Gary Taubes: MAHA, Ultra-Processed Foods, and Bad Science
Science journalist Gary Taubes discusses the MAHA Report, new dietary guidelines, and bad nutrition science on the latest episode of Just Asking Questions.
Can Robert F. Kennedy Jr. put the country on a diet and Make America Healthy Again? Just asking questions.
The MAHA Report is a 73-page document put out by a government commission headed by Health and Human Services (HHS) Secretary RFJ Jr. Its goal is to "study the scope of the chronic childhood disease crisis and any potential causes including the American diet." The report points out that childhood obesity rates in the U.S. remain higher than in other G7 countries.
We invited Gary Taubes to join us today because he's a science journalist and researcher who has spent more than two decades studying the American diet. His books include Good Calories, Bad Calories, Why We Get Fat, and The Case Against Sugar. He writes about nutrition science in his Substack, Uncertainty Principles, where he has tackled the question of whether ultra-processed foods are the likely culprit driving America's obesity problem.
We also discuss whether RFK Jr. is more likely to improve or derail U.S. health and nutrition research, the recent resignation of one of the National Institutes of Health's top nutrition researchers, and a challenge to Taubes' "sugar hypothesis" from the pseudonymous blogger Cremieux Recueil.
Mentioned in the podcast:
"The MAHA Report," by the White House
Spectrum of processing of foods based on the NOVA classification
"Are Ultra-Processed Foods the Problem?" by Gary Taubes
"Dietary Guidelines for Americans, 2020-2025," by the U.S. Department of Agriculture"
"Nutrition Beliefs Are Just-So Stories," by Cremieux Recueil
"Why Conventional Wisdom on Health Care Is Wrong (a Primer)," by Random Critical Analysis
"Top NIH scientist speaks out, says research was 'censored' under RFK Jr.," NIH researcher Kevin Hall on MSNBC's All-In with Chris Hayes
Chapters:
0:00—What does Gary Taubes think of the MAHA agenda?
2:10—What is wrong with the "American diet"?
6:48—Why is nutrition science so sloppy?
10:22—What's in the MAHA Report?
12:25—Have childhood diseases and disorders really been increasing?
17:02—How bad are "ultra-processed" foods?
27:42—Using "Occam's Razor" to figure out what's making Americans unhealthy
33:15—Taubes replies to Cremieux's criticism of his "sugar hypothesis"
40:42—Critiquing the "definitive" NIH study on ultra-processed foods
52:43—How much does willpower or self-control matter in controlling obesity?
1:03:14—Why a leading NIH nutrition researcher resigned from RFK Jr.'s HHS
1:08:22—Why Taubes thinks Jay Bhattacharya might make the NIH more functional
1:18:54—What's a question Taubes thinks more people should be asking?
Transcript:
This is an AI-generated transcript. Check against the original before quoting.
Zach Weissmueller: Can RFK Jr. put the country on a diet and Make America Healthy Again? Just Asking Questions. The MAHA Report is a 73-page document put out by a government commission headed by Health and Human Services Secretary Robert F. Kennedy Jr., which aims to, quote,
"Study the scope of the chronic childhood disease crisis and any potential causes, including the American diet."
The report points out that childhood obesity remains higher in the U.S. versus other G7 countries. And we've invited Gary Taubes to talk about this today because he's a science journalist and a researcher who's written about and studied the American diet for more than two decades with books like Good Calories, Bad Calories, Why We Get Fat, and The Case Against Sugar. He writes about nutrition science at his Substack, Uncertainty Principles.
Gary, thank you for coming on the show.
Gary Taubes: Thank you for having me, Zach.
Liz Wolfe: However you feel about him, you can't exactly deny that RFK Jr. is a pretty major disrupter to the medical establishment. What has your reaction been so far to what he's done at HHS, but also the broader MAHA agenda?
Gary Taubes: I mean, he is a major disrupter to the medical establishment. I have it fall kind of in the mid-range here in that all my research has been about nutrition and chronic disease, and I think that the medical establishment has made some significant mistakes that have led us to where we are today. And yet I tend to trust them more on other issues like vaccines than I do RFK Jr. and his colleagues. So I can comment on the nutrition. The many other ways in which he's disrupting the major medical establishment probably scare me as much as they scare anyone else.
Zach Weissmueller: So yeah, what is your take then on the nutrition aspect? Before we get into the specifics of this report, just your overview, your early impressions of what they're pursuing.
Gary Taubes: Clearly we have a problem in this country that's nutrition-related, a problem around the world, which are obesity and diabetes epidemics that begin to appear when populations Westernize their diets. So anywhere you look, from the Arctic on down, you'll find populations growing obese and diabetic in association with eating Western diets. And then the question is: What aspect of the diet is to blame? And everybody's got an opinion on that. And I think everybody has an opinion because the science has been so bad for the last century that it hasn't nailed anything down the way you would hope that science would give you unambiguous conclusions.
So we have a problem. RFK Jr. admits we have a problem. The previous administration's approach to this was to push for personalized nutrition—the idea that we could somehow identify in advance what everyone's ideal diet would be.
And I think that's a pipe dream.
So I completely support the context for these reports. My work has always been arguing for far better science than has been done by the nutrition research community, and I'm not sure that the MAHA contingent knows what that means or has any idea how to do it.
Liz Wolfe: Of the things that RFK Jr. in particular sort of fixates on, you hear a little bit of talk about eating whole foods—you know, a whole apple versus a bag of chips or apple juice or something like that. But there's also a lot of talk of avoiding ultra-processed foods. And then one of his sort of pet issues is avoiding certain food dyes. Of those sort of three claims, in a few seconds or less, what do you make of those claims? Is he getting it roughly correct, or is he really veering off course?
Gary Taubes: I have major issues with the concept of ultra-processed foods in general, I don't know if we'll get to that—I'm sure we will get to that. The food dyes…I just don't know. Like, if I think of a classic example, is Fruit Loops, which were in the news about six months ago because they're made in the U.S. with a dye that's possibly considered harmful. So here's a product where it's refined grain, full of sugar. I don't know what the sugar content of a Fruit Loop is, but I'm going to guess it's in the neighborhood of 40 percent of the calories come from sugar. And then it's dyed to be colorful so it can be sold with a toucan on the cover of the box.
What's the problem? The food dyes or the sugar? If you take out the dyes, now can whichever company produces Fruit Loops—Kellogg's, Post, whoever it is—can they now advertise this as healthy? Because it doesn't have deleterious, even though 40 percent of the calories come from sugar.
I happen to be one of these people, old-fashioned, who thinks that the problem is the sugar content, and maybe the refined grain itself. And maybe the combination is particularly bad. But if you focus on the food dyes, or the number of artificial ingredients—which is what the concept of ultra-processed foods is driving at. This is an ultra-processed food, it's got multiple ingredients in it that are made in factories. It's got ingredients that you can't find in your own kitchen. And this is what makes it deleterious?
