Given my general contempt for the government, I was absolutely startled at the change. The government actually being mostly correct for was a whole new concept.
Absolutely. I literally wasn't aware that the government could be correct about something. Guess that's due to growing up in the Clinton/George W. Bush era lol. "It's all lies, obviously" was just the default for my entire childhood.
Thank you for the analysis, and overall the response to the guidelines in the places I go to get information have been more positive than not. Regardless of where people may stand on seed oils, the fact that the guidelines are much shorter than in the past, and the focus on "real foods" are two salient points everyone can get behind without much debate.
I am a little weary of all the meat and diary being suggested, because in my view unless you are lucky to have the means to buy organic or direct-from-producer; the mega-farm factories where most chickens are housed are not a benchmark for good health. Beef is also being fed a lot of stuff that is highly problematic, plus antibiotics. What is wrong with pulses, plain old fashioned rolled oats, nuts? It is pesticides? Are carbs evil?
There is also a notable shift in tone towards personal responsibility -see how they changed their advice on alcohol- and "do your own research", which I do not mind, but it can be controversial to those in charge of public health. Can low-income people follow the guidelines? Not sure.
Bottom line: better than before, yes. But that was a low bar to clear. I'll pass up on the beef tallow, the full fat dairy and the red meat, thank you. Chicken breast, fish, pulses and rolled oats: you are still my best friends, LOL.
Good point on the brevity; the last guidelines were like 50+ pages when I opened them to compare.. most of it seemed like bureaucratic drivel. No civilian was ever gonna read it.
Can people develop personal responsibility? Maybe not everyone. But expert/institutional responsibility doesn't exactly have a great track record, either.
This bit on the guidelines is confusing to me: "prioritize oils with essential fatty acids, like olive oil." EVOO is ~10% LA ... and nothing else is "essential." They must mean "beneficial fats" and are implicitly referring to oleic acid?
> You can play this game at any Walmart or even truck stop: pick up random “food” items and guess if they contain a seed oil. Chances are, you win. My friends & family hate me because I constantly play this game with them and I never lose. NEVER!
I read this paragraph to my wife, who immediately rolled her eyes at me and sighed. Game recognize game.
> Just like with the old food pyramid, not 100% of the population are going to follow them 100%. But nudging some proportion of the population away from the cliff is great.
DGA is non-trivial for government-funded food programs, like school lunches and some welfare programs. The DGA gets codified into menus and educational materials. So they can actually have a broad reach. (I see you made this point later in the article.)
Yea, as a (spiritual, amateur) Austrian economist I can't not think in terms of things (products, services, ideas, laws, regulations, ..) making their way through the graph of production that is the economy.
Absolutely stellar analysis on the effects of institutions.
I would like to push back a little on your scorn for the NOVA classification. The way I understand it, it's analogous to how the Big Five model works for personalities, namely that it's derived empirically. In the same way that people do seem to naturally cluster along five dimensions when asked about personality stuff, there really seems to be a categorical difference in how sick NOVA 4 stuff makes you. Now, no one claims that that means that every item in that class is as bad as every other, or even bad at all! It's just that this happens to be the most predictive categorization of foods (with regards to health outcomes) humans have ever devised. Empirically that's true! The NOVA doesn't attempt to tell you why those foods are bad for you, statistically speaking. Maybe it's the seed oils, but the fact that it doesn't attempt to construct a causal theory isn't a bug; it's a feature!
That being said, I do think there is a good reason to believe there is something more going on. If you look into the food science (engineering really) behind how these foods are made, you'll see that ultra processed is a very misleading name. It makes it sound like it's extra milled, extra sliced, extra ground, or extra cooked. But that's not the case, it's more like: common ingredients like corn and soy have been decomposed at the molecular level and recomposed using chemical processes to form entirely new, never before seen molecules. *That's* what a lot of the stuff in NOVA 4 is. It shouldn't really be called Ultra Processed, it should be Chemically Synthesized Novel Compounds. I don't think it would be that strange if that is a large part of the cause of everyone getting sick. It's likely bad for the same reason PUFAs are bad: we were never designed to metabolize these molecules.
(I asked GPT-5 for an analogy:
Ultra-processed foods are only ‘processed’ in the same way plastic is processed wood: the raw material is there in a trivial sense, but the transformation crosses a boundary where the original biological object no longer exists in any meaningful way.)
