by Bee Wilson
The Americans and the British conducted their own reforms, but neither country wished to go as far as the revolutionary French. In 1790, President George Washington gave his secretary of state, Thomas Jefferson, the job of devising a plan to reform weights and measures. The United States already had decimal coinage, having thrown off pounds, shillings, and pence along with the British crown. But in the event, Congress couldn’t agree on either of Jefferson’s proposals for reform and spent several more decades failing to decide anything on the matter.
Meanwhile, in 1824, the British acted. There was no question at this time of following the French—national enemies against whom the country had only recently ceased to be at war—down a route of total metrification; the aim was simply freedom from the dark ages of multiple standards. In 1824, parliament voted to use a single imperial gallon for both dry goods and liquids. The new British imperial gallon was defined as “the volume occupied by 10 pounds of water at specified temperature and pressure.” This worked out as 277.42 cubic inches, which was close to the old “ale gallon.” Once the new gallon had been established, it was easy to readjust the pint, quart, and bushel measures to fit. The adage now went like this:
A pint’s a pound the world round.
Except in Britain where
A pint of water’s a pound and a quarter.
For Britain, read the British Empire. These new imperial measures were confidently promulgated wherever the British ruled. A pint of maple syrup in Colonial Canada was the same volume as a pint of whiskey in Colonial India.
Did this spell an end to confusion in measuring? Not at all. In 1836, the US Congress finally established American uniform standards and decided to take the opposite route to Britain. Instead of adopting the new single imperial gallon, the United States stuck with the two most common gallons from the old system, the Winchester (or corn) gallon for dry goods and the Queen Anne (or wine) gallon for liquids. It is not so surprising that America wanted different standards from Britain. The strange part is that the United States expressed its metrical freedom from Britain, not with its own modern measures but with quaint old British ones. When America sent a man to the moon, that man was still thinking in the pints and bushels of eighteenth-century London. Even now, in this age of Google searches, when the cook of the household is more likely to search for a recipe online than to scour the pages of The Joy of Cooking, the recipes flickering on the screen of American cooking websites are still overwhelmingly given in traditional cups.
The result has been nearly two hundred years of mutual non-comprehension in the kitchen between the two nations, made still worse since 1969 when Britain finally officially joined the metric nations (though many a British home cook still prefers imperial). The United States is today one of only three countries not to have officially adopted the French metric system. The other two are Liberia and Myanmar (Burma). To American ears, there is something cold, inhuman almost, about the European practice of quantifying ingredients in grams. To the rest of the world, however, American cups are plain confusing. How much is a cup, anyway? In Australia, the cup has been defined metrically, as 250 ml. But in the United Kingdom, it is sometimes translated as 284 ml, half a British pint. Canada weighs in with a 227 ml cup, corresponding to 8 imperial fluid ounces. As for the true American cup, it is none of these. The technical definition of a US cup is half a US pint, or 236.59 ml.
Given all this, why did Fannie Farmer in 1896, the “Mother of Level Measurement,” deem the cup system to be so superior and so exact? There was nothing inevitable about America’s preference for volume measurement over scales. If you look at earlier American cookbooks, the methods given are just as likely to be weighing with a scale as measuring with a cup. This is partly because many American cookbooks were in fact British—reprints of successful British recipe books such as Mrs. Rundell’s A New System of Domestic Cookery (1807). But even in authentic American books, most of the recipes imply the existence of scales in the kitchen. The first cookbook published by an American for Americans in America was American Cookery by Amelia Simmons in 1796. Simmons routinely deals in pounds and ounces. Her turkey stuffing calls for a wheat loaf, “one quarter of a pound butter, one quarter of a pound salt pork, finely chopped,” two eggs, and some herbs. She also gives the first American recipe for what would become a classic staple of the American kitchen: pound cake: “One pound sugar, one pound butter, one pound flour, one pound or ten eggs, rose water one gill, spices to your taste; watch it well, it will bake in a slow oven in 15 minutes.”
