Milk

by Anne Mendelson

  Until the actual onset of labor, pregnancy-sustaining hormones inhibit this future life-support system from producing true milk. In the last few days or hours, however, a kind of pre-milk equipped with various immunological weapons begins to be secreted. The start of labor triggers a cascade of drastic hormonal readjustments that produce more of this substance—a thick, serous yellowish fluid with a slightly laxative effect, known as colostrum. It is exactly what the newborn needs to get through the transition from placental to oral nourishment.

  Colostrum functions solely as an interim support, and doesn’t look or taste at all like any animal’s normal milk. Within a week or two in most species, the real milk supply appears in response to a sequence of hormonal commands. Mother and infant establish an intricate but (usually) efficient feedback in which a mouth clamping around a nipple causes the “letting down” of the milk that has collected in the alveoli and ducts since the last nursing.

  No animal except man has ever thought of decanting milk into receptacles other than baby animals, or of consuming the milk of any species other than its own. In fact, most interspecies substitutions would be disastrous for newborns because of the crucial matches between milk composition and the physiological needs of different animals. The milk of two kinds is no more interchangeable than the placental blood that supports life within the womb before milk takes over the job.

  Cast your mind over the many destined environments that infant mammals must confront during the period of adjustment made possible by nursing. Arctic waters, for instance, or grassy savannas. If your first duty in life is to grow an insulating layer of blubber, you would die on milk meant for creatures whose immediate task is to build the muscle power to run with the herd. The situation is less critical for humans past the age of weaning who choose to consume another creature’s milk, since it will ordinarily share attention with a number of other foods instead of being anyone’s only sustenance.

  If you know anything about the dairy industry, you know that the milk in one supermarket carton really started out inside not one cow but hundreds, whose milk has been pooled and sent to gigantic processing facilities where it is brought as close as possible to uniformity. Commercial milk is expected not to show any detectable variation throughout the 365 days of a year. A lactating mother’s milk, however, is anything but uniform, and isn’t meant to be produced over a whole calendar year. A natural breeding cycle would dictate that at a certain point the baby starts sampling other food. As it gradually loses interest in nursing and graduates to an adult diet, the mother’s milk supply synchronously changes composition and dries up, and the mammary system reverts to pre-pregnancy mode.

  In the smaller herd animals such as goats and sheep, mothers are ready to conceive again just when the young are fully functioning on their own. In fact, the spring lambs and kids will themselves be able to breed in autumn. The breeding interval is a year longer with cows, but a calf left to its own devices will be grazing and independent of the mother by eight or nine months. In all three cases nature has programmed the mothers to stop lactating before the time of poor grass or no grass and again come into milk when spring vegetation is at hand.

  If you could see and taste the milk of one cow’s, doe’s, ewe’s, or woman’s milking cycle, from the time she stops producing colostrum to the time when the young animal says farewell to nursing, it would be shot through with huge variations. Milk shifts in makeup not only throughout one lactation but from the beginning to the end of one day. Indeed, the first and last mouthfuls that an infant swallows at a single nursing ordinarily differ in composition (the final dribs and drabs being the highest in fat). And this is to ignore the question of how one individual cow’s, doe’s, ewe’s, or woman’s milk differs from that of others in her species, herd, or bridge club.

  We can imagine how visible such differences must have been to cottagers who milked only one animal. Composition could be averaged out to an extent when the milk of a whole herd was pooled. Still, the limits of interference with the biology of milk production were stringent in early times. The first way that prehistoric dairyists discovered to prolong lactation far beyond the needs of infants was to get lambs, kids, or calves weaned to some transitional ration as soon as possible and milk the dam twice (or more) a day as aggressively as the hungriest offspring. Usually she accepted this more willingly if the little one was kept close enough in sight to trigger the let-down reflex.

