« Does Your Morality Care What You Think? | Main | Is Ideology About Status? »

July 26, 2008


So can someone summarize the single strongest piece of evidence on this subject?

It's never enough for research to find that something has an effect. It needs to determine the magnitude of that effect. Confidence intervals are crucial. Scientists publish crap based on dumb null hypothesis testing methods (p-values) which only prove minuscule effects that are most likely due to systematic errors in their method. The reliance on p-values is what got us into this mess. Researchers should be required to proved that an effect is big enough to be practically useful, not just different from zero. IMO, with the irregularities and biases there are in research, a diet should increase lifespan by a large margin something like 2 to 5 years compared to a control group before it is promoted.

.Robert Adkins was a board-certified cardiologist who went against the teachings of his professors, his colleagues, and the American Heart Association. He further chose to aggrandize himself and his ideas, and he became wealthy doing so, at a time when any form of advertising was considered taboo for legitimate physicians. Consequently, his colleagues and the "official" organizations regarding diet information labeled him a "quack". He was not a "researcher" by trade, any more than Albert Einstein was a scientist by trade while working in the patent office and thinking about relativity. Adkins great thought was that most of the bad things that come from any type of diet involve the excess stimulation of insulin. The Adkins Diet is not a "craze"; it is a recognition of human evolution. Refined carbohydrates are a recent human invention for which humans have not evolved to handle in a healthy way. They, especially high-fructose corn syrup, cause an over-active secretion of insulin that converts excess calories to fat, is associated with atherosclerosis, Type 2 diabetes, and has a not-yet-understood correlation with some cancers. The physiology was there for all to see, but its significance was presented by an unpopular messenger, and thus rejected. It is a perfect example of harmful bias.

So can someone summarize the single strongest piece of evidence on this subject?

He does a pretty good job of summarizing his argument in this hour-long lecture at Berkeley.

What struck me is not any piece of evidence outside my own experience (the weight just falls off with no hunger and I simultaneously feel great) but his analogy with growing children. He says, we don't say children grow because they eat too much -- we say they eat a lot because they are growing. Obesity works the same way when insulin gets messed up. People overeat (and under-exercise) because their bodies are begging them to. In the absence of carbs*, weight is more or less self-regulating just as most other systems in the body are. On a low-carb diet, one doesn't have to consciously limit food. The body does it for you.

*carbs means significant amounts of starches and/or refined carbs.

I don't know if I'm qualified to determine what would qualify as the single strongest piece of evidence on this subject is, but I can tell you what struck me most forcefully: In his Berkeley lecture, Taubes discusses the prevalence of obesity among poor, developing-nations women--fat women who bring their skinny children to clinics to be treated for malnutrition and starvation (he's got statistics, of course). He says something along these lines: If obesity is due solely (or even mostly) to overeating, then these women must be literally taking food from their children's mouths. We have no known model of mammalian maternity that shows even a minority of mothers doing such a thing, much less a majority. Either these women are monstrous freaks of nature, or something else is going on.

The conventional wisdom is that the total number of calories is what is important - while the macronutrient ratio is not terribly important - provided it does not lead to malnutrition.

Demonization of one macronutrient or another is a common trait of diet fads.

History shows you can sell diet advice more easily if you claim that fats or carbs are the problem - while the (admittedly rather obvious) idea that calories are the problem seems to be something that few are prepared to pay for.

This article critiques Taubes' original New York Times Magazine report, with a number of scientists claiming that Taubes had misrepresented their views:


Here is a blog posting (on a generally pro-Atkins-diet site) with a review of Taubes' recent book by George Bray, one of the most respected figures in obesity research; and a rebuttal to Bray's criticism, by Taubes.

TimTyler:The conventional wisdom is that the total number of calories is what is important - while the macronutrient ratio is not terribly important - provided it does not lead to malnutrition.