I'm genuinely confused about where we disagree here. You write that "NOVA4 is completely different" from the other levels, but that's exactly my point! The naming is terrible (making it sound like a continuum when it's not), but the categorization captures a real thing.
On intellectual honesty: I think you have this backwards. The intellectually dishonest move in nutrition "science" has been to assert causal mechanisms prematurely. "Saturated fat causes heart disease," "cholesterol clogs arteries," etc. These confident mechanistic claims have been catastrophically wrong for decades.
NOVA's approach is epistemically virtuous precisely because it's purely descriptive. It identifies a cluster of foods that empirically predict poor outcomes without overreaching into causal claims. That's how science is supposed to work when you don't understand the mechanism yet: establish the phenomenon first, then figure out why.
And crucially, NOVA saying "we don't know why, but this category predicts outcomes" doesn't stop you or anyone else from proposing that seed oils are the culprit. It provides a well validated empirical foundation for that investigation. "Something about this category is making people sick" is a starting point, not a conclusion.
You might be right that it's seed oils. But how is descriptive agnosticism "whitewashing and preventing discourse"? It seems like the opposite: here's a robust correlation, now let's figure out what's driving it.
What would you have them do instead? Declare a specific causal mechanism before the evidence is in? That's exactly the playbook that gave us the lipid hypothesis.
My point was: if the NOVA/UPF people were honest, they wouldn't say "processing" is bad and imply that NOVA4 is just more of the same that made 1-3. They would say butter is totally fine, dried beans are totally fine, things made with seed oils and maybe corn syrup and other faux foods are not fine.
Currently, the implication is the opposite: the only reason compressed corn flour fried in seed oils with artificial flavor (doritos) is bad is that it's even more of sliced beef/dried beans.
Exactly, it's not a continuum. The implication of a continuum devalues & whitewashes what's actually bad about some of the foods in category 4.
My main quibble though is that even NOVA4 doesn't capture the thing well. If they actually did what you said, I wouldn't hate it so much. Packaged meat is in there, for heaven's sake! Canned noodles or soups. Chocolate!
If you wanted to get more orthogonal to health, I don't know how.
The lipid hypothesis isn't worng because they did causal methods, but because they knowingly lied. Incorrectly identifying & labeling things isn't better.
I'm glad you found my comment even though Substack apparently decided I wanted to reply to my own comment.
To be clear, I absolutely say that butter and dried beans are totally fine, and my main problem is indeed with seed oils and faux food. I just happen to think that NOVA4 and the people behind the system capture this intuition better than you seem to think. Although, as I said I agree the names for the categories are terrible because they provide bad intuitions to normal people discussing this, I don't actually think that intuition or bad understanding is the reasoning behind it.
The people who made it don't think NOVA4 is bad because it's "more sliced". For instance, the criteria for meats that go in NOVA4 is "reconstituted", which I think is a distinction that makes sense. It's meats that are chemically recomposed at (almost) the molecular level, such as nuggets and most sausages. I don't eat normal sausages any more, but I do eat artisanally made sausages where I can get them (I also happen to believe those would _not_ qualify as NOVA4, but of course I don't claim that NOVA is perfect and I don't take it as literal gospel for my own food choices). Further, bacon is another great example: traditionally bacon is supposed to be cured and smoked, but all bacon in any grocery store I have ever seen, has been infused with oils that have smoke flavor and are in fact not smoked or cured. That makes it NOVA 4, but I do eat traditionally cured, actual honest to god bacon. (I know a guy who does it by himself and sells it on Facebook, I guess you could call that artisanal bacon.) It also happens to look completely different and taste completely different from store-bought, what I would call UPF, "bacon". When everything about it looks, tastes, smells and behaves completely differently from real bacon, it makes total sense to me to call that Ultra Processed. Not because it's packaged in a plastic, vacuum sealed thing, but because of what they have done to it. It's legitimately a completely alien product!
