= 27d567bc2d48d9f40e9ccc20ea370b15)
Critiquing Gary Taubes, Part 1: Mainstream Nutrition Science on Obesity
Related: Trusting Expert Consensus
Lately, I've been thinking a lot about whether we can find any clear exceptions to the general "trust the experts (when they agree)" heuristic. One example that keeps coming up—at least on LessWrong and related blogs—is Gary Taubes' claims about mainstream nutrition experts allegedly getting obesity horribly wrong.
Taubes is probably best-known for his book Good Calories, Bad Calories. I'd previously had a mildly negative impression of him from discussion of him on Yvain's old blog, particularly some of other posts Yvain and other people linked from there, such as this discussion of Taubes' "carbohydrate hypothesis" and especially this discussion of Taubes' attempt to refute the standard calories-in/calories-out model of weight.
But I figured maybe the criticism of Taubes I'd read hadn't been fair to him, so I decided to read him for myself... and holy crap, Taubes turned out to be far worse than I expected. I decided to write a post explaining why, and then realized that, even if I were somewhat selective about the issues I focused on, I had enough material for a whole series of posts, which I'll be posting over the course of the next week.
The problem with Taubes is not that everything he says is wrong. Much of it is ludicrously wrong, but that's only one half of the problem. The other half is that he says a fair number of things mainstream nutrition science would agree with, but then hides this fact, and instead pretends those things are a refutation of mainstream nutrition science. So it's worth starting with a brief in-a-nutshell version of what mainstream nutrition science actually says about obesity.
(The following summary is drawn from a number of sources, including this, this, and this. Everything I'm about to say will be discussed in much greater detail in subsequent posts.)
Here it goes: people gain weight when they consume more calories than they burn. But both calorie intake and calorie expenditure are regulated by complicated mechanisms we don't fully understand yet. This means the causes of overweight and obesity* are also complicated and not fully understood. It is, however, worth watching out for foods with lots of added fat and sugar, if only because they're an easy way to consume way too many calories.
We currently don't have any great solutions to the problem of overweight and obesity. If you consume fewer calories than you burn, you will lose weight, but sticking to a diet is hard. It's relatively easy to lose weight in the short run, and it's possible to do so on a wide variety of diets, but only a small percentage of people keep the weight off over the long run.
As for low-carb diets, people do lose weight on them, but they do so because low-carb diets generally lead people to restrict their calorie intake even when they aren't actively counting calories. For one thing, it's hard to consume as many calories when you drastically restrict the range of foods you can eat. There's also some evidence that low-carb diets may have some advantages. in terms of, say, warding off hunger, but the evidence is mixed. There's certainly no basis for claiming low-carb diets as a magic bullet for the problems of overweight and obesity.
The above points are not the only issues at stake in Taubes' writings on nutrition. Admittedly, he covers a huge amount of ground, from the relationship between sugar and diabetes to the relationship between fat intake and heart disease to the alleged dangers of extremely-low carbohydrate diets. However, I'll be focusing on his claims about the causes of and solutions to the problems of overweight and obesity, because that seems to be the main thing people talk about when they talk about Taubes supposedly showing how wrong mainstream experts can be.
I'll also focus heavily on how Taubes misrepresents the views of mainstream experts on obesity. In the next post, though, I'll be temporarily setting that issue aside in order to look at what Taubes is proposing as an alternative. This will involve examining some claims made by Dr. Robert Atkins, whose ideas' Taubes champions.
*Note: if the use of "overweight" as a noun sounds weird to you, it does to me too, but I discovered as I researched this article that it's standard usage in the literature on the subject. I came to realize there's a good reason for this usage: it's inaccurate to talk about the problem solely in terms of "obesity," but constantly saying "the problem of people being overweight and obese" gets really wordy.
Next: Atkins Redux
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d){kind=logo}
Comments (99)
This is the kind of post that requires citations. Quote Taubes describing nutrition science orthodoxy and then quote an authoritative source from the establishment.
I say this not to just demand citations for citations sake-- it's just that it's easy to model how Taubes would respond to this post when all you're doing is asserting things. He would, at minimum, assert the opposite.