The "conventional wisdom" is wrong; that's what this is all about. There is a major difference between gaining weight and gaining fat. A local man who has been Mr. America, Mr. Universe, and Mr. Olympia stands five-feet, eight inches, and weighed 240 pounds at his competitive best. His body fat measured less than 5%. By every conventional index, he is obese. More calories than you need will indeed cause weight gain, but what kind of weight you gain is highly influenced by insulin. I was taught the conventional wisdom in medical school; to say again: it is wrong.

The conventional wisdom is that the total number of calories is what is important

It's diet and exercise that conventional wisdom says are important. A given intake of calories may or may not lead to obesity, depending on the level of activity.

"By every conventional index, he is obese."

Nonsense. The definition of obese is greater than 30% bodyfat. BMI and life insurance tables are just quick rules of thumb and explicitly warn that they are not useful for the extremely muscular.

Agreed with Tim. Personally, I ended up simply monitoring my calories, cutting my intake from probably 2500-3000 to around 1500-1800 per day. I did not follow any particular diet, in fact I had lots of fast food and hardly exercised. Still, I lost all my excess weight easily this way. Once at a healthy weight I found myself motivated to exercise and improve my diet.

My guess is a lot of people don't realize how many calories they are consuming. So this allows people to carry on imagining things like, "I eat very little but still gain weight" or for a thin person, "I eat as much as I want and never gain weight". And many are just unwilling to endure through any initial discomfort of being hungry when reducing consumption. Of course this is just a biased view based on my own experience.

billswift: "By every conventional index, he is obese."

Nonsense. The definition of obese is greater than 30% bodyfat. BMI and life insurance tables are just quick rules of thumb and explicitly warn that they are not useful for the extremely muscular.

Billswift, your comment is nonsense. Clearly, you are not a medical professional, nor have you ever had to represent an overweight person to their insurance company or their potential employer, as I have. The medical definition of obesity is: Obesity-the state of being well above one's normal weight. There is good reason for this: the increased weight puts a greater strain on the heart and the vascular system, regardless of its composition. See: Arnold Schwarzenegger's cardiac history. Some advice from the medical profession for some of the commenters on this blog: in terms of medical reliablity, anecdotal experience is equivalent to lies.

Taubes' hypothesis is that the hormones make you eat, or stop eating. The caloric explanation is pretty much like flogiston. It does explain everything but then it does not explain anything. Caloric explanation does not tell why a female has more fat than a male, why growing kids start eating more, or why expecting mothers suddenly become very hungry. Hormones, on the other hand, explain handily all those cases.

The critique linked to above seems pretty damming. So the fact that Taube goes on about bias so much leaves a bad taste in my mouth. Perhaps it is a bad sign when people go on at length about bias against their views - do such people just tend to be wrong?

Robin Hanson:The critique linked to above seems pretty damming.

That article was written in March 2003. I suggest you search for recent articles that have vindicated the Atkin's ideas about avoiding refined carbohydrates. Taubes may have written a biased book with inaccuracies, but that does not discount the underlying fact that humans have not yet evolved to handle refined carbohydrates in a healthy way. Atkins never said that you could eat as many calories as you want and lose weight, and yes it is true that people tend to feel satisfied with fewer calories when they produce less insulin. But the real point that readers such as Tim Tyler seem oblivious to is this: all the numbers are better and there is less atherosclerosis on a low-carb (refined carb) diet. Yes, Atkins was a showman; coincidentally, he was correct.

An interesting note: I read Robin's comment saying the critique was damning, but am currently in the middle of writing another post, so did not divert to read it. I initially found Taubes's view pretty plausible, but if Robin says there's a damning critique out there, there probably is.

So in my current state of mind, if someone asked me, "Do you believe in Taubes's view?" I would have to say "No. It seems pretty plausible on the surface, sure. But a trustworthy friend said in a blog comment that there was a damning critique of it, although I haven't read the critique yet."

It feels like a pretty unnatural state of mind and I have to work to maintain it, which may explain why there's so much disagreement in the world.

The latest large study of low-carb vs. low-fat (and vs. Mediterranean, which also did well) revealed low-carb to work significantly better for weight-loss and cholesterol:

Average weight loss for those in the low-carb group was 10.3 pounds after two years. Those in the Mediterranean diet lost 10 pounds, and those on the low-fat regimen dropped 6.5.

