Reinventing the Wheel: Milk, Microbes, and the Fight for Real Cheese 9780520964464

Table of contents :
Contents
Prologue: The Lost World
ONE. Ecologies
TWO. Real Cheese
THREE. The Third Rail
FOUR. Breed
FIVE. Feed
SIX. Microbes
SEVEN. Risk
EIGHT. Cultures
NINE. Families and Factories
TEN. Expertise
ELEVEN. Markets
TWELVE. Reinventing the Wheel
Acknowledgments
Appendix: How to Buy Cheese
Glossary
Notes
Index

Citation preview

R EI N V E NTI N G TH E W H E E L

c a l i for n i a s t u di e s i n food a n d c u lt u r e Darra Goldstein, Editor

g

re

n e t v in in

MI L K , M IC R O B E S , A N D T H E FI G H T FO R

bronwen percival francis percival

t

he

l

REA L CHEE S E

e e wh

university of califor nia pr ess

University of California Press, one of the most distinguished university presses in the United States, enriches lives around the world by advancing scholarship in the humanities, social sciences, and natural sciences. Its activities are supported by the UC Press Foundation and by philanthropic contributions from individuals and institutions. For more information, visit www.ucpress.edu. University of California Press Oakland, California © 2017 by Bronwen Percival and Francis Percival Library of Congress Cataloging-in-Publication Data Names: Percival, Bronwen, author. | Percival, Francis, 1978– author. Title: Reinventing the wheel : milk, microbes, and the fight for real cheese / Bronwen Percival, Francis Percival. Other titles: California studies in food and culture ; 65. Description: Oakland, California : University of California Press, [2017] | Series: California studies in food and culture ; 65 | Includes bibliographical references and index. Identifiers: lccn 2017006845 | isbn 9780520290150 (cloth : alk. paper) | isbn 9780520964464 (ebook) Subjects: lcsh: Cheesemaking. | Cheesemaking—Technological innovations. | Dairy farming—Technological innovations. | Cheese—Microbiology. | Raw milk—Microbiology. | Cheesemakers. | Cheese industry. Classification: lcc sf271 .p47 2017 | ddc 637/.3—dc23 lc record available at https://lccn.loc.gov/2017006845 Manufactured in the United States of America 25 24 23 22 21 20 19 18 17 10 9 8 7 6 5 4 3 2 1

If a Dedication or Introduction to the following Work should be thought necessary, I most humbly and justly address it to the excellent DAIRY-WOMEN, of Great Britain; duly sensible, that from them I received the first hints that led me to the performance, and without whose assistance and encouragement, joined with my own knowledge and experience, I should never have offered it to the Public. J O S I A H T WA M L E Y,

Dairying Exemplified, 1787

This page intentionally left blank

CONTENTS

Prologue: The Lost World ONE TWO THREE

Ecologies

•

FOUR

•

Breed

FIVE

•

Feed

EIGHT

68

Risk

93 119

Cultures

•

147

Families and Factories

•

TEN

•

ELEVEN TWELVE

•

27

42

Microbes

•

SEVEN

NINE

17

The Third Rail

•

SIX

1

Real Cheese

•

ix

•

Expertise •

Markets

175

192 218

Reinventing the Wheel

Acknowledgments 247 Appendix: How to Buy Cheese Glossary 255 Notes 261 Index 283

234

251

This page intentionally left blank

Prologue T H E LO S T WO R L D

A Lost World sits high in the mountains of the Auvergne in central France, a land of summer mists and long-dormant volcanoes. As every French child is taught in elementary school, it is here that the great French national myth has its origins, where the Gaulish chieftain Vercingetorix gallantly resisted Julius Caesar. In lulls between their skirmishes, the adversaries at the Battle of Gergovia may well have eaten the local cheese. Certainly, the cheese attracted the attention of the Romans. Pliny the Elder even mentions it in his Naturalis Historia. With a documented history of two millennia, it is cheese outside of time. We are here to visit Guy Chambon, one of only five remaining producers of Salers Tradition, and the cheese seems an appropriate object of pilgrimage. Made in huge near-hundred-pound wheels, Salers has a firm but supple paste and smells of straw, broth, and butter. Texturally, this is the closest that French cheese gets to Cheddar, but the flavor is free-spirited and feral, unlike any other cheese in the world. Salers is primal. But first we must get to the top of the mountain. A long and winding drive alternates dizzying hairpin turns with glimpses of the plateau of the Massif Central. It has been millions of years since the last eruption, but craters and the rounded domes of extinct volcanoes still scar the landscape. Once we have abandoned the car near the peak, there is still a brisk hike to Chambon’s buron, the austere mountain chalet where he makes cheese throughout the spring and summer season. It is nearly three in the afternoon, and we are just in time for the evening milking. Dairy farmers from biblical times would recognize Chambon’s method. The trappings of modern food production—of obsessive hygiene, efficiency

ix

of scale, and employee comfort—are nowhere to be found. Even a roofed milking parlor is conspicuously absent. Cows are milked in the field, without electricity or running water. The cows themselves are also from another era. Deep mahogany in color, with chunky bodies and impressive horns, Salers cows look like bovine triceratops. Almost all Salers are farmed for meat, and with their hardy bodies they require next to no supervision on the exposed mountainside. Only 5 percent of Salers cows are milked. Those that are give roughly a third the milk of a typical black-and-white Holstein. Notoriously surly, the Salers will not let down their milk in the absence of their young, so the calves are brought to a small pen near their mothers. Mother and calf share the same name, and one by one the calves are called forth and allowed to suckle for a few seconds before milking commences; dexterous tongues and saliva do the herdsman’s job of cleaning the mothers’ teats. Chambon’s family assist with practiced care, his wife sternly handling the mature cows while his teenage son chases their calves. Tall and rangy, the son’s adolescent physique matches that of the calves; when one escapes, he chases it down in a tangle of lanky teenaged limbs. Chambon is a stout man in a stained boiler suit, and he enjoys entertaining his audience. As he works, he keeps up a colorful commentary for the small group of curious hikers that has assembled on this sunny afternoon. It includes details about his herd—one cow, twenty years old, has had eighteen calves—but mostly consists of trash talk about the previous night’s rugby match. When he does the first milking in the early hours of the morning, or on those days when the weather is not so sympathetic, Chambon works without any onlookers. It takes him two hours to milk the entire herd of sixty cows. Once the calf has briefly suckled on its mother’s teats, he deftly prizes it away, tying it to her front leg so that it cannot reach the udder while he milks. A bit of salt placed on the back of the calf ’s neck encourages the mother to lick it, and mother and baby quietly bond while Chambon works. After a while, the tourists lose interest and head off back down the mountain, and the only sound is the musical ringing of the cowbells. As milking progresses, the warm milk is collected in wooden vats, known as gerles, which are used for years on end without ever seeing a cleaning chemical. Seeing the gerles, our host and guide gestures with enthusiasm. She turns to us and smiles: “Yes,” says Dr. Marie-Christine Montel, “wood is good!” x

•

PROLOGUE

S C I E N C E OV E R A B A R R E L

The cheese made in this part of the Auvergne survived the Romans, the Black Death, and the Nazis, but in 2004 its future looked bleak as it met its most implacable foes: public health officials unconvinced that a cocktail of milk and calf spit allowed to sour in a wooden bucket was fit for human consumption. The French food safety agency, the Agence française de sécurité sanitaire des aliments, was skeptical that it was possible for Salers to meet modern food safety requirements. Conflict between producers and public health bureaucrats is often couched in the apocalyptic language of clashing civilizations. In this case, it was quite literally true, with Iron Age Gaul struggling against postindustrial Europe. In part, the conflict stemmed from the internal political situation within the local cheese industry. The legislative and institutional framework within which Salers cheese is made is, to put it charitably, a mess. Looking back across the millennia, all of the cheese of the region was called Cantal, the name of the local mountain range and the modern administrative département—a name that carries great historical resonance. However, during the twentieth century, Cantal production began to change. As cooperatives ramped up their scale of production and began to farm more intensively, the old style of free-range mountainside cheesemaking on seasonal pastures was under threat. In response, in 1961 a new protected name, Salers, was created for producers who wanted to work within the free-range tradition. It would be the grand cru of Cantal. That, at least, was the idea. However, the rules that defined Salers cheese were lax and ill-drafted. The wooden gerles could be of any size, from the tubby two-hundred-liter barrels that Guy Chambon uses to thousand-liter vats employed by larger operations. Using milk from the indomitable Salers breed was optional, and many producers chose instead to use the milk of high-yielding, docile Holsteins. Salers Tradition was a protected term that could be used for the cheese of purists like Guy Chambon, but they are in the tiny minority. The conflict was as much within the cheesemaking community as it was with any external regulator, between the producers who felt that traditional practices were an impediment to progress and those for whom the old methods were the defining feature of the cheese. With the public health officials demanding that they modernize their production practices, the producers of Salers faced a stark choice. If they adopted the equipment and methods of twenty-first-century cheesemaking, PROLOGUE

•

xi

they would satisfy the authorities. However, abandoning the wooden gerles and joining the rest of the cheesemaking world in working with plastic and stainless steel would require them to buy commercial bacterial cultures to acidify their milk. Each gerle hosts its own unique microbial community that imparts a distinct flavor and personality to the cheese. Replace them, and the cheese would lose its heart: a standardized Salers would be no different from the semi-industrial Cantal churned out at a cheap price by large producers on the valley floor. For the tiny community of cheesemakers who were prepared to follow their cows twice a day into the fields to milk them and who depended on the premium price their cheeses commanded because of their uniqueness, this would be the end. Besieged and at the point of surrender, the Salers cheesemakers turned to their last best hope: Dr. Montel. Wiry and diminutive, Montel is an unlikely champion; with her diffident charm, she is more like a kindly French grandmother. Yet the government laboratory she directs at the Institut national de la recherche agronomique (INRA) facility in nearby Aurillac has become the call of last resort for traditional cheesemakers facing existential crises. A microbiologist by training, Montel applies her skills to testing the hidden and underlying assumptions of modern hygiene. Are consumers less likely to get sick if all of the work surfaces and raw materials are sanitized? That, says Montel, is an interesting hypothesis. And like any hypothesis, it can be tested empirically. As they studied the microbial communities of the gerles and the raw milk they contained, the scientists in Montel’s lab came to a startling conclusion. Not only was the wood of properly prepared gerles teeming with microbial life, but it also actively resisted attempts to contaminate it with pathogens. So lively was the ecosystem that these “natural starter factories” inoculated fresh milk within seconds of contact; the porous wood was not just safe, it was actively beneficial. Moreover, the system required raw milk. When Montel’s team attempted to make successive batches in the wooden gerles with pasteurized milk, the communities of the biofi lm changed. They soon became unbalanced, without the appropriate components for cheesemaking; somehow, the raw milk itself was keeping the communities stable.1 With Montel’s results in hand, the Salers cheesemakers easily won their reprieve. For a group that had been prepared for a fight to the death, it was a comic deus ex machina. After just a few months of scrutiny, modern molecular methods and some straightforward experiments had demolished mainstream notions of effective sanitation. xii

•

PROLOGUE

Salers was exceedingly lucky long before Montel and her team swooped in to save the day. By dint of its isolation and its avid consumer base, the cheese still had a continuous tradition left to save. But even a hundred years ago, it would hardly have seemed exceptional. Just like Salers, every cheese was once the product of its own indigenous microbial cultures, local breeds, and specialized knowledge. Today, the norm is industrial monoculture. DuPont, maker of the highly successful Danisco culture range, even boasts, “Every third cheese obtains its well-defined flavor and texture from our leading cultures.”2 But what if Montel had a time machine as well as a DNA sequencer? What might she have been able to save? This is the story of what has been lost and how scientists, farmers, and cheesemakers are working together to reinvent the wheel.

PROLOGUE

•

xiii

This page intentionally left blank

ONE

Ecologies

For her advocacy of small producers and raw-milk cheese, Marie-Christine Montel has become a minor celebrity within the cheese world. Her research is an inspiration to a generation of technically curious cheesemakers: she is the cheese geek’s cheese geek. When American cheesemaker Mateo Kehler talks about the on-site laboratory that he and his brother Andy have set up at the Cellars at Jasper Hill in Vermont, he jokes that “there is already a lab coat hanging up here with Marie-Christine’s name on it.” So there was great excitement when Montel was invited to deliver a presentation on her work with the Salers producers and their gerles at the 2015 American Cheese Society conference in Providence, Rhode Island. The annual conference is the single largest gathering of North American cheesemakers, importers, and sellers. There are educational sessions, industry briefings, and plenty of swag. Delegates wear intricately color-coded badges to denote their status, with categories ranging from members of the press to senior attendees, who are labeled “Aged to Perfection.” At the 2015 meeting, twelve hundred people packed into the Rhode Island Convention Center to learn, gossip, and do business. When it came time for Montel’s presentation, a large audience crowded into the seminar room. But as they heard the tale of the gerles and saw the experimental data, they sat unmoved. The elegance of the microbiology, the revalorization of ancient cheesemaking techniques, and the implications for understanding and enhancing their own cheeses were lost on the crowd. Instead, in the question and answer session at the end, it became clear that the research on gerles had seemed irrelevant to the majority of the audience. What most people wanted was advice on potential pathogens and handling issues with public health officials. 1

The audience had not been ready for her message, had not been primed for a world where microbial biodiversity could be the defining goal of good dairy farming. It was, said Montel afterward, as if she were “from a different planet.” It was true. She might as well have delivered a lecture on unicorn ranching. Despite our frustration, we could recognize and appreciate the concerns of the audience. Their problems were intimately familiar: this was the dairy world in which Bronwen’s family had lived for over a century.

YO U R G R E AT- G R A N D FAT H E R ’ S H E R D

A decade earlier, amid the cocktail chatter at our wedding in London, two dairy industries collided. Many of our guests were Bronwen’s colleagues from Neal’s Yard Dairy in London, a company that has come to emblematize the revival of British farmhouse cheese. Seated next to the company’s sales director was Bronwen’s uncle Eddie. We always thought of Eddie as a car enthusiast ex-NASA engineer, but he was also a farmer, fighting a last stand to maintain the viability of the California dairy that had been founded by Bronwen’s great-grandfather. At the time, Eddie’s farm in California was comfortably larger than any dairy operation in the United Kingdom. However, even with thousands of cattle, the farm was too small to prosper amid the vicissitudes of the American market for liquid milk. Despite milking high-yielding cows twenty-four hours a day, life was a struggle. But this was not a consequence of his poor management or lack of business savvy; rather, he was the prisoner of a market that was beyond his control. Bronwen’s great-grandfather, Fred Imsand, was a Swiss emigrant to California at the turn of the twentieth century. From a dairying family, he found work at a dairy while he romanced the chambermaids of San Francisco. He survived the 1906 earthquake by the simple expedient of standing in an empty lot during his milk-delivery run, and in the ensuing chaos he made his way to San Bernardino in Southern California, where through a combination of savings, resourcefulness, and gruff charm he acquired a farm of his own, Meadowbrook Dairy. The dairy of Fred’s era was diversified and complete. Beyond the cows and their milk, the family farmed chickens and pigs, smoked their own hams, and sold produce from their orchard. Yet Bronwen’s grandparents eventually rebelled against the unremitting toil of this system, embracing the progres2

•

ECOLOGIES

sive promise of scale, mechanization, and specialization. By the late 1950s, the dairy was a successful local business, with a fleet of milk trucks and—as the Inland Empire grew into Eisenhower-era suburban comfort—five little dairy drive-thrus. Fred had started out with a herd of twenty Holstein cows, but Bronwen’s grandfather Eddie Sr. recognized the direction that the milk market was going and made every effort to expand the business. In this environment, Bronwen’s mother’s involvement in the dairy as a young woman was restricted to some light bookkeeping as she studied for her medical school exams; a loathing for calves’ liver is the only legacy of her upbringing on a dairy farm. By the early 1970s, as supermarkets began to dominate retail sales, Eddie Sr. decided to concentrate solely on producing liquid milk and abandoned direct-to-consumer sales completely. With that decision came expansion: by the time Bronwen’s mother graduated from college, the herd numbered just under four hundred cows. It was ultimately a question of pragmatism, and decisions were made in order to survive. Eddie Sr. had spent a year studying at the University of California, Davis, in the 1930s before the worsening economic climate of the Great Depression demanded that he return to work on the farm, and at each stage of Meadowbrook Dairy’s expansion, he looked to the experts at UC Davis—one of the major American centers for industrial agricultural research—for advice. Each decision he made was based on progressive mainstream ideas about best practice. The pace at which Meadowbrook Dairy grew mirrors trends in dairy farming within the United States at large. We can see these changes reflected most starkly in US Department of Agriculture statistics. From 1970 to 2006, average herd size leapt from just 19 cows to 120 cows per farm. Hidden within the arithmetic mean is an even more significant change: small dairy farms are disappearing rapidly. The smallest class of farm, those with fewer than thirty cows, might still constitute nearly 30 percent of all dairy operations, but together this bottom third represents only 2 percent of all cows and 1 percent of milk production. In contrast, between 2000 and 2006, the number of farms with more than two thousand cows doubled. In 2006, almost a quarter of all milk production—and the majority in the western United States— took place at these megadairies.1 Europe is also seeing a steady consolidation of the dairy industry and growth of herd size. The number of registered dairy producers in the United Kingdom dropped by more than half from 1998 to 2013. From 2008/9 to 2012/13, the only UK dairy farms growing in size were those producing more ECOLOGIES

