Skip to main content

Oldest possible food: 20mya?

,

Looking into the oldest edible food suggests a range of 0.05–20mya food, after discarding myths like honey or woolly mammoth.

2021-06-22–3y2025-03-08 finished certainty: possible importance: 1

What is the oldest food you could potentially eat? I found myself wondering this while imagining the silliest possible gourmand, but found that a good answer is surprisingly tricky and wound up factchecking widespread myths and having to define what is “food”

Most foods spoil quickly:

  • the notoriously durable fruitcake lasts only a century or two

  • the most aged wines still become inedible after a few centuries (with an apparent record of drinking 184yo champagne).

  • Honey is supposedly good indefinitely, but the oldest edible honeys only go back a thousand years or so at best (despite widely-repeated myths to the contrary.1)

  • Eggs may go back 1,700 years.

  • Seeds can germinate after 2,000 years and so are surely still edible as well.

  • Bog butter can be found back to possibly 5,000 years, and descriptions of the oldest ones as waxy or having dairy smells implies that they might not be tasty, but are still edible (although see next section on adipocere).

Woolly Mammoths

There are many third-hand anecdotes or rumors of people eating ancient woolly mammoths, which are famously well-preserved (even to the extent of liquid red blood), but woolly mammoths are not necessarily that ancient (technically, woolly mammoths survived all the way up to about 4,000yo on Wrangel Island!) and at least two of the most famous cases of woolly mammoth banquets are bunk (eg. Tolmachoff1929).

Apparently, there are also practical challenges to woolly mammoth due to their preservation conditions: “Some paleontologists told me that they had tried to fry mammoth meat, but it had turned into a smelly liquid since mammoth meat decomposed into a particular type of matter over time (adipocere) that is inedible for humans but can be eaten by animals…”

Possibly various miners or permafrost prospectors have genuinely eaten woolly mammoth, but without reliable sources or scientific dating, none of them can establish a good ancient-food record.

Arctic Bison

The best verifiable anecdote I could find about the oldest thing ever eaten, permafrost meat, only takes you back like 0.05 million years; that was the 50,000kya bison “Blue Babe” (WP), eaten in 1984-04-06 in Alaska by paleontologist Dale Guthrie & guests (see Guthrie1990):

It had been done before. “All of us working on this thing had heard the tales of the Russians [who] excavated things like bison and mammoth in the Far North [that] were frozen enough to eat”, Guthrie says of several infamous meals. “So we decided, ‘You know what we can do? Make a meal using this bison.’” Guthrie decided to host the special dinner when taxidermist Eirik Granqvist completed his work on Blue Babe and the late Björn Kurtén was in town to give a guest lecture. “Making neck steak didn’t sound like a very good idea”, Guthrie recalls. “But you know, what we could do is put a lot of vegetables and spices, and it wouldn’t be too bad.”

To make the stew for roughly 8 people [Kurtén says 12], Guthrie cut off a small part of the bison’s neck, where the meat had frozen while fresh. “When it thawed, it gave off an unmistakable beef aroma, not unpleasantly mixed with a faint smell of the earth in which it was found, with a touch of mushroom”, he once wrote.2

…Guthrie, who is a hunter, says he wasn’t deterred by the thousands of years the bison had aged, nor the prospect of getting sick. “That would take a very special kind of microorganism [to make me sick]”, he says. “And I eat frozen meat all the time, of animals that I kill or my neighbors kill. And they do get kind of old after 3 years in the freezer.”

Thankfully, everyone present lived to tell the tale (and the bison remains on display at the University of Alaska Museum of the North). The Blue Babe stew wasn’t unpalatable, either, according to Guthrie. “It tasted a little bit like what I would have expected, with a little bit of wring of mud”, he says. “But it wasn’t that bad. Not so bad that we couldn’t each have a bowl.” He can’t remember if anyone present had seconds, though.

Permafrost Limits

But you could go further, of course. How much further can permafrost take us?

The oldest edible meat on Earth can be no older than 20 million years, because that’s the longest continuous glaciation/permafrost (in Antarctica). Any other meat has decayed, fossilized/mineralized3, been compacted beyond edibility, etc.

What Is Food, Anyway?

You can go older with pathological examples like ‘drinking glacial core water’ or eating ‘salt’ or ‘calcium carbonate (Tums)’ or iron or the oldest edible/nutritious element of all—lithium.4 But let’s define food as ‘organic’, in being made of non-metals & CHNOPS, which excludes all those.

What is the oldest possible edible organic biological food?

Ancient Seafood

I think that maybe there are extremely slow-living abyssal microbes or endoliths, which one could perhaps harvest in sufficient bulk to ‘eat’, but I am unsure if that’s feasible or if they would be considered to count. We know the abyssal microbes are alive and likely millions of years old, because they very slowly reproduce: “Atribacteria reproducing over millions of years in the Atlantic abyssal subseafloor”, Vuillemin et al 2020.