  A well-known illustrated encyclopedia of the world’s cheeses has an extraordinary photograph illuminating the point better than a thousand words. It was taken among the semi-nomadic Qashqa’i people of southern Iran, who still herd sheep and goats as people might have five or six thousand years ago. On an expanse of stony hardscrabble, a woman wearing the richly patterned Qashqa’i skirt and tunic sits hunkered down at the hindquarters of a shaggy, longhorned black goat that she is milking into a large pail. The animal looks trustingly up at the second member of this classic pre-industrial milking team: another woman who bends down to grasp the goat gently under the chin with one hand while using the other to restrain a tiny black kid, half hidden in the folds of her skirt.

  After some days or weeks the mother may be coaxed to let down her milk without the baby in sight. And if someone keeps milking her with might and main long after her real offspring would have gone on to other food, her body can be fooled into thinking that a young one still needs the peak milk supply that she was pumping out a month or two after giving birth.

  Of the Big Four milch animals, cows are the easiest to manipulate into prolonged lactation. And because they accept alterations of their natural breeding schedule more readily than the other three, they are the easiest to maintain on a year-round lactation (or as long as two or three years) if you can solve the formerly intractable problem of how to feed them through the winter with no pasturage. Today’s dairy-cow rations can be—and often are—bought and administered with no relation to local climate or vegetation, meaning that for many of the animals, twelve- or eighteen-month lactations are routine. Modern dairyists often actually inseminate cows a few months into lactation, allowing a drying-off period of only about six weeks before they “freshen,” or give birth and again come into milk. (Probably this strenuous regimen has much to do with the short life spans of modern cows. Ganmaa Davaasambuu, a researcher on hormones in food, has also discovered that routinely milking pregnant cows channels possibly harmful amounts of pregnancy-triggered estrogens into the American milk supply.) By staggering the breeding schedules of different cows in a herd, a farmer can achieve something like a uniform general output throughout every season. (Goats, sheep, and buffaloes obstinately prefer a more seasonal reproductive schedule.)

  The impersonally standardized form in which most of us now encounter milk tends to blind us to the realities of nursing for each and every mammal. At the onset of lactation any animal mother’s body starts channeling everything she eats into what is not just “nature’s perfect food” but nature’s only food for a newborn navigating the complex early stages of growth and development. Of course, nature never meant this precious substance to enter the outside world. It’s worth re-emphasizing that milk as it emerges from the nipple is as much a living fluid as blood, designed to go straight from the mother’s mammary system into the infant’s digestive system with no detours. Once it is sidelined into a pail or tank, its major components begin undergoing multiple interactions with the forces of irreversible chemical change. What we call cooking is one way of capturing and using these forces.

  A LITTLE HANDS-ON DAIRY CHEMISTRY

  In the age of “molecular gastronomy,” few food lovers will need to be told that every tastable quality in food rests squarely on chemistry. But in the case of milk, even the crudest chemical analysis is too long to fit between the covers of a cookbook. Rather than try to list the many thousands of substances present in any single arbitrarily chosen form of milk (for example, the homogenized and pasteurized cows’ milk usually, if incorrectly, labeled
“whole”), I would like to send you into the kitchen to conduct some simple experiments that show milk’s major components at work.

  You will get more out of these exercises if you first understand that, to food chemists, all animals’ milk is a structured system with three principal “phases.” Each phase, by the way, contains complexities that the utmost powers of science today can barely describe, much less duplicate. Put the three together, and you have something still more mind-boggling.

  All milk starts out with water, and the simplest of the three phases to visualize is an aqueous (water-based) solution with molecules of different substances dissolved in it. But milk is also a suspension, meaning that it contains minuscule undissolved solid particles floating in the aqueous medium. In the third place, it is an emulsion shot through with small dispersed globules of something that can’t be either dissolved or suspended: fat.