You're right. I'm planning on discussing relevant sources in much greater detail in future posts, but I'll add in some quick links when I get to my parents' house tonight.
(Sending this from my iPhone while on a plane.)
I would be much more interested in an attempted refutation of the Perfect Health Diet by Jaminet & Jaminet.
http://perfecthealthdiet.com/
Never heard of that one before. What's the reason for your interest in it?
Seems better researched than anything else I've ever read, and is equally indictive of the sort of medical consensus opinion which says saturated fat is bad for you.
I agree. I've read the Perfect Health Diet several times and it has had a big impact on my diet.
How's your weight and waist/neck/wrist measurments (assuming you don't have DEXA or immersion or something), now, compared to before you read the Perfect Health Diet?
I'm 5 foot 10 inches and have gone from around 170 to 150 pounds. More importantly, my Cholesterol has greatly improved going from
Whole Cholesterol 150 to 232; Triglycerides 98 to 69; HDL 33 to 76; LDL 98 to 142.
Despite what you might have heard, higher Cholesterol (if it's accompanied by more HDL) is better.
It's unfortunate that "calories in, calories out" and "saturated fats are bad" are both general medical consensuses (wow, that word is actually in dictionaries) - it seems very likely the first is true and the second false, but both issues have the same "medical consensus saying they're true vs fringe expert saying they're all wrong" dynamic.
I read some of the comments-- the carbohydrates recommended in the book work very well for some people (and better than a very low carbohydrate diet), but there are some people who don't do well on them.
From a cursory overview, that looks very similar to a standard paleo diet, but without the caveman verbiage or naturalistic fallacy.
Similar, but recommending white rice is utter heresy.
Edit: Looks like I was wrong about the heresy. I must have been confusing paleo with relatively mainstream worries about refined grains.
The Perfect Health Diet recommends you don't go zero carb, and it's very hard to find Paleo non-sugar based carbs (other than sweet potatoes) so white rice is recommended as being not as horrible as the alternatives.
It might be a gray area according to some, but heresy seems like a little much. A google search has only two results in the first page saying that it's bad, with the rest saying it's fine. Robb Wolf (a paleo advocate) says that rice is OK for active, healthy people (ctrl+f "rice"), and Mark says that white rice isn't bad.
FYI: I recently went to an endocrinologist to try thyroid treatment (synthetic T3).
She earnestly advised me to cut saturated fat from my diet.
This does not reflect well on your hopes for this series of essays.
I think it reflects very little. Let's see how cutting down on saturated fat works out for you. Even if it works out well (best wishes), you've got a weird metabolism, so what works for you is not a lot of information about what works for people in general.
"Cut out saturated fat" is the stereotypical horrible purely-made-up bad-science zombie-study opposite-of-smart advice of exactly the sort Gary Taubes was criticizing. It's one of the worst possible bits of advice you could give someone. It's probably not my comparative advantage to go into at length here, browsing paleo blogs (and I do recommend Perfect Health Diet despite the name) would fill you in.
Sorry, I clearly misunderstood your point.
On the other hand, I'm not sure what your point was. Would you be willing to expand on what you meant?
ChrisHallquist claims that Taubes is attacking a misrepresentation of what mainstream nutrition experts actually believe. Eliezer met an endocrinologist, and she just happened to spout horrible advice of the kind Taubes attributes to the mainstream. This is more likely to happen in worlds in which Taubes is correct about such advice being widespread among the mainstream than in worlds in which Taubes is exaggerating the spread of such advice.
Really? Is it that saturated fat is actually good for you, or that it isn't necessarily bad? Would you recommend more saturated fat to a thin healthy person who has no trouble with their food intake?
Yes, Paleo people and the Perfect Health Diet book would recommend that most Americans consume more healthy saturated fats such as the kind in butter from grass-fed cows. I put such butter in my morning coffee.
Most Americans get about the optimal amount of calories from protein, so if you cut out most sugar you have to replace the calories from somewhere else and the Perfect Health Diet suggests they come from healthy fats.
You know of any evidence replacing those calories with saturated fat is better for you then replacing them with (cis) polyunsaturated fats?
It's in the book the Perfect Health Diet.
What makes it a reliable source compared to others?
From what I can tell, yes, although I'm an economist not a life science person.
