Why Are We Defending Industrial Food?

When people argue about seed oils, the conversation almost immediately turns into whether they increase inflammation, whether they lower LDL compared with saturated fat, whether they are associated with cardiovascular disease, or whether there is enough evidence to say they are harmful. I understand why those questions get asked, but they are not really where my objection begins. I keep getting stuck further upstream at the fact that we're arguing over whether people should consume something that, in anything resembling the quantities we now consume it, essentially requires an industrial system to exist in the first place.

That doesn't mean I think processing automatically makes something bad. Humans have always processed food. We cook meat, grind grain, ferment dairy, press olives, dry fruit, salt food, smoke food, and do all kinds of things to make food safer, easier to digest, easier to store, or simply more enjoyable to eat. So I don't think the useful dividing line is processed versus unprocessed because that reduces too many obviously different things into the same category. What I am more interested in is what happens as processing takes us progressively further away from the original food and starts creating forms of exposure that would have been difficult or impossible through eating the food itself.

A soybean is food. Corn is food. Sunflower seeds are food. They all contain fat. But eating those foods and consuming a bottle of highly refined oil made from them are not the same thing simply because the fatty acids ultimately came from the same plant. Once we develop the ability to extract one fraction, concentrate it, refine it, deodorize it, stabilize it, and then distribute it throughout the food supply, we have changed the exposure in a way that eating the original food never could. That is the part I think gets skipped too quickly because nutrition tends to look at the resulting substance and ask what happens once it's inside the body. What does it do to LDL? What does it do to inflammatory markers? What happens when we replace saturated fat with polyunsaturated fat? Those are reasonable scientific questions, but they begin after we have already accepted the substance as food and moved on to comparing its effects with something else. I'm not sure I want to grant that first assumption so easily.

I think this becomes even stranger when nutritional authorities move beyond saying that these products appear safe and actually recommend that people consume them. The American Heart Association currently recommends non-tropical liquid plant oils and specifically includes oils like canola, corn, soybean, safflower, and sunflower oil among its preferred fat sources, while the World Health Organization similarly recommends replacing animal fats such as butter, ghee, and lard with unsaturated vegetable oils. I understand how they arrive there. If the question being studied is what happens when saturated fat is replaced by polyunsaturated fat, and certain cardiovascular markers or outcomes improve, then recommending that substitution follows from the framework being used.

But I think there is another question that gets almost no attention: why are we in the position of recommending a dietary staple that most people could never produce for themselves and that only exists at its current scale because an industrial system can manufacture it for them?

I don't mean that as some romantic argument that everyone should be able to walk into the woods and personally manufacture every ingredient they eat. I can't make most of the things I use in everyday life. It also doesn't mean that industrial production makes the resulting product harmful. My concern is that our dependence on increasingly sophisticated processing to create the thing itself tells us something about how far the final product has moved from the original food and about how historically unusual the resulting exposure is. That seems like something nutrition should at least take seriously before deciding that the product deserves to be actively recommended.

If industrial processing allows us to isolate one component of a food and consume it in amounts that would have been extremely difficult to reach through the original food, the fact that we can metabolize it or that one biomarker improves tells us something fairly narrow. I don't think those findings answer why that product should become a normal or desirable part of the diet. The research may establish that something appears reasonably safe, that it performs better than some alternative, or that it solves a practical problem. Those conclusions can all be true without establishing that the product itself represents something we should be encouraging people to eat.

Whey protein is a good example because I don't have some ideological objection to whey. I've used it before. I understand why athletes use it. It's convenient, high in protein, and can solve a very practical problem for someone who struggles to get enough protein through regular food. I still would not build a diet around whey protein, and I don't currently use it myself. Milk is the original food. Whey protein is one fraction of that food that has been separated, filtered, concentrated, and dried into a powder. There may be a reason to use it, but the fact that the final product has a favorable amino acid profile doesn't make it interchangeable with food in my mind.

I think that same question can be extended much further than seed oils or whey. Refined sugar, isolated starches, protein isolates, fiber powders, artificial sweeteners, concentrated extracts, and many of the ingredients that make up modern processed foods all involve some degree of taking something that originally existed inside a larger biological structure and separating it from that structure. Sometimes there is a clear reason for doing that and sometimes there may even be a benefit. My hesitation begins when the isolated product becomes normalized to the point that the burden suddenly shifts onto the person avoiding it to prove that it's harmful.

If I can get protein from meat, eggs, dairy, fish, or other foods, then I don't feel any need to justify why I'm not drinking whey protein. If I can get carbohydrates from potatoes, fruit, rice, oats, or whatever foods make sense for the person and the context, I don't need evidence that isolated starch is toxic before I decide it doesn't need to make up much of my diet. And if I can get the fats I need through whole foods and more traditional fat sources, I don't understand why the default position should be that I ought to consume a highly refined oil unless somebody can prove it hurts me. I would rather ask what problem adding it solves and whether there is a reason to move further away from the original food in the first place.