Chocolate is a similar example, I wouldn't eat most commercial chocolate, which is sometimes made more out of emulsifiers than cocoa. But I do eat artisanal chocolate made from only cocoa, milk, sugar and a few other ingredients. I'm not sure artisanal chocolate would officially qualify as NOVA 4, but I think it probably shouldn't under the stated criteria. There is such a thing as UPF chocolate, UPF bacon; and there is regular old school chocolate (hard to get and expensive), and there is regular old school bacon (almost impossible to find unless you know a guy).
I think the part where you and I depart a little in our views, is that I don't _only_ consider PUFAs (and by extension seed oils) bad. I do avoid seed oils like the frigging plague, but I also consider a lot of this reconstituted stuff bad, mostly for the same reason: we are not adapted to metabolize it.
And maybe we get tripped up a little because "packaged meat" sounds like the packaging is what _makes_ it bad, which obviously it doesn't. But there is a real and strong correlation! Chicken, full of PUFAs because it's fed on a terrible diet, is packaged. Fake Bacon is packaged. Reconstituted meats are packaged. Unfortunately minced meat is also packaged, which I think you and I agree is absolutely fine. That makes the packaged heuristic imperfect, so I don't follow it blindly, but I still think it is a good heuristic!
And I really appreciate the agnosticism of the NOVA people, there's no way we'd have such a good classification system if they tried to be causal about it. Also it would be dishonest, because no epidemiological or correlation study can ever claim to prove causal mechanisms. Yeah that makes the classification imperfect, but I don't think it's logically possible to do better, unless you want to do an intervention study (like an RCT), but we know that's never happening.
Substack is also behaving badly for me here. Somehow all your other comments it insisted on scrolling the comment out of view on every keystroke, meaning I had to type it completely blind lol. And also memorize what you said or keep scrolling up.. this one seems to work for now... phew!
I don't only consider PUFAs either. I eat very little of UPF even if they don't contain linoleic acid.
If NOVA didn't have the faux continuum issue, and if it actually mostly listed faux frankenfoods which I hate, I'd consider it.
The definition sounds like lab-created frankenfoods. And then under examples for NOVA4, they list: cookies, "fatty snacks" (nuts? cheese?), ice cream, chocolates, canned/dehydrated foods, fruit yogurts, dairy drinks (is milk a dairy drink?), packaged meat, fish & vegetables, pastries, cakes
Now it is probably true that terrible faux franken-foods exist for most or all of these categories, but listing them like this is criminal; it's the opposite of "true."
Almost all of these were/are around as old school, artisanal foods that our great-grandparents would've eaten. And they can be made totally healthy.
It's as if you recognized that Rolex watches are often counterfeited and you decide that means Rolex is fraudulent.
I don't appreciate the NOVA people, because the system is catastrophically wrong. I really have a hard time coming up with one more wrong in absence of having the correct information and deliberately saying the opposite (e.g. lipid hypothesis). This is just completely orthogonal, I'd say.
"Imperfect" isn't the word I'd use, I think there's criminal liability.
This is an interesting observation, I’m not sure what happened there. I hadn’t read those examples before! I was mostly familiar with the definition, which indeed captures most of what I care about.
Everything I’ve read about NOVA suggests to me that they do actually understand the problem and are indeed thinking of Faux Frankenfoods (that's a great name!). That’s what they’re trying to point to.
Maybe an intern wrote that list of examples?
Another way it can make some sense is that if you mentally insert “commercial…” in front of each example, it does kind of point to the correct thing. Like if I think of commercial ice cream I think of the ice cream I can buy at nearly any regular convenience store I know about. And that ice cream really is Faux Frankenfood. Maybe they're not saying ice cream has to be UPF, just that 99% of commercially available ice cream is.
The reason I’m defending them at all, is that I think the definition of NOVA4 really is pointing to a real thing, much more real than any nutritional science before it ever has. And that is confirmed by the empirical data. It’s a rough and crude approximation of a categorization, but interpreted charitably there really is something to it.
And also it finally puts to rest the old idea that all food with the same macro nutrient content is literally equal.
1. remove the implicit continuum problem (probably best to rename the entire thing from "processed food" and "UPF" to something else then)
2. add good examples
I might be for it.
Commercial itself is again technically wrong - nothing about the "commercial" aspect made these things unhealthy. People sold cookies and bread commercially 200 years ago.
If you're gonna define a problem, you have to get to the bottom of it. It's ok to do these broad incidental analyses when generating hypotheses, but not to state root causes.