More surprising were the measures of cholesterol. Critics have long acknowledged that an Atkins-style diet could help people lose weight but feared that over the long term, it may drive up cholesterol because it allows more fat.

But the low-carb approach seemed to trigger the most improvement in several cholesterol measures, including the ratio of total cholesterol to HDL, the "good" cholesterol. For example, someone with total cholesterol of 200 and an HDL of 50 would have a ratio of 4 to 1. The optimum ratio is 3.5 to 1, according to the American Heart Association.

Doctors see that ratio as a sign of a patient's risk for hardening of the arteries. "You want that low," Stampfer said.

The ratio declined by 20 percent in people on the low-carb diet, compared to 16 percent in those on the Mediterranean and 12 percent in low-fat dieters.

Re: Weight Loss with a Low-Carbohydrate, Mediterranean, or Low-Fat Diet

Is that relevant? The study was not isocaloric. Each group consumed a different level of calories. The calorie deltas after 6 months were: low carb -560 calories, Med diet, -254 calories, low-fat, -458 calories. The most obvious explanation for the varying levels of weight loss is that the participants consumed varying levels of calories.

Re: all the numbers are better and there is less atherosclerosis on a low-carb (refined carb) diet

As I understand it, this is not true. We still know relatively little about the effects of low carb diets - which is part of the reason why nutritionists rarely recommend them.

In any weight control discussion there are three considerations that rarely get mentioned, much less intelligently discussed. The first is biochemical individuality which amounts to genetically determined variations in the makeup of cells. The second is physiological variability that determines, to some extent, the absorption efficiency (most writers seem to believe that caloric absorption is 100% for everyone) of nutrients from the gut into the bloodstream. The third is unabsorbed calories. It's estimated that, on average, only 80 % of calories are absorbed. The remainder pass out of the body without contributing to the body's caloric energy needs.

As I seem to be the only one trying to drum up interest in the unabsorbed calories phenomenon, just Google "David Brown unabsorbed calories" or "David Brown calorie excretion" for further discussion of this physiological phenomenon.

Tim Tyler:Anyway, Taubes, seems to deny the basic thermodynamic perspective.

No he doesn't. Read the book. He's very clear on this point. Anyone claiming he is deny the basic thermodynamic perspective are making a mistake.

The hypothesis Taubes discusses as being open is that refined/starchy carbs encourage people to want to eat a calorific surplus.

Tim Tyler:The most obvious explanation for the varying levels of weight loss is that the participants consumed varying levels of calories.

Yes of course, but interestingly, from the method of that study, only low-carbers were allowed unrestricted calories, yet had a large deficit still. Taubes discusses this style of finding.

My only criticism of the Taubes books is that it is so detailed that it can be dull and hard to complete. He painstakingly covers the history of these issues, the studies involved, the science involved.

Let me aggravate Eliezer's difficulties by disagreeing with Robin about the critique being damning, for a trio of reasons that I will not disclose at this time, and say that I assign significant probability mass to Taubes' claims, but not as much as 25%, and that the probability mass that I assign to them is strongly correlated with that which I assign to other controversial medical claims, and more weakly correlated with that which I assign to some controversial non-medical scientific claims. I would also say that Taubes' claims being correct correlates significantly and negatively with UFAI risk and positively with weakly and positively with risk from gradual civilizational collapse. In any event I express sympathy for Taubes' being regularly and predictably misrepresented, and I believe, based on interviewing him, that he is actively trying to use the most sound epistemology he is capable of, while largely constrained by poor long-term memory.

Thanks for chiming in David Brown.

Foods high in specific kinds of carbohydrates are natural fuel sources for humans. We are not designed to be able to efficiently extract calories from non-carbohydrate sources. Meat makes us feel full, but doesn't provide calories efficiently.

Increasingly in human societies, vegetable and plant matter is expensive while starches and fats are cheap, and activity levels are far below what they once were. The obvious result is that people gorge themselves on nutrient-poor, calorie-rich foods and then don't burn off the calories, so they become fat.

