= 27d567bc2d48d9f40e9ccc20ea370b15)
Eliezer_Yudkowsky comments on Critiquing Gary Taubes, Part 4: What Causes Obesity? - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (110)
To me it feels like this whole series is simply failing to find a smoking gun of Taubes saying something false, or Taubes implying that the mainstream view is X when it is actually Y, because a bunch of individual papers saying Y does not mean that the mainstream view is not X. On a recent visit to an endocrinologist she still earnestly advised me that I ought to reduce fat in my diet because fat has 9 calories per gram. Recently glancing at a state-government handbook for pregnant women it contained the original food pyramid with half your calories supposed to be for grains (along with a recommendation to get folic acid which didn't distinguish folic acid from folate, no mention of iodine supplementation, and no mention of choline supplementation). This is what Taubes is criticizing and the fact that many experimentalists have found that this is terrible and published papers accordingly is part of his criticism, not a refutation of it; he is, precisely, accusing mainstream dietary science of ignoring its own better knowledge, and continuing to have endocrinologists and government pamphlets earnestly advising people that eating fat makes you fat.
The CDC diet and nutrition website at this very moment says:
To provide the context for those unfamiliar with Taubes' views on the polyunsaturated and saturated fats:
(Since most dietary fats are a mix of saturated and polyunsaturated fats, the layman accessible advice would be to cut down on fats, and on the other side, Taubes has to proclaim both to be ok).
Keeping in mind that 'saturated fat' is not fat with extra fat, it is the fat which doesn't have chemically unstable double-carbon bonds. I wonder how much of this idiocy is just because of 'saturated fat' sounding like fat with extra fat in it, and 'polyunsaturated fat' not being called 'poly-unstable reactive fat'.
That you would say this tells me your picture of how mainstream science works is even more distorted than I realized.
Studies have linked saturated fat to cholesterol levels and through cholesterol levels to heart disease. You said in the other thread that your nutritionist recommended you eat less saturated fat as a weight loss thing, which is wrong, because the real issue is cholesterol. But you shouldn't be dismissing worries about saturated fat as "idiocy."
Now, recently there have also been studies finding no link between saturated fat and heart disease. I haven't looked into the issue closely enough to tell what's going on, but it's possible nutrition science is in the process of realizing they made a mistake about saturated fat. Or maybe the experts have reasons I don't know about for discounting those studies, I'm not sure, which is why I've left the issue alone.
Yes, I agree that one needs to draw a distinction between nutritionism for weight loss and nutritionism for other health-related objectives. As far as I know, there is good scientific evidence backing up the latter.
That's why I asked Eliezer before what his doctor was hoping to accomplish by advising him to cut down on saturated fat. According to Eliezer, the doctor was indeed aiming for weight loss. Apparently the doctor in question is an endocrinologist. I would be curious to know what an obesity specialist would say.
The diet--heart attack link is debated as well, of course: http://en.wikipedia.org/wiki/The_International_Network_of_Cholesterol_Skeptics
Of course -- I am not surprised at all. But I still stand by my claim that there is good scientific evidence backing up nutritionism for health-related objectives other than weight loss.
Can you cite the one or two most convincing studies that saturated fat is a principle cause of heart disease? I have seen a lot from the other side and would like to get a fuller picture.
No, I can't -- not without researching it.
My claim is that there is good scientific evidence backing up nutritionism for health-related objectives other than weight loss. If you represent to me that you are seriously skeptical of this claim I will try to find some supporting evidence.
My skepticism at the moment is only about (quoting Chris from above) "Studies have linked saturated fat to cholesterol levels and through cholesterol levels to heart disease." (Well, and a bit with regards to cancer) I have seen such studies analyzed with alternative explanation convincingly by, for example, Uwe Ravnskov, Malcom Kendrick, Chris Masterjohn or Denise Minger, and wonder if there is more convincing evidence they atre excluding.
Nutrionism (meaning mainstream health science?) and health are of course much more broadly defined and not doubt contain many factual statements, for example on how to prevent scurvy, that don't necessarily protect the entire field from the possibility of error.
The fact that people with genetic hypercholesterolemia tend to ( as the cholesterol link would indicate) die of heart attacks fairly young is at least one strike against this view.
This is usually an acknowledged exception; whether the rationale for it to be an exception rather than counter evidence holds, I don't recall.
That you would say this tells me that your picture of how mainstream science works and the merits of Taubes' critique is even more distorted than I realized.
That was fun, but seriously. I posted it precisely because the nonsense about sat fat causing heart disease is one of Taubes biggest cudgels against nutrition science and it's something many experts are now admitting the medical establishment has been wrong about for decades. I'm confused how you came to a conclusion about Taubes without looking into it. It's probably what he deserves the most credit for.
E.g. Stephan Guyenet, whose arguments against Taubes account of carbs and insulin causing weight gain you posted earlier, thinks that no unbiased person who is familiar with the literature can believe there is a causal link between sat fat and cholesterol and heart disease.
It's possible Taubes is right about saturated fat, but he's sufficiently unreliable on other issues that I wouldn't trust a word he says about saturated fat.
And yes, this matters, because prior to this series Taubes was who everyone was citing for the "look how horribly wrong mainstream nutrition science is" claim.
FYI, I'm going to keep citing him.
Correlation is not transitive, i.e., X being linked to Y and Y being linked to Z does not necessarily imply X linked to Z. Unfortunately, biologists and health scientists make this kind of mistake all the time.
Weren't you trying to argue that mainstream science doesn't make mistakes?
One can also cut on fats to lose weight, if one doesn't proceed to eat more carbs and over-compensate in the end.
Normally, when someone is having trouble losing weight, they don't realise how many calories they consume, and errors with fat are the largest in terms of calories. When volition or judgement are impaired, that's when things get very messy and cutting down on fat may or may not result in eating more.
Didn't you say in the other thread that the advice was to reduce saturated fat? Is this the same doctor visit?
This is true. I wish we had surveys of the views of academic nutrition researchers on some of these questions, but as far as I can tell we don't.
That said, this isn't a random bunch of papers I'm citing. Rather, I'm citing the very same sources Taubes uses to show how absurd the views of people in mainstream nutrition science are supposed to be. Taubes is at the very least guilty of misrepresenting what his own sources say.
Furthermore, his choices of sources for the views of mainstream experts at least seem reasonable on the surface: FDA, NAS, Surgeon General, etc. Though I can't vouch for his description of them, Taubes describes the Handbook of Obesity as "edited by three of the most prominent authorities in the field”, and Joslin's as a "highly respected textbook."
Whether this is good advice will depend partly on what you replace the fatty foods you cut back on with, and how much fat you're currently consuming. And it's not entirely clear to me what level of fat consumption is ideal. That said, it sounds like she understands that calorie intake is ultimately more important than fat intake, and was not claiming eating less fat by itself was a sufficient condition for losing weight. What are you complaining about?
FWIW, research from the National Weight Loss Control Registry apparently indicates that the rare people who lose the weight and keep it off generally eat low fat diets.
Cite: http://www.nwcr.ws/Research/
My personal theory is that fat is not inherently unhealthy; it's just that if you eat a healthy diet you will naturally end up eating a diet which is low in fat compared to your typical American. In other words, I think fat is a somewhat useful proxy. But if you just focus on reducing fat intake, you will end up eating a lot of low fat potato chips and ice cream and drinking a lot of soda. Not good.