Honey does not spoil. You have heard this, and it is close enough to true that archaeologists keep turning up sealed jars still, in principle, edible. (The famous tomb finds are real jars but anecdotal science. Treat the specific ages as lore and the keeping as fact.) The usual explanation is that honey is too sweet and too acidic for anything to live in. That is right, and it hides the better story.

Nothing in honey is dead. I did not expect that.

Spoilage is microbes eating your food. To stop it you can kill them, or you can take away the water they need to work. Honey does the second. Every microbe runs on liquid water. The number that matters is not how much water is present but how much is free, unbound, available to a cell. Chemists call it water activity, written a_w, on a scale from 0 to 1. Pure water is 1. Honey sits around 0.6.

That number is not arbitrary. The yeasts that would ferment honey, the osmophilic ones built to live in sugar, can grow down to a water activity of about 0.61. Below roughly 0.60 they stop. They do not die. They drop into cryptobiosis: alive, intact, metabolism paused, waiting. Honey holds itself at 0.6 on purpose, one notch under the line, and the fermenters sit dormant on the other side of it.

So the jar in your cupboard is not sterile. It is a standoff. The microbes are there, viable, held still by a single number the honey keeps just below their threshold.

And honey defends that number with chemistry, not only dryness. As bees ripen nectar, an enzyme called glucose oxidase turns glucose into gluconic acid and hydrogen peroxide. The acid drives the pH down to around 3.5 to 4.5. The peroxide, in tiny continuous doses, suppresses what the osmotic wall alone might miss. Three defenses, layered: too little free water, too acidic, faintly antiseptic.

Here is the part I find lovely. The one thing that beats all of it is the one thing honey wants. Honey is hygroscopic. It pulls water out of humid air. Leave the jar open in a damp kitchen and it slowly takes up moisture. Its water activity climbs, and at some point it crosses 0.6 from below. The dormant yeasts wake. The honey ferments, which is to say it was always going to, the moment you let the water back in. (This is how mead begins, on purpose.)

That is why immature honey, capped by bees before it has dried down, spoils when sealed honey does not. Same sugars, same enzymes. A few points of water activity on the wrong side of the line.

I like preservation stories where the thing that survives is not inert. Honey does not keep because it is a wall. It keeps because it holds a balance, actively, at a number you can tip with a humid afternoon and an open lid. What is on the other side was never killed. It is only waiting for the water to come back.

Sources: the water-activity threshold for osmophilic yeasts (≈0.61), the cryptobiosis detail, and why immature honey ferments are from Immature Honey as a Quality Challenge in Global Apicultural Production (Foods, 2026). The glucose-oxidase to gluconic-acid and hydrogen-peroxide pathway and the pH range are from Biochemical Reactions and Their Biological Contributions in Honey (Molecules, 2022), with the layered-defense framing from Honey Bee Suite.