High-protein, low-carbohydrate diets aren't the solution. We know that they affect brain function and do not correspond with the vast majority of ancestral diets. Their only real benefit is that they let people lose weight while not having to restrain their urges, which most humans are incapable of to any real degree.

The critique linked to above seems pretty damming.

Robin, did you read the response from Taubes? If so, did you find it unconvincing? I thought it was pretty strong, especially the part about the USDA study.

Having read some of the back-and-forth and watched the Berkeley video it seems to me Taubes and his critics (other than Fumento) are often in violent agreement with each other on the underlying facts; the areas of disagreement are speculative and have to do with which of those proven facts are thought to be more relevant to the questions at hand. Or with which facts we might expect to be proven in the future, if we did suitable studies. I like that Taubes is careful to call his thesis "an alternative hypothesis" and wants to do randomized controlled studies to test it.

damning critique? have you bothered reading his response? oh i forgot, too busy writing new posts. way to contribute to the discussion. i'm horrified this nose snub 'scientist' gathering doesn't understand thermodynamics basics. watch the berkeley video and from the explanation of energy equation you'll understand why he is correct vs the scientist brethren working on fat grants all their lives. i didn't see in that video fat gorging promotion. maybe atkins was a kook, don't know didn't try his diet and banning fruits sound like a dumb idea to me but the premise of what taubes talks about looks logically sound. not bayesian of course because only a few chosen ones can comprehend that. and they're busy writing new posts and commenting bullshit on the side.

Uh: A major reason for the obesity epidemic is the widespread assumption that "it is a simple question of how many calories you consume each day and how many you burn." - David Brown

Re: Anyone claiming he is deny the basic thermodynamic perspective are making a mistake.

Well, it's an approximation. Taubes seems to think there is a metabolic advantage to low carb diets - at least he seems pretty critical of these comments - which present the conventional view that it is mostly calories that matter:

"the results of several of the [low-carb] studies actually refute the contention that low-carbohydrate diets, in the absence of energy restriction, provide a metabolic advantage for weight loss."

If you must get the message that the bottom line is calorie intake out by writing a book entitled "Good Calories, Bad Calories" for goodness sake, let the good calories be the ones in nutrient-dense foods, and let the bad calories be the empty ones.

OMG i just read bray's 'critique'. who are these people? are these the specialists that tell americans what to eat and what not? where do you find these geniuses? i wonder if he presented any papers with that logic. i might get on a deep fried chicken diet now just for spite.

Tim Tyler:

Is that relevant? The study was not isocaloric. Each group consumed a different level of calories. The calorie deltas after 6 months were: low carb -560 calories, Med diet, -254 calories, low-fat, -458 calories. The most obvious explanation for the varying levels of weight loss is that the participants consumed varying levels of calories.

The numbers are staring you right in the face! If the caloric hypothesis were true, we'd expect the low-carbers to do only slightly better than the low-fatters, with both being significantly better than the Med diet. That is not what happened.

Re: all the numbers are better and there is less atherosclerosis on a low-carb (refined carb) diet

As I understand it, this is not true. We still know relatively little about the effects of low carb diets - which is part of the reason why nutritionists rarely recommend them.

Did you not read even the summary of the study I linked to above? Cholesterol ratios were improved the most under the low-carb diet.


Their only real benefit is that they let people lose weight while not having to restrain their urges, which most humans are incapable of to any real degree.

Even if that were the only real benefit, it's an enormous one, and probably the single most important factor in long-term weight control. Almost anybody can lose weight once or twice. Keeping it off is a different story. People are too dismissive of those who can't restrain their urges, as if it's some sort of moral issue. Try holding your breath until you pass out sometime, resisting the urge to breathe, and you'll maybe get an idea of what those urges are like for many of the obese and the price they'd have to pay to resist those urges.