•

3

than two million liters of milk a year; based on average milk yields, the average herd size of these farms was approximately three hundred cows.2 With milk prices at historic lows, milk production at dairies with more than two thousand cows is becoming more widespread. Where the United States has led, Europe is following. Change came rapidly to Meadowbrook. A sudden series of family bereavements led to a swift generational succession, and in 1978, at age twenty-seven, Bronwen’s uncle Eddie took full responsibility. It was a tough time. The suburban expansion of San Bernardino was about to swallow the dairy, and strategic decisions had to be made. Again, the advice of the dairy extension program at UC Davis proved critical. Resisting a possible move to the San Joaquin Valley, Eddie relocated the entire operation fifty miles north across the San Gabriel Mountains to El Mirage. While the high desert did not afford lush pastures, a new five-hundred-acre alfalfa ranch provided vertical integration. The dairy’s systems were pared down for maximum efficiency, and an anaerobic digester was installed that converted manure into electricity. Again, scale increased. At its peak, the new operation was milking 2,200 cows, but life was still a struggle. For this most mainstream of dairies, there was no sense that processing the milk on the farm or attempting to produce a unique product could add anything to the sustainability of the operation. When family members talk about the dairy, their pride is palpable, but each commercial decision and each stage of growth is explained and rationalized as the inevitable consequence of market conditions. Over the more than thirty years Eddie spent running the farm, he was beaten down by a commodity market that he could not control. His is a common stoicism: “Every farmer goes through periods of up and down; it’s a cyclic business as far as profit and loss go. It’s never been a business where you can count on a percentage of margin.” The farm’s milk contract paid according to total solids—the number of pounds of butterfat and protein that the herd could produce—so the system was optimized for maximum production with maximum efficiency. In this model of dairy farming, controlling feed costs is everything, and Eddie was forced to become increasingly sophisticated at supplementing the silage and hay that he made himself with cheaply available commodity by-products. Technology helped. Eventually, it was simply a question of entering the details of the almond hulls, cottonseed, or citrus pulp into the computer to get the appropriate balance for a nutritionally optimized total mixed ration. To this extent, Meadowbrook Dairy was the diametric opposite of Guy Chambon’s operation. The entire conceit of Chambon’s Salers Tradition is to 4

•

ECOLOGIES

have cows who will thrive in a place where they will eat interesting food— hence the painstaking twice-daily milking in the pastures on the top of a mountain—and then let the cheese make itself, with a little help from the gerles. Uniqueness is fundamental. Taken together, these farming and cheesemaking practices, from the choice of the cussed and archaic Salers breed to the maintenance of the biodiversity of the mountainside pastures and the microbial biofi lms on the wooden gerles, make for the production of something that could not be achieved anywhere else. In contrast, Meadowbrook Dairy tacitly accepted that their output would be blended with milk from many other sources and that the route toward commercial sustainability was through efficiency and growth. Instead of keeping cows in a place where they would eat interesting food, Meadowbrook kept its cows where space was cheap and then fed them carefully calculated inputs to keep costs down and yields up. When none of his children expressed any interest in dairy farming, Eddie ultimately took the opportunity to divest himself of the business. When we talk with him now, his sense of relief is clear, even if it is bittersweet: “We ended up taking this as a time we could slow down and keep the [alfalfa] farm out in Inyokern, and we donated the land to the water district. It was an opportunity that comes along only once or twice in a lifetime.” Although the cheesemakers attending Montel’s presentation at the American Cheese Society conference almost certainly operated on a smaller scale than thousand-cow dairies, the mentality with which Eddie worked would have been familiar to them. It is true in much of Europe as well, where the scale might be smaller still but the same pressures toward consolidation, volume, and efficiency are being felt. For Bronwen, this too is familiar. When she was a teenager, she and her family had managed, by following received notions of best practice, to domesticate industrial dairying.

INDUSTRIAL WRIT SMALL

Unlike her mother, Bronwen did not grow up on a dairy farm. Visits to her cousins at Meadowbrook Dairy brought exotic new experiences, like the opportunity to climb mountains of fuzzy cottonseed feed, but the dairy and its evolution registered only vaguely on her consciousness. Her parents settled several hours further south in eastern San Diego County, in the foothills of the Cuyamaca Mountains. With hot, dry summers and mild winters, the ECOLOGIES

•

5

area was dominated by chaparral and horse fancy. Surrounded by equestrianism, Bronwen soon became obsessed by the desire to own a horse. And so as she began junior high school, Bronwen joined the local 4-H club along with her best friend Melody; it seemed more exciting than the Girl Scouts, and it would give her the chance to make a halter for the horse of her dreams. But Melody, whose family owned a billy goat called Buck Rogers, also convinced Bronwen to sign up for the dairy goat group. At the first meeting, she was exposed to the vision of baby goats frolicking in fresh straw, and all thoughts of horses were immediately banished. Bronwen’s parents recognized a modified win when they saw one, and within days a goat pen was being constructed in the backyard. When Bronwen’s parents, a musicologist and a physician, bought a house with land, they had no intention of dabbling in domestic dairying; they simply needed space where they could practice the violin without the neighbors complaining. Music loomed large in their lives—they had first met as teenagers playing in the same orchestra—and distance from other neighbors allowed assiduous practice to be combined with the odd hours of the work schedule at the hospital. While Bronwen’s mother had been raised on the dairy farm and her father’s family had dabbled in semisuburban farming when he was young, they had no firsthand knowledge of animal husbandry. Jerry Belanger’s book Raising Milk Goats the Modern Way (Garden Way Publishing, 1975) would be their homesteading bible. The goats, Natasha and Ginger, were to play a major role in the life of Bronwen’s family for the next six years. They grew into glossy, glorified pets, but a vague sense of paranoia surrounded their management. Their pen sat within a two-and-a-half acre goat smorgasbord packed with sumac, foxtails, eucalyptus, manzanita, and native sages. This type of Mediterranean scrubland is one of the classic environments for goat foraging, a way of exploiting marginal land that has been practiced for thousands of years. The goat book, however, suggested otherwise, claiming that a diet foraged from the native chaparral was poor in nutritional value and would lead to malnourished goats and problems with worms. The family deemed it much safer to buy in a mixture of hay, fermented alfalfa with molasses, and vitaminenriched, grain-based goat chow for them to eat. In an area famous for its raging wildfires, with strict requirements for brush control, the goats looked on quizzically every summer as Bronwen’s father scoured the property with a weed whacker and the family raked and bagged up the fallen weeds to take to the local dump. Ironically, within the world of forestry management, goat 6

•

ECOLOGIES

grazing is now regarded as one of the most effective weed management solutions for fire prevention; it is also inexpensive, nontoxic, and nearly carbon neutral.3 At its headquarters in Mountain View, California, Google now uses goats for just this purpose. Then there was the question of sex. There was no mistaking when the goats came into heat: the sex-starved animals would stand on the big rock in the middle of the pen and bleat loudly and incessantly. Were it not for this habit, they would likely have been bred only a couple of times. But as it was, they got their way year after year. When it came time to breed them, they were loaded into the back of the family’s dusty Chevrolet Suburban and whisked off to the goat breeder’s for a quickie with the buck. Bronwen’s father remembers setting aside his Messiaen program notes on a day when the rest of the family was gone, the kids all at school, and “driving the goat off to get nailed.” When the baby goats arrived, the family kept the females or gave them to other members of the 4-H club who wished to found their own goat dynasties, but their sentimental approach to animal husbandry left them in a quandary when it came to the boys. The first set of kids were both, tragically, male. They were sent off at a tender age to “eat grass” in a friend’s backyard. Another male kid followed; we learned in researching this book that he ended up as the main course at a family friend’s Easter celebration. Had Bronwen known the truth at the time, she would have been beside herself, but it highlights a perennial problem for sentimental domestic dairy farmers: males. Unless you are prepared to eat them, they have no value. Bronwen’s reservations about eating her pets were not remarkable. When we talked with Jeannette Beranger, senior program manager for the Livestock Conservancy in the United States, it became clear that many sentimentalists dabble in homesteading with rare breeds. Beranger launched into an anecdote about an enthusiastic couple with whom she worked who could not bear to see any of their males killed or eaten. The couple had the capital to keep them, ultimately tending a separate paddock of forty lost boys. Without those resources, even the ethically consistent lacto-vegetarian will eat veal. In the case of Bronwen’s goats, the kids were taken off their mothers immediately and raised on the bottle, making it possible both to collect the milk and to domesticate the kids. Bronwen’s mother recalled, “We pasteurized the milk because we hadn’t done all the tests. . . . There was some kind of goat virus, I think.” The cautionary tales of tuberculosis, brucellosis, Johne’s disease, and caprine arthritis encephalitis were enough to convince her it was the ECOLOGIES

•

7

right move. The family purchased a two-gallon, red-and-silver kitchen-countertop pasteurizer. To prevent any potential goat viruses from being transmitted, the baby goats drank pasteurized milk from their bottles too. The family drank the pasteurized goat’s milk with their meals, and Bronwen’s parents tolerated a few of her failed attempts at making goat’s milk fudge, but at almost two gallons per day, the goats were pumping out more milk than the family could ever use. The deep-freezer in the garage began to fi ll up with plastic containers full of goat’s milk. Bronwen decided that it was time to make cheese. She ordered a kit from the goat-supply catalog, and a collection of small silver foil packets, a vial of beige-colored liquid, and what looked like a set of plastic drinking cups with holes in them arrived in the mail. Bronwen had always enjoyed cooking, and the enclosed recipe was straightforward enough: warm the pasteurized milk, add the powdered bacteria and a few drops of the liquid enzyme, let the mixture sit overnight, and then spoon the resulting substance into the molds and let it drain. She followed the instructions with great care, scrupulously dipping all of the equipment in boiling water before using it for fear that bacteria other than those in the powder might make their way into the mix. Visions of giving the entire family explosive diarrhea were pushed to the back of her mind. The following evening, she presented her finished fresh goat cheese to the rest of the family. It was an off-white substance with a gelatinous texture and a sour flavor. Her mother spread a thin layer on a cracker and ate it quickly. Her father, likely having similar thoughts about imminent gastrointestinal consequences, suggested that he would “try some later.” Her younger brother and sister snickered and hid, too squeamish to go near it. While nobody was struck down, it was not a distinguished or delicious food. Hardly having begun, Bronwen’s first foray into cheesemaking was over. Soon after, her mother found an animal rescue shelter that accepted donations of pasteurized goat milk, and the problem of freezer space was neatly solved. In an environment ripe with potential for integrated goat husbandry, Bronwen’s family had exactly replicated the practices of industrial dairying. The difference was that at Meadowbrook Dairy, the cows had not even had the option of grazing diverse pastures. Bronwen’s teenage goat rearing demonstrated the obsession with absolute control that characterizes intensive agriculture: the brush was cleared by hand rather than by the grazing goats, and cheesemaking was driven by the contents of a packet. If there were no commodity by-products in the goats’ ration, it was because the goats were in 8

•

ECOLOGIES

a luxurious confinement. But the aim, encouraged by 4-H advisors and the wisdom of the manuals, was to remove the animals and their milk from the purportedly dangerous natural world and then precisely to control the inputs that were reintroduced to the system. There was no other way to imagine farming and absolutely no other way to imagine cheesemaking.

M A K I N G AG R I C U LT U R E WO R K

In this context, it is perhaps unsurprising that Bronwen never contemplated life as a dairy farmer. Instead, she followed a more recent family tradition and embarked on a path toward a career as a physician. However, just as she was about to head off to medical school, nagging doubts about the decision prompted a desire for some time for reflection. Two years as a Peace Corps health volunteer in Senegal provided that time. Living and working in a village of Fulani herders in the north of the country reawakened her curiosity about dairying: the rangy Senegalese cows were nothing like her uncle’s buxom Holsteins, and that in itself was intriguing. What were the differences in the relationships between the people, the animals, and the land? On her return from the Peace Corps, Bronwen searched for a path that would provide her with an opportunity to address these questions. She made cheese on a small dairy farm in New Jersey and then completed a master’s degree in anthropology at Oxford, which allowed for a more formal study. At Oxford, Bronwen’s thesis on the relationship between Protected Designation of Origin legislation and tradition gave her the chance to meet and interview the key players in the British artisan dairy industry. This is also where Francis enters the story: Bronwen’s thesis was the topic of the paper that she was presenting at the Oxford Symposium on Food & Cookery on the occasion that we first met. Ours is, for good or ill, a relationship founded on European Union food law. The academic study of cheese and culture was one thing—as we have seen with the relationship between Salers and Cantal in the Auvergne, appellation regulations are a manifestation of the politics and commercial realities of a region rather than a scholarly attempt at defining authenticity—but with a couple of months left on her student visa, Bronwen had the opportunity to start work at Neal’s Yard Dairy, a London-based retailer, wholesaler, and exporter of the cheeses of Britain and Ireland. Within a year, she had become their cheese buyer, and she suddenly had the opportunity to collaborate with ECOLOGIES

•

9

the best cheesemakers in the United Kingdom, to taste thousands of cheeses—indeed, every batch of production—and select the best for the company’s customers, and to understand the inner struggles and frustrations of artisan cheesemakers. As a job, it was part stock control, part technical guide, and part therapist. While Bronwen worked buying cheese in London and visiting cheesemakers throughout the United Kingdom, our dinner table conversation became driven by her experiences. It was a tense time. Never quite fights, but always on the verge of a rumbling marital discord, these conversations highlighted the differences in our backgrounds. Unlike Bronwen, Francis has no family heritage in dairying. A city boy who liked to eat, he was so disheartened by the food on offer at his college that he set about teaching himself to cook. What started as a hobby soon became all-consuming, and by the time he graduated he had determined that he wanted to write about food. His university experience shaped him in another way too: thanks to the mysteries of the room ballot, he spent his second year living next door to Sir John Plumb, the octogenarian former Master of the college. Never married, Sir John had led one of the great twentieth-century lives. He spent the war codebreaking at Bletchley Park and then, as an academic historian, managed successfully to combine scholarly rigor with the literary flair to sell books. The mentor to a generation of historians of the eighteenth century, Sir John’s commercial success had allowed him to indulge his love of high living. Along the way, he had accumulated an impressive wine collection, and as he approached his ninetieth birthday—and in the context of a tense relationship with the student body of the college—one of his greatest delights was “serving nineteen year olds wines that they will never again be able to taste.” For Francis, at that stage very much the ingénue, the opportunity to taste these wines was a formative experience. It was not so much the wines themselves as an introduction to a world that he had never contemplated might exist. Wine had always been just another alcoholic beverage to him, something that quenched a thirst and provided social lubrication. But these wines were different. Yes, the flavors were haunting, but so too were the questions that they inspired. How were the wines different? Why were they different? For a lifelong Londoner with precious little experience of growing anything, these were the first faltering steps in trying to explain differences in flavors in terms of the practice of agriculture. When, after spending five years as a line cook and eighteen months as a fishmonger, Francis started to write about food, it was to these questions that 10

•

ECOLOGIES

he kept returning. Wine was the lens through which to explore food and culture. And, to the detriment of domestic harmony, it provided a framework to ask uncomfortable questions about the cheese industry. Questions that seemed absurd when applied to cheese appeared self-evident in the context of wine, where an entire industry was already grappling with the consequences of the relationship between farming practices and flavor. It rapidly became our shared passion. When we were both given the opportunity to work a harvest with friends who produced wine in Burgundy, France, we leapt at the chance. The vineyards of Burgundy are a UNESCO World Heritage Site. Or rather, the system of climats, the precisely defined small parcels of vineyards on the slopes just south of the city of Dijon, is a World Heritage Site. These climats are the result of the interplay of natural conditions and over a millennium of interaction with human civilization: a combination of the viticultural decisions of medieval monks and the cultural consequences of the lack of primogeniture under the Napoleonic code has created a heavily codified system in which producers might own only tiny parcels in any given vineyard. The home of Pinot Noir and Chardonnay, this is the emotional center of the world for many wine lovers and natural territory for the obsessive: it is a region where even a small producer might make twenty different wines each year and where vineyard and producer share equal billing on the label. The thriving wine trade in Burgundy was in stark contrast to the struggling Anglo-Saxon dairy industry. The wines themselves were gorgeous, but even more striking was that this was a region where farmers were basking in commercial success: in Burgundy, small-scale agriculture offers material rewards, and farmers with tiny holdings become international celebrities. It was not always this way. Old-timers still watch their backs nervously, unsure how long these good times will last. They remember when, even as late as the 1970s, a grand domaine did not provide enough income to support a family. But now that the world has discovered a taste for their wines, the best Burgundian producers are plowing the money back into their work in the vineyard. The market for the wines of Burgundy explicitly values the work of the farmer-producer: wines made from bought-in grapes have less value, even when they are from the same appellations and vinified by the same winemakers. Burgundians recognize that value is created in the vineyard, and this has given them a route map for how aspirational producers might improve their fortunes. Embracing sustainable viticulture, eschewing systemic fungicides and synthetic fertilizers, and paying new attention to soil microbiology and ECOLOGIES

•

11

genetic diversity of the vines in the vineyard make for wines that are snapped up by a prowling international band of merchants and importers who are all on the lookout for the next rising star. Dropping yields is standard practice, and winemaking itself has become an exercise in sensitively shepherding the ferments rather than aggressively forcing them along. Good practice makes good wines and is rewarded in the marketplace because it can be tasted by discerning and passionate consumers. In a region on the very northerly margins of successful red wine production, every small accumulation of incremental advantage in the vineyard is vital, and so a great domaine shares its success among its farmworkers. “Good” wine tastes good. Giddy with the experience of the harvest, we could not but wonder: Why can’t cheese be like this? It is another primary agricultural product that can be processed at the farm. Yet Anglo-Saxon dairy farmers lack the road map, the path to take to improve their cheese. It was the search for that path that took us to the mountains of the Auvergne and our encounter with Dr. MarieChristine Montel.