And you can fix all of that—keep the sugar, keep the refined grain—and still sell a product that I think on some fundamental level, as I've written in my books and documented, could be the trigger of type 2 diabetes and obesity epidemics. But now it appears healthy because you've cleaned up the additives and the dyes in it.
Liz Wolfe: So your take, if I may summarize it correctly—hopefully accurately—is: The fixation on artificial ingredients and on food dyes in particular is just basically a red herring. We don't really know—jury's still out—on how bad those are, or whether they're bad at all. But people are just allowing themselves to be really distracted by that, when in reality, we kind of have established that that amount of sugar is pretty bad for you. And so you're concerned about the sort of red herrings that are offered by the RFK crowd.
Gary Taubes: Yeah, although those red herrings come from the nutrition crowd. So this is not distinct to MAHA. This is the nutrition research community in general. If you look at the history of the field—and my books are all, on some level, history-of-medicine books—beginning in the 1960s, we focused on dietary fat as a cause of heart disease, and that becomes sort of conventional wisdom. The reason people think butter will kill them is because of the saturated fat content.
Beginning in the 1970s and 1980s, we started putting out public health guidelines to lower the fat content of the diet and replace it with carbohydrates. Everybody kind of vaguely hopes that the entire country will start eating like they do out here in Berkeley, California, where I live, but we don't. Instead, we started eating junk carbohydrates instead of the dietary fats.
And this all coincides with this explosion in the prevalence of obesity and type 2 diabetes. If you are old enough in this country, as I am, you can remember a day when you were in junior high school and there was only one child with obesity in your junior high school class. And now it's 15 percent—something like that, 1 in 8 maybe.
We've had this explosion in these diseases that's coincided with this focus on dietary fat—and an argument about sugar. And in the early 2000s, the nutrition research community, prompted by these researchers in Brazil, say, "Well, we've been arguing about macronutrients. We've been arguing about carbohydrates, sugars, about types of fat. And we don't eat macronutrients independently; we eat foods. We eat types of foods. So let's embrace this idea of the amount of processing that goes into the food." People who are poor and suffer high levels of these chronic diseases tend to eat a lot of processed foods.
The wealthier you are, the more you can afford to be health conscious, the more you can avoid those foods.
So we end up with this idea that looking at the macronutrients—the carbohydrates, proteins, and fats—and the type of carbs, proteins, and fats is 20th-century science, and we have to leave it behind. And instead, we're going to embrace the processing, the ultra-processing of the food. And then you get this idea where you're treating basically all junk foods alike.
Zach Weissmueller: Before we start talking more about the ultra-processed foods issue, I want to, for the audience, I wanna make sure we're properly framing what the MAHA Report is. It's not coming to any particular conclusions necessarily. It's pointing out a bunch of different trends and then raising questions about what is causing these trends. I mean, they're kind of throwing a bunch of stuff out there. I'll just pull up some examples.
They show a rise in autism rates among children, a rate rise in depression among teenagers, prevalence of food allergies going up, life expectancy in the U.S. lagging behind other advanced industrial countries. Although, yeah, as I'm pointing out here, I think there's possibly some other explanations for that, because you tend to spend more on health care as your country gets wealthier.
But putting all that aside—your expertise, or what you spend a lot of time writing about, is nutrition. But they're really casting a wider net than that and saying these food additives and these other things—they might be causing some of these other issues.
Do you have any opinion on that and their kind of general sort of scattershot approach to things?
Gary Taubes: Well, again, anything's possible, right? So I tend to think that the autism numbers come from changes in diagnostic criteria and changes in focus and issues like that. I don't actually find it compelling that the explosion in autism diagnoses is a significant change in the prevalence of the disease. But even if—
Liz Wolfe: By that, really quickly, like, what do you mean by "changes in focus"?
Gary Taubes: You pay attention to a disease. You start providing funding for the disease in schools—I'm sure you guys have covered all of this—and you start getting greater cases that might have been diagnosed as another cognitive disorder or as not diagnosed at all. We all know the weird kid in the family—who you used to just consider weird—and now he's diagnosed as on the autism spectrum.
Zach Weissmueller: Yeah, I don't know why you have to talk about me like that, Gary.
ADHD is another example of this, where there's perhaps behavior that when I was growing up, people would have been like, "He's just a really wild, high-energy boy." And now it's kind of causing disruption in the classroom, and we need to get him an individual education plan. And for that, you need a diagnosis, things like that.
Gary Taubes: And then if the school gets funding for that individual education plan, there's more motive—so there's a lot of forces. I first came upon this in heart disease because I'm trying to understand—there, in theory, was a heart disease epidemic that exploded in America after the Second World War. And this was sort of a mythology in the nutrition community because you have this explosion in heart disease prevalence, diagnoses, and it goes along with the change in diet. And the idea was: America started eating more saturated fat, fewer fruits and vegetables, and this caused the epidemic.
And I'm talking to an administrator at the National Institute for Health Statistics, and he says, "You know, this heart disease epidemic that they're always talking about never happened." And I figured, well, he should know, right? He's a big deal with the National Institute for Health Statistics—whatever the technical name is. This is 25 years ago that I interviewed the guy. And so he gives me the gist of it. I look into it.
It turns out that the American Heart Association, which was founded circa 1918, is a small organization of doctors. In the 1940s, during World War II, decides to become a major fundraising operation for heart disease, which has always been the nation's number one killer of men. But now they want to make sure everyone knows that it's the nation's number one killer of men. So they start broadcasting the idea—they have TV specials and huge fundraising events in which they're communicating this idea that coronary artery disease is killing people, and it's killing men particularly.
And they get people to care. And among the people they get to care, are the doctors who examine patients. So as soon as you start getting people to care, you see an increase in the prevalence of the disease. And you actually see an increase in mortality of the disease because the doctors who have to sign off on the death certificates are more likely to put coronary artery disease as a cause of death now that they're aware that it is such a common cause of death.
You can argue that virtually the entirety of the epidemic was caused by these changes in patients and physicians paying attention to the disease, patients worrying about whether they had the disease, insurance companies reimbursing for the disease, and the change in diagnostic criteria on death certificates, and the likelihood that a physician will choose one cause of death versus another when he's filling out the death certificate.
The same thing happened in cognitive disease, as it clearly happened in Alzheimer's disease to some extent. We start paying attention to it as a public health entity, for the most part, in the 1980s. And as soon as you do, people who might have just been old, crazy Aunt Alice, who is demented, now become crazy Aunt Alice who has Alzheimer's disease. And the numbers go up.
Zach Weissmueller: Yeah, I mean, this is an important caveat to keep in mind as we proceed with this—both the level of uncertainty around this and some of these other factors that could be contributing to "line go up" on these various graphs.
I want to bring the conversation back to the ultra-processed foods in this document that we were talking about at the beginning of the episode. I've pulled a page from it where they reference it, and on the diet section, they write:
"The American diet has shifted dramatically towards ultra-processed foods (UPFs), leading to nutrient depletion, increased caloric intake, and exposure to harmful additives. Nearly 70 percent of children's calories now come from UPF, contributing to obesity, diabetes, and other chronic conditions."