I mean, just think about it. Being obese sucks, so there's a strong incentive not to be. Obviously, the urges are stronger still than that very strong incentive. Low-carb diets simply eliminate the physical, urgent urges, leaving only psychological ones, which are much easier (although still not easy) to control. You literally never have to be hungry to succeed on a low-carb diet, while restricting calories while eating carbs requires a constant struggle.

To be very clear, I do not claim to have mastered this literature. It does seem clear that academic consensus is against Taubes, and a cursory look doesn't tempt me to deny that consensus. Deviant prediction market estimates here would of course influence me much more.

Re: The numbers are staring you right in the face!

The study is totally hopeless at examining the role of calories in the results. Not only is the study not isocaloric - check out the error bars on the calories at six months:

Low carb -560 +/-1568, Med diet, -254 +/-761 , Low-fat, -458 +/-1413

Please wash your mouth out with soap before citing this irrelevant study back at me a third time.


The error bars is a good point, but low-carbers were specifically told they could eat as much as they wanted, and they still lost the most weight. I think that's a pretty important finding regardless of error bars.

A few points:

1. The article by Michael Fumento -- called "damning" above -- wasn't a review of the book; it was a response to the much shorter NY Times article that Taubes wrote back in 2003.

2. Taubes' response to Fumento seems to me to have the better of the case, particularly as to the insinuation that Taubes had misquoted various researchers. http://www.reason.com/news/show/28721.html

3. You can view Taubes' book as having two overall themes: A) The conventional wisdom on fat vs. carbohydrates has arisen largely via slender-to-nonexistent evidence viewed through confirmation bias by people who routinely ignored counterevidence; B) The better view is that carbohydrates are largely responsible for obesity and many serious diseases, and here's a thorough explanation of the biological mechanisms that might account for such a result. B is more speculative and contrarian, and people might differ as to whether Taubes makes the case convincingly. But to my mind, the book establishes A overwhelmingly.

From the comments, especially those of Tim Tyler, who seems mostly to enjoy seeing his name in print, most everyone seems to be overlooking or ignoring the fact that the "low-carb diet" is meant to be "low refined-carb diet". No matter what Tim Tyler says, avoidance of refined carbohydrates leads to improvement in blood profiles and reduction of risk-factors for cardiovascular disease, with or without weight loss. While in other academic pursuits it may not hold true, most of my physician colleagues subscribe to the cliche in medicine that those who can, do; those who cannot, teach. Or as one Italian professor told me: "I nefer learn anything; I teach."

Robin: It seems clear to me that academic consensus is against Taubes... and that academic consensus is against consensus. Consensus among sociologists is that economists are right wing tools of the man. Among economists its that sociologist don't know anything significant. To physicists there is Among medical researchers, its that medical interventions are highly useful, individually and collectively. Ditto among political scientists, but not among public health experts familiar with RAND and other data. Academic consensus is... great when there is a consensus and an associated track record of practical accomplishment.

You can't validly use the existence of a consensus to condemn a polemic whose stated thesis is that there IS a consensus, AND it is built on blind trust in some really old, really bad science. You aren't making any claim that he disputes.

Two things:

1) many of those who posted above have obviously not watched the UC Berkeley webcast as the issues are addressed and plausibly answered - it's far from an infomercial for the Atkins diet and anyone who says it is is trying to sell a competing point of view rather than critique the argument; and,

2) I'm surprised that Robin and Eliezer rush to the defense of the frequentists almost reflexively - did I miss a recent change of hearts? Taubes refers to reproducible science re: genomics and metabolomics - since when is 21st century science automatically dismissed when 20th century frequentist "science" is to the contrary?

Review the bidding.

Taubes says that the conventional wisdom on this topic is based on inadequate research. Does anyone disagree?

The claim that you will lose weight if you absorb fewer calories than you expend, and you will gain weight if you absorb more calories than you expend is clearly true. Further, it is more useful than the stock market advice to "buy low, sell high". But it might not be useful enough. So for example David Brown is clearly right that we do not absorb all the calories we eat. It isn't controversial that we don't absorb all the fat, for example. You might eat more calories with one diet and still absorb fewer. And figuring out how people can get themselves to eat less might be a big deal. Telling people to eat less is better than telling them to buy low, but maybe not enough better.