B I O D I V E R S I T Y M AT T E R S

The Anglo-Saxon world lacks anything like Montel’s INRA cheese laboratory at Aurillac. With its eight full-time members of staff, augmented by visiting students and academics, the laboratory is a fully equipped center of excellence in microbiology with a dedicated test creamery downstairs that is capable of making tiny batches of individual cheeses for experiments. The local cheese industry association has offices at the front of the building: the dialogue between practitioners and scientists is physically built into the space. In her office, Montel even has a tiny wooden model gerle that she uses as a wastepaper basket. In an effort to start a conversation between cheesemakers and scientists in the English-speaking world, Bronwen organized a conference, the Science of Artisan Cheese, with the help of Neal’s Yard Dairy. Hosted by Cheddar producer Jamie Montgomery, the conference was first held in 2012. It was a heady time. Organizing the presentations and liaising with the scientists reignited within Bronwen the spark of her teenage-scientist self, and she felt the desire to learn more about cheese in a rigorous environment. And so with Neal’s Yard Dairy generously granting her a two-month sabbatical, she made her way to the Dutton lab at Harvard University. 12

•

ECOLOGIES

Dr. Rachel Dutton has a talent for addressing big questions. As a PhD candidate working in the laboratory of Dr. Jon Beckwith at Harvard Medical School, she was dissatisfied with their choice of bacterium. Beckwith had spent forty years addressing the fundamental problems of biology through the study of Escherichia coli. In a world of discrete specialization, Beckwith’s was an E. coli lab. But Dutton was curious to work with other bacteria, ones about which nothing was known. Within the hierarchies of the academic world, in theory Dutton should have stopped there: the humble PhD student is a vehicle for his or her supervisor’s work, striving at the coal face to advance knowledge one experiment at a time. Not Dutton. She set about gradually winning over other members of the lab with her conviction that an alternative bacterium would be a compelling direction of study. After many meetings, Beckwith decided to give her a chance, and then her results swiftly convinced him that her ideas had merit. By the time Dutton left to establish her own laboratory, the entire Beckwith lab had moved over to the study of Mycobacterium smegmatis as a model for tuberculosis, based on Dutton’s idea. Straight out of her doctoral program, Dutton was awarded a Bauer Fellowship at Harvard’s Center for Systems Biology. It was an ideal fit. The Bauer Fellowships are designed to provide young researchers operating in interdisciplinary fields the chance to establish their own small laboratories for five years. They are self-consciously broad in the range of scholars that they attract, and the center employs everyone from microbiologists to mathematicians to engineers, all of whom are devoted to developing new experimental and analytical methods for solving biological problems. In Dutton’s case, her lab was to study real-world interactions in complex microbial communities. Their model system? Cheese. It was a great insight. Cheese rinds are ideal for a microbiologist looking to work on microbial communities. Unlike the anaerobic microbes of the human digestive tract, the communities on cheese rinds are easy to culture in the lab. They are also reproducible, and samples are readily available; Dutton first met Bronwen at the Slow Food cheese festival in Brà, Italy, where she had gone to collect as many cheese samples as possible. Lastly, the communities are just complex enough that their interactions are interesting without being so overwhelming that developing serious experiments becomes impossible. Dutton hired smart too. Her first postdoc, Dr. Benjamin Wolfe, is a mycologist of relentless intellectual curiosity and married to one of Boston’s top chefs. Amid the funky smells coming from the thousand or so experimental ECOLOGIES

•

13

cheese communities that they created and stashed in cupboards, on the top of shelves, and in the wine-cooler “cave” that they constructed, Dutton, Wolfe, and the rest of the lab team steadily teased apart microbial interactions using miniature “cheeses” made from freeze-dried curd. At the same time, they developed a reputation within the wider food world as the point of reference for questions of microbiology; an early interaction with David Chang of the Momofuku restaurant group brought them to the attention of chefs. (Wolfe would go on to write a successful regular column on food microbes for Chang’s magazine, Lucky Peach.) They were the microbiology lab that you went to not because you were worried about pathogens and safety but because you wanted to learn about the good microbes. Soon, food industry figures like Chad Robertson of Tartine Bakery in San Francisco and food writer Harold McGee were taking note. Dutton had become, in the words of the New York Times, “For Gastronomists, a Go-To Microbiologist.”4 When Bronwen arrived at the lab in January 2014, she was astounded. The Center for Systems Biology takes interdisciplinary collaboration seriously: the facilities are open-plan, with shared space and equipment. Lunchtime meetings encourage the cross fertilization of ideas, and lab members are regularly recruited as subjects for each other’s studies. The team in the next bay was studying the trillions of organisms that make up the human gut microbiome and how they affect drug metabolism and nutrition. Using themselves as guinea pigs, they had just completed a study, published in Nature, showing how changes in diet cause an immediate and pronounced shift in the balance of the gut microbial community. Bronwen’s years in the cheese industry had conditioned her to look at cheese microbiology from a quality assurance standpoint and instilled in her a love of supposedly sterile surfaces and an obsession with “avoiding contamination.” In the Dutton lab, she learned that the world looks very different from a microbe’s point of view and that health depends more on balanced ecologies than on obsessive protection from contact with the enemy. Any organism, even cheesemaking workhorses such as Lactococcus lactis and Geotrichum candidum, has the capacity to cause serious or even fatal disease if it gains a foothold in the wrong place.5 But as Montel demonstrated in her study on the gerles, robust microbial communities are formidable tools. Guy Chambon’s Salers is a cheese made in a porous wooden bucket in a room with a muddy floor. Pathogens never get a look in. Healthy microbial communities may be invisible, but they are not subtle; they are powerful in ways that we are only beginning to fathom. They—and 14

•

ECOLOGIES

cheese, by extension—lie at the center of a revolution in our understanding of the world around us and what it means to be healthy. The human skin microbiome differs between individuals, and some skin communities, like some cheese rinds, are capable of repelling pathogens, while others invite them.6 There is a growing body of evidence supporting the hypothesis that everything from the anticancer effects of cruciferous vegetables to mood and anxiety levels are mediated by communities of gut microorganisms. And it seems that it is not the number of microbes present but rather the composition of their communities that is crucial to their effects and their resilience.7 Dutton and Wolfe’s lab mates demonstrated brilliantly that species representing only a tiny proportion of the overall community in a certain set of circumstances can be catapulted into starring roles when their environment changes. In their experiment, the team bravely traded in their normal diet for a five-day protein fest of meat, cheese, eggs, and pork cracklings. The microbial communities in their guts immediately became enriched with an entirely different set of species specialized in digesting all-you-can-eat barbecue (hold the sauce). These altered gut communities returned to baseline soon after the subjects of the study went back to a normal diet.8 Human evolution takes millions of years, but microbes respond to a change of environment in a matter of days. Thus, the flexibility of the microbiome allows humans to adapt rapidly to radically different and highly specialized environments. In his book Missing Microbes (2014), Dr. Martin Blaser makes the case that the diversity within the human microbiome allows this rapid specialization to take place. Without “contingency microbes”—the understudies waiting patiently in the wings for their call to take center stage—once-harmless environmental stimuli may prove catastrophic for the host. When it was first described in the late nineteenth century, the bacterium Helicobacter pylori could be found in the stomachs of all members of the population. As a result of more aggressive sanitation practices and the bacterium’s sensitivity to ubiquitous antibiotics, H. pylori is now present in the stomachs of less than 6 percent of Americans born after 1995.9 And evidence is accruing that while H. pylori can have detrimental health effects, it also affects the immune response, conferring protection against allergies and asthma. How many of our “modern plagues,” from asthma to obesity, are linked to faltering microbiomes sapped of their diversity? Mirroring the multiple scales of farming across which cheese is made, from the microbial to the mammalian, exactly the same message about the importance of diversity comes from Dr. Paul M. VanRaden, a cattle geneticist. As ECOLOGIES

•

15

he puts it, “preserving genetic variation is the key to adapting to different definitions of perfection.”10 Biodiversity is essential to future-proofing the cattle population. VanRaden is no fringe radical. He is the US Department of Agriculture research geneticist employed to study the economic value of genetic traits in cows. He spends his time perfecting the statistical models to help improve—by which he means boost the economic return from—the national and international dairy herd. The scientific community is also slowly appreciating the importance of biodiversity in arable farming systems. Studying corn farms across the Northern Great Plains of the United States, Dr. Jonathan Lundgren and Dr. Scott Fausti demonstrated that biodiversity “performs critical ecosystem functions that cannot be replaced indefinitely by technology such as pesticides and herbicides.” Lundgren and Fausti identified all of the insect species in the plant foliage for fift y-three different corn farms. They found that more biodiverse cornfields had fewer pests. Most intriguingly, it was not simply the number of species and the abundance of individual insects that correlated with pest abundance. Rather, it was the balance of species within these communities that seemed to be connected with lower pest populations.11 Just as with the microbial communities of the wooden gerles and the human gut, it was the strength and resilience of the ecological network that predicted the health of the system. What makes cheese unique is its capacity to link the biodiversity of these three different worlds: flora, fauna, and microbiota. Moreover, cheese has the capacity to do this in a form that the consumer can taste. The uniqueness of Guy Chambon’s Salers—the reason that it is a product worth hunting down and worth the extra cost—is in the unique experience that it offers. It relies on the diverse mountain flora, but to exploit those interesting—and otherwise entirely marginal—pastures, Chambon must use a rare breed of cow that has the genetic traits not just to survive but also to prosper on the mountainside. His processing of the milk follows suit: there is no pasteurization to destroy the native microbial communities, nor is there any inoculation of starter cultures other than contact with the biofi lms in the gerles. In the rounded depth of flavor of the cheese, we as consumers can experience the totality of his farming system. When it comes to biodiversity, cheese is the essential food, the tool that allows the consumer to trust, but verify.

16

•

ECOLOGIES

T WO

Real Cheese

The French word for the maturation of cheese, affinage, has been adopted wholesale within the Anglo-Saxon world simply because we lack an existing term of our own. “Cheese maturer” lacks the romantic ring of affineur as a job title, so the French terminology has largely won out. It conveys a sense of both the aging and the refining of cheese as it matures. But in some important ways, it does not properly embrace what is happening to the milk in the course of the entire cheesemaking process. Through making cheese and then aging it to maturity, we have the opportunity to discover the unique character of the milk itself. Cheese is not so much milk’s leap toward immortality as its passage to adulthood. In this respect, it is interesting to compare the language used for supervising the aging of wine. In French, the process of looking after wine from fermentation to bottle is not affinage but rather élevage. The literal translation would be “raising,” in the same sense as rearing an animal or bringing up a child. It is a useful idea: not only does it capture the relationship between milk and cheese but it also gives us a tool with which we can imagine the connection between raw materials and finished product. There is something appropriate in thinking about milk in terms of childrearing metaphors. Consider a nursery full of newborn babies. As hospitals are all too aware, it is extremely easy to get babies mixed up. Glance in on a nursery of neonates, and they all look like Winston Churchill; if they are swaddled, then it is impossible even to guess their sex. This, then, is the state of milk: it has much potential, but this potential is as yet unrealized, making milk appear interchangeable and substitutable. Even milk as distinctive as that from Guy Chambon’s Salers cows would taste “milky” if it were sold as a pint of 2 percent. Moreover, looking at newborns—or tasting fresh milk— 17

does not give us much of a clue as to their finished character. Yes, we can spot certain defects, but there is nothing to say that the child screaming the loudest is going to be the most garrulous adult. Even birth weight does not significantly correlate with eventual adult size. Then there is the impact of parenting. Take any of those little bundles of genetic potential and subject it to a dehumanizing upbringing, and you will see the consequences in the person’s character as an adult. The greater the structured brutality, the more consistent the output. In situations where institutions like the military require cadets to subsume their individuality to an unquestioning collective identity, great emphasis is placed on dismantling the sense of self. So too for cheese. The greatest raw milk, with an ideal balance of microbes and perfect physical properties, can easily lose all sense of its own character through insensitive cheesemaking; pasteurization is not even required to wipe the microbial slate clean. At the other extreme, if a child is raised in an environment entirely devoid of structure, the resulting adult will be a feral mess, having developed little or no language or other prized facets of our humanity. Likewise, leaving milk entirely to its own devices is not the route to great or distinctive cheese: in that direction lies moldy yogurt. The challenge is to provide the child—or cheese, as the case may be—with just enough structure to nurture the development of a full-fledged individual.

K N OW I N G YO U R M I L K

So how do you know the potential of your milk? What, for that matter, is good milk? When dealing with fresh liquid milk, these are difficult questions. Across the course of the nineteenth and twentieth centuries, scientists and government regulators attempted to define milk quality, to make milk knowable beyond its milky flavor. Milk adulteration was a significant concern, and in the nineteenth-century market for liquid milk on both sides of the Atlantic, something as elementary as diluting milk with water was a common trick used by milk traders to stretch profits. The problem for authorities was that it was impossible to tell the difference between water that had been added for nefarious reasons and water that was simply a constituent of what had come from the cow. With the variation in milk composition across the cycle of a season, any definition of “natural” would be arbitrary at best. A procession of techniques was employed, from monitoring the specific gravity of the liquid to 18

•

REAL CHEESE

studying its chemistry, its boiling point, and its capacity to refract light. In each case, both a laboratory and credentialed scientific expertise were required. If unfortunate consumers do not have their own fully functioning laboratories at home, whom should they trust? It is not surprising that external— often government-sanctioned—marks of quality have entered the market. Organic certification and the geographical indication of protected food names bring yet another layer of extra knowledge that is required of beleaguered consumers. With no hope of verifying any of these claims themselves, consumers must trust the efficiency of the certifying body’s audit and the rigor of the specification. Yet even as nineteenth-century scientists and dairy technicians were squaring off over the best tests to verify the naturalness and authenticity of liquid milk, consumers were keenly aware of the variable quality of the cheese put before them. As early as the seventeenth century, the insatiable demand for butter in London was prompting dairy farmers in the surrounding counties to cut corners. Cream would be skimmed off to make highly profitable butter, but this left the problem of the remaining skimmed milk. It was made into cheese so consistently bad that it inspired its own doggerel. The cheese: Mocks the weak effort of the bending blade, Or in the hog-trough rests in perfect spite, Too big to swallow, and too hard to bite.1

At least it was cheap. Unpalatable as it was, the cheese from the counties of Essex and Suffolk did find a market as provision for ships on long voyages, where its capacity to resist heat was useful. It was still universally detested. Even now, centuries later, there is very little cheese production in the counties near London. Cheese was—and still is—something that defined particular places through the experience that it gave the consumer. In hindsight, thinking of her childhood, Bronwen cannot help but be amused. Natasha and Ginger, her glossy and healthy pet goats, lived in a dry environment and were milked carefully by hand into spotless utensils; they were in a position to produce exceptionally low-risk milk. But she never had the chance to discover what the milk was like, to taste the character of the land on which she grew up. Without a clear understanding of the risks, and terrified of the miasma that might have been lurking within the goat shed, when it came to making cheese she had opted to wipe the microbial slate clean and rely on a freeze-dried packet of purified bacterial strains. REAL CHEESE

•

19

More than twenty years later, this destroy-and-replace approach to milk microbes is still the default mode for most cheesemaking. Safety through sterility is an approach that is so comfortingly obvious that we have taken it for granted for more than a century, both in our food production and in our daily lives. The common term for microbes in everyday, nonspecialist language is “germs.” In many ways, this terminology is sensible, as it reflects their status as organisms capable of growing and developing, like the seeds of a plant. But the difference between “germ” and “pathogen” in the collective imagination is ill defined. Language is powerful in shaping our ideas and prejudices. And we have no lay word for microorganisms that isn’t wrapped up with fear, fi lth, and contagion. But is aiming to eliminate all microbes actually the most effective approach to fighting pathogens and staying healthy? “Farms and dairies are not hospitals,” according to Dr. Montel. And indeed, there is more to the microbiology of milk and cheese than marauding pathogens. Raw milk is a tremendous reservoir of microbial diversity, with many of its residents yet to be identified and understood. Hundreds of species of bacteria, yeasts, and molds have been found in raw milk, and each of those species evolves within its farm ecosystem, adapting itself to fi ll a specific ecological niche. These microbial communities represent complex and ever-evolving systems that combine the microbial residents of the fields, the bedding and the feed, the animals, the equipment, and even the milkers themselves. However, if you just drink liquid milk, you are never aware that they exist. It is the interaction of the two scales of farming—the macro level of the animals and plants and the microbial level within the milk—that makes (or rather can make) cheese that tastes of the farming system within which it is produced. There is no way that Guy Chambon, even with all the technology and investment in the world, could make his cheese from a large herd of animals fed on industrial by-products. Reflecting on her own homesteading experience, Bronwen came to a realization. As she tasted cheese every day at work, it became clear to her that whereas milk is difficult to “know,” since fault-free milk all tastes more or less milky and sweet, cheese is dramatically more articulate. It is fully formed and speaks to you. In cheese, the aspects of milk that are most challenging to monitor, from the level of protein to the balance of microbial inhabitants, are made manifest. You are encountering an adult rather than a mewling, puking infant. Furthermore, this knowledge is accessible to anyone who tastes the cheese. Cheese is democratic. There is no requirement for superhuman tast20