Amelia Simmons’s pound cake is not a great recipe. The very short timing, of 15 minutes, must be a typo (pound cake in my experience takes around an hour), and Simmons doesn’t tell us how to mix the batter (do we add the eggs one at a time to avoid curdling? Or in one great swoop?). However flawed, it does show that in 1796 at least, Americans were not averse to putting butter and flour on a scale. Pound cake remained a favorite long after cups took over. Even Fannie Farmer includes a pound cake, not dissimilar to Simmons’s, except that she had replaced the rosewater and spices with some mace and brandy; and she plausibly says that it will take one and a quarter hours in a “deep pan.” And she has replaced the pounds with cups.
By the middle of the nineteenth century, cups were taking over from pounds altogether in America. At first, the cup would have been any breakfast cup or mug that was on hand. This is how traditional cooks still do much of their measuring in countries from India to Poland. You take a glassful of this and a cupful of that and it works just fine, because you’ve made the dish a hundred times before, using the same glass or cup. The problem comes only if you attempt to instruct others outside your family or narrow community in how to make a certain thing, when the recipe gets lost in translation. What was different about the cup measurements of nineteenth-century America was the shift from using cups to using the cup—a single standard with precise volume.
Why were Americans so attached to their cups? Some have seen it as a feature of pioneer life, when those traveling west would carry makeshift kitchen utensils on the wagon, but wouldn’t want to be encumbered with heavy scales. There must be some truth in this. In a far-flung frontier settlement, a local tinsmith could rustle you up a cup, whereas scales were an industrial product, made in factories and sold in towns. Besides, frontier meals tended to be ad hoc, such as johnnycake, a stodgy mess of cornmeal and pork fat thrown together from cupfuls and handfuls of this and that.
Yet the frontier mentality cannot entirely account for America’s wholesale adoption of the measuring cup. The evidence from cookbooks is that measuring cups were being viewed not as an inferior substitute for scales, but as better than them. The cup was used in fancy well-equipped kitchens in the cities as well as in creaky wagons. Catherine Beecher, whose sister Harriet Beecher Stowe was the best-selling author of Uncle Tom’s Cabin, wrote a cookbook—Miss Beecher’s Domestic Receipt Book, published in 1846. Beecher notes that “it saves much trouble to have your receipt book so arranged that you can measure instead of weighing.” She assumes that her readers will have scales as well as cups, but views cups as handier. She advises weighing each ingredient the first time it is used, and measuring the volume of the weighed ingredient in a “small measure cup.” Beecher’s idea was that next time the ingredient is required, the cook will be able to bypass the scales and use only the cup.
The cup’s ascendancy was helped by the kitchenware itself: the gradual emergence of specially manufactured measuring cups with gradations for half cups, quarter cups, and so on. Catherine Beecher talks of ordinary teacups and coffee cups, but in 1887, Sarah Tyson Rorer noted the recent appearance “in our market” of “a small tin kitchen cup.” These cups were sold “in pairs, at various prices . . . one of the pairs is divided into quarters, and the other into thirds.” This is recognizably the measuring cup as it continues to this day.
By the 1880s, it had become common for cookbook authors to give cup conversions, so that cooks could do without scales altogether
. Maria Parloa, a popular cookery teacher based in Boston, gave the following conversions in 1882, using a “common kitchen cup holding half a pint”:
One quart of flour one pound
Two cupfuls of butter one pound
One generous pint of liquid one pound
Two cupfuls of granulated sugar one pound
Two heaping cupfuls of powdered sugar one pound
One pint of finely chopped meat, packed solidly one pound
The problem with all these conversions was how to interpret them. Exactly how solid is “solidly packed” meat? How do you distinguish a “generous” from a “scant” pint of liquid? And what on earth is a “heaping cupful”?
Mrs. Lincoln, another Boston cook, and Fannie Farmer’s predecessor as the head of the Boston Cooking School, attempted to weigh in with some qualifiers. A spoonful, Mrs. Lincoln noted, not altogether helpfully, was generally supposed to be “just rounded over, or convex in the same proportion as the spoon is concave.”