It's obvious that not all foods provide the same health benefits. If you were to go on a vinegar diet, and you drank enough vinegar to gain weight on that diet.... Is it plausible that perhaps refined foods with a lot of carbohydrates might tend to be unhealthy, eaten to excess? Sure. How do we find out how much is excessive? Maybe the controlled studies that Taubes wants to do and which have not been done?

There's no particular reason to decide any known diet is right at this point. If refined carbohydrates are particularly bad for you, then an Atkins diet might be better even if it's bad for you in some other ways. Something different might be better still. We won't find out until we look.

Have I said anything controversial? If not, what is the argument about?

A common problem is that experts lie about consensus. I'm not talking about competing communities of academics, the way michael vassar is, but smaller groups that might be expected to have a consensus. Often equal sized groups claim consensus without acknowledging the other's existence. They might do it to improve the credibility of their claims to outsiders, but they fail to acknowledge the other group even within the community, so it may be simple ignorance. Stuart Buck's pair of claims represent two such communities, I believe.

A common problem is that experts lie about consensus.
An even commoner problem is that people pay attention to consensus when they shouldn't. Without sufficient evidence to support the position in question, a consensus is just a majority making an error. Who cares what the experts say if they cannot back up what they say with reasoned argument?

One thing that Taubes is right about is that a lot of research into diet is poorly done and the consensus is not always supported by science. Studies that support the consumption of whole grains compare eating whole grains to eating processed grains, without considering potential selection biases. The consensus in favor of numerous small meals each day seems unaffected by growing evidence (at least from experiments on animals) that eating as irregularly as every other day, even holding calories constant, can be beneficial.

I suspect there's a bias towards telling people to do things they don't want to do. Part of it is that if there's no drama in what you're saying, who's going to listen? And I suspect that cutting people off from their choices and pleasures is a status marker, and has its own satisfactions.

If I'm right about any part of this, then there will be a bias towards recommending extreme dietary changes.

And what if fatness isn't a good measure of health or the lack of it?

I remember the first time I read the theory that arteriosclerosis might be a matter of infection and inflammation. The idea that it might *not* be a moral issue about people living wrong was surprisingly hard for me to assimilate.

The health effects of how much fat you eat and what kind may be an open question, but I've never seen convincing evidence that weight gain/loss isn't in the end a matter of number of calories (in and out). Studies that suggest otherwise are usually based on subjects self-reporting their intake. I'd bet that, in terms of comparing low-carb vs. low-fat diets, the subjects on the low-fat diets would be more likely to underreport (quite possibly unawarely) because they're hungrier than the people on the low-carb diets.

Hi Folks, This is the first time I've seen your website and it's fascinating. I've read the Taubes book about five times (no kidding) and I am conducting an online discussion about the book with a bunch of friends: Professor heart patients, Diabetic scientists, High cholesterol engineers, obese librarians. Smart, well-educated people suffering from what Taubes calls the Diseases of Western Civilization. Having read the book so many times, having conducted so many discussions about the book and having my own experiences this is what I can tell you: 1) It's an amazingly well-researched book: thorough, comprehensive, critical, relevant. A beautiful example of research. 2) Taubes anticipates and has an answer to every one of your comments about calories, nutrients, metabolism, body-type, will-power, hunger, genetics, environment, evolution, scientific rigor and credibility, the appropriateness of epidemiological studies, researcher bias, specialization within medicine, diet(s), conservation of energy, exercise, cell structure and energy, adipose tissue vs muscle tissue, inflammation, the "Moral" issue of "not living right," homeostasis, scientists afraid to argue with consensus, government policy and funding. Really. All of that and more. 3) I'm frankly appalled at the number of people who are so fast to attack Taubes without having actually read the book. Skepticism is essential. But being dismissive without actually doing the reading? Well, that's just . . . stupid. And a horrifying example of bias. You have questions about the book, about the quality of the research, about the argument, well, then READ IT. --Katherine


If there were a God, I'd ask him (her) to bless you.

