Obesity: Meal by Meal, Year by Year! (Hypothesis v. Hypothesis?)

Listening to Gary Taubes' recent interview, I found this song popping into my head from Godspell.

and this one

Why?  Well there's something about the whole "day by day by day"  thing from Godspell, and "weeks turn into years how quick they pass" line from Do You Know that popped into my head as I heard Taubes droning on about how obesity is not a meal by meal thing but something that develops slowly over years and decades even.

Of course he's right about that to large extent.  It's not what you eat for dinner tonight or isolated night, or any given day, or even any given week that ultimately results in obesity.  But it sure as heck is the cumulative effect of our short term choices.

Although he never directly addresses Stephan Guyenet's Food Reward series, he's clearly referring to various studies that look at how the quantity we eat at any given meal may be determined by what we ate at a previous meal, etc.  Apparently in defense of his abhorrent cherry-picking, Gary with one sweep of his forked  golden tongue, dismisses all such research as irrelevant.  

Well, I would ask Gary the following:  
  • If how much we eat at any meal is of no relevance, what relevance is the insulin spike from any given meal to our insulin levels over the years?   
  • If how much we eat is of no relevance, why do you repeat in this interview what you've stated many times previously (and LC'ers constantly tout) that folks eat less on low carb?
  • Why all the satiation talk then?  Who cares if carbs or fat or protein is satiating then ... right?  It's all irrelevant, right?
If the long term development of obesity is not about how much we eat in terms of calories at any meal, then it equally cannot be about how much we eat in terms of carbohydrates at any meal.  Insulin released in response to a meal is not forever.  The effects on lipolysis/esterification in adipose tissue are relatively fleeting.  In other words, this nonsense argument argues against TWICHOO as much as it argues against any other hypothesis.  

As I discussed in   Stick a Toothpick in It? Taubes' 4-pronged Carb/Insulin "Fork" Loses its Tines -- hmmm ... his TICH-fork? -- the immediate effects of insulin are the only remaining part of the "evidence" put forth in GCBC as to why carbohydrates are uniquely fattening that have stood the test of time.  The rest have all fallen by the wayside.  For this "toothpick" to hold any water, you HAVE to extrapolate "meal by meal" to "year over year".  Thus, Gary has given us permission to dismiss all of his cherry-picking as irrelevant.  Thanks Gary!

It seems to me that Gary Taubes is hopelessly mired in the terminology of statistical hypotheses.  He wants to couch his disagreement with Stephan in the context of a pair of competing hypotheses for a single unifying hypothesis on the cause of obesity.  First, there will never be one mechanism, therefore a single hypothesis, that will ever explain obesity in all cases.  This is silly.  Secondly even if Stephan's FR is a total load of bunk, this does NOTHING to validate TWICHOO.   The two hypotheses are not even remotely a competing pair of complementary hypotheses.  

Therefore, if Gary wants to school Stephan in how to do good science, and wants to support his accusation against Stephan of cherry picking, he needs to do it by falsifying Stephan's hypothesis specifically with studies and observations that must be ignored for it to be valid.  I think Gary tried to do that in his AHS (amazingly hypocritical snarking?) bout with verbal incontinence as he rattled off a litany of remote cultures that FR supposedly does not explain.   As if not being familiar with the BMI's of Boy Scout Troop #206 in East Bumfrig Idaho somehow discredits Stephan/FR.  It is all the more galling and frustrating that so many of these frequently cited groups utterly fail to support, and often outright contradict, TWICHOO.  I have one word for you:  Pima.  Ahh but he knows more about them than just about everyone on the planet ...  Not to mention that there are large, well known, carb eating cultures -- billions of humans -- including even his own generation  of Americans who did not become fat eating carbs.  

One of these days I would like to see Gary address the meat of Stephan's falsifications (and mine and Kreiger's and others') of TWICHOO.  Because it is Taubes' Wrong Insulin-Carbohydrate Hypothesis of Obesity.  Even if Stephan was a cherry-picker extraordinaire, it wouldn't excuse Taubes' continued inability to address the science that refutes his fantastical theories.  That's what this discussion should be about moving forward.  The science that supports, or rather the lack thereof, the TWICHOO.

To use Taubes' words, there is overwhelming evidence relating to his hypothesis.  It overwhelmingly refutes it, however.  

I see defenses of Taubes these days to be similar to someone attempting to defend Bernie Madoff.   In reality, Madoff is almost universally seen as a "bad guy" because of his Ponzi scheme and the ultimate decimating effect it had on so many lives.  You don't see people jumping to his defense because even if in the end he was a criminal, he (a) made a lot of people a lor of money along the way, (b) employed a lot of people between his own enterprises and in support of his luxurious lifestyle, and (c) probably gave a lot to charities over the years.  Similarly, these defenses of Taubes' so called contributions to the field of nutrition/obesity don't justify his professional malfeasance.  It matters not that (a) many who have gone LC because of it have lost weight, etc., (b) he's caused folks to kick the tires, or (c) he was right all along about cholesterol.  It doesn't excuse his behavior in not setting the record straight, in straightforward (written form) manner.  

Address and discuss the SCIENCE Mr. Taubes.  Unless and until you can do that, it might behoove you to rein in the ego and not presume to tell actual scientists how to do their jobs.  


M. said…
As to your third bullet point (satiation), I believe in the last Jimmy Moore interview Taubes said specifically that satiety signals have nothing to do with his insulin theory (the same interview where he said that some people have to be careful about the carbs in green leafy vegetables.)

It is funny to see how his insulin hypothesis actually puts him at odds with many other low-carbers (don’t forget that cheese and nuts are fattening because of their carbs/insulinogenic properties…), but these same low-carbers insist that Taubes’ insulin hypothesis is the one true theory of obesity.

The way all these contradictory ideas get blended together into a monolithic amalgam of “low-carbism” with Taubes as the guru is the worst sort of “science” possible.
Anonymous said…
I watched Guyenet's talk on vimeo last night. I thought everything was extremely sensible - except the apparent implication that set-point pertains to fat only. From my experience it pertains to the whole body, including muscle.

I would have preferred that Taubes wasn't included in the video at the end - but that did give me a chance to see for myself how snooty he is, as well as witnessing the ludicrousness of Taubes telling Guyenet or anybody else not to ignore evidence. I expected Rod Serling to pop up :)

I also noticed as you did that Taubes was criticizing Guyenet for not being versed in obscure every population study out there. Weird stuff.

Let me ask a question. Kindly tell me if the following is correct, about history of a few months ago. The Potato Diet story came out roughly at that time. Trying to defend Taubes' theory, others scrambled to posit that the whole insulin-makes-me-do-it theory therefore really only applies to people whose metabolisms have been deranged by years of overeating. (And that might very well be true.) The prime person behind that amendment to the theory was Peter from hyperlipid. It was never Taubes' own original theory.

Is that correct?