•

REAL CHEESE

ing ability; anybody with a degree of curiosity and a modicum of experience can easily learn to make sensible judgments about the finished product. Cheese is milk made knowable. T H E R E WA R D S O F DA I RY I N G E X E M P L I F I E D

If cheese makes milk knowable, how do we then know if the cheese itself is any good? Faced with constant invocations of the past and tradition, we were both intrigued by the history of British cheese and cheesemaking. This was the basis for Bronwen’s thesis, but she had not had an opportunity to genuinely investigate historical practice. And so the great exploration began. One of the first historical figures that we encountered is now Bronwen’s greatest cheese crush, Josiah Twamley. Perhaps it is not surprising that Bronwen finds Twamley such a resonant figure: he too was a cheese buyer. Whereas Bronwen visits cheesemakers and selects cheese in the twenty-first century, Josiah Twamley was a London cheese factor doing exactly the same thing in the late eighteenth and early nineteenth centuries. Like Bronwen, he was also determined to improve his industry. In two books, Dairying Exemplified, or The Business of CheeseMaking (1784; a revised second edition was published in 1787) and Essays on the Management of the Dairy (1816), he shared his experiences selecting cheese across England and put forward a trenchant vision of best practice. He does not mince his words. Twamley’s discussion of cheesemaking in his era has been the subject of study by historians interested in the competing expertise of the male London cheese factor and Enlightenment improver and the dairymaids whose products he bought. Read by a fellow cheese factor two centuries later, his concerns are familiar. (Twamley even has an obsession with dairy hygiene, albeit from a premicrobial base of knowledge. He describes dirty dairies as bastions of “sluttish nastiness,” a term that has become Bronwen’s default description of poor cheese-test results.2) More to the point, Twamley has a precisely articulated philosophy of the power of uniqueness in cheese and its capacity to give both cheesemaker and factor a sustainable livelihood: Such a dealer [cheese factor] is very certain, that in a large connexion of trade, he will find some very good judges, who know how to prefer excellence in quality, and are well acquainted with the perfections required in the article, and perhaps from their situation in life, are enabled to get a much higher, REAL CHEESE

•

21

than a common Market-price, for a superior Article: Such a Person will have such goods, in what place soever they can be met with, and knows also that in order to procure them he must give a superior Price.3

In the late eighteenth century, Twamley understood the essential covenant that still drives ambitious producers today: make a better product, a unique product, and the market will reward you with more money. For the modern food movement, the loose collection of activists and advocates who define themselves in opposition to the industrial food system, this invocation of the power of big-spending, high-status consumers to reward the exceptional artisan smacks uncomfortably of elitism. Self-consciously complicated idiosyncrasies of production, as those involved in the making of Salers Tradition, do not help either. “Not all cheese,” they might say, “can be like Salers, either in terms of its methods of production or its price.” That is unfortunate. One of the achievements of the twentieth century was to give us the illusion that we have successfully defeated nature when it comes to limitations on food production. Food—or at least the handful of commodities to which modern agriculture devotes its attention—is unprecedentedly cheap. We can see that quite clearly in the case of cheese. In the United Kingdom in 1865, the wholesale price in the London market of the top Cheddars reached 112 shillings per hundredweight. Measuring this worth by looking at the income value, the percentage of relative average income that would be used to buy the cheese, it becomes some £112 per kilogram (roughly $75 per pound) in 2015 prices.4 At wholesale, that represents roughly ten times the modern price. We live now with the consequences of this apparent cheapness as we grapple with the costs of the unpriced externalities of the intensification of agriculture, from environmental damage to the epidemic of chronic diet-related diseases. As we discuss in chapter 5, on feeding ruminants, a low sticker price does not mean that food is genuinely cheap. Moreover, not all cheese should be like Salers Tradition. Salers is a highly specific set of solutions to the problems posed by farming and cheesemaking in the mountains of the Auvergne; other cheeses find their own paths without resorting to the same techniques. But we are certain that cheese itself should be more expensive. Or rather, cheese should demand more resources of us as a society, as should all other forms of animal protein. How those resources are distributed throughout our society is a separate political decision, one that in democracies we make at the ballot box. 22

•

REAL CHEESE

Those factors that allow for the production of “cheap” cheese—the standardization, pasteurization, and blending of milk from potentially many thousands of individual farms—are also those that pose the most danger to rural communities. By removing the identity of the farmer as producer and replacing it with a carefully constructed corporate brand, they destroy the capacity of the cheese to have a social life, that is, its ability to reach across the culturally distinct and diverse chasms between producer and consumer. This matters now more than ever. The tumultuous politics of 2016, with the Brexit referendum in the United Kingdom and the presidential election in the United States, underline the divide between urban and rural on both sides of the Atlantic: we now have two distinct Great Britains and two distinct Americas that are struggling to communicate and understand each other. This divide makes for awkward pauses and tense moments around farmhouse kitchen tables when we visit from London and conversation turns to politics, but cheese has the capacity to bridge this abyss. Cheesemaking is a brilliant technology to allow remote dairy farmers to access distant markets. It takes something inherently heavy and intensely fragile, in the form of milk, and transforms it into an artifact of remarkable robustness that is only 10 percent of the original weight. Cheese is designed for cultural exchange, and it is no coincidence that those cheeses with the most ancient names, like Parmesan or Gruyère, are hard cheeses that are designed to travel. Lorraine Lewandrowski is a fourth-generation dairy farmer and farmer advocate in upstate New York. She is an attorney representing farmers and landowners in her rural community in Herkimer County, New York, but she also, together with her sister and brother, farms sixty Holsteins and a few pet Jerseys on some thousand acres of meadows and rough hill country. It is a place with a strong dairying heritage: this is the New York milkshed, the traditional source of the milk consumed in New York City, and it is the place that drove the first great nineteenth-century expansion of the American cheese industry. But now it is hurting, and Lewandrowski feels that there are not many in the food movement who feel their pain. There are many large milk processors in upstate New York, such as yogurt behemoths Chobani, Fage, and Dannon. But the milk that they use is essentially interchangeable, and it does not matter if it is sourced locally or from as far afield as Tennessee. For Lewandrowski, the tensions within the system came to a head in 2009, when the milk price crashed globally. In rural New York state, it was a time of desperation and increasingly angry rallies. “I went REAL CHEESE

•

23

to one rally in the summer of 2009 where I feared for my life,” said Lewandrowski. “I decided that I would do what I could from my law office by making calls to people I found on the Internet who might be interested in rural Northeast farmers. Virtually every group that I spoke to said that they were really more interested in ‘local’ food, ‘good food,’ fruits, vegetables, the Hudson River Valley.” The disconnect between the pain of the rural dairy farmer and the urban food movement that was increasingly skeptical about milk itself was palpable. If this was the alliance on which farming was to depend, then the future looked bleak. “Other farm women started phoning NYC cheesemongers to at least tell them what was going on,” Lewandrowski recalled. “It seemed so desperate, but to our surprise, the cheesemongers told us to a person that they were doing all that they could to inform the public of the desperate situation. They told us that artisan cheesemakers were trying to speak on behalf of those of us commodity farmers, their neighbors in the Northeast.” Cheesemongers were the natural allies for the farmers, their voices in communicating with an urban consumer base. During the Occupy movement in 2011, Lewandrowski presented a show on the Heritage Radio Network with New York City cheesemongers Tia Keenan and Anne Saxelby that they called “Occupy Dairy.” It allowed for a wide-ranging conversation on air about the state of the average Northeast dairy farm, from the dangers posed by the consolidation of buyers and the associated antitrust concerns to milk price volatility: “The global-scale milk processors would never encourage frank conversation.” When, in 2013, a group of dairy farmers decided to submit proposals to speak at an urban conference called Just Food, it was again the cheesemakers who backed them up, supplying coolers of cheeses. “Our neighbors [were] saying,  . . . ‘Give ’em hell! Tell them about the land,’ ” Lewandrowski recounted. “Sharing cheese with the people who attended our presentation, we were able to say, ‘This is what the landscapes taste like, this is our local dairy in central New York.’ ” This is why cheeses made with integrity, those that retain their sense of place, are not elitist but rather subversive. Even when they are bought from a large supermarket, these cheeses demand of the retailer such care and attention that they must be cut carefully to order, a job that requires a face-to-face interaction with a dedicated cheesemonger. On multiple levels, cheese has the capacity to subvert industrial food systems and connect the urban and the rural. 24

•

REAL CHEESE

REAL CHEESE

There is great precedent for the cheesemonger as activist and advocate. In the mid-twentieth century, the dark days for the British cheese industry that saw the retreat from farmhouse production and the rise of the supermarkets as the dominant force within the industry, there was one retailer who stood out. Major Patrick Rance had fought the Nazis, but in peacetime he found himself engaged in another existential struggle, this one based out of his modest cheese shop in Streatley, not far from London. With his monocle and aristocratic connections, he cut an eccentric figure, but he had a clear vision of his aspirations for cheese. In his Great British Cheese Book (1982), he outlined the problem: from 1948 to 1974, there had been a drop in the number of farms making Cheddar in the southwest of England from sixty-one to thirty-three. (As we write this, in 2016, there are only five.) He also proposed a solution. Taking his inspiration from the Campaign for Real Ale (CAMRA) and the Campaign for Real Bread, he noted that “fortunate and few are those within reach of Real Cheese to go with it”5 and called for a similar movement for cheese. Rance’s fight was with factory production. For him, the enemy—the unreal cheese—was the mountain of vacuum-packed blocks that was steadily replacing the United Kingdom’s established territorial cheeses. We agree with his sentiment, but following his inspiration, this book pushes further. The risk of defining and essentializing the real, of claiming to separate the authentic and the fake, is that it becomes an exercise in arbitrary exclusion. CAMRA, as successful as it has been at celebrating and reinvigorating British cask ales, also stands accused of stifling the British craft brewing scene with its insistence that only cask-conditioned ales are “real.” That is not our intention; it would be absurd to claim a single style of cheese as uniquely real. Rather, for us, real cheese is a manifestation of wider biodiversity, a food that exploits all of the resources and raw materials of the farm, from the botanical to the microbial. It is an acknowledgment that dairy farming and cheesemaking are one and the same process, and of the moral hazard that comes from any intervention—whether it be aggressive use of fertilizers, pasteurization of milk, or insensitive use of microbial cultures—that obliterates the link between the cheese and the environment from which it is fashioned. At best, these interventions are simply a patch; at worst, they threaten to undermine the sustainability of the entire industry. An opportunity lies before us. Advances in our understanding of biology have given us the tools to begin to understand and work with natural REAL CHEESE

•

25

ecosystems at every level. Evidence is accruing of the social and environmental benefits associated with food systems that look beyond the production of faceless commodity outputs. Real cheese is subversive in its simplicity: it reunites farming and flavor. And in doing so, it rewards diversity and sustainability at every level. What follows is a journey through the cheesemaking process, beginning with the very first decisions that a farmer must make about what and how to farm and continuing through each of the key stages of cheesemaking itself. We start with a primer on the mechanics of cheesemaking, to provide some context for these decisions, and then follow the life of the cheese from its earliest inception. At every stage, we have deliberately chosen not to segregate the book like a textbook: the reader will encounter issues at the same time that they would trouble the farmhouse cheesemaker rather than according to a pedagogical schema. We know that cheesemaking is inherently interdisciplinary: it is equal parts art, science, and intuition. As we track the path from field to consumer, we want to convey the pitfalls and complexities facing the producer. The same is true of the geographical scope of the book. We discuss the issues involved in the context of a cheese or region where they are of particular concern. There is no absolute Real Cheese, only real cheeses made in the context of specific places.

26

•

REAL CHEESE

THREE

The Third Rail

There is a term in politics that is reserved for the issues that are too difficult and emotionally charged for mainstream politicians to touch: they are the “third rail” of a nation’s politics. One of very few public transportation metaphors in American political life, it refers to the third rail of a railway track, the one that conducts the electricity to power the train: touch it and you will get zapped. In the United States, a third-rail topic might be Social Security reform; in much of the rest of the world, the administration of public health care. The issues involved are so complex and provide so little reward within the present electoral cycle that they are avoided by politicians. The same phenomenon is at work within the cheese industry, where gatekeepers are deeply reluctant to communicate to a popular audience anything more than the most basic level of understanding of how milk and cheese work. There is a collective sigh, the gentle suggestion that the ideas might be “too technical,” and earnest explanations that such details are “very interesting, but just not right for our audience.” The technical aspects of cheesemaking are the third rail of the cheese industry. But fear of the third rail creates problems of its own. As a teenager, Bronwen did not have access to information that would have allowed her to imagine a more holistic approach to cheesemaking. Although she was fascinated by biology and had taken a summer internship in molecular biology research, she had not been exposed to anything that might have triggered her curiosity about the inherent potential of the milk she spent hours stripping from her goats’ udders, lugging back to the kitchen, and pasteurizing. For all she knew or cared, even as she read papers on signaling pathways and learned to pour agarose gels, the bacterial cultures and enzymes that she had used to make cheese at home might as well have been 27

mystery powder and magic potion. Science was complex and fascinating. Cheesemaking was a cheap kit that made inedible white stuff. This lack of knowledge is a common problem when it comes to cheese, a food that is shrouded in mystery. Despite the fact that the average American eats over thirty-three pounds of cheese per year, few understand the process by which liquid milk is transformed into solid mass. Revisiting the intellectual world of Bronwen as a teenage homesteader, the chapter on cheesemaking in Belanger’s Raising Milk Goats the Modern Way makes for fascinating reading. It is entirely devoid of any discussion of cheese as a fermented food. No suggestions are given on how best to monitor the development of acidity in the curd—not even trial and error through tasting is advocated. Where microbes feature, they are foes to be avoided through fastidious hygiene practices and the sterilization of equipment. If a book that has sold hundreds of thousands of copies and is considered the bible for domestic goat farming can forget that cheese is a fermented food, what help is there for the consumer? Lack of the most basic understanding of the cheesemaking process among consumers is the single most perilous threat to the future of cheese of integrity. All other potential challenges are rooted in this original problem, from the prospect of overzealous government regulation to the impact of ever more intensive farming. Even the seemingly straightforward categories that we use to divide and organize cheese provide room for shortcuts and fakery. Take the example of Camembert. The defining feature that most of us associate with Camembert is the snowy white coat of mold that forms the rind: inside, the paste is seductively smooth. A cheese professional might describe the cheese as a soft-ripened bloomy rind, but we would all know one if we saw it. Or would we? With our focus on that exterior coat, we miss that there are actually two completely different types of Camembert sold as one and the same. When the original method—an approach still used to make the archetypal Camembert de Normandie and many smaller-production cheeses of this style—is used, the curd is initially chalky and acidic, and it is the slow growth of the mold on the rind that breaks down the curd into its characteristic silky texture. A young Camembert produced in this way will have a tart, flaky core until the cheese slowly ripens through to the center. The alternative style of Camembert is favored by producers concerned with minimizing the time between vat and supermarket shelf. With this method, the curd is never allowed to acidify fully, so it is elastic and even textured from day one, with no need for time to ripen. Making this “stabi28

•

THE THIRD RAIL

lized” curd is a shortcut designed by technologists. The mold for the rind just needs to be sprayed on, and as soon as it has begun to grow, the cheese is ready for sale. It will never develop the same depth of flavor as a cheese made using the classic method, but the throughput at the factory will be much greater and the costs lower. Unfortunate consumers, looking at the smooth texture and white, fluff y rind, may be disappointed by the insipid flavor of the stabilized cheese, but they will be none the wiser. The label affi xed to the cheese is not there to help. Within the AngloSaxon cheese world, the term “artisan” can be and is used to describe just about any cheese short of the Kraft Single. In France, a legal classification is used to distinguish between cheeses that are fermier (made by the owner of the animals that produce the milk), artisanal (made on a modest scale using bought-in milk), coopérative (made by dairy farmers who collectively own a facility that processes their milk), or industriel (made from the pooled milk of many producers). At the consumer level, very few within the Englishspeaking cheese industry dwell on these critical distinctions and their ramifications. Just as the term “artisan” is a moveable feast, so too words like “farmstead,” “traditional,” and “handmade” are bandied about as mood music, detached from real meaning. At every turn, the customer is fed marketing banter rather than given the tools to put cheeses in perspective. In the unscrutinized territory between popular conversation and technical realities, shortcuts are taken and customers are duped. The point of this book is to demonstrate that all is not equal when it comes to cheese, that the decisions made—both virtuous and cynical—during the process of farming and cheesemaking are writ large on the cheese that results: not on the label or in the flowery rhetoric used to market it, but in the actual substance and flavor of the cheese. With the right knowledge and a bit of experience, anyone can learn to taste and recognize the difference. Without it, anyone can be taken for a fool. This is why technical understanding is the third rail of the cheese industry: it is where the power is.