What are your objections to this category that we call "ultra-processed foods" that they are particularly fixated on?
Gary Taubes: Okay. So let's say the numbers—everything they're saying—is right. I find them a little extreme, but who knows? Then the question becomes: Why? That's what we want to know.
We're not going to get rid of ultra-processed foods. We're not going to get rid of processed foods. We're not going to make the entire country eat nothing but broccoli and wheat pilaf and pick your choice of low-fat proteins. So they're always going to be there. They're always going to have benefits. We need foods that have shelf lives, and that's going to require processing. We need food that don't go bad if you leave it out in the cabinet for two weeks.
So why are they bad? What's causing the obesity, diabetes? And then the question becomes: Is it the ultra-processing? Which is, again, defined by the number of ingredients on the label, the place where the ingredients are made, the place where the food is made?
For instance, a pizza that is made at home with four ingredients and the pizza that's made in a factory with six ingredients—one is ultra-processed and the other is not.
Zach Weissmueller: This slide I've pulled up is the NOVA classification of what constitutes going from unprocessed in Group 1 over to ultra-processed in Group 4.
For Group 1—unprocessed or minimally —they say fresh, dry, or frozen vegetables or fruits, grains, legumes, meat, fish, eggs, nuts, and seeds. The kind of whole foods Liz was talking about at the beginning.
Group 4, on the other end of the spectrum, sugar-sweetened beverages; sweetened, savory packaged snacks; reconstituted meat products; pre-prepared frozen dishes; canned instant soups; chicken nuggets; or ice cream.
And then, like right next to that would be just "processed foods," not ultra-processed, which is like canned vegetables, and meat, or cheese.
Liz Wolfe: Cheese has to be cured, hermetic…
Zach Weissmueller: Yeah. So are these categories that you are saying are not—it's not clear that they're extremely meaningful nutritionally?
Gary Taubes: Well, so if a product becomes harmful because it's ultra-processed—the idea is, like I said, if you look at the pizza on the ultra-processed menu and the ultra-processed slice—if that product is harmful because it's ultra-processed, because of where it's made and the industrial ingredients, the fact that it's not made at home. Is that product literally more harmful than if I just—you know, we all have this image of the Italian family cooking at home. They're making the pizza dough and flattening it out, and we've got tomato sauce that Salvatore made with sugar and salt in it and garlic and tomatoes, and we've got pepperoni that we bought at the local market on Third Street in Little Italy. That's not made in a factory—it's made by Uncle Joe out in Staten Island. And that's a healthy pizza.
So if I eat that pizza, I'm good. But if I eat the DiGiorno's pizza that I buy frozen in a box from the DiGiorno factory—that's also in New Jersey, I'm going to guess—that's an ultra-processed pizza.
And they have different health effects?
Zach Weissmueller: And you're skeptical of that idea.
Gary Taubes: Well, first of all, we should be skeptical. So the question is: Can we demonstrate that they have different health effects? And if they have health effects, that would be wonderful to know. Because now we could start identifying—we've got the gist that the pizzas are the same—the macronutrient content of the pizza, the fat content, the cheese, and the carbohydrate content in the dough, and the tomatoes, even the sugar content in the tomato sauce—that's all constant. We can leave that out.
And we could identify the cause of why this ultra-processed pizza—the frozen pizza—is bad by some scientific study. We could figure out, "Well, it's this 12-syllable chemical preservative, and that emulsifier, and this food dye," and we've established in some kind of scientific manner that that's what differentiates the home-cooked pizza from the deleterious ultra-processed pizza. Now we can make the ultra-processed pizza in a way that's no longer harmful.
Zach Weissmueller: So in that experiment, you would just feed people the DiGiorno's pizza for a week and feed another group the homemade, you know, Salvatore homemade pizza?
Liz Wolfe: That's how I source all my…
Zach Weissmueller: Maybe longer than a week—maybe a month. I don't know how long that would have to be. How would you design—
Gary Taubes: First of all, it's got to go for more than a week. One of the fundamental problems in this whole field—in the epilogue of my first book, Good Calories, Bad Calories, my first nutrition book—I made this point: that the nutrition research community, on some profound level, doesn't seem to understand what science is and how to establish reliable knowledge. It's like they're playing science. They dress up in white coats and they use multisyllabic words, but they don't fundamentally understand science.
So the problem is you're studying chronic disease. You want to know what causes obesity, type 2 diabetes, heart disease, cancer, even autism. Those diseases tend to develop over years to decades—not one week or two weeks. So you can't do studies for two weeks and expect them to tell you what's happening over 20 years.
I mean, a classic example: We could make you start smoking cigarettes. One group can smoke tobacco, and the other group can smoke something more benign. And in two weeks, you're not going to see any evidence of lung cancer.
On some level, you have to do this—and this is where we run into the problems that RFK Jr. and the MAHA crowd will run into. Because if you want to study the effect on chronic diseases, you'll have to take a long time to do it. And if you're studying it in foods—like just the pizza example—in an ideal world, we would make home-cooked pizza and the factory frozen pizza, and we would box them in such a way that our participants in our study didn't know which they were eating. Because you don't want them to think they're getting the killer ultra-processed food or the healthy home-cooked pizza.
We would feed it to them for a year. We'd randomize, say, a thousand people to the home-cooked or the frozen pizza. Again, they can't know which they're getting. We deliver them the food. And at the end of the year, we would hope that we would see some difference in health status between the two.
I was going to say, if we didn't, of course people who believe ultra-processed foods are killing people would say, "Well, that's because we only ran the study for a year. We didn't run it long enough." Or, "We didn't power it. If we had 100,000 participants getting pizza instead of just 1,000, we'd have seen the difference." And they'd be absolutely right. Maybe we didn't run it for long enough.
So this is really difficult stuff to do. When you're dealing with the emulsifiers, and the food dyes, and the chemical additives, you can do toxicity studies in rodents. Rodents because they have shorter lifespans you could at least take up a far greater proportion of their life in the study. Run it for a year, you get a third of the mouse's lifespan.
You don't know if what's bad for mice is bad for humans or what's good for mice is good for humans. So you run into other problems. But you could conceivably identify those problems.
Okay. Go ahead, Liz. I'm sorry.
Liz Wolfe: Well, I'm just really curious. With this discussion over ultra-processed foods that RFK Jr., he's not starting it, but we're very much going through a moment where that is a trendy thing to talk about. Is this all a distraction from what could be a simpler and more effective conversation, which is just: Care about your macros?
Like, the American public needs to consume more protein on average than they are, a little bit less processed carbohydrates, and less sugar. Is there a way for this conversation to be done much better, where we focus on people's macros versus whether or not people are consuming ultra-processed foods, which, as you know, is a sort of fake category?
In a sense.