What Fannie Farmer did was to take these measures and remove all interpretation from them. The knife with which she leveled the top of her cups eliminated all doubt, all ambiguity. Cups must not be generous or scant, heaping or packed. “A cupful is measured level. A tablespoonful is measured level. A teaspoonful is measured level.” This exactitude offered the cook a sense that cookery had been elevated to the level of science.
Farmer’s method was indeed a huge improvement over the heaping and scant measures of previous writers, so perhaps we can forgive her for failing to spot the fact that the entire system of cup measures was flawed.
Fannie’s fixation with level measures in the kitchen reflected how late she came to cooking. She was born in Boston in 1857, one of four daughters of a printer (a fifth sister died in infancy). She never did much cooking at home. Fannie would probably have become a schoolteacher, like her three sisters, except that while still at high school, she was struck down with an illness, probably polio, and after a period of paralysis, was left permanently weakened, and with a limp. It looked for a while as if she might never leave home. In the 1880s, aged twenty-eight, she took on a job as a mother’s helper in the home of a friend of the family. There she developed an interest in cooking. In 1887, she enrolled in the Boston Cooking School, one of several new schools across the country aimed at teaching middle-class women how to cook. She must have done something right, because seven years later, she was running the school, dressed in a white cap and a white apron that stretched all the way down to her ankles.
The Boston Cooking School taught Fannie Farmer to cook with the purpose-made measuring cups that had lately become available. And she in turn recognized no other method. Her entire approach was about offering cooks a sense that they could do anything, so long as they obeyed the rules and followed her instructions to the letter: absolute obedience would lead to absolute proficiency. As a latecomer to the kitchen, Farmer had none of the natural instincts to fall back on about how much of any ingredient was needed and how long to cook it. Everything had to be spelled out. She would go so far as to stipulate that the pimento garnish for a certain dish be cut three-fourths of an inch long and half an inch across.
The idea was to create recipes that would be absolutely reproducible, even if you knew nothing about cooking: recipes that “worked.” She inspired the same kind of devotion as Delia Smith in Britain today. (“Say what you like about Delia,” people often remark, “her recipes work.”) Evidently, plenty of people found Farmer’s level measures comforting, given her colossal sales (her 360,000 copies sold put her in the same league as Uncle Tom’s Cabin, which had sold more than 300,000 in the months after publication). So long as you had your cup measures and your case knife, these were recipes you could trust, and the admirable thing about a Farmer recipe was that you could repeat it time and again with roughly the same results.
Whether we would want to achieve Farmer’s results today is another matter. Her tastes have not worn well. She was fond of such things as spaghetti timbales (soggy cooked pasta reheated in a mold with salmon forcemeat) and avocados filled with oranges, with truffle decoration and a condensed milk sauce. This brings to mind the food writer Elizabeth David’s comment: “What one requires to know about recipes is not so much do they work as what do they produce if they do work?”
Some of Fannie Farmer’s faith in her own system came from the fact that she had rejected entirely the archaic instructions based on analogy that had constituted almost all kitchen measuring up to her lifetime. Since medieval times, recipe writers had dealt in such currency as fingerbreadths of water and butter the size of a pea, a nut, or an egg. The most universal analogy seems to have been that of the walnut. For Farmer, cup measures were superior to fingers and walnuts because they were both more accurate and more precise. In many ways she was right. Instructions such as “an egg-sized lump of butter” drive lots of rational people to despair. Today, the cookery forums on foodie websites are full of frustrated home cooks trying to ascertain the exact dimensions of a lump of dough the size of a walnut. Is it one tablespoonful? Or two?
Yet for hundreds of years, these comparisons were the main idiom of measurement in the kitchen. Here is Hannah Wolley, author of the Queen-like Closet, in 1672, with a recipe to make “pancakes so crisp as you may set them upright.” The recipe in its entirety reads: “Make a dozen or a score of them in a Frying-pan, no bigger than a Sawcer, then boil them in Lard, and they will look as yellow as Gold, and eat very well.” This is not a recipe at all in Fannie Farmer’s terms. Wolley does not tell us how to make the batter, or how long to cook it. How hot is the lard? How much of it should we use? How many pancakes do we “boil” at once? And how do we drain them?
