does age affect one’s’ diet? Does it have an effect at all? If so, how?

GT: Many of us have this observation that we could eat anything in our youth. When I was in my early teens I played American Football in college and tried to get as large as humanly possible because that is what you want with people who play my position in football. Despite eating constantly and massive amount of foods I could not get over about 235 pounds. By the time I was 35 years old I could get to 235 pounds effortlessly and the battle was to not weigh 235 pounds. In fact, I got an email a couple days ago from a Green Beret (a Marine) who described his struggle with his weight and how despite being one of the most fit army/military men you can find he said that as he got older he kept getting bigger and bigger. This is a pretty common observation. The conventional wisdom is your metabolism slows down so you expend less energy and therefore the same amount of calories consumed comes in excess to the energy expenditure. The argument I make in my books and that I find more reasonable is that we change the way we partition our fuel. With the fuel we do consume, we go from burning to storing it. It is probably a result of a change in insulin signaling which is the primary hormone involved with this fuel partitioning. We thus go on to become more and more insulin resistant or our fat tissues become more and more insulin sensitive. Some combination of this and you could easily imagine a situation where the biggest proportion of the food you eat goes to being stored and a smaller proportion goes to energy expenditure. You perceive this as a decrease in metabolic rate and impulse to exercise, while in reality is it is simply a change; an effect of this change on your fuel partitioning.

R: Does the question then become more of a comparison argument where you are comparing, say, the natural metabolic derangements that accompany aging versus the chronically elevated levels of insulin which come with the Western Diet throughout a lifetime? Is it a mixture of both then?

GT: Well, again I would argue that it is the Western diet that is changing the insulin signaling and the carbohydrates in it changing the insulin signaling which is in turn changing the metabolic profile. It’s kind of interesting because you could imagine a decrease in energy expenditure in all people as we age. I doubt that 70 year-old hunter-gatherers (if they lived that long) were as active as 2 year-old hunter-gatherer’s or 4 year-old hunter-gathers. If you think of lions and predators, the adults sit around all day when they are not hunting and the children are playing and frolicking. This would all be a manifestation of how their bodies partition fuel and which energy is being used for expenditure. So when expenditure comes down, intake should come down to meet it. If somebody gets bed ridden there is no reason why they should continue to eat as much as they otherwise would. If a lumberjack had an accident and he is in bed for 6 months you would assume that his appetite is going to be less than when he was working as a lumberjack and cutting down trees all day long. That is kind of the issue where if you are gaining fat then there must have been some change in your regulatory system regulating fat accumulation. It’s not just enough to imagine you decrease expenditure because decreasing expenditure should go along with a decrease in intake.

R: Do you think that men and women require the same diets? Could you generalize or not?

GT: Would men and women both benefit from eating less refined grains and sugars? Yes absolutely. Is there such a thing as an ideal diet? That is I guess the question you could ask. What would that diet be? I guess it would maximize health and longevity and perhaps energy expenditure so that would mean that you are physically active. For women that would also mean fertility and the ability to nurse and so on. There are a lot of different things you would have to imagine this ideal diet could do. I would argue that if you remove the refined grains and sugars and easily digestible starches then you are pretty much along the way. You know you have done about most of what we know or what I think we know you could do to improve your diet. Although that does depend on what you replace it with. You could argue that women sort of evolved consuming a different diet if there were any evidence from hunter-gatherer populations or any evidence that they sort of naturally consumed more gathered foods and perhaps ate slightly less more hunted foods (laughs)…But I’m not a big fan of evolutionary arguments other than that it makes sense that we did not evolve to eat the foods we are eating now (laughs). One of the reasons that this makes sense is because we have all the chronic diseases that they seem to cause. So it is not just that a new food is inherently bad but if a new food goes along with new chronic diseases and those can be pretty easily explained by the presence of the new food, then there is a pretty good chance that the new food is bad. Whether certain combinations of old foods or foods that we did evolve to eat are more or less ideal than other combinations of these and whether this change is gender specific, who knows really.

R: Yes, the required studies have not been done yet.

GT: They have not been done. They probably never will (laughs).

R: (laughs) Ah, to hope! In your opinion what should concern us about genetically modified foods? Should we be concerned in the first place?