I think you misread my question.
I think this is an important question and I would refer you to the concept of "nutritionism" as described by Yoni Freedhoff, a Canadian obesity researcher. He defines it as "the notion that specific properties of foods are sufficient to make them healthy"
I've been thinking about this stuff a lot, and it occurs to me that a lot of diet thinkers are guilty of the sin of nutritionism.
To illustrate, everyone knows that doughnuts are unhealthy, but one can ask why exactly they are unhealthy.
Is it that they contain a lot of fat?
Is it that they contain a lot of carbohydrates?
Is it that they contain a lot of sugar?
Is it that they are heavily processed and not what cavemen would have eaten?
It's dangerous to answer these questions incorrectly. If you assert that muffins are unhealthy because they contain a lot of fat, it won't be long until someone shows up with low fat muffins and people start pigging out on them.
Perhaps the most extreme example of nutritionism is diet soda. It contains, no fat, no carbohydrates, and no calories. So you would think people could easily achieve significant and lasting weight loss success by switching from regular soda (and juice) to diet soda. But it doesn't seem to work.
Let's make this a bit more complicated by adding a few questions. By nutrients I will refer to both micro- and macronutrients. You may allow this to refer to indigestible substances and artificial flavors as well.
Is it that they replace healthier food?
Is it that they regulate appetite differently?
Is it that they regulate digestion differently?
Is it that they regulate metabolism differently?
Is it that they regulate where the fat is stored?
Is it that they regulate physical activity differently?
--
I'm sure people can add even more relevant questions to this bunch, and I would be interested to read them.
Yes those are good questions. In answering them, I think one should keep in mind that nobody has ever invented an effective diet doughnut. This suggests to me that it's not a matter of adding some nutrients to the doughnut recipe; that the problem is inherent in the doughnut.
By analogy, one can look at the attempts to create a non-addictive morphine. Which resulted in heroin. Oops! Probably it is impossible to create a non-addictive morphine because the analgesic aspect of morphine is exactly what makes it addictive.
Although AFAIK there is not scientific consensus on this point, I'm pretty confident it's a similar problem with foods like doughnuts. They just make you feel too good. And that screws up something in your brain.
Opiates are relatively non-addictive if their use is regulated intelligently, which for some reason all people can't reliably do themselves. I think the same might make sense for certain foods. I can eat a doughnut every once in a while just fine, but if I started binging them it might be difficult to stop. Same applies to nicotine, caffeine, alcohol, video games and movies for me too.
The question of what is or isn't addictive and why is important but it's a little beside the point. I am happy and perhaps eager to discuss addiction in the context of dieting and obesity, but at the outset we need to agree on a definition of "addiction."
My point in bringing up heroin is that it's basically impossible to separate the good aspects (analgesic) and bad aspects (addictiveness) of morphine because both qualities are the result of the same mechanism. Analogously, it's (in my opinion) impossible to separate the good aspects (tastiness) and bad aspects (fatteningness) of foods like doughnuts because both qualities are evidently a result of the same mechanism.
Actually I am tempted to go further than that and hypothesize that in both cases a big part of the problem is the part(s) of one's brain which process pleasurable experiences.
Well what was the doctor intending for you to accomplish by cutting down on saturated fat? Weight loss or something else?
Weight loss.
Oh wow. Based on everything I know about weight-loss, I'm pretty sure that isn't right.
Yes, typically saturated fat consumption (but ironically not cholesterol anymore) is used to explain heart disease in mainstream nutrition, not necessarily obesity, but the argument there may be the 'more energy dense than other macronutrients' one.
Especially as someone who hasn't read Taubes, I would have preferred this post to be longer and combined with some more substantial discussion. It feels weird that you're recapping what the mainstream nutritionists believe but not actually contrasting it with the target of your criticism: it left me with a slightly frustrated feeling of "okay, so this is the mainstream wisdom, but what about it?"
Thanks for the feedback.
I was going to post this as one monster article, but as it approached 4,000 words with a couple long sections still to be written, I decided it needed to be broken up. Also, I've posted monster articles before on LessWrong, but in a couple cases I ended up feeling like the length turned off people from reading them.
