A Recipe Is a Point, a Technique Is a Space
in my twenties i worked at one of the big national pizza delivery chains. i grew up around living systems, plants, gardens, animals, so the idea that something alive runs on its own schedule was not news to me. what was new was being responsible for one on a clock. the dough came in green off a truck two or three times a week, and the job was to manage it: proof it, hold it, read it against how busy we expected to be. push it too fast or too slow and you paid for it at the dinner rush, and you paid with your hands. under-proofed, cold, green dough has no give at all, and my fingers would go numb and then start to hurt trying to stretch it. over-proofed dough surrenders the other way: that distinctive smell, and a sticky mess clinging to your fingers while the pizza tries to tear itself thin. either way the trays told you what you had done to them three hours earlier.
i already knew the word âfermentation.â what i did not have was the working version: the dough was alive, it responded on its own schedule, and temperature was the lever. everything i understand about cooking since then is downstream of learning that lever early, with money on the line.
the question is when you find out
temperature was just the first mechanism i learned. ratios and order of operations matter everywhere in the kitchen, at the stove as much as in a bake, and you are going to break them, everyone does. the real difference is when you find out. at the stove the feedback is immediate: taste, adjust, throw the recipe out halfway because the pan told you something. a bake answers you once, at the end. supporting my wifeâs cake business is where that became concrete: creaming, reverse creaming, and straight blending mix the same butter, sugar, and flour in a different order, and each bakes into a different crumb, denser, more tender, more open, but you cannot feel that difference in the bowl. you find out after it has cooled, once the ratio and the order have already decided the outcome. buttercream works the same way from the other side: American, Swiss meringue, Italian meringue, the method is the recipe, not an ingredient list.
the dough job sat at the far end of that spectrum: three hours between the decision and the verdict. i think that is why it taught me so much. when feedback is instant you can get away with never really understanding, you just keep nudging until it tastes right. when the verdict does not come for hours, you have to actually know why, because the moment to fix it is long gone before you see the result. the slower the feedback, the more you are forced past nudging and into actually understanding what you are doing.
curry is not a dish, it is a set of axes
the thing that broke it open for me was curry, and specifically the moment i understood that âcurryâ is not a dish at all. it is a multi-axis style of cooking. i found a table, on Wikipedia of all places, that laid it out along five axes: how much heat, how watery or creamy the sauce runs, how dry or wet the dish sits on the plate, how far it leans sour against sweet, and even the cooking method itself, a quick stir-fry against a slow simmer.
that table did something no recipe ever had. it handed me the generator instead of an instance. once i could see the axes, i did not need the recipe anymore, i could just place a dish somewhere in the space and cook toward it. i spent the next few months making dozens of variations with no recipe at all, going by feel and spice combination, moving along the axes on purpose. none of what i made had a name, and that was the point. i just learned where i liked to live in the space, which turned out to be the dry, hot corner: sauce that clings instead of pools, heat that does the talking.
learning a place through its food
i married an Afghan woman, and cooking became the way i learned a culture from the inside. i taught myself barbari bread and a stack of traditional Persian and Afghan dishes, kashk-e bademjan, ghormeh sabzi, khoresht gheymeh, abgoosht, each one a small education in what grows where, what keeps through winter, which humble dish became the beloved one.
and that is where the axes come from. nobody sat down and invented a cuisineâs space, the place drew it: the climate, the soil, the trade routes, the long winters. the dishes carry those constraints inside them, which is why food is a window into a place. you learn the geography and the history by cooking the constraint.
a shepherdâs pie that took the long way home
here is where all of it came together. i started with the shepherdâs pie i grew up with and pulled it apart into its ideas, then rebuilt it out of three cuisines.
the meat layer became a mushroom duxelles, because my wife is vegetarian, spiced like keema matar, the beef and pea mince curry. French technique, Indian and Afghan flavor, no beef. over that, a bright layer of blanched peas with butter and mint. on top, mashed potato piped in ridges, its flavor coming from saffron-infused milk, a Persian note in a British dish. all of it in flaky pastry, baked in a cupcake tin.
and here is the detail that still gets me: the piped saffron potato came out looking exactly like duchess potatoes, a classic French preparation i did not know existed until after i had made it. i was not following anyone. i understood what the potato wanted, what the fat was for, why the ridges caught more color, and i arrived at an established technique from first principles. duchess potatoes were already sitting there, worked out by someone else a long time ago, and i got to the same place without anyone telling me the name for it.
it tasted like a culmination. British, French, Indian, Persian, folded into one thing, and none of it possible if i were still cooking from recipes instead of from understanding.
close
the thread through all of it is not temperature, and it is not the curry table. it is what the dough job forced on me first: when the verdict comes three hours after the decision, you cannot nudge your way to competence, you have to actually know why. and knowing why is what turns a recipe transparent. the mechanisms underneath it, temperature, ratio, order of operations, stop being instructions and become axes, and once you have axes, a recipe is just one point in a space you know how to move through. move through the space long enough and you start arriving at places other people already named. that is all the duchess potatoes were: proof the space is real, and that other people mapped it a long time before i got there.
this is also exactly how i build software: an opinionated taxonomy, authored up front, the data model, the names things get, the handful of constraints that are not allowed to move. the near-infinite flexibility comes after, and it only works because the layer underneath it holds still. the constraint is what buys the freedom, in a curry or a codebase.
recipes are for the part of the journey before the primitives click. after that, the fundamentals are the tax you pay once for a lifetime of improvisation.