My problem is, why can't I trust the people I pay to be experts on this topic? My taxes go towards all kinds of scientific research and grants that are used to fund researchers who spend their entire professional lives developing expertise in this area. Why should I have to second-guess them, to become an instant expert myself, to seek out and read scientific papers, review articles, discussions of people's personal results?

Why is it that I, or Gary Taubes, can read the literature and learn the truth, while a professional with far more experience and knowledge than either of us is unable to do so? This is the fundamental paradox I run into again and again on controversial topics.

Two possible answers are, A, you can't, so trust the experts; or B, the experts are committed to the conventional wisdom and institutional forces prevent them from acknowledging the truth, while you are free from such prejudices. All I can really say is, I'll be really angry if B turns out to be true. Why should I have to do every damn thing for myself? Why can't I live in a world where people have a reasonable level of competence, where experts actually have expertise? For now, I just hope that A is correct.

Mr. Finney: There is also answer C: I am unable to understand the topic, so I must rely on the experts.

I came to this blog relatively recently because of a referral from the Singularity Institute, where the opening page reads: "All it takes is one technology – Artificial Intelligence, brain-computer interfaces, or perhaps something unforeseen – that advances to the point of creating smarter-than-human minds. That one technological advance is the equivalent of the first self-replicating chemical that gave rise to life on Earth."

If this statement is accurate (under answer C, I am forced to accept it), and if the leading (and only) research fellow at the SI is busy participating in morality blogs and deciding whether participants are trolls, I certainly hope that this seemingly irrelevant activity will somehow lead to the solution of the most important dilemma to occur since chemical replicators. And I hope it does so rather expeditiously, since I'm almost 62. I'd do it myself, but I don't know how.

Crush, I believe that you are right that weight gain is a matter of calories in -- calories out.

But let me make a metaphor -- stoichiometry in chemistry always works. Aftera chemical reaction you always have the same numbe of atoms, and you always have the same number of each kind of atom.

It's always true. But nobody thinks that's the whole story.

If you need 1200 calories to maintain your weight, you have to get them somehow or you'll lose weight. If you aren't losing weight, somebody who follows you around and records what you eat will find you're eating at least that number of calories. But it makes a difference to your health what you eat. 1200 calories of salad dressing is different from 1200 calories of wheat, or 1200 calories of cabbage, or 1200 calories of cow muscle tissue. The number of calories is not the whole story.

Mr J Thomas asks:
"Have I said anything controversial? If not, what is the argument about?"
Perhaps it is about Incommensurability? Per Thomas Kuhn: the ovelapping of successive paradigms.
It is my personal belief that we're approaching such a shift in paradigm with the publication of GC/BC.

"It isn't controversial that we don't absorb all the fat." I'm not trying to pick on Mr Thomas but only use this to make a larger point:

Perhaps it's true for Olestra (an artificial fat) but people adapted to a high fat diet (the Inui, the Masai ot the Kaniewski's Optimal Diet) have very efficient conversion rate for fat intake.

Conversely,excess protein intake beyond what's needed for cellular repair/rebuilding must be converted to glucose by the liver through a process called gluconeogenesis before the body can use as energy. This is an inefficient conversion process with significant amount of energy wasted as heat (thermogenesis). The reason behind eating protein promotes satiety?

Additionally, a large portion of the carbohydrates we take in (especially "indigestible" fiber) is actually "processed" by the gut microbiota and converted to fat for storage and our later use - an extremely efficient energy conversion process.

From PubMed:
"The gut microbiota is an important environmental factor that affects energy harvest from the diet and energy storage in the host."

In short, people who rely on the all-too-convenient and mis-applied Laws of Thermal Dynamics make some critical false assumptions about how the body (our body, as the host, and the trillions of microbes that colonize our intestines) process the food intake.