  The dissolved substances include minerals such as calcium and potassium, several proteins known as lactalbumins and lactoglobulins, and the special sugar of milk, lactose. The suspended particles, which are larger than even the large globulin and albumin molecules, are called micelles and are intricately cobbled together out of more calcium as well as casein, a unique form of protein that furnishes the main protein reservoir for the newborn. The still larger emulsified fat globules are made up of countless different lipids (fat-related substances) linked up in still more innumerable configurations, each globule being surrounded by a delicate but sturdy membrane that keeps the contents from spilling into the milk.

  The three fundamental phases can be clearly seen in any sample of milk under a microscope. But people actually knew of them long before microscopes. The kitchen-lab experiments that I suggest re-enact some major pre-industrial discoveries of dairying peoples. (A note to people who have trouble digesting lactose: Unless you suffer from extraordinarily severe lactose intolerance, you should be able to taste small samples of full-lactose milk without ill effect. In this sequence of exercises, the two things most likely to cause symptoms are the unsoured milk and whey. Little of the original lactose will remain in the soured versions, and the cream in several experiments has much less lactose than milk. Skim milk has proportionally more than whole milk.)

  EXERCISE 1 Begin by getting enough pasteurized, unhomogenized whole cows’ milk to work with, preferably a gallon but at least half a gallon. (I’m all in favor of rescuing raw milk from the public-health doghouse, but will leave that battle to others.)

  What if you hunt frantically through a dozen stores and can’t find unhomogenized milk? Or are appalled at how much it costs compared with “regular” homogenized milk? Well, look on the bright side—you’ve already learned something! Without spending a cent on anything except gas and shoe leather, you have just discovered one of the true idiocies of the American milk industry. Illogical as it may seem, milk that retains its three basic phases in unmonkeyed-with form is—when you can find it—usually at least three times as expensive as milk that has been put through a complicated, energy-intensive alteration of the original structure. Go figure. But if you completely strike out, buy a gallon or half a gallon of skim milk (preferably without added milk solids, another distorting bit of interference) and half a pint of heavy cream (preferably unhomogenized—simply look for a label that doesn’t say “homogenized”—and non-ultrapasteurized). If you have to use these expedients, skip the first two exercises and go on to the third (this page).

  Pour the cold milk into a glass or stainless-steel bowl that is deep enough not to spread out the contents in a big lake but wide enough to admit a scooping/skimming tool such as a shallow ladle or large spoon. You may find the mouth of the bottle or carton partly stopped up with what used to be called a “cream plug,” a thickened blob testifying to the fact that your milk probably took several days to get to the store. (When unhomogenized milk was more common, it moved much faster and was likelier to reach the family doorstep with a liquid cream layer than a definite plug on top.) Break it up, if necessary, by gently whisking until it is smoothly recombined with the milk.

  PITCHER TO MEASURE THE DEPTH OF CREAM, WHICH USED TO BE THE MEASURE OF QUALITY IN UNHOMOGENIZED MILK

  Before going any further, taste the milk. Concentrate your attention on what’s in your mouth: something ethereally subtle but concretely there. This milk has a kind of roundness or depth that the homogenized equivalent doesn’t. The reason is that the contrast between its leaner and richer components hasn’t been ironed out but remains just delicately palpable. Its flavor is not so much flavor as a sensation of freshness on the palate that scarcely translates into words. “Sweetness” is as close as anything, but it’s an elusive note on the thin edge of perception rather than sugar-in-your-coffee sweetness.

  What you have just sampled is, for humans past the age of weaning, one of the world’s oldest beverages after water. It was never drunk everywhere in the fresh liquid form you are encountering; the reasons, as we have seen, vary, from human digestive quirks to the ubiquity of milk-souring bacteria. You should also recall that goats’ and sheep’s milk, with their more distinctive flavors, are still older in human experience, and that your sample has undergone a few significant changes through pasteurization. Even so, you are getting close to something primeval.

  EXERCISE 2 Let the milk sit in the refrigerator, covered, until a well-defined layer collects at the top. Layering will start within hours, but it may take from twelve hours to two days before the “top milk,” as people used to call it, is really well separated.