When I'm done with this series, I'll probably do a post-mortem on how I could have broken it up differently.
By the way, is there a consensus view on low-carb diets and water retention? I've read in more than one place that reducing one's carbohydrate intake can cause fast temporary weight loss due to the body retaining less water. Not sure if this is a consensus view though.
This seems to be widely-accepted, but I couldn't tell you for sure.
It's not a huge point, but I think it's important because part of the hype of low carb dieting comes from the promise of rapid weight loss at least at the beginning.
It's true. It's said that for every gram of glycogen, you need three grams of water to store it. The average person has about 500g of glycogen, so you'd have around 2kg weight loss just in glycogen and water from starting a low carb diet.
It's a psychological effect. People start the diet, immediately lose 5lbs over a week, and think "OH MY GOD, this is working, I can stick to this!" They establish the habits and systems of losing weight. Then when weight loss slows down to the more reasonable 1-2lbs per week, they aren't bummed out because they know it's working.
When they cheat, their weight immediately shoots up -- and then when they stick to it, it immediately goes back down. This helped me a lot with getting back on track when I used a low carb diet to lose weight.
I basically agree, I'm just wondering if there is expert consensus on this chain of reasoning, i.e.
Low carb-dieting tends to produce stunning weight loss results in the first few weeks with far less actual fat loss.
As a result, many people tend to feel falsely that they have found the silver bullet for weight loss in low-carb dieting, i.e. a way to circumvent the calories in/calories out paradigm.
I personally am pretty convinced of the above two points, but I am wondering if there is expert consensus.
How popular is Taubes and in what kind of a crowd? Google didn't help much.
I tried to read GCBC about a year ago since it was recommended to me here but never finished it, because I wouldn't have had time to check the sources and continuing without checking would have been quite useless. I hope you do a good job with the sources and people jailbreak them if necessary.
He's referenced a lot by paleo and ketogenic diet bloggers. Not sure if he holds much of an audience outside of there.
Wise, you are.
Luckily, with a couple exceptions, the key sources have turned out to be available online.
When you accuse Taubes of misrepresenting others' views, you touch on an important point. Before you can trust the expert consensus, you need to determine (1) what the proper class of experts is; (2) what exactly the issue in controversy is; and (3) how the experts actually stack up on that issue.
If it's a controversial issue, you can bet that the above 3 meta issues will also be controversial. If you have the intelligence and critical thinking skills necessary to resolve the 3 meta issues, then you can probably make a good assessment of the underlying controversy independently. On the other hand, if you are self-deceived or stupid about the underlying issue, then you will probably fool yourself on the meta issues too.
So "trust the expert consenses" doesn't seem all that helpful to me.
All this is true. However, these variables are not always perfectly correlated. It is important to recognize cases where some or all of (1), (2), or (3) are easier to answer than the object level question. That is when trusting the expert consensus is a good idea.
Well I think all 3 need to be satisfied. But that only happens when the issue is non-controversial. In which case you can just look up the answer on Wikipedia.
If an issue is controversial, it's a pretty safe bet that one or both sides will (1) distort the issue to make the other side seem less reasonable; (2) exaggerate the extent to which supposed authorities agree with them; and/or (3) attempt to choose a class of supposed experts which is most favorable to their position.
Will this be a steelmanning? I don't much care about the quality of Taubes' arguments unless they're the best arguments to be had, and I'm pretty sure they aren't.
I definitely think it is interesting criticised as it is, without steelmanning.
I agree, if only because Taubes' arguments are probably more widely known by a large margin than the steelman arguments.
If your goal in reading these posts is to (possibly) come away with the best possible argument against the Taubes-esque position on nutrition, then a steelmanning is what you should desire.
If your goal in reading these posts is to help others understand why they should or shouldn't listen to Taubes' advice, than a critique of Taubes' specific arguments will likely be helpful.
Of course, this probably depends on how close Taubes' arguments are to a steelmanned version...
Right, there are two interesting questions that steelmanning would obscure:
I'm mainly interested in Taubes because of what people cite him as supposedly showing, which requires dealing with his actual arguments. I will to an extent end up looking at actual evidence that supports the effectiveness of low-carb diets. I'm not sure how to steel an the wilder accusations Taubes makes against mainstream nutrition science, except maybe to frame it as a case of bad science communication, but... actually, come to think of it, I should probably talk about that.