Experts are necessary because the world is simply too complicated for one person to understand everything that we need to know to be competent citizens. But that doesn't mean that we stop thinking critically. If anything, it means that we must be really, really good at critical thinking. If I listen to a bunch of "experts" -- let's say nutritionists, dietitians, doctors, advertisements for billions of dollars worth of low-fat foods, newspaper and magazine reporters, government recommendations and policies -- and they all say the same thing: fat is bad for you; meat is bad for you; high-fructose sugar is better for your blood sugar levels; yadda, yadda, yadda, then there appears to be a consensus. And consensus can be awfully compelling. But what if all that expert advice just doesn't seem to working in the real world?

This is what a good friend of mine eats: cold cereal with fat-free milk for breakfast. An "energy" bar and an apple (plus a banana if he's really hungry) for lunch. A salad with some chicken or salmon for dinner. That diet is almost nothing but carbohydrates: Grains, fruits and vegetables. He plays softball every summer. He golfs and runs. He is not overweight. He had a heart attack at age fifty. Everyone I know drinks low-fat milk and eats low-fat yogurt. They're all eating fiber. And they all have at least one symptom of Metabolic Syndrome: high blood pressure, diabetes, hyperinsulinemia, heart disease, obesity, high LDL levels, low HDL levels, inflammation. These are all good people who are trying their best.

And, so I start my critical thinking engine: it can't simply be a matter of will-power; these folks have plenty of will-power. It can't be about eating fat and meat because they aren't eating fat and meat. It can't be about lack of exercise because they exercise. It can't be about genetics because they had slender grandparents. Okay. I'm not an expert on this stuff, but it sure seems to me that the so-called experts aren't either. Maybe there's some other theory out there that I haven't heard about? Maybe it would be worth my time and trouble to read what the fringe has to say. And, so, I pick up a copy of "Good Calories, Bad Calories" and my entire understanding of human metabolism changes. AND, I see Thomas Kuhn's "The Structure of Scientific Revolution" at work. But I am, by nature, a skeptical person. I really did read Taube's book five times. I really did do other reading on the same subject. I really did ask some very pointed questions of my doctor. I really did try to think of all the objections to Taubes' argument and see how he dealt with those objections.

I'm convinced that Taubes is right. But that conviction is based on critical thinking. The problem with a lot of experts that they don't do enough critical thinking about the basis of their own expertise. Well, the problem with the human race is that we don't do enough critical thinking.

I think that experts are necessary: I don't know how to build a computer, I don't know how to feed the poor, I don't know how to make solar energy economically viable. And I depend on experts to provide me with information. But that doesn't mean I'm allowed to shut off my mind. Just the opposite, it means that I am obligated to think critically about that information, to make sure that it's "good:" accurate, relevant, reliable, timely, unbiased. There's no shortage of information, folks. It's always been a question of quality rather than quantity.

The thing is, I'm not saying anything original. Critical Thinking should be the goal of every single class, subject of inquiry, college course, book, scientific research project, blog, artistic endeavor, newspaper, website, human mind in existence. Everything else is just noise.

Thanks, Sophie's Dad, for the blessing. My particular brand of critical thinking has lead me to be an Atheist. But I think blessings come from people anyway.

I realize that I've taken an awful lot of time and space for someone who's just joined the discussion. I don't get to discuss these matters very often, so I'm, Uh . . . enthusiastic.


J Thomas, I agree that what makes up those 1200 calories makes a difference, absolutely, for long-term health; I was just questioning the claim that people could eat more calories on a low-carb diet and still lose more weight. (Although, eating carbs does cause you to retain more water, so you can lose a little more "weight" on a high-protein diet, but it's just water weight.)

Katherine, in Okinawa the average diet is naturally low-calorie and low-fat, and Okinawans have some of the longest life spans on the globe. So I don't think low-fat diets cause heart attacks. Michael Pollan has written interestingly on the matter of why our grandparents were slender and we are not.

It hasn't been proven with humans, but calorie restriction lengthens life span in every other species tested (provided adequate nutrition).

I certainly hope it doesn't turn out that saturated fat and high-protein diets are required for heart health, because I eat a vegetarian diet for environmental/sustainability reasons.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30