CHEESEMAKING 101

It is perhaps easiest to imagine cheesemaking as farming on two separate scales. On the one hand, there is macro-scale farming, the keeping of ruminants for milk. But on the other hand, turning milk into cheese is also THE THIRD RAIL

•

29

farming, just on a microscopic scale. Cultivating microbes is the very essence of almost all cheesemaking, which relies on the action of lactic acid bacteria to preserve perishable milk. These bacteria—whether native to raw milk or added as a starter culture—digest the lactose in the milk and produce lactic acid. Where they come from and how they are harnessed are decisions as important as the breed—or even species—of animal that the farmer is milking. Fermenting foods using lactic acid bacteria is an ancient practice used not only for cheese but also for many foods including dill pickles, sauerkraut, and raw-fermented sausages, such as salami. It had several big advantages in a preindustrial, prerefrigeration society. The first was preservation: using this method, a perishable source of nutrition—a bucket of milk, a cabbage, or a freshly slaughtered pig—could be set aside in times of plenty and returned to the table months later when food was scarce. How does fermentation work its magic on these delicate fresh foods? The answer is that lactic acid bacteria are not so much preservers as they are transformers. A cabbage is no more sterile than a glass of raw milk, and in some cases much less so. Studies on the microbial load of fresh vegetables show that they host a stunning variety of different bacterial genera, many of which—such as Pseudomonas and Yersinia—are associated with spoilage and even alarming diseases. (Pseudomonads are a group of “soft-rotting” bacteria associated with vegetables; the genus Yersinia includes the bacterium that causes plague.) Mixed in among this unsavory bacterial cohort are a few lactic acid bacteria, often vastly outnumbered by the undesirable members of the community. Leave a cabbage to its own devices for too long, or worse, rip or bruise it so that the leaves are damaged, and these latent spoilage bacteria will spring into action: soon, the cabbage will be exuding slimy, smelly, rotten vegetable juice. Making sauerkraut, however, creates a set of conditions that select for the growth of lactic acid bacteria: chopping the cabbage finely and mixing it with salt pulls moisture out of the crunchy leaves, and stuffing the mixture tightly into jars excludes oxygen, which spoilage organisms like Pseudomonas need to grow. Within a short time, the lactic acid bacteria respond to this custom-made environment. As they digest the nutrients in the cabbage slurry, they release lactic acid as a by-product, making the environment even more hostile to any competitors. Within a few days, the partial digestion of the cabbage by these beneficial bacteria has created a pickle that is extremely resistant to further microbial growth, and the undesirable bacteria, once so numerous, have been obliterated. 30

•

THE THIRD RAIL

The second advantage of lactic fermentation is that it releases intense flavors from bland raw materials, not just through the production of tart acidity but also through the action of enzymes released by the bacteria on the building blocks of the food itself. As it ferments initially and is subsequently stored, the food breaks down. Large molecules that have very mild or no taste are cut down into smaller ones that are piquant, sweet, or savory. Insipid vegetables are transformed from edible chore into delicious condiment. Cheesemaking involves a very similar fermentation to that of pickled vegetables, but with an added layer of sophistication and complexity. Cheesemaking combines fermentation with concentration. Liquid milk is transformed into a solid, and water is extracted. What is left at the end is essentially a fermented concentrate of the nutritious elements of milk: the fats and the proteins. This curd then goes on to break down (and get more interesting to eat) as it ages. Why not approach milk like cabbage and simply ferment it without draining? Go ahead: this is called yogurt, and it’s a wonderful food. But yogurt also has some drawbacks. For one thing, it’s bulky, both to store and to eat: yogurt takes up almost ten times as much space on the shelf (and in the stomach) as the equivalent nutritional package of cheese. Yogurt, unless it’s strained, also has the same amount of moisture as the original milk: 80 to 90 percent of its volume is water. Because microbes are more able to thrive in a moist environment, yogurt is also intrinsically less preserved than cheese is. It may keep for several weeks, but not much longer, particularly without refrigeration. When it comes to preserving milk for the long haul, removing moisture is a necessity. The removal of water from milk to make cheese—and the complex biochemistry of milk that makes it possible to do this in various ways—also gives rise to fabulous potential diversity. To achieve different types of pickled cabbage, flavorings are a necessity: distinctiveness comes from a dash of caraway, some grated apple, or even red chili flakes and shrimp paste in the case of kimchi. Cheesemakers conjure an even greater range of tastes and textures from a single set of ingredients. Depending on the process used—and there is a wide variety of potential combinations of time, temperature, cutting, stirring, and pressing that one can inflict on a vat of curd—the same milk can be transformed into rubbery-sweet Appenzeller, crumbly Caerphilly, or a silky-smooth, spoonable Mont d’Or. When you stand before a shelf full of dairy products, it is easy to forget that milk evolved to feed baby mammals and that cheesemaking is a THE THIRD RAIL

•

31

remarkably clever manipulation of its attributes. But in order to understand cheesemaking in the vat, we need to take a step back and look at the characteristics of milk itself. While milk is 80 to 90 percent water, it is also nutritionally dense, containing all of the nutrients needed to support a fast-growing baby animal during the first months of its life. Milk contains lactose, or milk sugar, which dissolves readily in this aqueous solution, but milk is particularly notable for its ability to hold in suspension a large quantity and variety of other nutritional compounds—fats, proteins, and minerals—that would not normally be soluble in water. In order to do this, it packages them up in special ways. We all know that oil and water do not mix. Melt a stick of butter into a saucepan of skim milk, and the mixture does not become whole milk but rather a greasy slick of liquid fat atop some watery milk. Fat in fresh milk is enclosed in microscopic globules, stable droplets of fat surrounded by membranes that sit comfortably within the aqueous solution as an oil-in-water emulsion, a bit like a vinaigrette in reverse. There are over a trillion of these milk fat globules in a liter of whole cow’s milk, bringing the total percentage fat in cow’s milk to 3.5 to 5.5 percent, depending on the breed of the cow. Protein is another major milk nutrient, and here again nature relies on clever packaging. About 80 percent of the protein in fresh milk takes the form of the most plentiful milk protein, called casein. If the total amount of casein in milk were simply added as individual protein chains, the milk would be so viscous it would be impossible to get it out of the udder. Instead, many thousands of these casein proteins are packed together into microscopic spheres called micelles. While the size of casein micelles may vary, they are generally about one-hundredth the size of the average fat globule, and there are about a quadrillion (that’s 1015) of them in a liter of milk. Even though they are visible only under an electron microscope, the way they refract light is what gives milk its opalescent whiteness. There are several different varieties of casein, and the composition of the different casein proteins within the micelle depends on many factors, especially the breed of animal that produced the milk. However, a feature of all casein micelles is a layer of water-loving protein filaments, or “hairs,” that protrude from their surfaces. This hairy coat attracts and gently binds to water molecules, holding a layer of water in between adjacent micelles and thus allowing them to freely slide past one another, keeping fresh milk in a liquid state. Countless advertising campaigns have made sure that we all know milk is full of calcium, an essential nutrient for the growth of healthy bones and 32

•

THE THIRD RAIL

teeth. Everyone remembers the Got Milk? campaign, and it is true that baby mammals need a lot of calcium. Unfortunately, calcium is not particularly soluble; as with caseins and milk fat, if you tried to add the required quantity of calcium directly to water, you would run into problems. In this case, it would form large crystals of insoluble calcium phosphate. Imagine a child shoveling sugar on their cornflakes, and you can picture what would happen to the calcium. Before long, it would just sit, undissolved, at the bottom of the bowl of milk. To avoid this, nature has evolved yet another clever delivery system: in addition to their scaffold of proteins, casein micelles are held together by a calcium phosphate “glue.” This allows an enormous amount of protein and minerals to be packaged up and delivered in liquid form to the baby. Not only is calcium phosphate great for building bones and teeth, it also has a pivotal role in determining the texture of different cheeses. For the first few weeks of a ruminant’s life, it digests food like a singlestomached mammal. When it drinks its meal, the milk bypasses its first three stomachs and goes straight to the fourth stomach, also called the “true stomach” or abomasum, where digestion begins. There, it meets an enzyme called chymosin, which is produced by the stomach lining. The first phase of milk digestion is the formation of a clot, or curd. This happens with the help of the chymosin and hydrochloric acid, also secreted by the stomach. Chymosin selectively clips off the water-loving hairy coat of the casein micelles. Without their watery shields between them, the micelles come into contact and stick together, forming a jelly-like curd. The curd itself is a three-dimensional net of tiny, sticky casein micelles, which trap the much larger fat globules and any bacteria present (such as the lactic acid bacteria) within it. This clot forms within just a few minutes, as any parent who has had to deal with the curdy burp-ups of freshly fed babies will know. The acidic environment of the stomach also has a pronounced effect on the curd and its casein micelle scaffold. First, the presence of acidity causes the casein network to contract and begin to squeeze out moisture. At the same time, the acid begins to dissolve the calcium phosphate “glue” that holds the micelles together, releasing the minerals locked within them. The calcium seeps out of the clotted curd, along with the water, sugars, and a few soluble (noncasein) proteins, which all make their way into the small intestine, where they are absorbed into the bloodstream. Meanwhile, the partially dehydrated curd, composed of the casein network and the fat globules trapped within it, remains in the stomach, where it is digested more slowly. THE THIRD RAIL

•

33

Cheesemaking takes this process out of the stomach and into the vat. We have already introduced the basic outline of the process: milk is clotted under the influence of chymosin—also known to cheesemakers as rennet—and often with the help of acid, which, in the case of cheesemaking, is produced by lactic acid bacteria rather than supplied by the stomach. A certain amount of water is drained away to make a solid mass of some description. Salt is added, and the fresh cheese may be eaten right away or left to age. Here, we see how suburban squeamishness prevented Bronwen’s teenaged experiments from exploiting the holistic resources of her little dairy farm. Although these days almost all cheese producers rely on the consistency of commercially produced rennet, the original source of this chymosin was the stomachs of the male offspring. If only sentimentality had not prevented her from enjoying—or even knowing about—the festive Easter roast, there was absolutely no need to send away for a special cheesemaking kit. With a few healthy animals, she had all of the resources she needed to make cheese.

R O U T E S TO D I V E R S I T Y

The key to determining the basic character—the structure and texture—of a cheese is the order in which the acidity develops and the moisture is removed. It is the minerals in milk—particularly the calcium phosphate locked into the micelles of fresh milk—that have the most pronounced effect on cheese texture. Cheeses with the calcium phosphate left in place are supple; cheeses from which the minerals have been leached away are brittle. At the first extreme, if the cheesemaker takes the moisture out of the curd before the acidity develops, the minerals stay locked within the curd, leaving it with an elastic framework and a smooth texture. At the other extreme, if acidity develops in the curd before the moisture drains away, as it does in a baby ruminant’s stomach, the minerals dissolve out of the tart curd and are carried away with the liquid (also known as the whey), leaving the curd “demineralized” and resulting in a brittle texture. At one end of this cheesemaking spectrum are the classic goat cheeses of the Loire Valley of France and their variants: small, tartly acidic cheeses with a brittle texture. These cheeses put acidification first and drainage second. Warm milk is poured into buckets along with some starter cultures (or whey from the last batch of cheese), which contain microbes to start the fermentation, and a tiny amount of rennet is added. The milk is left to acidify and set for anywhere 34

•

THE THIRD RAIL

FIGURE 1. Left, yogurt-like lactic curds being molded; right, the resulting brittle texture of lactic cheeses. Photos by Bronwen Percival.

from twelve to forty-eight hours, by which time the curd resembles—and tastes like—a big bucket full of set yogurt. This acidic curd is deceptively delicate, as the casein micelles have unfurled in the high-acid environment and let go of all of their mineral glue. The moist curd must be scooped and placed very gently into molds to avoid shattering it (see figure 1). When the curd is transferred to the molds to drain, the whey flows out freely, carrying the minerals with it, while the curd settles into a brittle, flaky mass that firms up only with subsequent air-drying. Within the cheese industry, these are known as lactic cheeses. At the other end of the spectrum, Alpine cheeses like Comté, Gruyère, or even generic Swiss put drainage before acidification to achieve an elastic texture. They are made with extremely fresh milk in which no acidity has yet developed. The milk is heated, cultures and rennet are added, and we are off to the races. For those used to the sedate pace of lactic cheesemaking, the production of mountain cheeses can seem like a violent blur of activity. Moisture is extracted using every means available: the curd is cut into a slurry of tiny pieces and stirred constantly while being heated to a high temperature, which causes the curds to contract and drive out whey. Because the process is so quick, very little acidity develops before the curds have expelled their moisture. The mineral glue stays put within the micelles, even as the curd is stirred and cooked. The dry and rubbery grains of curd that result are scooped out of the vat, transferred to large molds, and then pressed while their acidity level is still very low, resulting in a firm cheese with a pliant texture (see figure 2). In soft mountain-style cheeses like Mont d’Or or Winnimere, the process is exactly the same as for their hard counterparts, but more moisture is left in THE THIRD RAIL

•

35

FIGURE 2. Left, small grains of sweet curd used to make Comté; right, the supple texture of the freshly molded cheese. Photos courtesy of Le Comité Interprofessionnel de Gestion du Comté.

figure 3. Left, glossy, low-acid soft cheese curds; right, the smooth, spoonable texture of Winnimere cheese. Photos by Bronwen Percival (left) and courtesy of the Cellars at Jasper Hill (right ).

the curds. Unlike the hard-cheese curd, which is pulverized and then heated to high temperatures, soft cheese curds are cut into larger pieces and are not heated. But like the hard mountain-cheese curds, these soft cheeses are transferred to their molds before acidity has had a chance to develop, so both moisture and calcium phosphate are retained, giving the finished cheese a glossy, sometimes even spoonable, texture (see figure 3). The cheeses we have looked at in the first three categories neatly separate the processes of drainage and acidification: one always comes before the other, giving the cheese either a very brittle texture or a very smooth one. The 36

•

THE THIRD RAIL

FIGURE 4. Left, Appleby’s Cheshire curds acidifying as they drain in the vat; right, the open, friable texture of the finished cheese. Photos by Bronwen Percival.

making of English-style cheeses, or a cheese like Salers, involves drainage and acidification at the same time. The ways that these two processes are coordinated can give a range of results, from brittle and flaky Cheshire to pliant yet crumbly Cheddar to succulent and pillowy Lancashire. When making English cheeses, milk is often allowed to begin to acidify before the rennet is added, and the process of cutting and stirring happens as the acidity continues to develop. The whey is drained away, and the curd particles that have settled stick together and are cut into blocks (see figure 4), which may be stacked, pressed, or further cut down into smaller pieces over the course of several hours or even left until the next day. Finally, the curds— by now both dry and fairly acidic—are ripped into small pieces, often with the help of a toothed machine called a curd mill. The curds are then mixed with salt and pressed into molds. Depending on the contours of the process and how drainage is aligned with the development of acidity produced by the lactic acid bacteria, more or less calcium phosphate may be retained, resulting in cheeses with a range of different textures and consistencies. A handy way to visualize the structural differences between families of cheese is to look at the moisture and acidity level at the moment that the curds go into the mold. Caught at that instant, the world of cheese can be divided into four quadrants, which illustrate some intriguing relationships between various styles of cheese (see figure 5). For instance, the texture of industrially produced, stabilized Camembert is very similar to that of Reblochon. Whether the cheese has a rind dominated by white mold or THE THIRD RAIL

•

37

Stilton

Valençay Époisses Saint-Marcellin

Lancashire Wensleydale Acidity

Cheshire Cheddar

Camembert Roquefort Stabilized Camembert

Comté Emmental

Ossau Iraty

Reblochon

Mont d’Or

Moisture FIGURE 5. Cheeses organized according to the moisture and acidity of their curds at the point of molding.

orange bacteria is superficial rather than structural. The diagram also reveals an important truth: blue cheese is not a category. The blue mold found in these cheeses, Penicillium roqueforti, makes for a highly distinctive set of flavors in the finished product, but there are diverse cheeses that happen to be blue. Stilton sits in the high-acid, low-moisture quadrant of the diagram alongside all of the other classic British territorial cheeses, whereas Gorgonzola, another classic blue, is the result of low-acid, high-moisture curds. Indeed, if we look back one hundred and fi ft y years, all of the British cheeses had highly prized variations featuring blue mold. In the English county of Wiltshire in the nineteenth century, it was Vinny (blue) Cheddar that commanded the highest price.1

C H E E S E M A K I N G A S PA R E N T I N G

These are the origins of cheese texture, but that is only half the story. Where does the flavor of cheese come from? A few cheeses taste directly of the milk used to make them, particularly cheeses like fresh ricotta and young, low-moisture mozzarella. Likewise, the tart, yogurt-like acidity produced by the lactic acid bacteria during the 38

•

THE THIRD RAIL

cheesemaking process is a primary flavor in many acid-coagulated fresh cheeses. The vast majority of cheeses also have salt added to them, both as an aid to preservation and as a seasoning. For every one of the cheeses in figure 5, the curds at the moment of molding would taste of some combination of milk and acidity. And they would all taste pretty boring. Fresh, young cheeses are cakes of curd composed of varying amounts of milk protein, fat, and moisture. That curd is the raw material on which microbes and their enzymes go to work. The microbial enzymes break down the tasteless fats and proteins into small, volatile molecules, transforming mild milk into a vast range of compounds with vivid flavors, from fruity esters to savory amino acids to obnoxious sulfur compounds. This is an important point: the flavor of cheese is not the taste of the molds or bacteria growing on it. A bacterial colony plucked from a petri dish or freeze-dried blue mold spores sprinkled from a packet have practically no flavor. But allow those same microbes the proper conditions and time to grow in mild and milky curd, and they will digest it to release a riot of flavor, revealing the inherent sensory potential of the milk’s own building blocks. Whether—and how well—ripening microbes grow depends on the characteristics of the young cheese, which are determined by the cheesemaker’s choices during the making process outlined above. The curd’s salt and moisture content, its acidity level, and the accessibility of important nutrients will determine its ability to support the growth of various microbes and regulate the activity of enzymes released as the cheese ripens. These microbes and enzymes are exquisitely sensitive to small environmental differences, and even a tiny shift can send a cheese careening off its proper path. Each organism has its own set of optimums. Some may be lethargic at cold temperatures and vigorous when warm; others may respond to a tiny difference in moisture content. A minute boost to an organism’s competitive advantage can easily make the difference between it remaining an insignificant, invisible presence or growing to dominate a microbial community. The effects are pronounced. Ostensibly similar Cheddars can develop flavors that evoke a meaty roast dinner, a box of Band-Aids, or a decomposing cauliflower. A batch of goat cheese with a pristine rind can sit unblemished directly beside or below a batch marred by dusty blue spots or ugly puffs of dank grey mold. Blue cheeses can develop pink or brown patches within the curd or fail to turn blue at all. In these circumstances, we are primed to blame contamination: How did this offensive microbe get into the cheese? Was the THE THIRD RAIL

•

39

herdsman careless when he was milking? Was a window left open in the creamery, allowing black mold spores to float in? While rogue spoilage microbes can lead to quality problems, most of the time phenomena that look like microbes gone wild have more mundane physical explanations. A small imbalance in moisture or acidity level has much more power to wreak havoc on a cheese’s development than a microbe out of place. Likewise, the maturation conditions—such as the amount of air movement or humidity—further influence the activity of the microbes on the rind, pushing the development of the cheese in one direction or another, speeding it up or slowing it down, and leading to the evolution of the desired communities and flavors as the cheese matures.