Unless you were already very confident in the art of making pancakes, you would get nowhere with Wolley But assuming you did have long experience in batter making and frying, it is an interesting recipe. The fancy imagery—“no bigger than a Sawcer” and “yellow as Gold”—made perfect sense if you knew your way around a kitchen. The end result—for twice-fried pancakes—sounds unusual: a bit like a pancake-doughnut hybrid; a cardiologist’s nightmare, but genuinely useful for someone who wanted to make pancakes “so crisp as you may set them upright.”
Before the nineteenth century, almost all recipes dealt with measurements much as Wolley did. They were aide-mémoires for those already skilled in the kitchen rather than instructions in how to cook. This is part of the reason old recipes are so hard to reconstruct: we have no idea of the quantities; we do not know the rules of the game. Take this one, from the Roman Apicius. It is for “another mashed vegetable” (the capitals are his):
COOK THE LETTUCE LEAVES WITH ONION IN SODA WATER, SQUEEZE [the water out] CHOP VERY FINE; IN THE MORTAR CRUSH PEPPER, LOVAGE, CELERY SEED, DRY MINT, ONION; ADD STOCK, OIL AND WINE.
Not to put too fine a point on it, this sounds disgusting: slimy cooked lettuce with two applications of onion, one at the beginning and one at the end. But the quantities and cooking times could make all the difference. Lovage, celery seed, and dry mint are all pungent, aniseed-y seasonings; a pinch of each might be acceptable, a whole spoonful would be horrible. Defenders of Roman cuisine say that there would have been a fine balance among all the strong flavors. We have no way of knowing if they are right.
Compared to this Apician type of recipe, which gives no quantities at all, a “piece of butter the size of a walnut” was a huge improvement. It sounds vague, but actually it isn’t, relatively speaking. Measurement is always a form of comparison—between the fixed standard and the thing being measured. In ancient societies, measuring began, naturally enough, with the dimensions of the human body. Sumerians in Mesopotamia invented units of length based on their own hands: the width of a little finger; the width of a hand; the distance from the tip of a little finger to the tip of the thumb on an outstretched hand. The basic Greek measure was the daktylos, the breadth of a finger. Twenty-four fingers made a cubit. The Romans took the Greek daktylos and made it
a “digit.”
Cooks in the kitchen did exactly the same thing. The finger was a measure that was ever-present. It was literally handy. “Take four fingers of marzipan,” says Maestro Martino, the most renowned cook of the fifteenth century. Artusi, the Italian best-seller of the late nineteenth century, begins one of his recipes invitingly: “Take long, slender, finger-length zucchini.” Using fingers to measure reflected the tactile nature of kitchen work, in which fingers were used to prod meat, form pastry, knead dough.
If there were fingers, there were also handfuls. To this day, many Irish cooks make soda bread using handfuls of flour and refuse to do it any other way. It sounds like it wouldn’t work because human hands are so variable in size. But the great thing is this: an individual cook’s hand never varies. The handful method may not work as an absolute measure, but it works very well on the principle of ratio.
A ratio is a fixed proportion of something relative to something else. So long as one person uses a single hand, to pick up the flour and other ingredients, the ratios are constant and the soda bread will rise. Some nutritionists today still use the human hand as a unit for measuring portion sizes: a portion of protein for an adult might be the palm of your hand (minus the fingers); for a child, the portion is the palm of a child-sized hand. In many ways, ratios work better in cooking than absolute measures, because you can adapt the recipe to the number of people you are cooking for. Michael Ruhlman recently wrote a whole cookbook founded on the principle of ratio, arguing that when you know a culinary ratio, “it’s not like knowing a single recipe, it’s instantly knowing a thousand.” Ruhlman’s ratio for bread, for example, is five parts flour to three parts water, plus yeast and salt, but this basic formula can be tweaked to become pizza, ciabatta, or sandwich bread, or it can be scaled up from one loaf to many. Unlike an Irish soda bread maker, Ruhlman constructs his ratios from precise weights, not handfuls; but the idea is the same.