Gary Taubes: So that's the problem. The nutrition community has embraced ultra-processed foods because they failed to solve the macro problem. So the research is—
Zach Weissmueller: You said in your writing that they call this 20th-century thinking. So it's like, we've gone beyond macros, and that's an outdated way. But you're saying that it was never actually settled.
Gary Taubes: Well, this is the thing. So, you think of science—famous principle in science, right? Occam's razor, which is: Don't embrace a complicated hypothesis if a simple hypothesis will explain what you're dealing with.
And so the idea is, you start with a simple hypothesis and you generate evidence testing that hypothesis. And if it turns out that hypothesis cannot explain the evidence, you either get another hypothesis or you complicate the hypothesis. You start adding other parts.
So the simplest hypotheses originally were: maybe it's just sugar. We're loading these kids up with sugar. Worse than that, we loaded their mothers up with sugar when they were young. And so, if this sugar idea is right, their mothers grew up to become more likely to be obese and diabetic when they were of childbearing age. And then they pass it on to their children. And their children are born more predisposed to become obese and diabetic when they get older. And all of this is triggered by the sugar content of the diet.
How would you study that? Well, then you have to tell people to get off sugar and basically see what happens. That's the easiest way to do it.
But these hypotheses have been around since the 1950s, 1960s. There had been people arguing for sugar, and arguing for refined grains in general, and people arguing it's all about the fat and the calories. And again, it was because the research community never got together to solve that problem, to really test those hypotheses.
And then around 2010, 2015, they just said, "Oh."
Liz Wolfe: That's such a frustrating thing. Because when I look at what has improved my own nutrition and how friends of mine have lost weight, it hasn't been through—by and large—fad diets. It's been through embracing a relatively simple set of principles.
Which is just like, "Wow, I really need to boost my protein content. Wow, I'm a little overnourished. I'm snacking, and I'm eating at times when I don't actually need that. And if I eat at discrete mealtimes and have a substantial amount of protein and a bit of fat at each mealtime, I notice that I'm kept full for longer. And so the temptations to drink my calories or to snack are just diminished.
And so I'm able to get into a state where food isn't just this crutch that I use in states of boredom." You combine that with being aware of how much sugar you're drinking—whether that's alcohol or sodas. I don't mean to oversimplify the field that you have devoted your life to understanding and explaining, but when I think about what's made a difference in my own life, it's just that.
And what if the MAHA messaging were something along those lines, versus trying to figure out whether Red Dye 40 is a big problem or not?
Gary Taubes: The short answer is yes.
Long answer is, we are going to get into an entirely different line of discussion. One of the issues we've always had in nutrition research is the leanish, healthy people tend to assume that the way they eat keeps them lean and healthy. And therefore, if everyone ate the way they ate, everyone would be lean and healthy. It's sort of a natural assumption.
The more likely scenario is that people who are born with a predisposition to become obese and/or diabetic would become obese and/or diabetic if they eat like you, Liz. And if they make the changes that you made, they might be a little bit less obese and diabetic—or might delay their progression to obesity and diabetes—but not enough.
And so the reason the fad diets exist is because those people find that they—and I used to be one of them—get heavier even when they're eating the whole foods, and healthy foods, and in moderation, and avoiding sugary beverages, and counting their calories.
And so, because the conventional wisdom isn't working for them, they go looking for something else. And everything else in that world is a fad diet. But everything else in that world kind of divides down to: avoid sugars, or avoid carbs, or avoid fats, or eat a Mediterranean diet, or eat lots of plants and vegetables, or eat only plants and vegetables. So you have various buckets you can put those into.
Somebody like you can get by with moderation. Somebody who's predisposed to being obese and diabetic cannot.
Zach Weissmueller: They might need to take more extreme measures in cutting out all carbohydrates and sugars.
Gary Taubes: The reason ketogenic diets, Atkins, even carnivore, work for people is because they cut out all carbs. And people are predisposed to have to cut out all carbs. I could be wrong.
Zach Weissmueller: Yeah, let me ask you about some of the recent challenges to that hypothesis. Because as this, I think, in large part thanks to your influence, this way of thinking has gained in popularity and maybe even found some foothold. You've kind of at least nudged the government dietary guidelines away from what they used to be.
But there are some critics out there. And I'm going to pull up one of them—it's a pseudonym. Goes by the pseudonym "CREMU." Pretty well known on Twitter and Substack. He posted a piece responding to your hypothesis that includes this graph, which we will link in the description.
The yellow line is the share of sugar that has been taken in by Americans over time. This is all according to the— it's American sugar consumption down, according to WHO and NHANES—I'm not sure what agency that is. And then the other line is showing obesity: the share of overweight and obese Americans continuing to go up over that same time period.
And CREMU here has even marked the second dotted line as "Gary Taubes coverage in like 2012," so therefore, the sugar hypothesis is invalid, and we need to move on. What's—
Liz Wolfe: It's kind of an asshole way to make the chart, huh?
Gary Taubes: Leave this up for a second. Because first of all, whoever CREMU is—he points out in this blog post that he only takes an hour to do the post. He actually has a timer. If he doesn't finish at the end of an hour, his program throws the post out.
And I had a comment on this—on X—was: "Let's be serious. The guy needs to take more than an hour to do his work." I mean, I admire the time-to-hourly-wage. He's probably doing very well for himself, turning out a post in an hour. But unfortunately, things are more complicated.
Liz Wolfe: Maybe he's hopped up on sugar. It's a natural stimulant for him.
Gary Taubes: I mean, I actually, in my early years, in my second book which was on this scientific fiasco called cold fusion. I ate Danjou pears, five a day. I would basically have a cup of coffee, I would eat a pear, I would get a sugar rush, I would write for an hour. Then I would take a walk, smoke a cigarette, come back, have more coffee, eat a pear, write for an hour. It was all the sugar rush from the pear.
Liz Wolfe: Cigarettes and sugar are the secret to your success?
Gary Taubes: Yeah, I also gained 15 pounds. And back then I thought it was—because back then, this was the early '90s—we believed that if a food didn't have fat in it, it could not make you fat. That was one of the messages we were getting from the nutrition community. Turns out that, at least for me, that was inaccurate.
But if we go back to that chart—can we show that chart for a second? If you look on the left-hand side, this is fascinating because it's sugar as a share of carbohydrates. So what it's saying is that from 1990 to 2000, 46 percent, roughly, of the carbohydrates we consumed were sugar. Which is fascinating, because carbohydrates also make up usually 50 to 55 percent of the diet. So roughly a quarter of the calories we were getting were from sugar.
And then, because of the way he has it graphed, it looks like sugar consumption plummets in the early 2000s, which is pretty much coincident with the awareness of an obesity epidemic. And actually, if that were 2002, it could be my first sort of infamous New York Times Magazine cover story.
But then it drops to only about 41 percent of all carb calories. So what he's implying is that sugar consumption, if we convert this to total calories, that sugar consumption dropping from, say, 24 percent of all calories to 21 percent of all calories should have been enough to blunt the increase in obesity prevalence in this country.