I'm also not trying to create some purity test where foods become morally worse every time another machine touches them. That's too simplistic and it creates obvious problems. Olive oil, butter, yogurt, and flour all require processing. There are plenty of foods humans have eaten for a long time that don't appear in nature in their final form. The issue is how much the process changes the thing being consumed and whether that process creates an exposure that bears much resemblance to what eating the original food would have produced.

That is why concentration seems important to me. If I eat corn, I'm eating the starch, water, fiber, protein, vitamins, minerals, fat, and physical structure of the corn together. I'm not extracting one relatively small fraction and consuming it by the tablespoon. Industrial processing allows us to do exactly that, and once we can do it cheaply enough, that isolated fraction can appear in dozens of foods across the day without the person ever consciously deciding to consume very much of it. At some point I think it is reasonable to ask whether we are still talking about processing food in the ordinary sense or whether the technology has created a dietary exposure that didn't meaningfully exist before.

That still doesn't mean the exposure must be harmful. This is where people usually try to force the argument into a place I'm not necessarily taking it. I don't need every highly processed ingredient to be dangerous for the broader point to hold. Some may be harmless. Some may be useful. Some may be better than what they replace in a particular context. I'm only saying that the further the final product moves away from the food it came from, especially when technology allows us to consume amounts that would otherwise be unrealistic, I want to understand why we are adding it rather than treating its absence from the diet as something that needs to be defended.

There are plenty of reasons that could justify processing. Convenience, cost, shelf life, food security, athletic performance, and medical necessity can all matter depending on the circumstances. A person who can't eat enough protein may find whey genuinely useful. Someone living somewhere with limited access to fresh food may need foods that can survive storage and transportation. Processing solves real problems and I don't think pretending otherwise makes the argument any stronger. What I question is how quickly solving one of those problems becomes a nutritional argument in favor of the resulting product.

I think some of this comes from the kinds of questions public-health nutrition has to answer. Public-health authorities are dealing with entire populations living inside an existing food system. They have to consider price, availability, scalability, shelf life, agricultural production, and what people are realistically going to substitute for something else. If millions of people are already eating a particular amount of fat, asking which inexpensive and widely available fat produces better cardiovascular outcomes is a reasonable population-level question. An industrial oil may work very well within that problem because industry can produce enormous amounts of it cheaply and consistently.

That explains part of why nutritional institutions might recommend it. However, it doesn't completely answer the question I am asking about food.

If I were starting from the other direction and asking how I would construct a diet from foods humans can recognize, obtain, prepare, and eat with relatively little transformation, I don't know that I would ever arrive at soybean oil unless the industrial food system first made soybean oil available to me. That's an odd thing to ignore when we are talking about what people should eat.

The same issue becomes even easier to see as food technology moves further in this direction. Cultivated meat is a useful example, although I want to be precise here because the FDA's review of cultured animal-cell products is a safety process; the agency explicitly describes its consultations as evaluations of companies' safety conclusions rather than recommendations that people should eat the products. That is exactly the separation I think needs to remain intact. A product can meet whatever safety standard we decide is appropriate for entering the food supply and I can still ask why I should choose it over the animal food it's intended to replace.

If a company can grow animal cells in a controlled facility and demonstrate that the resulting product isn't harmful, that answers an important regulatory question. It doesn't settle the nutritional question for me. If someone wants to make an environmental argument for it, make that argument. If the argument is about animal welfare, make that one. If it eventually becomes cheaper or more accessible than conventional meat, that may become relevant too. But "we haven't shown that it is bad for you" is a remarkably low bar for deciding what we should encourage humans to eat.

This is probably the broader issue I have with the direction food is moving. Industrial capacity allows us to create substances, concentrations, combinations, and eventually entire foods that no individual human could produce without a complicated technological system behind them. Some of those products may turn out to be perfectly safe. Some may solve legitimate problems. I just don't think safety alone creates an affirmative case for eating them, and I don't think nutrition should abandon the question of what food actually is simply because chemistry allows us to reproduce some of its measurable components.

That is where seed oils keep bringing me back to the same problem. I can grant, for the sake of the argument, that some of the research on them is neutral. I can even grant that certain substitutions may improve certain outcomes. I still don't understand why that automatically moves them into the category of something worth advocating for when the entire product, at least in the form and quantities characteristic of the modern food system, depends on taking a food, breaking it apart, extracting a concentrated fraction, and making that fraction available in amounts and contexts that wouldn't otherwise exist.

Maybe that process turns out to be completely benign. I am open to that possibility. But if I already have access to foods that meet the same nutritional needs without going nearly as far through extraction, refinement, concentration, or manufacture, I am going to start there. As processing takes me progressively further from the original food, I want a progressively clearer reason for why I should follow it.