Is that really the mainstream claim? Because it doesn't make sense from the engineering standpoint, given that no burner is 100% efficient. Or does "burn" misleadingly include "expelled without burning"? Are there discussions/measurements of the human burning efficiency and how it depends on metabolism and why?
I feel like the calorie issue is a red herring. Diets that require expenditure of willpower almost always fail, regardless of details. The new diet has to be the new normal, without being a source of stress.
You should separate out the need to spend willpower to limit superstimula foods, with that required to reduce your total caloric consumption to where you are always hungry. Any sane eating plan for humans requires the former, whereas a diet requiring the latter is probably doomed to fail.
{Edited in response to below.}
Reasonable, but a fair number of people seem to be able to eat at least some superstimulus foods in moderation, and it might be more sensible to include the possibility of limiting superstimuli rather than eliminating them.
Agreed and edited.
I've never had a problem with weight loss, but on reading on the issue, one insight I've gained is that if you are always hungry on a diet that maintains a healty weight, that is strong evidence that something is misaligned hormonally, be it due to food you are now eating, genetics, some past syndrome developed by diet or exercise, etc.
Perhaps it would help to have a clear definition of "willpower." Does getting up and going to work every day require "willpower"? Based on my understanding of "willpower," I would say it does, although perhaps not as much willpower as sticking to a diet.
Diets that require expenditure of willpower fail because the willpower expenditure indicates that something is going wrong with your body, which may well be shutting down or slowing down. Strangely enough, these same weak-willed dieters often seem to have little trouble exercising hard - for years on end, maybe - though often to no avail, of course.
Diets that "don't require willpower" are the diets that are not wrecking your metabolism, it may be as simple as that.
I seem to be able to come up with various willpower draining activities that are useful, so how should I proceed if I wanted to know a particular diet is such an activity instead? Evolutionarily speaking hunger is clearly an indicator that something is wrong, but I'm not sure how that translates to values of a modern obese human.
Well, exercise is known to have other benefits than weight loss, so I wouldn't encourage people to stop moving their bodies on those grounds. Also, this seems to be the kind of useful activity I was talking about.
It's probably more fair to say that diets which require willpower indicate that something may be going wrong with your body.
The body has a semi-reliable warning system.
I've often wondered, since first learning the little factoid that "the stomach has almost as much neural complexity as the brain", whether we've mis-located the source of the 'willpower' problem.
If our digestive tract is a fully-trainable neural net with direct two-way feedback to our brains (and it certainly seems like it is), then it might be fruitful to start figuring out how to re-train it directly, or short-circuit it. Concepts like "reward circuits" and "willpower" almost invariably focus on the brain in our heads; has anyone done any studies on what the stomach's internal reward circuits look like?
I'm not familiar with much of the rigorous science on gut stuff, a lot of it is fairly new. I will say that the Cue Action Reward model of habit formation works the best when you use the strongest cues and rewards, i.e. sex or food. I started training fasted a couple years ago. I didn't realize that I was reprogramming my cue action reward from
to
to the degree that when hungry my first thought is now to go exercise.
Also, many people who don't succeed at losing weight have notable professional accomplishments.
There are people who maintain weight loss through will power. I don't know whether it's actually good for them.
You only have one pot of willpower. If you use up all your willpower doing professional tasks, exercising, etc then you'll have less willpower to resist over eating. If you use up all your willpower for weight loss, then you'll have less willpower for other tasks.
This article (and other research done by the authors) talks about that sort of thing. This is why diets that work are ones that reduce the amount of willpower required, or shift the willpower requirements to situations where it is easier to control your decision making. For example, a low carb diet works because you simply don't buy carbs at the store, so when you're at home, you don't have to expend willpower when you're hungry and deciding what to eat. Intermittent fasting works because you simply don't eat breakfast.
I eat for pleasure and I have a hard time moderating my intake, so I've worked around this by reducing my breakfast to coffee, yohimbine (caffeine/yohimbine is great for energy and appetite reduction), and fish oil (~2g). Lunch is two cups of milk and two scoops of protein powder as these are all that I have at work, so it is the only option -- no willpower used. This puts my daily calories low enough that I can eat as much as I want for dinner and still lose weight consistently.