GT: I don’t really know in that sense. This is not one of my areas of expertise. My gut feeling is that they are relatively harmless. The foods we are modifying are foods that I’m arguing we probably should not be eating to begin with. So there you are.

R: Especially when you think about all the corn and grains.

GT: Yes. I will make one comment about this argument though. The primary problems come from genetically modified grain – wheat in particular and hence the wheat belly argument. The thing that I point out in the first chapter of ‘How We Get Fat’ is that all these populations that had high levels of obesity and diabetes were not eating genetically modified wheat. Usually it was diabetes researchers who were measuring weights in these populations that assessed this. The idea that you could still get significant levels of obesity and diabetes without any kind of genetically modified super wheat to do it is there. These were populations before the 1980s for the most part and some of them pre 1970s.

R: Historical basis.

GT: Yes, white flour and sugar were probably sufficient [to explain it]. Whether the way we make our wheat today makes it worse is possible but I would kind of doubt it on principle until I saw really compelling evidence to back it up.

 R: A simpler question I guess. What should the 3 most important points on a food label be?

GT: Contains sugar, contain HFCS, contains sucrose. Basically, CONTAINS SUGARS. I would not actually worry about 2 or 3.

R: The other people I interviewed are so quick to label as many as they can.

GT: I have these ongoing discussions with my wife. We have a 3 year old and a 6 year old and like with many kids it is hard to get them to eat and it is hard to get them to eat the foods you want them to eat. We live around Berkley California which is a kind of health mecca and the supermarkets are full of all kinds of supposedly healthy foods and healthy probiotic yoghurts that will improve my child’s gut biota. So the argument that I keep making to my wife in is that if it makes a health claim, look to see how much sugar it has in it.

R: I see, we should keep sugar as a point of reference.

GT: Yes. It is the same thing you see now with the gluten issue. It is very easy to make, or I assume it is relatively easy to make gluten-free products and then advertise them as gluten-free. However, if you then look for the thing that says that they are also sugars-free or sugar and HFCS-free, you are not going to find it because it is a lot harder to make a tasty food without sugar than it is without gluten. I feel like the gluten movement has been embraced by industry because it is a relatively easy way for them to look like they are producing a healthful food when in fact they are actually producing a food that still has sugar in it that their clientele and the public will buy now that we have become used to high levels of sugar in everything. It makes it a little bit more difficult.

R: It has skewed the norm.

GT: It has skewed the norm, yes.

R: Ok. What role do you think the probiotics have in healthy diet? Do they have any role at all?

GT: That is not an area of expertise for me…[sighs] I just have not really looked into it. I’m quite a bit of a one-trick pony. I spent all my effort learning enough to write Good Calories Bad Calories. By that time I had my second child and since then I have spent most of my effort and energy trying to get people to accept and understand these ideas. I want the medical research establishment to take them seriously and test them, but I literally don’t have time to or perhaps even the intellectual capacity anymore to take on a completely new subject. With the probiotics in general I could probably bet you that there is not enough randomized controlled evidence to suggest that they are actually doing something meaningful.

R: Yes.

GT: If you always asked yourself “what would it take to test this idea?” and this idea was for example that they are supposed to get rid of a symptom, then for instance you go on antibiotics, which tends to wipe out your gut biota. So you then replace it with the healthy bacteria in the probiotic. I can see the logic in that and I can imagine it would be pretty easy to test it either with an N=1 [type of experiment] or a relatively short experimental trial. Whether or not you are going to be healthy, live longer and have less chronic disease simply through a change in your probiotics is harder to accept. This is because you can imagine what it would take to test that. I can imagine that it has never happened [the experimental trials].

R: Yes. Another person I interviewed was telling me that out of the 800 health claims that probiotics seemed to have, all 800 had not yet been proven.