T H E H A R D YA R D S

Making great cheese is not easy. It requires compelling raw materials and skills that take years to perfect. In this sense, cheesemaking is like many other activities that demand hard work in exchange for deferred gratification, such as learning to play a sport or a musical instrument or learning a language. The path to technical competence is long and harrowing, but because these pursuits are hard, the rewards are commensurately great. Deep down, we all want to believe that the ultimate victory will go to the person who tries the hardest and succeeds through integrity and determination without taking any shortcuts. They are the sports-movie clichés: we know in our hearts that a low-tech regimen of dragging laden sledges through arctic snowdrifts will trump a pharmacopeia of anabolic steroids administered by white-coated lab technicians. Or will it? Read through the winners list of any cheese competition or survey the counter at a local cheese shop, and—universally—the selection comprises some of each: the products of graft and integrity sit alongside the deceivers, whose appearances belie their cynical methods of production. These pretenders are like lazy students who shirk long hours of study but then cheat on their final exams and walk away with full credit. Their success through the use of shortcuts makes a mockery of those who don’t cut corners. Anyone who has ever put in the hard yards knows how that feels. Nowhere was this predicament clearer than when we were confronted by a table laden with prizewinning American Cheddar. Tasting the first cheese, 40

•

THE THIRD RAIL

we were amazed by its intensely sweet caramel flavors; it was a cheese with an appeal that transcended conscious thought. In his book on Cheddar, Gordon Edgar describes many American Cheddars as having an “unexpected, almost sugary-sweet crunch.”2 But by the time we had sampled our way through a dozen of the cheeses, we were thoroughly fatigued, worn out by the cloying sweetness. All of them tasted the same. Those Cheddars owed their sweet character to the addition of a sachet of Lactobacillus helveticus, a bacterium that has recently become the culture of choice for those seeking to make cheeses with broad mass-market appeal. With a powerfully sweet flavor profi le, Lb. helveticus has become ubiquitous among the better block Cheddars and is creeping its way into larger-production clothbound cheeses as well. Making Cheddar with Lb. helveticus is the equivalent of an athlete using performance-enhancing drugs: it is a shortcut to big, sweet, easy flavors—milk doping, if you will. We are deeply intrigued by the milk from small American dairy farms, many of which utilize highly extensive, low-input farming systems. But we never get to experience the unique character of the milk when it is hidden behind such a dominant set of added flavors. Whether we are sampling a cheese made with so much starter culture that all you can taste is yogurt-like acidity or one whose overt sweetness comes entirely from the cultures that are added, we find ourselves in a place where the character of raw materials and the hard work of farming well and making cheese with delicacy and care don’t matter. But when cheese is made with milk that has been farmed with intention, and when that milk’s character is magnified rather than obscured by the cheesemaking process, the final product’s unique flavors take on a deeper meaning. They connect us to centuries of history and faraway places, to the ecosystems of small farms, and to communities of plants, animals, microbes, and people living in balance with one another and the environment. The integrity of the system is made manifest in flavors that can be achieved in no other way, and the virtuous circle is complete.

THE THIRD RAIL

•

41

FOUR

Breed

It is hot, and we are bickering as we attempt to navigate the freeway somewhere to the south of Boston. We have just arrived from London, and August in New England is sweltering and humid, never the best weather for marital harmony, especially when combined with a complicated combination of tunnels and off-ramps. We have been driving for an hour, and every sign seems to be pointing back toward Boston. With only minutes to spare, we finally arrive at our destination, Plimoth Plantation, the living-history museum in Plymouth, Massachusetts, that attempts to recreate the life of the Mayflower settlers. The museum is closing, and other visitors are already making their way out of the model village. Our appointment is with Norah Messier, the director of living collections. A bundle of energy, Messier has been working in living-history museums for over fi fteen years, often playing the role of a seventeenth-century colonist. Her husband also works at Plimoth Plantation, and at present he is cast as Myles Standish, the Pilgrims’ military advisor. We greet him as he swigs from an illicit modern water bottle, but we are here to talk to Messier about her greatest love: the animals of seventeenth-century Pilgrim farms. A committed homesteader, Messier chats with us about small-scale agriculture. We mumble some pleasantries; the drive from Boston has left us annoyed at each other and in generally foul moods, and we have to work hard to be polite. As we walk through the farming facilities, Messier points out the projects that she has been working on. Her aim is to recreate seventeenth-century farming practices, and much of her work is in deliberately unraveling four hundred years of progress. The museum has been back-breeding its corn and planting it with seed from the previous year’s harvest to try and get something as user-unfriendly and low-yielding as the corn that a seventeenth42

century English settler would have struggled to farm. There are Arapawa Island goats, derived from animals left to run wild on a Pacific island by Captain Cook in the eighteenth century. They remain a work in progress: after two hundred years of no human supervision, their udders are almost invisible, and their milk yields would be considered pathetic by even the most desperate Pilgrim. Messier blushes at the historical inauthenticity of Hyacinth, the livestock guardian llama—she is a necessity to fend off attacks from predators but would thoroughly bemuse a seventeenth-century Englishman—and then leads us up toward the paddock we have come to visit. There, sheltering in the shade of a tree, is Jesse, the American Milking Devon ox. His tail swishes to keep off the flies. He is gorgeous. We stop at the gate, and he comes forward to say hello. Deep red in color, with mighty horns atop a small, muscular body, Jesse leans in to be patted and stroked. We exchange a look, all of the tension of the journey gone. Our experience has been that new entrants to dairy farming almost always settle on goats. Just as Bronwen found as a teenager, goats make for glorified pets, their babies even more adorable than a litter of kittens. For novice dairy farmers, cows are more difficult. They are bigger and more intractable, and they have more potential to be a threat. Jesse is different. At the ripe old age of sixteen, he still looks extraordinarily youthful and vigorous. We are smitten. In 1623, Plimoth Colony received two cows and a bull, brought across the Atlantic from the county of Devon in England. The animals were aboard the third ship to arrive at the colony, and they became the original bovine Americans. When we show a picture of Jesse to a dairy farmer in modern-day Devon, she immediately recognizes him as a North Devon but is not intrigued by his dairy possibilities: the North Devon breed that remains in the United Kingdom has been bred for beef, and she tells us, “Nobody milks those.” But in the seventeenth century, their ancestors were an ideal choice for the challenges of life in the colony: according to Messier, American Milking Devons were “the triple threat of the bovine world.” In England, the cows from Devon had established a reputation as active, hardy animals, able to thrive in the harshest of environments, pull a plow, give milk, and fatten on poor forage for beef. Ironically, given the American Milking Devons’ modern-day popularity for dairy, the original cows’ value to the Pilgrims as draft animals and as providers of fertilizer far surpassed the value of their milk. For dairy, the colonists kept goats. When Jesse bends slightly to reach some long grass, Messier points out his distinctive features: “In the bone structure, you can see the beef qualities, but BREED

•

43

you can [also] tell how hard they work. They are fabulous for being trained, very bright, very obedient. You can handle them right from the beginning.” This combination of attributes made the breed a fundamental part of farming life in colonial America. As late as the mid-nineteenth century, their capacity to draw a cart all day and thrive on very little forage made Devons the principal draft animal on the Oregon Trail. However, by the beginning of the twentieth century, their position was under threat. To try and understand this decline, we approached John Hall, the secretary of the American Milking Devon Cattle Association. Hall’s family has a long history with the breed in its New England heartland—his parents met when his father went to visit the only other local family with Devons and found his future wife doing the milking—and he now has a herd of fi ft y-six cows on his farm in Connecticut, making it one of the largest holdings of American Milking Devons in the world. For Hall, it was the period of growth that followed World War II that almost killed the breed: “Production ag[riculture] came into play, and you had to milk one hundred cows to be profitable rather than twenty; you had to have a Holstein that gave you one hundred pounds of milk a day rather than a Devon that gave you twenty.” It was the same for beef, with the introduction of large feedlots out West destroying the economics of the Devon: “They could grow Angus and Herefords that they could feed up in eighteen months. Devons take you five or six years to grow a full-size animal.” The third string to the Devon’s bow, its efficiency as a draft animal, was obliterated by the ubiquity of the mechanical tractor. Amid the conditions of the postwar boom, the Devon felt like a historical curiosity that belonged in a museum rather than on the dinner table. By the 1970s, their total number had dropped to around two hundred animals worldwide. It was a breed on the verge of extinction. Things look better now. There has been a concerted effort to rebuild the population, including the creation of a semen bank from six bulls, and homesteaders and other very small-scale farmers are attracted to the low inputs that the breed requires. The American Milking Devon Cattle Association now has around 1,200 animals registered on its books, although the Livestock Conservancy still rates the breed as “critical,” its most endangered ranking.1 Given these numbers, what happens on our second day in Massachusetts is a stroke of great serendipity. Waiting in the line at a barbecue stand, we meet a young woman who seems intrigued by our voluble argument about rare breeds. That curiosity in itself is rare enough. But we soon find out that 44

•

BREED

Elsie Gawler, who runs the tiny North Branch Farm in Monroe, Maine, with her partner and another couple, not only farms American Milking Devons but has also just embarked on a cheesemaking project with their milk. We very much doubt that she is prepared for the enthusiasm of our reaction to this information: here at last is the chance to talk with someone about using Devon milk for cheesemaking. Once she recovers from the excess of our initial reaction, Gawler tells us about her farm and her new creamery. The organic farm is on 330 acres of Maine forestland and is run as a highly diversified endeavor. There are four acres of vegetables, a five-acre orchard, an acre of blueberries, and a herd of just seven cows. Not just that, but the cows (mostly American Milking Devons but with some Canadiennes and Jerseys thrown in too) are used for both dairy and beef. As Gawler says, when an animal misbehaves in the milking parlor, “there is always the option to eat it!” With its minute scale—the farm has only just acquired a milking machine, its cows having previously been milked by hand—and heavily diversified structure, North Branch Farm sounds like something from another era. It fits. Gawler herself is a prominent folk musician, a stalwart of the Maine Fiddle Camp, and she met her partner, Tyler Yentes, through music. But the farm also illustrates the sort of highly extensive pasture-and-hay-based system in which the American Milking Devon prospers, especially when the return from the carcass is also part of the equation. Gawler’s first cheeses are not yet ready to taste. She is committed to making a raw-milk cheese, so she has built her own small aging facility, and the wheels are maturing in the cave. She is, however, enthusiastic about the milk from her Devons. It is rich, certainly, but never as laden with fat as that from the Jerseys. She cannot understand why more cheesemakers are not making use of the breed. Suddenly, we are served our pulled pork, and the hot sandwiches drag us away from our topic of conversation. The combination of meat in a bun and discussion of multipurpose breeds is a good one. With dripping sandwiches in hand, we discover that we all very recently attended weddings in Aveyron, in the south of France. The region, a rural backwater at the very southern edge of the Massif Central, is home to the Aubrac breed of cattle, another beautiful multipurpose breed. While Gawler waxes lyrical about the charm of her Devons, the pulled-pork sandwiches immediately remind us of something else. In the city of Rodez, the capital of the département of Aveyron, there is Aubrac Burger, a fast-food joint dedicated to the pleasures of the BREED

•

45

Aubrac cow as a source of both meat and cheese. There, we had tucked into burgers made from Aubrac beef, garnished with salty, tangy Laguiole cheese, and two sides of aligot, the local cheesy mashed potatoes. It was an integrated farming system in a takeout box. Why is there no American Milking Devon Burger to celebrate this great American cow? When the time comes to leave New England, Jesse has left a huge impression. We wonder aloud to each other on the plane: perhaps it is partly due to the similarity of their red coats, but we both think that he resembles a smaller, friendlier Salers. He was a polite New Englander rather than a peasant of the Auvergne, but by all accounts, American Milking Devons can prosper in similarly rugged climates and with minimal inputs and supervision. Considering this, it seems bizarre that Elsie Gawler’s as-yet-unready cheese at North Branch Farm was the first cheese we have encountered made with Devon milk. The real question becomes: What do the bodies of today’s dairy cows say about the real values of the dairy industry, and is there any room for the multipurpose animal?

COWS G O N E W I L D

Our focus here is on the dairy cow, simply because its history over the course of the last three hundred years is, in itself, the history of the development and organization of the dairy industry. But first, we must go back to the beginning: to understand breeds, we need to understand what came before. Or rather, to understand how humans have shaped the bodies of their animals, we need to see what happens when cattle are left to their own devices. Chillingham Castle, in the English county of Northumberland, sits only twenty miles from the Scottish border within a stunning landscape of rolling hills, ancient oak trees, and craggy vistas popular with tourists looking for a William Wallace fi x. Indeed, the castle occupied a crucial strategic position during the thirteenth- and fourteenth-century Wars of Scottish Independence. The purportedly haunted castle and gory dungeon exhibit are only part of Chillingham’s attraction. It also boasts something much more unusual: a herd of wild white cattle that have lived completely enclosed within the castle’s 360-acre private park for over seven centuries. There is a rich mythology surrounding these wild cows, which by the nineteenth century had become famous for their purity and unbridled ferociousness, so different from mundane domestic cows. Even the young Walter 46

•

BREED

Scott, in his romantic ballad “Cadyow Castle,” chronicled the thrilling imagined hunt of the “Mightiest of all the Beasts of Chase . . . the Mountain Bull . . . of Caledon.”2 The genetic origins of the Chillingham cattle are far from clear. The estate itself claims that the animals are “the only survivors of the wild herds which once roamed Britain’s forests,” although there is stronger evidence to suggest that they are no more than domesticated medieval cattle gone wild.3 Either way, the herd has run feral for many centuries, completely cut off from other cattle, and offers a fascinating case study of what happens to animals when they are left to get on with things in a state of nature. The promotional flyers produced by Chillingham Castle still play to a similar formula as the romantic poets, promising those embarking on its cattle encounter an “untamed spine-tingling experience” featuring plenty of goring. So it was with a fair amount of excitement, mixed with a bit of trepidation, that we set out to meet the Chillingham herd face to face. Approaching the enclosure through a grassy meadow punctuated by massive oaks, visitors must step through a series of disinfectant foot baths designed to avoid diseases being brought into the park. Nearing the perimeter, consecutive high-tensile fences come into view, designed both to prevent escapes and to prevent any fraternization between the wild Chillingham cattle and the domesticated modern cows in neighboring fields. Park warden Ellie Crossley meets us at the “hemmel,” a wooden shed papered with posters depicting the history of the Chillingham herd, and we venture forth into the enclosure, ready to meet King Bull. After this tremendous buildup of anticipation, it’s hard not to be underwhelmed when we meet the beasts incarnate. A group of greyish-white cattle stand chewing their cud and enjoying the sunshine. Their graceful horns are impressive—particularly for anyone used to dehorned domestic cattle—but they possess none of the threatening grandeur of the romantic poets’ or the advertorial copywriters’ creative imaginations. It is not a trick of perspective or distance: the Chillingham cows are actually shockingly small, less than half the size of a modern Holstein. The female cows’ udders are tiny, so small and tucked up that they are barely noticeable. These girls might have gone wild, but their mammaries remain well hidden. The bulls, meanwhile, are much lighter in their hindquarters than modern breeds. Crossley, who speaks affectionately of her charges, compares them to deer, explaining that they are faster and much more nimble than domestic cattle and able to accelerate quickly when surprised or threatened. Despite their diminutive stature, they look surprisingly healthy and fi lled out for wild BREED

