Comments on “Perfection Salad”
Comments are for the page: Perfection Salad
Food science since 1988
Hi, thanks for the comment!
Remember this was written in 1988, and I’ve reprinted it unaltered, because it’s interesting to see what has changed since then.
You are absolutely right that there’s been extensive advances in food science, and some of it is cool and useful. I mention that in the 2015 follow-up about nutrition science, although I don’t go into it in detail.
Where I would disagree with you is “modern science actually has quite a lot to say about nutrition” (taking “nutrition” to be the same as “knowledge about the health effects of differences within a vaguely normal range of diets.”) The follow-up is about that.
Although I don’t mention them specifically there, I don’t believe that there’s strong evidence for the relevance of fiber or glycemic index. These are more things you can measure, so people want them to be meaningful, but the available methods are unable to show that.
I don’t want to argue the details, because it will come down to “these 23 studies say glycemic index is important”; “yes, but 21 of them have obvious massive methodological flaws, and the other two show small effect sizes, and here are 17 studies that show there is no effect”; “yes but 14 of those studies were funded by the sugar industry, so we should discount them”; “well the two reasonably-good studies in favor of glycemic index were done in China, and their studies are almost always somehow bogus”; etc.
The broad point is at this meta level: individual nutrition studies virtually always turn out to be worthless, and when you do a proper meta-analysis, it turns out you know nothing. Ioannidis’s recent papers, which I link there, make this point strongly, for example.
Heh - I stand corrected . . . (but strangely unrepentant)
I owe you an apology, for taking this single post out of context. If I’d done my homework and read all the posts in their proper order, maybe I wouldn’t have jammed my foot so far down my throat. (Look on the bright side - maybe I’ll ingest some toe-jam bacteria, and accidentally improve the health of my intestinal microbiome.)
Having said that - I still think you’re overstating the case when you say science knows “nothing” about nutrition. A lot of data has been collected, but people have used that data to make wild generalizations. They’re sort of papering over the huge blank spaces on the map, instead of waiting for the research to slowly fill in the gaps. The human metabolism is wildly complex, and there really is a lot of individual genetic variation - as well as the good old intestinal microbiome, which apparently clusters in three or four main varieties, with individual variation.
I think the movie “Super Size Me” is an example of a good, old-fashioned, true scientific experiment. Here is before, here is after, and here is a single variable (diet) that led to a massive observable change. Admittedly, that single variable (diet) can be broken down into lethal overdoses of fat, sugar, preservatives, and what-have-you; if the guy was suicidal enough, he could have repeated the test adding nothing but massive quantities of sugar, massive quantities of fat, etc., etc., to further isolate the effect; but he did clearly establish the data that eating that particular combination of pseudo-foods had a rapid and observable effect on his physiology.
I don’t actually approve of that style of testing. It’s bass-ackwards. Why would you deliberately inflict harm on yourself, or another living being, trying to prove something is harmful? You could expend the same time, money, and effort testing something that might help. How about “Salad Bar Me”? I’ll eat nothing but salad for six weeks and see how my health is then. You could give your lab mice artichokes and tofu, and see if they do better on the treadmill, instead of force-feeding them fat and trying to measure the threshold where a heart attack is bound to occur.
Anyway -
I see that you scoff at glycemic index, but I have to say this. When I took my first nutrition class, 25-30 years ago, mostly what we were taught was “fat-carb-protein,” and how many calories we could eat based on gender-weight-activity level. Just in the last few years, I became aware of glycemic index, and I must say I’ve noticed better health since I’ve paid more attention to ingredients and not just the “carb grams” and “calorie” numbers on the label. I don’t have the dedication to document every moment of my time to turn it into a truly scientific study, but my body is my laboratory, and I’m in it 24/7/365.
Definitely, I agree with your main point that all the “food pyramid” and associated hype is nothing but dogmatism; it tries to co-opt science, but is not truly scientific. Now I’ll have to go back and read other posts I missed, to pick up the full thread of your argument.
Physics Envy
The tendency of imprecise disciplines to mimic the methods and standards of physics is comical.
Molecular Gastronomy & Procrustean Beds
I look forward to seeing you expand your argument, giving specifics from cognitive science and connectionism. I love the way you’ve chosen something completely left-field, like culinary science, to illustrate the underlying dynamics of a social movement, and point up the absurdity of its underpinnings.
Having said that, I do have to disagree with your statement that science no longer has to say much about cooking. For example, check out this link:
http://discovermagazine.com/2006/feb/cooking-for-eggheads
This so called “molecular cuisine” seems to be a legitimate discipline, as it uses what we know about physics to improve the texture, taste, and predictability of kitchen outcomes. It’s not blandly reductionistic, but is a tool for improvisation - though I still prefer just throwing stuff in a pan and using the old “trial & error” to see how it comes out. “Molecular gastronomy” has an entry in Wikipedia, so it “must” be real - right? ;)
Modern science actually has quite a lot to say about nutrition. In addition to calories, proteins, carbohydrates, and vitamins, you’ve got soluble and insoluble fiber (those good old fruits and vegetables, and your whole grains). Perhaps most importantly, you’ve got glycemic load (the speed at which different foods increase your blood sugar. Even the pH of foods turns out to be important: For people whose intestinal microbiome has been compromised by overuse of antibiotics, low pH foods can be disastrous because they let yeast expand dramatically and choke out the helpful bacteria.
If you’re developing these ideas for a book, you’ll want to incorporate these new findings, and clarify your argument in regard to the old “home ec” curriculum. It’s not that “culinary science” doesn’t exist - your argument seems to be more along the lines that “rationality” is a completely nonscientific believe system, that tries to cloak itself in the guise of “science” to influence social outcomes and control academic curriculums. In your essay about domestic science, you portray it as a kind of Procrustean bed that ignored all the data outside its agenda, and stretched the available data to fit - and this is obviously not what “science” is about.
I hope you’ll forgive me for kibitzing. I think you’ve got the bones here for a really great chapter, and I’d love to see how you develop it in terms of cognitive science vs connectionism (fields I know a lot less about). Thanks for another great post. I agree that “rationality” (as opposed to true experimental science) is responsible for much of what is twisted in society and our personal lives. I certainly was warped at an early age by rationalists, and I think I’m still recovering!