Vegetarianism Is a Modern Achievement

What Vegetarians Owe to 1948

Let me be honest about my position before I make an argument, because the honest version is more complicated than the polite one.

I do care what you eat. Not in the sense that I want to police your plate, but in the sense that I think what a hundred million people eat has produced a metabolic crisis that everyone now pays for, and I have spent a book arguing exactly that. It would be false modesty to open by pretending otherwise.

I also mean the title without irony. A diet that excludes animal foods entirely, sustained across a lifetime without deficiency, is a genuine achievement. It is just a recent one, and the date matters more than almost anyone wants to discuss. That is what this piece is about.

What I object to is not being disagreed with. It is a specific move that arrives whenever I write about human metabolism — not disagreement, but accusation. Not your evidence is weak but you are complicit in cruelty. Not a counterargument but a verdict.

That is the thing I push back on. And by the end I want to show you why it is a different kind of claim than the one I am making, even though both of us apparently care what other people eat.

First, the part that can be checked.

The chemistry

Vitamin B12 does not occur in plants. Not in trace amounts, not in poorly absorbed forms. It is not there. Bacteria make it; humans get it from animal tissue. Without it you develop pernicious anemia and irreversible damage to the nervous system.

The rest is less absolute and points the same way. Heme iron from animal tissue absorbs several times more efficiently than the non-heme iron in plants, and the phytates in grains and legumes inhibit what absorption remains. Zinc, the same. Preformed vitamin A exists only in animal foods; plants offer beta-carotene, which must be converted, and conversion efficiency varies enormously between individuals depending on which version of a particular gene they carry. Some people convert poorly and have no way of knowing. The long-chain omega-3 fats used structurally in the brain and retina come from marine sources; the plant precursor converts at roughly five to ten percent, worse for the longest chain. Creatine, carnosine, taurine, most dietary choline — predominantly or exclusively animal.

None of this is contested. It is in every nutrition textbook. It is not a position; it is an inventory.

The timing

The standard reply is that all of those gaps are addressable. That is true. B12 costs a few cents a day. Algal oil supplies the long-chain omega-3s directly. Iron and zinc can be managed with attention or a pill.

B12 was isolated in 1948.

Before that there was no supplement, because no one knew the substance existed. Pernicious anemia was a fatal disease of unknown origin. You cannot correct a deficiency that has not been discovered, in a nutrient that has not been named, with a compound that has not been synthesized.

So “it’s all addressable” is true and it is seventy-eight years old. For the several hundred thousand years before that, it was not addressable. It was a deficiency, arriving slowly, presenting as exhaustion and then as nerve damage, and no one on Earth could say why.

This is not a small footnote. It means the diet became viable at a specific moment, and that moment is when industrial analytical chemistry arrived.

Look at the traditional populations people cite as proof of ancient sustainability. Indian lacto-vegetarian practice retained dairy. Often eggs. That is not incidental — that is the animal-food input that made the pattern survive across generations. Strict exclusion of all animal foods, sustained over generations, has essentially no historical precedent anywhere on the planet. The absence of the natural experiment is itself the finding.

What this does not establish

I am going to be as hard on my own argument as I am on the other one, because that is the difference between an argument and a verdict.

It does not establish that vegetarians today are worse off. Cohort studies generally find vegetarian populations with equal or better all-cause mortality and lower rates of type 2 diabetes and heart disease. That is the honest state of the outcome literature, it is cited constantly, and it does not point my way.

But it is worth being precise about what those studies are actually comparing, because almost nobody who quotes them has looked.

Consider who becomes a vegetarian in a wealthy country. Someone who has thought about their diet at all — which already separates them from most of the population. They smoke less. They drink less. They exercise more. They are, on average, wealthier, better educated, thinner at baseline, more likely to have a physician they actually see, more likely to take a supplement, more likely to read a label. Every one of those variables independently predicts living longer. Now assemble them into a cohort and compare them against the general public, whose diet is fast food and soda and whose lifestyle is everything the first group avoids, and ask what the resulting difference in mortality tells you about vegetables.

It tells you something. It does not tell you what it is usually said to tell you.

The most-cited evidence, the Adventist cohorts, carries this problem in its purest form. Seventh-day Adventists are a religious population that abstains from tobacco and alcohol as doctrine, observes a weekly day of rest, holds strong community ties, and eats a largely plant-based diet. Comparing them to the American average and attributing the survival gap to the absence of meat requires ignoring four other variables, each of which has a large independent literature behind it. It is not that the studies are dishonest. It is that they cannot isolate the thing everyone wants them to isolate.

And here is the part that rarely gets quoted. When researchers have compared vegetarians not to the general population but to health-conscious non-vegetarians — people matched for the surrounding behaviors, who exercise and avoid smoking and pay attention, but who eat meat — the mortality advantage largely narrows or disappears. That comparison is the one that isolates the diet, and it is the one that most cleanly implicates everything else.

There is a further wrinkle. “Vegetarian” is not a diet, it is the absence of a category of food, and it spans an enormous range. A person eating lentils, eggs, yogurt, and vegetables cooked at home and a person eating pasta, refined bread, seed-oil-fried snacks, and processed meat substitutes are both counted as vegetarians in these cohorts. They have almost nothing in common metabolically. Averaging them together produces a number that describes neither.