I don't know how they came up with NOVA. I just think it's distracting if we're looking at the ACTUAL cause of obesity. Sure it's helpful to nudge people away from a group of foods that includes many of the worst offenders, but let's not pretend the list makes sense in detail.
Level 3 is qualitatively different from level 4, they're not just "more processed." And items in all/most levels are orthogonal to actual health in my opinion.
It might be empirically derived, but it's largely wrong on the details. Its main categorization ("processed") is mostly meaningless orthogonal.
> It's just that this happens to be the most predictive categorization of foods (with regards to health outcomes) humans have ever devised. Empirically that's true!
I think this is bizarrely wrong. It's only true if you purposefully blind yourself to other ideas, which, yes, they do. I think being aggressively & intentionally ideological and narrow-minded is not a virtue.
The feature is what makes it so bad; its main strategy is intellectual dishonesty.
If they were honesty, they'd say that NOVA4 is completely different, and that it's mostly due o industrial foods & ingredients.What they're actually doing is whitewashing and preventing discourse.
You're right on the added protein junk food. I see "protein milk shakes" and "protein bread" and even "protein cookies" the other day lol. The emphasis on "whole foods" and big pictures of steak will hopefully nudge people in a positive direction?
But we'll see.. unintended consequences and all that.
I think it'll takes decades to reverse the metabolic damage, even if companies responded quickly & drastically. But maybe it'll slow down or arrest the trend for now, lol.
Given my general contempt for the government, I was absolutely startled at the change. The government actually being mostly correct for was a whole new concept.
Absolutely. I literally wasn't aware that the government could be correct about something. Guess that's due to growing up in the Clinton/George W. Bush era lol. "It's all lies, obviously" was just the default for my entire childhood.
I wonder how much of the rest of the world follows suit and starts recommending a similar food pyramid.
>My friends & family hate me because I constantly play this game with them and I never lose. NEVER!
For me people just give me labels to read for them or laugh when I roast some new products they bought to eat, lol.
Ha same. People blush when they mention what food they bought and hand it over for me to "check."
I heard Bitten Jonsson, RN (she is a sugar-addict expert) say, "I am an opposite -er: I do red meat, salt, sun, and butter." I kind of agree with her!
If I take sugar from my diet (carbs), I can't be afraid of saturated fat because what else will I survive on?
Just seed oils :D
Thank you for the analysis, and overall the response to the guidelines in the places I go to get information have been more positive than not. Regardless of where people may stand on seed oils, the fact that the guidelines are much shorter than in the past, and the focus on "real foods" are two salient points everyone can get behind without much debate.
I am a little weary of all the meat and diary being suggested, because in my view unless you are lucky to have the means to buy organic or direct-from-producer; the mega-farm factories where most chickens are housed are not a benchmark for good health. Beef is also being fed a lot of stuff that is highly problematic, plus antibiotics. What is wrong with pulses, plain old fashioned rolled oats, nuts? It is pesticides? Are carbs evil?
There is also a notable shift in tone towards personal responsibility -see how they changed their advice on alcohol- and "do your own research", which I do not mind, but it can be controversial to those in charge of public health. Can low-income people follow the guidelines? Not sure.
Bottom line: better than before, yes. But that was a low bar to clear. I'll pass up on the beef tallow, the full fat dairy and the red meat, thank you. Chicken breast, fish, pulses and rolled oats: you are still my best friends, LOL.
Cheers!
Good point on the brevity; the last guidelines were like 50+ pages when I opened them to compare.. most of it seemed like bureaucratic drivel. No civilian was ever gonna read it.
Can people develop personal responsibility? Maybe not everyone. But expert/institutional responsibility doesn't exactly have a great track record, either.
This bit on the guidelines is confusing to me: "prioritize oils with essential fatty acids, like olive oil." EVOO is ~10% LA ... and nothing else is "essential." They must mean "beneficial fats" and are implicitly referring to oleic acid?
I think the word "essential" turns people's brains into mush. They literally think it means "good."
Sort of like people can't grasp that "marginal" doesn't just mean "a little" and "exponential" doesn't just mean "a lot."