•

47

animals. “They naturally keep themselves in check, they just don’t get a lot of the diseases you see in cattle,” Crossley tells us. “It’s just survival of the fittest. They do it themselves—they’re incredible.” Physically, the Chillingham cattle are entirely unexceptional, but from a genetic standpoint, they are fascinating. The Chillingham herd is one of the most extreme examples of inbreeding, in cattle or otherwise, in the mammalian world.4 Over hundreds of generations, and through several genetic bottlenecks—the herd size has fluctuated dramatically over the years, even dropping to only thirteen cows during a particularly harsh winter in 1947—they have become for all intents and purposes genetically identical. These cows pose something of a conundrum, as even moderate levels of inbreeding in domesticated animals are often associated with decreased fitness. Highly inbred animals often experience a range of serious health effects, including infertility. A research team headed up by Peter Visscher at the University of Edinburgh’s Institute of Cell, Animal and Population Biology studied the genetics of the Chillingham herd and concluded that when intensive inbreeding occurs in an environment that is highly selective for overall fitness, in this case a remote Northumberland moor, it may “purge” the genome of weak genetic material.5 Under the wrong circumstances, inbreeding can be a source of weakness. It is also an inextricable part of creating a breed. Talking and working with farmers, it is very easy to accept the concept of breed without question. One farmer keeps Jerseys; another has pedigreed Holstein-Friesians. All modern European cattle breeds, from the hulking Italian Chianina to the diminutive Dexter, belong to a single species, Bos taurus. By definition, any two members of this species are capable of producing fertile offspring with one another. Within the enormous range of genetic diversity encompassed by Bos taurus, random chances, different climates and agricultural systems, and more than a dash of human intervention have led to cows that share certain defined and valuable characteristics. These similar groups of more closely related animals have over time been described, and honed, first as characteristic of “cows from a certain area” and subsequently as breeds. But there is no scientifically reductive, genetically based definition of “breed.” As the United Nations Food and Agriculture Organization’s Commission on Genetic Resources for Food and Agriculture makes clear, “breed” is a term of art. Peel away the layers, and the concept is no more than a shared social construction, an artistic description that has far more to do with the way that humans classify the world than the essential properties of the animals themselves. 48

•

BREED

T H E LO N G R E AC H O F T H E S H O R T H O R N

The improvement of livestock has been going on since the dawn of domestication, many millennia in the past, but modern livestock breeding and the codification of its systems are far more recent. Coincidentally enough, the birthplace of modern cattle breeding is in County Durham and North Yorkshire, less than a hundred miles south of Chillingham. By the late eighteenth century, agricultural reformers and Enlightenment thinkers began to apply their energy to developing a science of agriculture. Most agriculture at the time was carried out in extensive mixed-farming systems in which cattle played a crucial role not only for dairy produce but also as draft and meat animals. And it was in this environment that the Shorthorn cow first came to public attention. The Shorthorn was a cow that seemed to have it all: it was big framed, had a docile temperament, and put on weight easily and quickly, making it ideal for the efficient production of high-quality beef. Additionally, it was a prodigious dairy animal, giving copious amounts of milk. It could even be harnessed and put to work as a draft animal if required. By 1834, when the Society for the Diffusion of Useful Knowledge published their guide to cattle farming, the reputation of the Shorthorn was such that it merited special attention: “It is the combination of perfections which has conferred, and will perpetuate, the superiority of this breed of cattle.”6 With such a long list of attributes making it well suited to the mixed farming systems popular at the time, the Shorthorn spread like wildfire. Cattle were driven south along the Great North Road from Yorkshire toward London and were bought by farmers on the road or purchased at markets, to the extent that “before they had been many days on march their owners often found themselves like generals without an army.” 7 By the 1830s, the Shorthorns were becoming more and more widespread, and by the 1850s, there were Shorthorn breeders all over the country, from Cornwall in the extreme southwest to the Orkney Islands off the northeastern coast of Scotland.8 Meanwhile, the first Shorthorns had been exported to North America in 1817, and the breed rapidly gained a foothold in the United States and Canada as well.9 For modern readers sensitive to the threat of losing rare and unique genetic resources, there is a poignancy to contemporary accounts of the breed’s rise to dominance. Today, the Gloucester breed is critically endangered. But in the 1830s, the dairy farmers of Gloucestershire, who at that time were farming BREED

•

49

the local breed, were “so much alive to the superiority of the short-horns, that they [laid] hold with avidity of anything which approach[ed] them in colour, or [was] called by the name.”10 Shorthorns were on the march, and local breeds were being pushed to the margins. By the time that the making of Cheddar cheese was being codified in the southwest of England in the 1860s, the Shorthorn was enshrined along with it as the classic breed for producing the milk for Cheddar. How quickly the old red North Devons and West Somerset middle-horns, with their creamcolored heads and hindquarters and deep-blood-red barrels, had been brushed aside.11 The rise of the Shorthorn breed was accompanied by the rise of another social animal: the breeder of pedigreed Shorthorns. The Shorthorn was created as an institution and ruthlessly marketed by brothers Charles and Robert Colling, who recognized the potential of their local stock and set about developing and purifying it using a system of intensive inbreeding developed only a few years earlier by another Englishman, Robert Bakewell. Today, the concept of breeding is inextricably linked to improving the conformation and performance of stock, but the Collings believed that breeding meant finding good foundation animals and using inbreeding to copy them. It made for single-line family trees. The most famous bull of the early nineteenth century, the Collings’ Comet, had a father (or sire, in the language of cattle breeding) who was also his grandfather (grandsire), and only one grandmother (granddam), who was also his great-grandmother (great-granddam). Quick-maturing, docile, and ideal for both meat and milk production, their line of improved Shorthorn stock possessed all the qualities necessary to take the cattle world of the early nineteenth century by storm. As the breed began to take off, next-generation breeders such as Thomas Bates bought animals from the Collings’ line simply because of their immaculate heritage. For Bates, quality was not determined by performance but by the genealogical purity of the lineage, and this was reflected in his choice of names for his cows, which were all Dukes and Duchesses. Bates started a program of even more intensive inbreeding than that practiced by the Colling brothers (perish the thought!) and was closely involved in the publication of the first public pedigreed Shorthorn herd book in 1822. Thus commenced an inevitable contest over whose lines were the purest, with many breeders “disdaining to cross” their stock with any animal that might prove less exalted.12 Pedigreed “fancy” Shorthorn breeding took hold as a pastime of the moneyed elite and went hand in hand with social standing. Unlike horses bred for 50

•

BREED

racing, whose aristocratic owners derived prestige and prizes from their animals’ ability to run fast, the Shorthorns were all for display. It was a display that was popular with the public too. When the massive Shorthorn known as the Durham Ox was brought to London in 1802, the single day’s gate receipts totaled £97, the equivalent of £7,735 ($12,900) in 2014, adjusted according to the Retail Price Index. Not just that, but more than two thousand people bought prints of a painting of the obese beast.13 The painting, which emphasized the grotesque and hulking body against broom-handle legs and a tiny head, was self-consciously exaggerated, but it showed the aspirations of the breeders clearly enough. The painting bore as little relationship to the dimensions of a real cow as the pages of Vogue—or perhaps more appropriately, pornographic magazines—do to normal humans. The historian of the diffusion of the breed, John R. Walton, refers disdainfully to “over-fed showyard Shorthorns, their genealogies as much a matter of scrutiny and debate as those of their owners.”14 For these breeders, it didn’t matter if an animal had no functional utility as long as it had the right pedigree. This aristocrat’s pastime was not limited to Great Britain. In the 1860s, cattle directly descended from Bates’s stock were being imported to the northeastern United States and traded for extravagant sums. Perhaps the definitive moment of the craze came at a New York auction in 1873, when a seven-year-old cow named the Eighth Duchess of Geneva was sold for the breathtaking price of $40,600, the equivalent of around $830,000 today.15 She died not long after.16 According to breed historian John Walton, foregoing the pursuit of “dual-purpose rent-payers” in favor of “no-purpose parasites” had tarnished the reputation of the Shorthorn industry.17 However, even by the mid-nineteenth century, registered breeders accounted for only a tiny minority of farmers with Shorthorns. The herd book might have been the equivalent of haute couture, extravagant and impractical, but the diff usion of ideas and the stock that fi ltered down to more tentative and conservative farmers had a lasting effect on the farming community and primed it for what was soon to come. The Shorthorn was the first—but certainly not the last—improved dairy cow.

T H E WO R L D I N B L AC K A N D W H I T E

Across the North Sea from the home of the Shorthorns, the Netherlands has a long history of dairy farming. As early as 1640, the city of Gouda made five BREED

•

51

million pounds of cheese a year.18 The Dutch lowland pastures were good for grazing, but the limits of space drove the local farmers to ever-increasing lengths to improve productivity. At first, this involved the obsessive collection of cow manure as fertilizer for the fields, but eventually, attention turned to the animals themselves. By the late nineteenth century, the black-and-white Dutch cow was sufficiently globally renowned for milk production as to attract serious interest from the United States. It would be American breeding with Dutch genetics that dominated the world of dairy in the twentieth and early twenty-first centuries. We are living in the era of Holsteinization. Dutch cows first landed in North America in 1613, imported by the Dutch East India Company, but they did not gain anything other than local prominence until 1852, when Winthrop Chenery, a Massachusetts breeder, purchased a Friesian cow from a Dutch sailing master who landed cargo at Boston. Chenery was sufficiently pleased with this single cow that he made subsequent imports of Friesians in 1857, 1859, and 1861; he also established a herd book to keep track of the pedigree of these Dutch cattle.19 At the same time, the breed became known as Holstein—a place in Germany, not in the Netherlands— through a mistake by an American official overseeing the import.20 In the United States, which until then had been a country whose cattle industry was driven by the demands of beef rearing, the productivity and milk yields of the Holstein were received with enthusiasm. Not just that, but the breed seemed capable of miraculous improvement by assiduous breeding.21 In the Netherlands, extreme “dairy-type” Friesian cattle fell out of favor in the early years of the twentieth century: their body shape was considered ungainly, they required a large amount of food yet produced somewhat watery milk, and furthermore, out-of-condition high-yielding animals were suspected of being responsible for an outbreak of bovine tuberculosis. Dutch breeders had concentrated on making their Friesians animals that would prosper in grazing-based systems and ensuring that they retained plenty of body characteristics appropriate for beef. In the densely populated Netherlands, where space was at a constant premium, there was no room for the luxury of separate beef herds. The resulting animals were good milkers, but they were also hardy. When Dutch cattle breeders visited their American counterparts in the 1950s, they were at once impressed and revolted.22 To Dutch eyes, the Americans seemed to have lost the intuitive aspect of breeding. In the United States, breeding was not done by look or feel; rather, it was science that was tracked through statistics.23 Ruthless monitoring of milking data, ultimately 52

•

BREED

assisted by the widespread adoption of artificial insemination, gave spectacular returns. Between 1944 and 1997, the milk yield from the average American Holstein increased by a staggering 369 percent.24 The numbers are impressive. In the United States, the average Holstein gives around 23,000 pounds of milk (10,432 liters) per lactation, which works out to about 75 pounds (35 liters) of milk per day. At the time of writing, the individual record annual milk yield for a cow is held by a Holstein bearing the name Ever-Green-View My Gold-ET, who produced 77,480 pounds of milk in 2016.25 To put that in context, it is roughly ten times the average yield from a Salers cow. While in the United States the Holstein is now utterly ubiquitous—in 2002 just 5 percent of the US dairy herd was composed of animals other than pure Holsteins or unregistered Holstein “grades”26—in Europe it was not until the 1980s and 1990s that full-throttled Holsteinization took over. In the British context, we see interesting ebb and flow to the idea of the “progressive cow” of any given era. Whereas the Shorthorns dominated the midnineteenth century, to the extent that they are still remembered as the traditional Cheddar cow, the thirty years after World War II was the era of the British Friesian. After the struggle for wartime survival in the face of German U-boats in the Battle of the Atlantic, progressive agriculture was the order of the day. And progress meant yields. Friesians offered increased yields compared to the Shorthorns, and advisors from the British Ministry of Agriculture counseled conversion. The records of the Torrington Artificial Insemination Centre in the West Country county of Devon provide a great example of the pace with which the Shorthorn disappeared. In 1950, 21 percent of the center’s membership farmed Friesians, 15 percent Ayrshires, and 28 percent Shorthorns. By 1980, 55 percent farmed British Friesians, and there were no remaining Shorthorn farmers.27 Any breed that lives and trades on its high-yielding, progressive character is immediately vulnerable when a new star arrives with better metrics for production. The Friesian had replaced the Shorthorn in the United Kingdom and Holland, but in the 1980s and 1990s, both the British and Dutch dairy industries recognized the power of the American cow for sheer yield. With the growth of artificial insemination and a global market for bull semen, the tall “coat-hanger” Holsteins reached across the Atlantic. The Holstein is a cow that has been designed for a very specific farming system: it is an extraordinarily efficient converter of bought-in feed into milk with relatively low solids. The derisory term for Holsteins within the dairy industry is “white water factories.” Without the high-energy inputs that BREED

•

53

come from a diet rich in concentrates, the Holstein suffers and loses body condition. Holstein USA, the American breed society, does not pretend that it is a breed used by homesteaders or micro dairies. In their publicity materials, black-and-white cows dance with flying dollar signs; in another image, an inquisitive Holstein looks around a corner as a tanker truck passes by. According to Holstein USA’s own data, test Holsteins averaged 3.69 percent fat and 3.09 percent protein in their milk. This is in contrast to a Jersey herd, where the figures are 4.82 percent fat and 3.65 percent protein. The trick, however, is that the Holsteins each produced an average of over eighty pounds of milk per day, compared with a mere fi ft y-seven pounds per day from each of the Jerseys. With milk contracts for cheese-component pricing written for the total number of pounds of butterfat and protein delivered, rather than the relative proportion of the yield, the Holstein is the more profitable breed.28 But the Holstein is optimized for liquid milk production rather than for cheesemaking. A study from the University of Padua showed that the variants of casein expressed in Holstein milk coagulate poorly compared to those of other breeds, leading to loss of cheese yield.29 The Holstein is the ultraspecialized physical incarnation of an industrial system: its milk is just good enough for the money. Or rather, as expressed by the blunt tagline favored by Holstein USA, “More milk equals more money.” Relentless genetic selection, largely driven by milk-production metrics, has resulted in the Holstein’s incapacity to do anything other than eat and make milk. Fertility is a problem. In the United States, the conception rate among Holsteins (the percentage of cows that conceive after a single insemination) fell dramatically from 66 percent in 1951 to 40 percent in 1997.30 The sheer size of the animals, and especially their long spindly legs, make calving a struggle. Despite all of the Holstein’s brilliant production metrics, the breed is ill-adapted for the day-to-day challenges of existence.

O LY M P I C AT H L E T E S A N D N O B E L P R I Z E W I N N E R S

In the dairy industry, margins are low, and an incremental improvement in efficiency and production can mean the difference between success and bankruptcy. The development of better stock is key to survival. And all serious farmers, whether or not they operate within an intensive system or are under intense financial pressure, are interested in selecting and propagating the best animals, thus enhancing the productivity of their herd over time. 54

•

BREED

At one end of the spectrum, this might simply involve selecting fertile and low-maintenance bulls from within a herd and using the best and healthiest cows to breed replacement stock, as has been done since time immemorial.31 But new technologies have given farmers the tools to develop their herds with much more speed and efficiency. The first great leap forward in cattle breeding was artificial insemination (AI). The idea of removing sex from the equation and impregnating animals with donor semen was first studied in Russia at the very beginning of the twentieth century, but the phenomenal growth in AI occurred in the United States in the 1940s. The founding of the New York Artificial Breeders Cooperative Inc. allowed for close collaboration between farmers and researchers from Cornell University, and the techniques developed in the United States eventually became established worldwide.32 The markers along the way as the technique was developed are testament to human ingenuity. In the early twentieth century, Danish researchers from the Royal Veterinary University in Copenhagen had established the method of rectovaginal fi xation, whereby the technician guided the delivery of semen into a cow’s cervix by steering with their arm deep in the animal’s anus. In 1914, an Italian team developed the first artificial vagina for the collection of dog semen, which inspired the Russians to build a sturdier vessel for the use of bulls, stallions, and rams. These days, bulls might mount a cow sex doll made of cowhide stretched over a special frame, complete with rubber tubing for the donation, or they might be tempted to mount other males known as “teasers,” with technicians waiting to jump in with a plastic bottle at the appropriate moment; some bulls do actually have sex with a female, only for the semen subsequently to be reclaimed from her vagina. The semen is then frozen in liquid nitrogen in “straws.” Glass vessels had proved problematic, as they tended to shatter at the extremely low temperatures required to preserve the semen, and it was not until 1940 that the Danish researcher Dr. Eduard Sørensen chanced on the ideal packaging. Watching his daughter drink punch through a cellophane straw at a children’s party, he realized that it would make an ideal vessel for deep-frozen semen.33 The power of AI is simple and compelling: it gives every farmer equal access to the best bulls on Earth. So-called progeny-proven sires are ranked according to the merits of their offspring for a dizzying array of traits, with scores given for everything from the normal conformation attributes to the length of productive life, the number of somatic cells and the percentage of fat and protein in the milk, calving ease, and on and on. Using AI, a farmer BREED

•

55

could easily select a different bull to impregnate each of his cows, depending on their individual strengths and weaknesses, something that would previously have been unimaginable. There is the convenience factor as well. Rather than the expense and commitment of maintaining a dangerous bull, breeding one’s cows becomes as easy as picking up the phone. A friendly technician is on call to handle the liquid nitrogen and do the dirty work. For a man, the whole exercise is a humbling glimpse of potential obsolescence. AI conjures visions of a world where the right to reproduce is reserved for Nobel Prize winners and Olympic athletes. And in the current market, nobody is interested in a bronze medalist. A little bit of semen goes a long way, with a single ejaculation able to be broken down to provide anything from fift y to five hundred straws. “Semen extenders” are mixed in to bulk out the dose and help it survive its time in liquid nitrogen. Initially, in the 1940s, this was a mixture of egg yolk and sodium citrate, but modern extenders include a cocktail of glycerol, egg yolk (or possibly milk), and citrate mixed with antibiotics to guard against venereal disease. With extenders, one ejaculation can be stretched up to twenty-five times further. Powerhouse “millionaire bulls” are becoming more common. The record holder, a Holstein bull named Toystory, delivered over 2,415,000 straws’ worth of semen and fathered over half a million animals in fi ft y countries over the course of his lifetime, according to Wisconsin-based genetics company Genex. This highly selective market holds fantastic genetic rewards for an infinitesimal few. Bull calves from Holsteins hold little value as beef animals, and farmers may choose to spend more on sex-selected semen to increase the chance of female calves. Farmers who wish to speed up the genetic progress of their herd even further may also choose to harvest eggs from their best cows and use their less-valuable cows as embryo-transfer recipients, maintaining the farm’s milk supply and providing replacement calves while ensuring that the weaker cows’ genes are expunged from the herd as quickly as possible. The last decade has seen another technological leap forward, one that at face value is more subtle than sexed semen and embryo transplants but that promises to transform breeding: the advent of genomics. Nate Zwald, US sales manager for Alta Genetics, based in Watertown, Wisconsin, describes the power of this new technology with infectious enthusiasm: “With genomics coming on board, we’ve seen as much dramatic change in the last eight years of breeding as we did in the previous thirty or forty.” Genomics has already entered the popular consciousness through the work of companies such as 23andMe, which offer human customers a detailed 56