None of this means the outcome data is worthless, and I am not going to pretend it favors me. Confounding is not refutation, and a critic who dismisses every inconvenient study as confounded is doing the same thing I complain about elsewhere. What it means is narrower: the epidemiology does not currently support a strong claim in either direction, and anyone who tells you it settles the matter — whichever way they lean — is describing a literature they have not read carefully.

Nor does the timing argument single out vegetarianism. Agriculture produced its own invisible deficiency diseases — pellagra, beriberi, rickets — undiagnosable until someone identified the missing compound. That is consistent with everything else I write: a great deal of what we eat is historically novel and we understand it poorly. The point cuts wider than one dietary choice.

And the obvious rejoinder deserves an answer, because it is the one I hear most: we also lacked refrigeration and sanitation, and no one argues we should live without those. True, but they are not the same kind of thing. Refrigeration and sanitation remove external hazards — spoilage, contamination, pathogens. They protect food that was already adequate. Supplementation does something categorically different: it supplies a nutrient the diet cannot provide at all. One lowers a threat from outside. The other fills a hole on the inside. A diet that needs the first is a diet exposed to a dangerous world. A diet that needs the second is incomplete on its own terms, and no amount of hygiene will finish it.

What is left

This. You learn something real about a species by asking what it needed before it knew anything at all. Humans needed animal foods. Not preferred, not thrived on — needed, in the specific sense that without them a predictable set of deficiencies arrives and no one could do anything about it. We are the only species that has solved that problem, and we solved it in a laboratory, within living memory.

That is a fact about human biology.

Why I care, and what kind of caring it is

Now the part I promised.

I think we are in the middle of a metabolic catastrophe. Type 2 diabetes, obesity, fatty liver, cardiovascular disease — conditions that were rare and are now ordinary, arriving earlier in life every decade, consuming a larger share of national wealth every year. I do not think that happened because people got lazy. I think it happened because the food supply changed in specific, identifiable ways and we were told the changes were improvements.

That is a cost we all carry. The insurance pool is collective. The hospital capacity is collective. The productivity loss and the caregiving burden and the tax base are collective. So no, I do not think diet is purely private, and I am not going to pretend I believe something I don’t in order to sound tolerant.

But notice what my claim actually is, and what it is not.

It is not about vegetarians. The pattern driving the crisis is refined carbohydrate, industrial seed oil, and ultra-processed food — and an omnivore living on those is in exactly as much trouble as a vegetarian living on those. The failure mode is not the absence of meat. It is the presence of things we did not evolve to eat, in quantities we did not evolve to handle. A person eating vegetables, legumes, eggs and dairy prepared at home is not who I am worried about, whatever the label on their diet.

And more importantly: my claim is the kind that can be wrong. It says this input produces that outcome at this scale and here is what it costs. Every step of it is checkable. Show me a cohort matched for smoking, drinking, exercise, income, and food quality that contradicts it and I have to move. I have already conceded, above, one place where the data does not favor me.

That is the whole distinction I have been building toward.

Two kinds of caring

If you have weighed the suffering of animals against the metabolic case and decided the suffering matters more, that is a coherent position and I am not going to pretend it isn’t. Moral reasoning is not obligated to answer to biochemistry. You are entitled to name a cost and accept it. I would rather talk to you than almost anyone else in this argument.

That is not who shows up.

What shows up is someone who cannot make the metabolic case, so makes the moral one instead — and then behaves as though the moral one had answered the metabolic one. It didn’t. It changed the subject. And the change of subject is the point, because on the biology there is nothing to say. B12 is not in plants. That is not a debate; it is an inventory. So the conversation gets moved to territory where evidence has no jurisdiction, and where the person who brought evidence can be recast as the person with the character defect.

Notice the shape of it. I make a claim about human physiology. The reply is not your data is wrong but you are complicit in cruelty. That is not a rebuttal, it is a reclassification — of the argument into a trial, and of me into a defendant. And it is efficient, because once the frame is moral, disagreement becomes evidence of bad character, and the more carefully you argue the more you look like a man constructing an alibi.

I would find this more impressive if it were harder to do. Anyone can escalate to morality. It requires no reading, no citations, no exposure. It cannot be checked, which is precisely its appeal. The metabolic case takes work to make and can be shown wrong; the moral verdict takes a sentence and can only be resisted. When someone reliably chooses the second, I stop assuming it is conviction and start suspecting it is convenience — a position adopted because it wins arguments, not because it survived one.

You can watch the same instinct operate on the evidence itself. The cohort studies get cited with total confidence and zero curiosity about what is inside them — because the number points the right way, and interrogating it might cost something. That is not how a person behaves when they are trying to find out whether they are right. It is how a person behaves when they already know.

And there is a smaller irony worth naming. The moralist is usually the one taking the B12. Standing on ancient principle while depending on a molecule that did not exist in usable form until 1948, in a bottle manufactured by the same industrial apparatus he is arraigning me for participating in. I say that not to score a point about hypocrisy but because it illustrates my actual argument: this diet is a modern achievement, and its most confident defenders are the least willing to say so.

So: I care what you eat, and so do you. The difference is that my caring makes a claim about consequences that you can test and overturn, and the version I keep meeting makes a claim about my character that no evidence can touch.

Only one of us has brought something that can be argued with. I notice which one of us keeps being asked to defend himself.