Original Cin

I think it’d be useful to step back for a minute, and talk about the logic of what Taubes says in Good Calories, Bad Calories. It’s nothing shocking or amazing, but there are a few small insights.

First of all, even though Taubes never says as much, the following is true over the long term.

Calories Stored = Calorie Income – Calories Out

Taubes is very, very upset at the “Calories in, calories out” mantra that’s brought up time and time again in debates on obesity. But, that is what the above says, right? Well, sort of. Obesity as a problem is too much fat, not just too many calories stored, so we need to recognize that calories can either be stored as fat or anything that’s not fat in the body (I’ll say muscle, even though that’s really far from correct). Also, the mantra implies that both calories in and calories out are variables that can be controlled directly by the individual.

Now, where you go from here depends on your point of view.  Just to get it out of the way, you could suppose muscle stays constant and choose an unorthodox way of increasing calories going out; this is the “Olestra and Bulimic” viewpoint.

(Sidenote, feel free to skip: Taubes very briefly speculates that anorexia might have a hormonal component to it, similar to the one he believes is behind obesity, but operating in reverse. The only evidence he gives for this is the fact that insulin was occasionally administered as appetite stimulant way back when. It’s very plausible — eating disorders in general seem to run against the grain of what people are normally driven to do.)

Supposing that there weren’t any changes in calories consumed (not a safe assumption), you could increase calories burned through exercise, which also ought to increase muscle mass, giving you an even better reduction in fat. Exercise almost certainly has health benefits, but weight loss is probably not among them. Taubes touches on this very briefly, and there was also a previous blog post on it, so I won’t spend a lot of time on it: what most likely happens when you increase calories burned through exercise is that calories consumed increases to compensate; also, activity at other points in time might decrease, lowering the number of calories burned.

Now, the conventional wisdom that Taubes is against; eat fewer calories. This is normally combined with exercise, above, but let’s consider it by itself.  If you hold muscle mass and calories burned constant, fat should decrease with calories consumed. If you want to be more sophisticated, you might say that calories consumed is a function of willpower (the set point hypothesis?), or stomach capacity (Taubes counters this by pointing to experiments where rats had their food cut with water and clay, but call it the high-fiber approach), or taste (the Shangri-la approach, I’ll blog about this later), or even the price of food (very important for understanding the politics and economics of Calories). If calories consumed and burned can be controlled directly, or at least not too indirectly, this is the correct approach; but this probably isn’t true, or people wouldn’t have nearly so much difficulty with weight control. Also, the existence of yo-yo dieting seems to indicate that it’s not safe to assume muscle mass remains constant.

Now, the story that Taubes outlines is fairly complex compared to the above.  He basically says that calories consumed is a function of appetite, appetite is a function of the energy available to the body, which in turn is a function of the behavior of the fat that’s already in the body, which is a function of insulin levels, which a function of carbohydrates eaten. Fat levels are a function of fat behavior too. Stating it this way, it seems kind of unlikely due to the long chain of causality. If there weren’t any evidence already on the table, I suppose I’d say, “let’s test everything else first.” But the orthodox view has had a lot of testing already, and been found wanting by most; the exercise view and the olestra view have their own problems. In light of the mountain of evidence that Taubes points to, again, I think it’s reasonable to say that fat levels are probably a function of carbs eaten somehow, even if not exactly as described above, and less certainly but still probably, so are both calories in and calories burned. Let’s call carbs “c”:

Muscle + Fat(c) = CaloriesIn(c)+ CaloriesO(c)

It’s a lot more than the conventional view, but just as you can’t run a multi-national corporation (or, indeed, even most smaller companies) just by chanting “Equity = Assets – Liabilities”, so might saying “Calories In, Calories Out” be insufficient for weight loss.

A few closing notes; I once saw someone argue against low-carb diets on the basis that muscle mass might be a function of carbohydrates; specifically, that muscles “need” carbohydrates to function and muscle will decrease with decreasing carb consumption. Taubes makes it sound very much like muscles will run on triglycerides very well (which I didn’t even realize were another “fuel source” in the body before reading Good Calories, Bad Calories). Also, Taubes makes a note that weird things can happen that make it appear like the above equality doesn’t hold, so it’s not actually all that useful for determining what to do; if you incorrectly assume that calories burned remains constant, and you don’t seem to lose muscle, and you eat more and you’re losing weight, the result is puzzling.