  Why such wide variation? Take this as the first concrete illustration of what you just read about the non-uniformity of milk in a state of nature—something you should get used to if you plan to work much with milk in your own kitchen. If your sample came from Jersey cows, it will separate faster and more distinctly than Holstein-Friesian milk. Goats’ milk would take several days for a somewhat incomplete separation, while water buffaloes’ milk would separate quite clearly in much less time than cows’ milk.

  Use your skimmer to remove the top layer to a smaller nonreactive container such as a small glass bowl or measuring cup. You won’t be able to get all of it without a little remixing of top and bottom. Another lesson: The two layers can’t be fully separated by hand (it takes a mechanical centrifuge). But you have now performed your first feat of applied dairy chemistry. You have used gravity to isolate (though incompletely) the emulsified phase from the other two milk phases.

  Of course you know that the thicker liquid in the smaller container is cream. It rose, or “creamed,” because it is lighter than the rest of the milk. (“Heavy cream” is a misnomer as regards specific gravity.) It contains nearly all the milkfat, still emulsified in a small amount of the original water-based solution; traces of the suspended micelles also remain. The reason for differences in creaming time is variation in the size of milkfat globules; larger ones, like those in buffaloes’ milk, rise faster than the much smaller ones in goats’ milk. With cows’ milk there are well-known differences among breeds.

  When cream separates more promptly there’s also more of it. A gallon of good rich Jersey milk may give you as much as two or three cups of cream. But don’t be surprised if your sample yields less than half that amount—again pointing to the unpredictability of mothers in comparison with machines. Taste it. Maybe if it were shocking pink or moss-green, one wouldn’t have the same reaction, but its ivory sheen and caressing smoothness suggest some stunning union of virginal and carnal. Put it back in the refrigerator for the nonce.

  EXERCISE 3 Your original container of milk is now a container of “skim milk,” which means the solution and suspension parts of the three-phase system that you started out with. Stir any vestiges of cream back into the whole, pour out a small amount, and taste it. It will be very slightly sweeter than the milk you sampled before. (The cream will be sweet, too, but the two kinds of sweetness are indefinably different.) Notice how much richer and finer this feels on the palate than any commercial skim milk,
even or especially the kind enriched with intrusive-tasting milk solids. In hand-skimmed milk the residual trace of cream creates an effect you wouldn’t guess from its minuscule volume.

  If you weren’t able to obtain unhomogenized milk, proceed with a gallon or half a gallon of skim milk. In any case, pour half of the milk into a saucepan, half into a nonreactive bowl or wide-mouthed jar. Stir a little cultured “buttermilk” into the second half—we’ll come later to the reason for the quotation marks. Plain yogurt is not as suitable because yogurt cultures really require special cosseting. But it will work as a second choice. Whichever you use, the label should unmistakably say that it contains active cultures.

  Relative amounts aren’t terribly important, since different samples will vary with the strength of the culture. Half a cup of “buttermilk” or plain yogurt should be more than enough for two quarts of skim milk. What you are doing is “inoculating” the milk with microorganisms that will convert some of the lactose into lactic acid. Cover the container and let it sit undisturbed at room temperature until it is slightly soured. This happens faster in a warm room, but timing will be unpredictable no matter what—eight hours, twelve, sixteen. You will get a thriving colony sooner or later and shouldn’t worry if it’s later. (The only reason for complete failure would be antibiotic contamination of the milk—illegal and rare, but not absolutely unheard of.)

  Keep tasting the milk at intervals until it has a perceptible sourness and a bit of body. Now pour yourself a little and drink it. You are tasting something that most of the world’s milk users are far more familiar with than fresh sweet milk. Plain soured milk brings history to life, or more accurately, prehistory. It harks back to the earliest chapters of human culinary discovery—the knowledge of how to change one flavor into another.