> You can play this game at any Walmart or even truck stop: pick up random “food” items and guess if they contain a seed oil. Chances are, you win. My friends & family hate me because I constantly play this game with them and I never lose. NEVER!
I read this paragraph to my wife, who immediately rolled her eyes at me and sighed. Game recognize game.
Think like an Economist (everybody HATES it!).
Friend: Hey let's go out for dinner
Economist: Compared to what, and at what cost?!
Ha! I can see in your many writings that you have a mind for economics.
Mises' Human Action has been one of the most influential books I've ever read.
I'm going to add that to my list!
> Just like with the old food pyramid, not 100% of the population are going to follow them 100%. But nudging some proportion of the population away from the cliff is great.
DGA is non-trivial for government-funded food programs, like school lunches and some welfare programs. The DGA gets codified into menus and educational materials. So they can actually have a broad reach. (I see you made this point later in the article.)
Yea, as a (spiritual, amateur) Austrian economist I can't not think in terms of things (products, services, ideas, laws, regulations, ..) making their way through the graph of production that is the economy.
Absolutely stellar analysis on the effects of institutions.
I would like to push back a little on your scorn for the NOVA classification. The way I understand it, it's analogous to how the Big Five model works for personalities, namely that it's derived empirically. In the same way that people do seem to naturally cluster along five dimensions when asked about personality stuff, there really seems to be a categorical difference in how sick NOVA 4 stuff makes you. Now, no one claims that that means that every item in that class is as bad as every other, or even bad at all! It's just that this happens to be the most predictive categorization of foods (with regards to health outcomes) humans have ever devised. Empirically that's true! The NOVA doesn't attempt to tell you why those foods are bad for you, statistically speaking. Maybe it's the seed oils, but the fact that it doesn't attempt to construct a causal theory isn't a bug; it's a feature!
That being said, I do think there is a good reason to believe there is something more going on. If you look into the food science (engineering really) behind how these foods are made, you'll see that ultra processed is a very misleading name. It makes it sound like it's extra milled, extra sliced, extra ground, or extra cooked. But that's not the case, it's more like: common ingredients like corn and soy have been decomposed at the molecular level and recomposed using chemical processes to form entirely new, never before seen molecules. *That's* what a lot of the stuff in NOVA 4 is. It shouldn't really be called Ultra Processed, it should be Chemically Synthesized Novel Compounds. I don't think it would be that strange if that is a large part of the cause of everyone getting sick. It's likely bad for the same reason PUFAs are bad: we were never designed to metabolize these molecules.
(I asked GPT-5 for an analogy:
Ultra-processed foods are only ‘processed’ in the same way plastic is processed wood: the raw material is there in a trivial sense, but the transformation crosses a boundary where the original biological object no longer exists in any meaningful way.)
I'm genuinely confused about where we disagree here. You write that "NOVA4 is completely different" from the other levels, but that's exactly my point! The naming is terrible (making it sound like a continuum when it's not), but the categorization captures a real thing.
On intellectual honesty: I think you have this backwards. The intellectually dishonest move in nutrition "science" has been to assert causal mechanisms prematurely. "Saturated fat causes heart disease," "cholesterol clogs arteries," etc. These confident mechanistic claims have been catastrophically wrong for decades.
NOVA's approach is epistemically virtuous precisely because it's purely descriptive. It identifies a cluster of foods that empirically predict poor outcomes without overreaching into causal claims. That's how science is supposed to work when you don't understand the mechanism yet: establish the phenomenon first, then figure out why.
And crucially, NOVA saying "we don't know why, but this category predicts outcomes" doesn't stop you or anyone else from proposing that seed oils are the culprit. It provides a well validated empirical foundation for that investigation. "Something about this category is making people sick" is a starting point, not a conclusion.
You might be right that it's seed oils. But how is descriptive agnosticism "whitewashing and preventing discourse"? It seems like the opposite: here's a robust correlation, now let's figure out what's driving it.
What would you have them do instead? Declare a specific causal mechanism before the evidence is in? That's exactly the playbook that gave us the lipid hypothesis.
My point was: if the NOVA/UPF people were honest, they wouldn't say "processing" is bad and imply that NOVA4 is just more of the same that made 1-3. They would say butter is totally fine, dried beans are totally fine, things made with seed oils and maybe corn syrup and other faux foods are not fine.