And the graph on the right—well, it's not even worth getting into that.
So, when I was first confronted with this kind of argument, it was made by this fellow Stephan Guyenet in a debate I had with him hosted by the Cato Institute. I pointed out that this is the equivalent of saying: Imagine we've established that cigarettes cause lung cancer. And if you actually look at the data on cigarettes and lung cancer, you have the same kind of chart—exactly the same kind of chart.
Because per capita smoking in America starts going down in 1965—so it goes up, and then in 1965 it comes down—and lung cancer rates continue to go up for the next 30 years. What folks like CREMU and Guyenet were arguing was the equivalent of saying: If we went from an entire nation smoking a pack of cigarettes a day—20 cigarettes—to 17 or 18 cigarettes a day, that would stop the lung cancer epidemic and we would attempt to see it turn downward.
Obviously it wouldn't. I mean, who knows? Maybe it would. Maybe it's such a tenuous epidemic that only smoking 17 cigarettes a day is benign, but smoking 20 causes lung cancer. But it's hard to believe that's true.
And you have the same sort of phenomenon here. So, you go from getting 46 percent of your carb calories from sugar to 41 percent. And then you're assuming that 41 percent is benign. And there's sort of no reason to assume that.
And then all kinds of other effects come in. The reason lung cancer rates went up for 30 years after per capita smoking started to turn down is because lung cancer has a threshold to develop of 20 to 30 years. So merely quitting smoking doesn't necessarily—even if you do quit entirely—it doesn't necessarily lower your risk. It takes about 20 years for your risk of developing lung cancer to get back to the risk of a nonsmoker.
Zach Weissmueller: So another reason why the field seems to be interested in ultra-processed foods is that there was a large study that's often cited as the landmark study showing that ultra-processed foods are harmful compared to the alternative.
You've had some critiques of that as well. In fact, you posted on your blog—a picture of basically they fed people for, I believe it was, two weeks?
Gary Taubes: Two weeks on each diet.
Zach Weissmueller: They fed the people on the ultra-processed menu stuff like what's pictured on the left here. And then the unprocessed menu—so on the left you've got, you know, burger, fries, bunch of lemonade. And on the other side, looks like some salmon and green beans and sweet potatoes or something like that.
This is often cited as, like I said, the landmark study. What is your criticism? What's wrong with the design of the study?
Gary Taubes: Well, I mean, you think about it, "everything," would be the short answer. Again,I'm not defending ultra-processed food. Just step back for a second. When we were young, ultra-processed food was known as "junk food." And the reason it was known as junk food is because if you came from a health-conscious household, you weren't supposed to eat it. Because you knew it was bad for you. You didn't know why it was bad for you. It was junk.
Then Michael Pollan comes along with his book In Defense of Food in the early 2000s, and he relabels junk food as "food-like substances." And you're not supposed to eat them because they're not really food—they're food-like substances. It's all the same argument. Now you get the "ultra-processed food."
But again, we want to compare. We want to find out if the ultra-processed foods—they definitely associate with disease. If you look at the people who eat more ultra-processed foods, they have higher rates of obesity and diabetes than people who don't. But the people who don't are health conscious. And the people who don't tend to have higher socioeconomic status. They can afford to cook for themselves. They can afford to eat out at fancy restaurants and not go to McDonald's for dinner. They can prioritize their children's health over convenience.
They have better doctors. They have better health care. There's a lot of reasons why they could be healthier, and the ultra-processed foods could just be a canard—something that happens to associate with being wealthy and health conscious.
So now you come along and you do this study. And you put people into a metabolic ward—and this is why everyone loved Kevin's study. Because he has a metabolic ward, which is sort of a hospital ward but with a kitchen and equipment to monitor what they're eating. And also, you could measure their energy expenditure and things like this. And now you could feed them precise foods, and you could quantify exactly how much you eat. And nutritionists love this because now they can say, "This study is controlled."
Zach Weissmueller: I have to wonder who signs up for these studies to be sort of locked in a ward for two weeks of being fed.
Gary Taubes: It's a critical factor when you do this research, right? Because the people who don't have families, or because they don't have family responsibilities. And often who don't have jobs—because they'll typically get paid for these studies. If they do have jobs, they have to be able to leave the jobs, right, for a month. It's not two weeks on each diet. They have to be able to leave for a month to do the study.
There's actually a population of people who basically do these studies for a living. But they're not typical humans. They're not typical Americans because they can step out. And that's a problem—another challenge to doing this research.
So the point is—think about a drug study. When you do a drug study and you want to tell if the drug is harmful or beneficial, you randomize people to the drug or a placebo. And the reason you randomize them is you don't want a situation where only the rich, health-conscious people take the drug and only the poor people who can't afford to be health conscious take the placebo. Because then, if you see an effect of the drug, you won't know if it's because of the drug or the type of people who are taking it.
So you randomize them so you get a little of everything in both groups. And then you give them the drug and the placebo. And the reason you give them the placebo is so they don't know that they're on the drug. Because if you know you're on the drug, you'll get placebo effects. And not just placebo effects—whatever subconsciously might happen that makes you healthier. People will also act differently if they think they're doing something helpful, like they might eat less.
Zach Weissmueller: So you're saying there's not a placebo—or that's not really possible—when these are your options?
Gary Taubes: When these are your options, you know when you're eating the unprocessed menu, even if you don't know why you're eating the unprocessed menu.
And then there are other issues. Like, there's that reason hamburgers were invented. McDonald's, they know everybody likes a good hamburger. As long as they're willing to eat animal products. Salmon, green beans? Not so much.
So if the endpoint of your study—what Kevin Hall found in this study—was that people ate considerably fewer calories of the salmon and the green beans than they did of the hamburger and the French fries, maybe they just didn't like the salmon and the green beans.
I have two sons. One likes salmon, one doesn't.
Liz Wolfe: But that is, you know, like a useful dieting strategy, right? Like sometimes, if I'm going out or whatever, I force myself to—it's such a dumb example—but to drink glasses of white wine or whatever. Because when you drink a sugary cocktail, you accidentally drink six of them, and you get more drunk than you want to, and you have more sugar than you meant to.
And so like, I don't know—some of the healthy-person psyche, as I understand it, is like: force yourself to eat the thing that will have you consume more slowly, and ultimately less of the substance, right?
Gary Taubes: Yeah. Again, a longer conversation we could have about this. It just reminds me—I had a friend decades ago who got acupuncture to quit smoking—in the earlobe. I remember—a very successful New York editor. And I asked her what the acupuncture was supposed to do. And she said, "It makes me not like the cigarettes." And I said, "Well, why don't you just smoke a brand you don't like? Instead of buying Marlboro Lights, buy Winstons."
But it doesn't work because it's addictive.
So again, it's sort of—they're not addictive to everyone, but they're very addictive to the people who end up smoking their whole lives. The point here is: yes, it could be a valuable strategy to only eat foods you don't like.
Although then I would ask the question: Why is it some people have to only eat food they don't want if they want to fight obesity, and other people don't?