Maybe this is just a verbal quibble, but how can someone "consistently lose weight"? Losing weight, if successfully done, results in reaching some target, and then you don't lose weight any more.
I'm training for strength sports, and this involves cycles of gaining weight to add muscle and losing weight to cut fat. As such, I've done a number of weight loss diets in the past two years, and this is what I've found to work every time for me (or, consistently).
Of course for some people it seems to take heroic amounts of willpower not to go shopping for more unhealthy food or go to a restaurant during the lunch break.
I'm experimenting with VLCD right now, and it doesn't seem to take much willpower either. I think the calories are low enough that my body thinks it's starving although it gets everything it needs, so I'm not constantly receiving hunger signals.
Danger! The book the Perfect Health Diet has some very negative things to say about this. As I recall, this kind of diet greatly increases the risk of some diseases.
Danger noted. I'm doing this only a month or two, and don't recommend this to anyone else.
As a side note, will you agree to update everyone in 6 months or a year with how your experiment went?
I doubt I will do it longer than a month or two, but sure, success will be measured in a longer time span. I think losing weight is significantly harder than maintaining it, it's just that most people fail to include maintaining weight to their plans.
Starving for a year doesn't sound like a great dieting plan either.
This is not true. Willpower expenditure is required for all sorts of decision making activities, so that a diet requires willpower expenditure is not evidence that it is damaging your body. A diet that requires significant willpower is severely suboptimal, which is why very few people report success with intense calorie counting and moderation, and why very many people report success with "trick" diets like Shangri-La, paleo, keto, intermittent fasting, etc.
As I understand, outside pathological cases human digestion pretty consistently manages to absorb 95% of the calories we consume (excepting dietary fiber, which we can't digest at all). Calories lost as heat count as burned. There is one complication to this that I'll discus in the next post, but it turns out to be a minor factor.
I've seen a plausible claim that people with lactose intolerance get fewer calories out of dairy products.
The lactose is digested by gut bacteria, the byproducts of which are the problem. By Chris's definition, that counts as digested.
In this claim, "consume" means the number of calories absorbed by the body, and "burn" means the amount of calories used (or wasted).
You can measure you resting metabolic rate. Most people are close enough to what formulas report that actually measuring your caloric expenditure isn't terribly useful, but some people (myself included) have slower metabolisms than calculators and formulas predict.
Are you referring to the energy in chemicals expelled from the body? Once through the digestive system (absorbed in the intestine) there are actually remarkably few ways for chemicals to exit the body. The main route is through respiration, through which CO2 is released. Urine expels liquid-soluble wastes. The energy contained in these wastes (that is, the free energy available from reacting with oxygen) is very low, so it's usually ignored. There are other minor routes (such as sweat and milk), which, aside from special circumstances (lactation) don't expel much energy either.
That might be because protein stimulates the production of insulin as much as, if not more, than carbohydrates, and for people who don't have a fucked up grelin/leptin feedback loop, insulin is a powerful appetite suppressant.
Are you stating this as fact or just hypothesizing? If as a fact, do you have a source?
A couple, actually:
- Here for the satiety effect of high protein intake;
- Here as just one of the several example of proteins inducing a high insulin response.
Low carb diets also restrict refined carbs, so people are less likely to bounce their blood sugar.
If you were looking for examples of expert madness to support your thesis and wrote up a counterexample instead, well done for going so far to avoid confirmation/'publication' bias.
The first post I was originally neutral on and negative on the second for the same reason as Jack. But now you've started to demonstrate some of Taubes' misrepresentations in the third post, I am more positive on the second post. I'm not sure whether it was avoidable (for any reasonable amount of effort) for the posts to be so non-sequentially dependent in their standing. I shall probably hold off on voting on the posts until everything's in.
I'm looking forward to the rest of 'em!
I do think the third post is somewhat dependent on the second one, as a large part of the problem is Taubes trying to make mainstream diet recommendations look like a mirror-image of the Atkins diet.
In fact, I wonder if I should have combined the two posts into one.