Currently, the implication is the opposite: the only reason compressed corn flour fried in seed oils with artificial flavor (doritos) is bad is that it's even more of sliced beef/dried beans.
Exactly, it's not a continuum. The implication of a continuum devalues & whitewashes what's actually bad about some of the foods in category 4.
My main quibble though is that even NOVA4 doesn't capture the thing well. If they actually did what you said, I wouldn't hate it so much. Packaged meat is in there, for heaven's sake! Canned noodles or soups. Chocolate!
If you wanted to get more orthogonal to health, I don't know how.
The lipid hypothesis isn't worng because they did causal methods, but because they knowingly lied. Incorrectly identifying & labeling things isn't better.
I'm glad you found my comment even though Substack apparently decided I wanted to reply to my own comment.
To be clear, I absolutely say that butter and dried beans are totally fine, and my main problem is indeed with seed oils and faux food. I just happen to think that NOVA4 and the people behind the system capture this intuition better than you seem to think. Although, as I said I agree the names for the categories are terrible because they provide bad intuitions to normal people discussing this, I don't actually think that intuition or bad understanding is the reasoning behind it.
The people who made it don't think NOVA4 is bad because it's "more sliced". For instance, the criteria for meats that go in NOVA4 is "reconstituted", which I think is a distinction that makes sense. It's meats that are chemically recomposed at (almost) the molecular level, such as nuggets and most sausages. I don't eat normal sausages any more, but I do eat artisanally made sausages where I can get them (I also happen to believe those would _not_ qualify as NOVA4, but of course I don't claim that NOVA is perfect and I don't take it as literal gospel for my own food choices). Further, bacon is another great example: traditionally bacon is supposed to be cured and smoked, but all bacon in any grocery store I have ever seen, has been infused with oils that have smoke flavor and are in fact not smoked or cured. That makes it NOVA 4, but I do eat traditionally cured, actual honest to god bacon. (I know a guy who does it by himself and sells it on Facebook, I guess you could call that artisanal bacon.) It also happens to look completely different and taste completely different from store-bought, what I would call UPF, "bacon". When everything about it looks, tastes, smells and behaves completely differently from real bacon, it makes total sense to me to call that Ultra Processed. Not because it's packaged in a plastic, vacuum sealed thing, but because of what they have done to it. It's legitimately a completely alien product!
Chocolate is a similar example, I wouldn't eat most commercial chocolate, which is sometimes made more out of emulsifiers than cocoa. But I do eat artisanal chocolate made from only cocoa, milk, sugar and a few other ingredients. I'm not sure artisanal chocolate would officially qualify as NOVA 4, but I think it probably shouldn't under the stated criteria. There is such a thing as UPF chocolate, UPF bacon; and there is regular old school chocolate (hard to get and expensive), and there is regular old school bacon (almost impossible to find unless you know a guy).
I think the part where you and I depart a little in our views, is that I don't _only_ consider PUFAs (and by extension seed oils) bad. I do avoid seed oils like the frigging plague, but I also consider a lot of this reconstituted stuff bad, mostly for the same reason: we are not adapted to metabolize it.
And maybe we get tripped up a little because "packaged meat" sounds like the packaging is what _makes_ it bad, which obviously it doesn't. But there is a real and strong correlation! Chicken, full of PUFAs because it's fed on a terrible diet, is packaged. Fake Bacon is packaged. Reconstituted meats are packaged. Unfortunately minced meat is also packaged, which I think you and I agree is absolutely fine. That makes the packaged heuristic imperfect, so I don't follow it blindly, but I still think it is a good heuristic!
And I really appreciate the agnosticism of the NOVA people, there's no way we'd have such a good classification system if they tried to be causal about it. Also it would be dishonest, because no epidemiological or correlation study can ever claim to prove causal mechanisms. Yeah that makes the classification imperfect, but I don't think it's logically possible to do better, unless you want to do an intervention study (like an RCT), but we know that's never happening.
Substack is also behaving badly for me here. Somehow all your other comments it insisted on scrolling the comment out of view on every keystroke, meaning I had to type it completely blind lol. And also memorize what you said or keep scrolling up.. this one seems to work for now... phew!
I don't only consider PUFAs either. I eat very little of UPF even if they don't contain linoleic acid.