GT: [laughs]

R: I haven’t gone to check that for myself yet, although I wouldn’t be surprised if that were the case. [laughs]

GT: You get these huge efficacy effectiveness studies with pharmaceuticals because they can get patents out of them and there are people who stand to make a lot of money if they could come up with the drug that does something beneficial without an equal amount of harm. However, with naturally occurring bacteria there is nobody who is really motivated to spend the money necessary to find out about them and do the experimental trials to establish if they are really doing something. The fact that they might seem to make a difference if you were to take them yourself is interesting. But the history of science is a history of how easy it is to be deluded. I’m always amazed when people tell me – well first let me tell you, I have a friend who does a lot of, what is it called, ‘Self experimentation’. It is one thing when you are looking at the blood risk factors for disease and do something like measure your LDL particle number. However if your influences are subjective experiences like how much sleep you got, how smart you might feel that day, how energetic you are, even whether you have more or less gastrointestinal upset and so on, it then becomes extremely easy to be fooled. There could even be some kind of psychosomatic placebo effect, so you just never know.

R: Ok. Do you have time for a last question?

GT: Yeah sure.

R: I just had a cholesterol blood test recently and I was very curious to see my results after having significantly adapted my diet. What are the primary parameters that one should focus on when receiving a blood test/cholesterol analysis?

GT: I have almost been convinced that the most important number is the LDL particle number. There is a whole host of studies now that show that if you measure LDL particle number you can get a pretty good assessment by measuring Apo-B, the protein component of LDL particles. When we actually look at the number of LDL particles, the size and density issue kind of washes out. The idea that small dense LDL particles were bad – something that I wrote about in’ Good Calories Bad Calories’ – appears to be so because when you do have a lot of small dense LDL particles, you then also have a lot of LDL particles.

R: Ok, I see.

GT: So the total number of LDL particles is the factor that best tracks with heart disease and atherosclerosis. The hypothesis behind it is then driven by the triglycerides, which the state of the science today suggests is in turn driven by the carbohydrate content of the diet. It does not change the ultimate end point, which is if you want to improve that number as best you can, go on a low-carbohydrate high-fat diet. Although it does suggest that if there were one thing you wanted to measure that would be it, the LDL-P or Apo-B, both of which will give you the number. If you were to get the test for your particle size and density, you would probably see that [tracking] 199 times out of 200.

R: Ok, so are you saying that the LDL particle [composition] itself is less important than the actual number of the LDL particles (or specifically the Apo-B lipoprotein)?

GT: Yes. For Apo-B you will get LDL plus VLDL and IDL, but LDL makes up about 95% of it as I understand. Again, this is something where I have not spent a lot of time looking into so I’m relying on people who have and whose judgment I trust, including my colleague Peter Attia. What would the single best number be? Well, it is interesting when you talk to people who swear by LDL particle numbers. They will dismiss anything you could learn from any tests like these VAP tests which will give you size and density [of LDL particles] or simply getting triglycerides and HDL measured. You could say that in most cases you would learn enough from knowing if your triglycerides were high and your HDL was low and be pretty confident you are somewhat insulin resistant. I find it a bit odd to be dismissive about that. The way to deal with this [insulin resistance] is to get rid of the carbohydrates and replace them with fat. If your LDL particle number is very high, in most cases what that is telling you is that you are insulin resistant. The few times it is not, it is probably telling you that you have got what is called hypercholesterolemia which is a genetic defect. That is one of the reasons why you can argue that it is particle number and not size and density [that matters] because people with familial hypercholesterolemia have large fluffy LDL particles but just have so many of them and get heart disease at relatively young age. I think that if I were to pick a number circa 2012, that [LDL-P] would be it.

R: After following a relatively high-fat and high-protein diet I have to say how impressed I was by how closely my figures tracked to what the theory in ‘Good Calories Bad Calories’ suggested.

GT: Well thanks!

R: It was quite amazing. I had high total cholesterol, a pretty low Apo-B protein count that was within the ranges and good HDL. It is always nice to see that confirmed on an individual basis.

GT: I get emails everyday now from people discussing their LDL numbers or how their weight has changed. Some people like the Green Beret discussed how these books finally helped him understand what had been happening to him his whole life. It is one of the reasons why when I do get optimistic (I’m not a naturally optimistic person) it is because it seems so obviously right. It is hard to imagine that.

R: I think it really is paradigm shifting.

GT: Yes.

 R: You cannot really escape that conclusion.

GT: Yes.

R: Well it certainly was very interesting to speak with you.

GT: Thanks!

R: And thank you for the opportunity.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s