What we're trying to establish is that the ultra-processed foods somehow are bad for you. And the assumption is they're bad for you because they make you eat too much. That was the conclusion of Kevin's study.
Are we saying they're bad for you because we like them better? Because if you eat too much of a food—or you eat it too quickly—I would argue on some fundamental level, you like it more.
But there's actually evidence in rodents—significant rodent evidence—that how much an animal likes a food is independent of whether or not it gets fat.
You know, the best-tasting food in the world, from my opinion, has always been in Paris. I had the opportunity to live there for a couple of years. The food was delicious. And yet the Parisians have a lower level of obesity than most other industrialized nations.
So the question is: If we can disassociate taste from obesity, and if we're doing a study where ultimately the answer is, "Well, I like this food more, that's why I ate more of it," we're not learning that much. Maybe we're not learning anything we didn't know before we started.
And if this is considered a landmark study—and there were other issues with it—but as we said, they're only feeding people for two weeks. And at the end of two weeks, they are starting to eat less of the ultra-processed food. And so the effect that they report is vanishing.
It may not vanish entirely. There's no way to know, because it's only done for two weeks. But you're always left with this question.
Obesity is a physical and psychological burden. I understand why people who have obesity have to learn to accept and embrace their body size regardless. It's funny—it's difficult even to talk about this without fear of being canceled. But for the most part—and studies have shown this—obesity, in particular in children, is a burden. It's a psychological burden. It's a physical burden. And it comes with increased risk of chronic disease.
And so if it's caused merely by these people eating too much of the processed food—ultra-processed food—then the question becomes: Why don't they eat less of it and prevent the obesity from occurring?
That's what we think we would do. And maybe that's why we avoid ultra-processed foods.
If I eat one McDonald's Double Quarter Pounder with Cheese, I'm not satisfied. But if I eat two, I am. But if I knew that it was making me fatter, I think I could live with one.
So you end up with a whole series of questions that imply that the difference between lean people and obese people, when confronted with ultra-processed foods, is that the lean people will resist eating too much, and the people with obesity—or predisposed to become obese—will not resist it.
And then there's all kinds of technical mumbo jumbo that people try to evoke so that they don't blame that failure to resist on a lack of willpower. Because the one thing all obesity researchers say they agree on at the moment is that obesity is not an issue of character or willpower.
But if they end up saying that the problem with the foods is that it makes people eat too much—and the difference in the people who stay lean in an environment full of UPFs and obesogens is how much they eat—then why don't… You get into these thought patterns.
And part of what I end up arguing in my Substack and my books is that the obesity research community doesn't think about this very much. They get to the, "Oh, look, we have a two-week study showing people eat more of this food, we're going to embrace it because it agrees with all our preconditions."
Liz Wolfe: It's frustrating because I understand the impulse to treat a demographic—especially when there is still a lot of unexplained science as to why some people are so predisposed to obesity and diabetes—treating this crowd with extraordinary empathy. That's, I think, very appropriate and kindly.
But there is also a little bit of this challenging dynamic where it's like: if we talk about self-control, and if we talk about willpower in attempting to erase that from the equation…Which, by the way, Ozempic just kind of might do for us. Right? Technology might do that for us.
But it ends up selling short the people who have worked really hard to use their self-control and their willpower to improve their circumstances in life. And so it ends up being very benevolent to one group, but ultimately kind of insulting to the other group.
Of course, they're the group that you're allowed to insult, right? Versus the group of people who are struggling with obesity. Right?
Like, I mean, I naturally—I want to eat endlessly. I love food. I'm a baker. But I make choices to not do that. And it's not necessarily something that comes naturally to me, but it's something that I care about. And so I prioritize it.
And we also do know that people have some amount of variance in terms of how they're naturally wired in self-control—in all areas, in all areas of vice.
But ok, all of this is a little far afield. I love how you've detailed some of the issues with actually conducting good studies in this area.
But the thing that I'm really, really curious about is: If you replaced RFK Jr. and were in charge of the sort of new MAHA movement, what would you prioritize and focus on? Because we know the food pyramid and the standard nutrition guidelines have been pretty bad when run by the federal government for a while. But now the sort of alt-RFK Jr. movement isn't really doing a great job either.
So if you were doing it, how would you craft a MAHA moment?
Gary Taubes: Well, first of all, one of the joys of being a journalist is you can criticize people without actually having to offer up solutions.
Liz Wolfe: That's why I got into it. Same with you.
Gary Taubes: Let's be serious—offering up solutions… I mean, this is a very, very difficult problem. But it's true: all my books ultimately are calls for better research.
So the first thing I think they have to do is—because again, the nation is polarized in a way that it doesn't feel like it's ever been polarized in the past—what I want… I started researching the nutrition–chronic disease connection in the late '90s for a series of investigative articles for the journal Science. And that led to my books and these New York Times Magazine covers.
We need some agreement on what the problems might be. The way the nutrition community has embraced this is—they act as though they know the answers. And they clearly don't.
Even with ultra-processed foods, they're all willing to say ultra-processed foods are the problem. But they've never thought deeply—I mean, some people have, but they tend to ignore those people—about the problem with the ultra-processed food definition. What it sweeps under the rug issues like: Is it the ultra-processed foods, or is it the food dyes? Or is it the sugar? Is it the refined grain? Is it the seed oils?
There's a whole world of people, including RFK Jr., who are saying seed oils are toxic. Are they? I don't know. I haven't seen compelling research to convince me of that, but…
Zach Weissmueller: If I could pull up one thing—I think it illustrates the point you're making. I came across this in your Substack. You compare—this is on the left, unprocessed oatmeal; on the right, processed oatmeal. I believe this was pictured in a Wall Street Journal article about all the processed foods. And then you point out—if you look at the ingredients on the back, what's the first ingredient?
You know, they put these red boxes around these various processed additives, but you point out that the second ingredient behind whole grain oats is sugar. So does your argument boil down to: Maybe we should be looking at the second ingredient in the list, as opposed to…
Gary Taubes: So the Wall Street Journal article was about how to identify ultra-processed foods in the supermarket. Those are their boxes.
And when you're looking at ultra-processed foods, you care about things like the whey protein concentrate and the emulsifiers. But ultimately, what you have is a healthy food—a benign food—oatmeal. And that was "Old-Fashioned Oatmeal." So you get the full oats. Takes longer to cook.
And then you refine the oats so it's easier to cook. And then you add sugar and artificial flavoring so it tastes like banana nut. Although they call it natural flavor, so "natural flavoring." Then you worry about, like I said, the whey protein concentrate.
Whereas when I was growing up, my mother would not have let us eat the banana nut cereal because she would have recognized that as falling into the category of junk food.
The question becomes: Why? How do we identify that?
So again, now you can do a study where you could create the minimally processed version of that banana nut oatmeal that doesn't have whey protein concentrate, that doesn't have emulsifiers—but does taste like banana nut, whatever that tastes like.