But while verifying that they are alive & edible, the reproduction poses something of a ‘Ship of Theseus’ problem in that, if they’ve necessarily repaired themselves & reproduced over those millions of years, and likely all the atoms have ‘turned over’, perhaps many times, you have to ask if they are the ‘original’—if arbitrary amounts of repair don’t matter, then you yourself are flesh that is billions of years old… Indeed, on this view, isn’t every life-form descended from LUCA (ie. all the ones we know of, like yourself) ‘actually’ something like 3–4 billion years old?

But if we are willing to include those by arguing for enough continuity to count, spores are also a possibility; there are claims of million-year-old spore revivals, but apparently doubt about the more extreme claims like reviving spores from 25–400mya. Perhaps some day?

Jurassic Park

Finally, a wild card is amber (fossilized resin). Amber famously traps many insects, and could carry us back to 320mya easily. (Apparently plants were not creating resins before that.)

Animals trapped in amber look well-preserved, but I’m unsure if they would be edible—similar to petrified wood or the dinosaur collagen example.

Unlike abyssal bacteria or spores, where we can be sure of their biological integrity because they are still alive, few or no complex biological compounds like DNA have been recovered from the oldest ambers. They could be too heavily chemically transformed by the resin, fossilization, and decay process, akin to chemical fixation. The preservation process necessary to survive hundreds of millions of years at room temperature may have to be intrinsically destructive of all normal chemical properties.

Record: 20Mya

So while it is possible we could someday push the oldest-food record far past 20mya to as far as 320mya, 20mya seems like the safest answer. There is nothing impossible or all that dubious about 20mya, and there are two different ways to get there: the very oldest possible permafrost preservation, and abyssal microbes.

If you insist on a documented verifiable example, then the best we can do right now is 0.05mya (Blue Babe).

Appendix

Rarest Simplest Natural Food

Another interesting question might be, “what is the rarest simplest human-made food in Nature?”

Things like ‘cooked meat’ would not count because animals are cooked in wildfires all the time and predators eat them—some raptors even deliberately spread fires to hunt! Products like wine have natural analogues: overripe fruits or honey caches can ferment & produce alcohol on their own, and inebriated wild animals have been often documented, so that category is broadly out.

Things like baked goods would be highly unnatural, but are also quite complex—one doesn’t simply grab some grass and bake some bread. Bread requires extensive agriculture & processing: there is nothing in the wild like domesticated wheat, much less ground up, fermented into dough, and baked. (Wild grains would be frequently burnt in fires, and perhaps occasionally mechanically damaged, but never the full baking process.)

Vegetables, likewise: highly unnatural, but usually extremely selectively bred or hybridized over centuries and hardly ‘simple’.

Milk & eggs are interesting because they are so versatile, but are still ‘simple’ in that you can get recognizable moo-juice out of wild livestock and wild bird eggs are perfectly edible. Some derivative of them might be simultaneously rare and still simple.

Forest fires might fry eggs, for example. But that might be too common. What about dairy?

Butter is just milk which has been physically shaken or stirred for long enough that it solidifies, no additional ingredients required; having made some in elementary school by shaking a Tupperware container for a long time, I have a hard time imagining any natural process which might accidentally result in the creation of butter. (I could be convinced, though: perhaps a cow could drown in a river and be tumbled around for hours before washing ashore…?)

I would suggest as a better dairy candidate, cheese: it is made with milk, which is common due to mammals, and requires only a simple acidification or digestize enzyme step to solidify milk into cheese, but probably exists nowhere in the wild except for posssibly occasional instances of a young nursing mammal being killed while not eaten or scavenged before their stomach enzymes can digest their last meal into something vaguely remiscent of ‘cheese’. This would be both simple and rare.

  1. The widespread claims that ‘tomb honey’ has been recovered from Egyptian tombs and is still liquid and edible after several thousand years turn out to be false.↩︎

  2. This was actually written by Björn Kurtén, pg194, in his 199134ya book, The Innocent Assassins: Biological Essays on Life in the Present and Distant Past, describing Blue Babe & the dinner, who also gives the exact date of 6 April 198441ya.↩︎

  3. Depending on how trace a residue is acceptable as ‘edible’, it is possible collagen might count, as collagen has been detected in 96 million year & 195 million year old dinosaur bones, due to unusual chemical stability; but I think these are far too trace to count as potentially being a meal.↩︎

  4. If we did choose to allow even elements as food, then the answer would seem to have to be, because the temporal order of the nucleosynthesis of elements proceeds in roughly size order, lithium.

    The only two elements smaller are hydrogen and helium, and I’m not aware of any possible way in which a human could be said to ‘eat’ or ‘drink’ them, as opposed to a molecule requiring other, younger, elements like oxygen, while lithium plays many biological roles—famously, in psychiatry—and may even be a nutritional need of humans. So if elements like iron or copper or zinc count as ‘food’, lithium has to as well.

    And because at least some lithium was created a few hundred seconds after the Big Bang (“trace amounts (on the order of 10−10) of lithium”, citing Fields2012), any human-scale quantity of lithium carbonate etc will almost surely contain at least some lithium atoms from the Big Bang.

    So, at least some of the lithium you eat as a supplement, or in your drinking water, or for psychiatric treatment, would be 13–14 billion years old.↩︎