If NOVA didn't have the faux continuum issue, and if it actually mostly listed faux frankenfoods which I hate, I'd consider it.
But look at this for example: https://ecuphysicians.ecu.edu/wp-content/pv-uploads/sites/78/2021/07/NOVA-Classification-Reference-Sheet.pdf
The definition sounds like lab-created frankenfoods. And then under examples for NOVA4, they list: cookies, "fatty snacks" (nuts? cheese?), ice cream, chocolates, canned/dehydrated foods, fruit yogurts, dairy drinks (is milk a dairy drink?), packaged meat, fish & vegetables, pastries, cakes
Now it is probably true that terrible faux franken-foods exist for most or all of these categories, but listing them like this is criminal; it's the opposite of "true."
Almost all of these were/are around as old school, artisanal foods that our great-grandparents would've eaten. And they can be made totally healthy.
It's as if you recognized that Rolex watches are often counterfeited and you decide that means Rolex is fraudulent.
I don't appreciate the NOVA people, because the system is catastrophically wrong. I really have a hard time coming up with one more wrong in absence of having the correct information and deliberately saying the opposite (e.g. lipid hypothesis). This is just completely orthogonal, I'd say.
"Imperfect" isn't the word I'd use, I think there's criminal liability.
This is an interesting observation, I’m not sure what happened there. I hadn’t read those examples before! I was mostly familiar with the definition, which indeed captures most of what I care about.
Everything I’ve read about NOVA suggests to me that they do actually understand the problem and are indeed thinking of Faux Frankenfoods (that's a great name!). That’s what they’re trying to point to.
Maybe an intern wrote that list of examples?
Another way it can make some sense is that if you mentally insert “commercial…” in front of each example, it does kind of point to the correct thing. Like if I think of commercial ice cream I think of the ice cream I can buy at nearly any regular convenience store I know about. And that ice cream really is Faux Frankenfood. Maybe they're not saying ice cream has to be UPF, just that 99% of commercially available ice cream is.
The reason I’m defending them at all, is that I think the definition of NOVA4 really is pointing to a real thing, much more real than any nutritional science before it ever has. And that is confirmed by the empirical data. It’s a rough and crude approximation of a categorization, but interpreted charitably there really is something to it.
And also it finally puts to rest the old idea that all food with the same macro nutrient content is literally equal.
If they come out and
1. remove the implicit continuum problem (probably best to rename the entire thing from "processed food" and "UPF" to something else then)
2. add good examples
I might be for it.
Commercial itself is again technically wrong - nothing about the "commercial" aspect made these things unhealthy. People sold cookies and bread commercially 200 years ago.
If you're gonna define a problem, you have to get to the bottom of it. It's ok to do these broad incidental analyses when generating hypotheses, but not to state root causes.
What is this empirical data?
I don't know how they came up with NOVA. I just think it's distracting if we're looking at the ACTUAL cause of obesity. Sure it's helpful to nudge people away from a group of foods that includes many of the worst offenders, but let's not pretend the list makes sense in detail.
Level 3 is qualitatively different from level 4, they're not just "more processed." And items in all/most levels are orthogonal to actual health in my opinion.
It might be empirically derived, but it's largely wrong on the details. Its main categorization ("processed") is mostly meaningless orthogonal.
> It's just that this happens to be the most predictive categorization of foods (with regards to health outcomes) humans have ever devised. Empirically that's true!
I think this is bizarrely wrong. It's only true if you purposefully blind yourself to other ideas, which, yes, they do. I think being aggressively & intentionally ideological and narrow-minded is not a virtue.
The feature is what makes it so bad; its main strategy is intellectual dishonesty.
If they were honesty, they'd say that NOVA4 is completely different, and that it's mostly due o industrial foods & ingredients.What they're actually doing is whitewashing and preventing discourse.
You're right on the added protein junk food. I see "protein milk shakes" and "protein bread" and even "protein cookies" the other day lol. The emphasis on "whole foods" and big pictures of steak will hopefully nudge people in a positive direction?
But we'll see.. unintended consequences and all that.
I think it'll takes decades to reverse the metabolic damage, even if companies responded quickly & drastically. But maybe it'll slow down or arrest the trend for now, lol.