That's what you feed people for breakfast. One group gets this. And then in an ideal world—because time is limited—we're going to do this for breakfast, we're going to do the pizzas for lunch, and something similar for dinner.
Dinner's probably harder to identify a minimally processed dinner that looks like an ultra-processed dinner. But if we put our heads together—we have the smartest people in the nation working on this…
Anyway, if I were in RFK Jr. 's shoes—and leaving all the vaccine stuff out of it and the defunding of various medical groups out of it—and just focusing on nutrition…
Unfortunately, I think we have to go back to kind of first principles. We have various hypotheses about what is causing the obesity and diabetes epidemics.
My first book suggested refined grains and sugars. A lot of people think that's the problem. There are people who think seed oils are a problem. There are people who think just eating too many calories is the problem. And then there's food dyes and additives.
So what I would want to do is put together a committee—because I want this to be bipartisan to bring people across. I don't want congressmen involved. I want the smartest scientists in the world.
So we're going to get 20 retired Nobel laureates who are still young enough to be intelligent, but old enough to have free time. And we're going to put them on this committee. And then we're going to give them a staff of very, very bright postdocs.
You could think of this as similar to the Supreme Court, which has its staff of clerks working for every justice—who are the best in the nation.
They're going to collect all the information. They're going to read up on all these hypotheses. And these people are going to decide what's meaningful and what's not. Is there enough there that we should worry that sugar is uniquely toxic? Is there enough there that we should worry that seed oils are uniquely toxic?
If seed oils—are there specific seed oils we should worry about?
And if we worry about them, what research do we need to do to begin to establish whether they are or are not? And what should we recommend to the public in the meantime?
Zach Weissmueller: So you almost need a kind of meta discussion here about what it is we actually need to study first—whereas they're coming in with a little bit of a preconceived idea.
Gary Taubes: Like, the reason I'm on your podcast, right? Is because I wrote some books that you think are intelligent. Or your editors think are intelligent. They said, "Have Gary Taubes on. He's not an idiot."
Zach Weissmueller: That was me. I had to fight for it, Gary. I had to fight for you.
Gary Taubes: I remember. Thank you, Zach.
So the question is: If I'm right—it's not the fat, it's not the saturated fat, it's the carbs. I wrote the book assuming that smart people would read this and say, "He makes some very good points. We really have to address this."
And actually, some very smart people have read it and said, "He makes some very good points. I wish we would address it."
But in general, let's get some really smart, unbiased people together. Is the hypothesis that I'm saying—that's been there since the 1960s—are the points valid?
Because if they are, this has to be tested. And then we can design experiments to test it.
In the 1970s and '80s, the NIH spent half a billion dollars—actually, in total probably a billion dollars—testing the idea that dietary fat causes chronic disease. And as I document in my book, they basically failed to confirm that hypothesis.
So, if you spent that much money testing whether sugar and refined grains cause these chronic diseases that are now epidemic, you might also fail.
That's the problem with testing any hypothesis. Failure is a very likely scenario. But you have to do it.
Zach Weissmueller: Well, I'm curious, though—do you think there's any hope of an RFK-led HHS going down this path?
And in particular, to get some insight into how they're thinking about all this—and what's going on with the internal politics of all this—you know, Kevin Hall, who designed that ultra-processed food study we were talking about earlier, has since resigned from the NIH.
And after he resigned, he went to the press and publicly gave his reasons for resigning. He appeared on MSNBC on All In With Chris Hayes in April. I just want to play a clip from that interview where he explains why he resigned from the NIH after being there for years, and why he believes that there's a kind of anti-scientific attitude within the current administration.
Roll that, please, John.
(Kevin Hall Clip)
Kevin Hall: What we were interested in doing was understanding: When people consume ultra-processed foods, do they really hijack the brain's reward system in a way that we know addictive drugs do—drugs like methamphetamine or cocaine?
They cause this really large increase in dopamine in the brain.
And we published this result, which was interesting. It seemed that these foods—while they may be addictive in many people—didn't cause the outsized increase in dopamine that many drugs that are highly addictive do.
And it really challenged some of the narratives behind how these foods actually work to change your brain chemistry. And it was published.
And we had a reporter ask for an interview to discuss the findings of that paper for a piece she was writing in The New York Times. And that interview was denied. Our press release was also quashed—for no reason given.
And finally, the reporter at The New York Times asked, "What is it? Why can't I have an interview? I've actually interviewed Dr. Hall before. We'd really love his input on this paper."
And then she actually received a telephone call from the HHS communications director, asking whether or not this was really a piece about RFK Jr.—because the paper didn't seem to fully support the preconceived narrative that ultra-processed foods are addictive.
And so I was shocked to hear about this. And after that conversation, the reporter was given the ability to ask written questions—to which I responded through our press office and gave approval to some slight edits about whether drugs were "drugs of abuse" or "addiction."
And then what I saw was fascinating. Because the responses that were then sent to the reporter had been edited—trying to downplay the study, saying it was very small, when it was actually the largest of its kind. And that the participants had only been in the hospital for five days—when this was actually the only study that had that level of control.
And it struck me as really concerning. Because if people were willing to say, "There's a little bit of daylight between our preconceived notions about how the basic science of food addiction might work and the narrative that's going on," what is it going to be like if they want to start actually meddling in the research itself?
That was my concern.
Zach Weissmueller: So he's saying that even with this mildly… this study that cast just the tiniest bit of doubt on the ultra-processed food hypothesis—in that it found that ultra-processed foods don't affect the brain in the same way that crack does—was then met with a lot of resistance within the RFK-led HHS.
And therefore, he's dubious of the idea that anything other than a finding supporting this hypothesis is going to be advanced.
I know you've worked with Kevin Hall in the past. What is your impression of this whole kerfuffle?
Gary Taubes: Tempest in a teapot.
And I have worked with Kevin in the past, and Kevin and I have had some notorious fights. I think we kind of like each other, but professionally, I'm not a fan of his science and he's not a fan of mine. And that can get in the way of personal affections.
First of all, this all happened pretty much before Jay Bhattacharya took over as head of NIH. And Jay was famously the target of censorship and attempts to cancel his career because of what he believed were his scientific obligations. And so I would have confidence that anything Jay Bhattacharya— the last thing Jay Bhattacharya would do—would be to censor somebody's results.
So, you know, I listened to what Kevin said. I mean, one of the issues here—and I can see this happening. I do know people in the MAHA administration. Jay Bhattacharya and Marty Makary, the head of the FDA, both blurbed—endorsed—my last book, Rethinking Diabetes. I think of both of them as friends.
The MAHA Report has been very friendly to the sort of notion that carbohydrates are the problem—the refinement and the sugar content of the carbohydrates. You start talking to people who are advocating for tallow as a healthy fat and that we should not worry about our meat consumption—those people tend to be people from my side of the nutritional aisle. And so in my world, they're probably likely to do more good.
You know, again, I have spent my entire career reporting on terrible science coming out of the NIH.
It's funny, because I probably agree with Kevin's conclusions about the food addiction work. I had a colleague who, for decades, every time I told them that sugar triggered the same dopamine response in the nucleus accumbens of your brain—the reward center—as sex and drugs do, my colleague would say, "Yeah, but so does buying shoes. Are you going to say buying shoes causes obesity?"
He was absolutely right. This part of the brain has been designed to maximize things that, evolutionarily, we might have wanted to feel pleasure to repeat as a species.
I think a five-day study is not long enough. A small study—even if it's the biggest in the world—might still be a small study.
So the point—and I've argued this before to public health people—is, if I were running NIH, I wouldn't want my press office writing press releases, and I would not want my researchers speaking to the press. Because the high likelihood is that anything they are talking about that they've done is either meaningless or wrong. That's the nature of science.
And so you get this constant stream of meaningless or insignificant research being communicated to the media. I'm not sure I would disagree with the press office that said we have to rein these people in. But I don't think it says anything about the level of science going forward.
The question always with RFK Jr. and his allies is: Are they willing to actually do the science, or are they just going to assume—or pretend—that they already know the answers?
Zach Weissmueller: So, I mean, it sounds like—it's kind of like directionally, this might be an improvement, especially with people like Bhattacharya in there, even if it's kind of a mess with all these ancillary issues and focus on ultra-processed foods. That might be a distraction. But at least there are some bright spots, and they might start to look at what you're more interested in.
I guess my last question for you is about: How hopeful are you actually about the prospect of being able to figure this out?
You've been writing about it since 2002. You've outlined the problems—the huge challenges with studying this—probably the biggest one being how long such a study would need to be in order to tell us anything useful, since these are cumulative effects over time.
I mean, how optimistic are you, epistemologically, that it's even possible to design the kind of study—and what kind of expense it would take to execute something?
Should we be optimistic about that?
Gary Taubes: There are specific questions you can ask that would be important to know. And the studies would be expensive by the standards of typical NIH funding—but not expensive by the standards of pharmaceutical industry research investments.
So my hope is that Jay Bhattacharya, with RFK Jr.'s oversight, will say, "Look, there are studies we just have to do that are going to cost us hundreds of millions of dollars each. They're going to be difficult, but we have all these brilliant scientists…"
What brilliant scientists do is they overcome challenges to the research methodology. You figure out how to get these questions answered. I think that could be done. In an ideal world, I hope it would be done.
Even if it's done, it would take a long time. So even if you do a study, say, for only five years—although I could imagine studies that would give us important, meaningful answers in one to two years—it still takes a year or two or three to set the study up, and a year to do the analysis.
Five years would come…
Zach Weissmueller: I mean, you would be surveilling what people are eating—to the calorie—for five years? Is that what you would pick?
Gary Taubes: But again, it depends on the question you're asking. What level of knowledge you need about the calories consumed.
And I'll give you an example—this is how I think about it. So, probably the last time we spoke, in 2016, I had a book, The Case Against Sugar, and I wrote the book because I felt that sugar and sugary beverages are the most likely explanation—I think are the most likely explanation—for the obesity and diabetes epidemics.
But in the beginning of the book, I said, if this were a legal case, I would have enough evidence to indict. I would not have enough evidence to convict.
So I can make the argument that sugar is the primary culprit for the obesity and diabetes epidemics—and maybe sugary beverages, particularly—that they are to obesity and diabetes what cigarette smoking was to lung cancer.
I can't prove it because I don't have the experimental evidence.
So now, let's think of how we're going to get that evidence.
Well, he's talking about obesity and diabetes. Let's get 10,000 or 20,000 people, and we'll randomize half of them to eat a sugar-rich diet—what they've been eating all along—and the other half to avoid all added sugars.
And after five years—he's saying the half that avoids all added sugars—they're allowed to eat fruit, but no sodas, no fruit juice, no sugar on their cereals, no sugar in their tomato sauces—they're going to be healthier.
But in the process of avoiding their sugars, they're not going to have liked their food as much. We know from watching The Godfather that putting sugar in tomato sauce makes it taste better.
So they're going to eat more of the food that has sugar in it. So telling them to avoid sugar—they're probably going to eat fewer calories.
So maybe the benefit we saw only came from the fewer calories, not from the less sugar.
So now we have a study that may have only told us what Liz believes: that if they just ate less, they'd be healthier. Sorry, Liz.
Have we learned anything we didn't know—or we didn't think we knew—all along? The answer is no.
That's when it starts getting complicated. You need people who are smarter than I am to say, "Ok, how are we going to fix this?" So we make sure they eat just as much as the people who are eating their usual sugar-rich diet, but one diet has sugar and the other doesn't?
So we can rule out the possibility that they just ate less because they didn't like it as much. And that gets complicated.
And the more you think about it, the more complicated it is.
And like I said, science moves forward when people who are smarter than I am think of ways to overcome the challenges to doing the research. And we need people doing that in nutrition.
In general, the nutritionists have made precious little effort. They give up. They say, "Let's just blame all ultra-processing," and then we don't have to worry about any of this stuff.
But they do. And I'm hoping MAHA does. I know Jay Bhattacharya would.
So I do worry about anything they do lasting more than three more years—as much as I worry about it not lasting more than three more years, either way. I get worried in this day and age.
Zach Weissmueller: Yeah. Well, I hope that he gets this message. We've had him on Reason TV several times before, and I generally agree with you that I think he takes a careful approach to science, and maybe he can have some positive influence here along these lines.
I want to bring the conversation to a close by asking you the question that we give to all of our guests. Very appropriate for Gary Taubes, you have lots of questions and uncertainty about the world.
So what is a question that you think more people should be asking?
Gary Taubes: What's a question I think more people should be asking?
You couldn't give me this one in advance so I could think of something meaningful to say?
Liz Wolfe: Everyone else comes up with something super profound, so we expect you to.
Gary Taubes: Like I said, we live in a polarized world. Virtually all the issues that tend to interest us—that obsess us—tend to be issues that we almost can't talk about with some of our friends, because we differ so importantly.
So I always think people should be asking themselves: "What is it that this person sees that makes him or her disagree with me as much as they do?"
There's this concept from Daniel Kahneman, the Nobel laureate—cognitive psychologist, behavioral… whatever you would call him—"what you see is all there is". He writes about it in Thinking, Fast and Slow.
And it really resonated with me. People make their judgments based on what they see.
So if you find people who disagree with you so profoundly—as we all tend to do, living in our silos—I always want people to say, "What is it that person is seeing that I'm not?" Because my assumption is, if I could see what they're seeing, I would understand why they disagreed with me.
Zach Weissmueller: That's such a great question. That's very much in the spirit of this show. So thank you for that, Gary.
You can find Gary Taubes's thought-provoking books on Amazon. You can read his work at the Uncertainty Principles Substack.
Thank you for coming on the show.
Liz Wolfe: Thank you so much, Gary.
Gary Taubes: Thank you so much. Thank you, Liz.