•

BREED

report of their genetic makeup, identifying which variants they possess of hundreds of thousands of different genes. A single company, Illumina, makes the DNA chips used for both bovine and human genomic analyses. “In humans,” Zwald tells us, “the only data is associated with people who choose to be tested and who choose to put their data into a database. In cows, it’s as if there were a database with all the health and performance data of almost every person in the world. That’s what we have for dairy cows.” It is at once a terrifying and fascinating thought. Even more impressive is the fact that genomics makes use not just of current data on animal performance but of historical records and samples as well. “Not only do we have the genetic material on those [long-deceased] bulls, but even in the 1960s and 1970s, if a bull had ten thousand daughters, and they were all on milk recording, we have all that historical data,” Zwald explains. “We can say, for example, that a given bull’s daughters produced two thousand pounds more milk than their herdmates, and they lived longer and had less mastitis, because that information is all recorded. Now, with genomics, we know which genes made that happen, and we can use that information to shape the future.” Breeders working with goats, sheep, pigs, chickens, and beef cattle are all using genomic technology, but dairy cattle, with monthly samples taken for milk recording and meticulously kept herd records, have far and away the most impressive phenotypic dataset. Now, rather than having to wait for four to five years for potential new bulls to be evaluated on the basis of their daughters’ performance, calves are screened, and their “genomic proof ” immediately known. Ultimately, as the bovine genome is subjected to more analysis, and links between genotype and phenotype are even better established, farmers will be able to selectively breed with a much higher level of precision and control. So what about novel attributes, traits that might be interesting only to cheesemakers operating in extensive and low-input systems—could genomics help them as well? Zwald is pragmatic but optimistic. For any given characteristic, “the only question is: How do you measure it for an individual cow, to know which cow and therefore which sire does a better job? You need the ongoing phenotypic data to make genomics work.” As the cost of the technology decreases, one can only imagine that these sorts of bespoke analyses will become more accessible. In the meantime, genomic-proven sires are already being selected for new traits, such as compatibility with robotic milking machines. The robots work best when cows have longer teats with a bit more distance BREED

•

57

between them, which makes it easier for the laser-guided milking units to locate them. Likewise, selecting for genes associated with a faster milking time means that cows never have to wait in line too long before a robot becomes available. These new technologies are clever. They are also morally neutral. Genetics companies have become successful by giving farmers great service and the tools they need to get where they want to go. The question becomes: What does the market reward? This is not a vague existential daydream but rather a very specific question. For many, the answer can be found in merit indices published by various national institutions, particularly the US Department of Agriculture. Economists and cattle geneticists analyze macroeconomic trends and assign weighted values to each trait.34 Bulls’ scores on these indices are published in their lonely hearts advertisements in the semen catalogs, a useful, one-stop quality mark to help harried breeders make quick judgments about which lines to introduce to their herds. Even at today’s pace, markets change more quickly than genes do, and a reissued index that revises the weight of health versus production scores can instantly send an individual bull’s ranking hundreds of points down the charts or launch previously middling stock into the genetic stratosphere. Even economists admit that the indices are based on forecasts pushed disturbingly close to the point of guesswork—hardly a rock-solid foundation for black-and-white decision making. Merit indices are just one model of value.

BREEDING FOR CHEESE

Suffolk, England, is a lush county of green marshes and river valleys that looks out across the North Sea toward the Netherlands. While the historical reputation of the cheese from the county is very poor, it is now home to one of England’s most enterprising young dairy farmers. Jonny Crickmore is in his early thirties, energetic and intense and perhaps even a little impetuous. He has been part of the family’s farming business since he was four, but when he and his wife, Dulcie, took the reins at Fen Farm Dairy in 2010, they quickly realized that if they continued in the highinput, low-margin direction they had been headed, they would soon be out of business. Crickmore recognized that selling his milk on contract was denying him an opportunity to add value. They contacted French cheese consultant Ivan Larcher in the hope that his Gallic sensibility would guide them toward making the United Kingdom’s first respectable raw-milk Brie. 58

•

BREED

Fen Farm Dairy had started with Friesian cows in the 1960s and, like so many other farms, followed the trends over the years. By the late 1980s, they were breeding to Holsteins. “The first cross was fantastic,” Crickmore explained, “because you had the hardiness of the Friesians plus all this extra milk. Unfortunately, everyone got greedy and bred more and more towards Holsteins.” The result was a predictable ocean of somewhat watery milk, hardly ideal for cheese. Larcher said he would only work with the Crickmores if they agreed to work with a breed that produced milk suited to cheese. He refused to work with “those black-and-white cows.” Instead, Larcher suggested Montbéliardes. In France, the Montbéliarde has become synonymous with cheese. Originating in the Alpine east, the breed rose to prominence for its association with France’s bestselling Appellation d’Origine Protegée cheese, Comté, and has now been adopted for the production of many others, from Reblochon and Mont d’Or to Saint-Nectaire, Cantal, and Salers. This mass conversion was not arbitrary. The Montbéliarde is the result of a carefully planned campaign of over fi ft y years to develop a cow perfectly adapted to the needs of the cheesemaker. It is now the most profitable cow in France. Whereas the Holstein is a cow bred for quantity, the Montbéliarde has been sculpted to a different set of requirements. First and foremost is the focus on the level of protein in its milk. While the amount of fat in a cow’s milk is based on a combination of genetics and feeding, the level of protein is heavily influenced by genetics. Montbéliarde breeders set out to create milk that hovers around a highly unusual fat-to-protein ratio of one to one, a bespoke milk for cheesemaking. As the Holstein shows, not all milk proteins are created equal when it comes to cheesemaking, and the Montbéliarde association has also sponsored research on which variants of casein most influence cheese yield. Even corrected for total solids, milk from cows that carry a specific genetic variant, Kappa-casein B, yields 8 to 12 percent more cheese than that of cows without it—a massively significant difference when considering total yield from a cow throughout the year.35 Once this gene and its function had been identified, the Montbéliarde breeders worked hard to disseminate it through their breeding stock, to the extent that over three-quarters of semen in the Montbéliarde AI bank is from bulls with the variant.36 Many of the high-quality French cheeses made with Montbéliarde milk are made from raw milk. The breeders decided on a further goal: decreasing the breed’s susceptibility to mastitis. This is important both for making good BREED

•

59

cheese and for cow health and longevity. Cows prone to mastitis are among the first to be earmarked for a euphemistic “early retirement” at the abattoir. In 1981, Montbéliarde cows averaged just over 5,500 liters of milk per lactation, hardly enticing for dairy farmers used to Holsteins, which gave almost 40 percent more milk on average. Twenty-five years later, the average Montbéliarde has caught up with the pedigreed Holstein of the 1980s in terms of total milk production, but the breed also boasts a stunning combination of hardiness, disease resistance, easy calving, and milk components. No wonder its population almost doubled in France over the same period.37 Jonny Crickmore proved just as receptive to the breed as his counterparts in France had been. “Everything seemed to make sense about the Montbéliardes,” he told us. “And, unlike the Holsteins, they are good for beef as well. They just ticked all the boxes.” He contacted the official Montbéliarde breeders’ cooperative and signed up for their animal sourcing service. He and Dulcie boarded a plane a few weeks later. “The French representative picked us up from Geneva airport, and we weren’t beating around the bush. We had thirty-five dairy farms to see in two and a half days, so we went straight to the first farm,” he recounted. “I was expecting the farmer to take us to a field with a bunch of heifers [young cows who have yet to calve], and we would pick some out. We arrived at the first farm, and there was just a single Montbéliarde and a donkey! He asked me what I thought about it. We took some notes, as if we were paying close attention, and we said, very seriously, ‘We’ll come back to that one.’” It got better from there. In all, they selected seventy-two cows, which they loaded up and trucked ten hours north, the animals trading in their alpine vistas for Suffolk marshes. Crickmore had been preparing himself to initially lose money on the Montbéliarde cows in exchange for the better quality of milk required for his cheese. And it was true that the French cows only produced about threequarters as much milk as his Holsteins did on the same system. But when he crunched the numbers after a few months, he was intrigued. His Montbéliardes were maintaining their level of production, but they were eating far less food than the Holsteins and staying in better condition at the same time. “They’ll only produce so much milk, and then after that, if you overfeed them, they’ll just get fat,” he explained. “With Holsteins, the more food you can get them to swallow, the more milk they’ll give.” That wasn’t all. Crickmore found that despite the stress of their voyage and having to adjust to life at sea level, the Montbéliardes were also healthier, just 60

•

BREED

as promised: “They’re so much more durable, they just seem to last. We have practically no mastitis. All of those things are where you really start saving the money. All right, you need to keep more of them to get the same amount of milk. But if you look at it in a different light, it is a better quality milk too, and if you use it for what it is, you also get your money in a different sort of way.” The Crickmores’ Baron Bigod is an excellent cheese, full of milky layers and complexity, but a startling demonstration of the superiority of the farm’s milk came a year later. With excess milk on their hands, the Crickmores invited another cheesemaker to make her cheese at Fen Farm Dairy. Her previous milk source had been a mixed herd of cows supplying the liquid milk market, and nobody—least of all the cheesemaker herself—was expecting much difference when she made the switch. But the cheeses in her first batch made from the Montbéliarde milk were massive: she had placed the normal amount of wet curds in the molds and had ended up with almost 50 percent more cheese. It wasn’t simply a question of yield. While the cheeses made from the previous farm’s milk had had a tendency to be wet and spongy, the Montbéliarde milk cheeses held their moisture differently; they were both better drained and fluffier.

BREEDING IN AND OUT

The attributes that farmers prize in their herds come at a price. Thumb through an issue of Dairy Herd Management or Farmers Weekly, and there it will be: a vaguely alarmist article warning that random use of Holstein semen on Holstein females is becoming risky due to inbreeding levels in the national herd creeping ever higher. Inbreeding: it is the dark accusation that enthusiasts of one breed hurl at those of others. Holstein USA emphasizes the number of bulls available due to Holstein being the most numerous dairy breed, but members of rare breed societies mutter darkly about the route that brought Holsteins their high yields. The numbers alone are startling. In 2008, it was estimated that the size of the Holstein population—globally some thirty million animals—was effectively, from a genetic perspective, the equivalent of between only thirty-five and sixty individuals.38 Furthermore, all of the Holstein bulls whose semen is available for AI in North America are descended from only two bulls, and there is not much difference between them; their Y chromosomes are almost completely identical. The other Holstein male lineages that once existed are practically extinct.39 BREED

•

61

Cows, like humans, have two sets of chromosomes and therefore two copies of most of their genes, one from each parent. Variants of these genes are called alleles. Some alleles are the same throughout a population; other genes have many diverse alleles carried by different individuals. If an organism has two copies of the same gene rather than two different variants, it is referred to as being “homozygous” for that allele. As cows become more inbred, they inherit two identical copies of more and more of their genes. Put simply, the inbreeding level of an organism is the percentage of the locations on its genome where the genes from both its parents are identical. Many breeders have drawn an arbitrary “do not cross” line at 6.25 percent inbreeding. To put this in context, a father-daughter or brother-sister pairing would produce an offspring with a level of 25 percent inbreeding, while a level of 6.25 percent is equivalent to an offspring of first cousins. By mid-2015, the average level of inbreeding in the American Holstein population at large had already hit 6.25 percent, and a random Holstein mating would be expected to lead to offspring with an inbreeding level of 7.5 percent, prompting a slew of articles about the impending apocalypse: poor health, infertility, and lost profits were clearly just around the corner.40 While the health and fertility effects of inbreeding on Holsteins are perhaps more pronounced because selection has been focused primarily on total production rather than overall fitness metrics for so long, the problem of rising inbreeding levels is hardly unique. In part because their total populations are so much smaller and the pool of bulls more restricted, other well-known dairy breeds are equally, if not more, inbred than the Holstein. The Jersey, Guernsey, and Brown Swiss breeds are all more than 7 percent inbred on average.41 As we saw with the Shorthorn, the very concept of a breed requires genetic convergence sufficient to achieve a “true breeding” type, and despite the convenient, if arbitrary (and already largely surpassed), line drawn in the sand at 6.25 percent, there is no real evidence that there is a true upper limit to what is achievable. Nate Zwald points out that with the extensive genomic data now available and selective emphasis placed on health and fertility alongside productivity traits, inbreeding levels of over 10 percent are sustainable, and even desirable. Much better, the theory goes, to mate one’s cows with excellent closely related stock than to introduce weaker—if unrelated—genes into the herd.42 The line between inbreeding and line breeding to enhance desired traits is a very fine one. The most significant risks from high inbreeding levels come into play with breeds that have gone through population bottlenecks outside of a highly 62

•

BREED

selective environment. The Gloucester breed is historically notable as the source of the first anti-smallpox serum in 1796, but it gradually collapsed over the following hundred and fi ft y years as it was supplanted first by the Shorthorn and then by the Friesian and the Holstein. By 1919, there were only 130 animals left, and the breed flirted with extinction through much of the twentieth century. A recent genomic study revealed that the inbreeding level of the Gloucester is between 12 and 14 percent, but it also helped to identify the individual animals with the genetic potential to stimulate production and health improvements.43 Breeders and geneticists have long been aware that crossing inbred lines with very different genetic material can increase the fitness of the offspring. Because of the greater number of differences between the genes carried by the sperm and the egg, these first-generation offspring have much less chance of the same genetic material being inherited on both sides. First-generation crosses from complementary inbred lines almost universally exceed their parents in overall health and fitness. In the 1950s and 1960s, a great deal of research was done in the United States on crossing different breeds, but despite the hybrid vigor of the crossbred cows and their impressive yields of milk fat and protein, when the numbers were crunched, milk volume won the day, and the prolific Holsteins consistently came out on top.44 Half a century later, with Holsteins’ health and fertility faltering and a market that has started to prize components above volume, those calculations have changed, and the practice of crossbreeding is being explored by an increasing number of milk producers and cheesemakers. Advocates of crossbreeding are keen to emphasize the health benefits— which include better fertility, easier calving, and greater longevity—that come from lowering the level of inbreeding, without putting too much emphasis on hybrid vigor. As any seed breeder knows, it’s easy to get impressive results with first-generation hybrid crosses; the challenge comes when it’s time to propagate the herd further. You can cross these offspring to a third breed—and many crossbreeding systems do use three different breeds—but ultimately, the level of heterosis (the performance bonus borne of high levels of heterozygosity) never reaches the giddy heights of those first one (with a two-breed system) or two (with a three-breed system) generations. And there is an irony at the heart of crossbreeding: its effectiveness depends on the maintenance of pure breeds and the gains achieved through line breeding. In the same way that breeding a superstar pedigreed Holstein to an unrelated but average Holstein is likely to produce a less impressive— BREED

•

63

albeit less inbred—calf, simply outbreeding to a genetically distinct but unimproved mate will quickly lead to the high-performance genes becoming diluted. Crossbreeding isn’t a free-for-all; it doesn’t work without obsessive purebred line breeding from which to feed the system. Breeder Jon Lundgren sums it up succinctly on his popular Dairy Crossbred Blog: “Instead of worrying about inbreeding, I embrace it. If genomics is increasing the rate of genetic gain at the expense of inbreeding, as a crossbreeder using completely unrelated bloodlines in a rotation we see only the good side that comes with genetic progress and none of the ill effects of inbreeding.” While the principle is counterintuitive, farmers practicing crossbreeding can still experience the negative effects of inbreeding if they breed their crossbred cows back to pure sires that they are too closely related to. Rather than stepping away from highly regimented pedigree systems, crossbreeding programs add a layer of complexity. Keeping track of which crosses must be bred with which pure breed and which crosses the daughters and granddaughters of these matches are lined up for, with multiple generations running alongside one another in the herd, is a challenge that can be dealt with through obsessive record keeping and clever tricks like colored ear tags. But it is hardly closer to a state of nature. While they produce less milk than their Holstein antecedents, crossbred cows often thrive on lower-input systems, particularly grass-based systems such as those that dominate the dairy industry in New Zealand, where crossbred Holstein-Jerseys make up over 25 percent of the milk-recorded cows.45 Depending on what breeds dairy farmers are using, for instance if they are crossing their Holsteins with Montbéliardes, they might—in addition to higher milk solids and better health statistics—reap the benefit of increased value for their bull calves. But despite the growing conversation about crossbreeding—and its popularity in Europe and the Southern Hemisphere—the US dairy industry remains a stronghold of specialized single-breed dairies.46

B E YO N D B R E E D

At first sight, the modern world of cattle breeding appears to be a testament to scrupulous meritocracy. In place of a system that breeds animals simply to look like exaggerated paintings of the perfect bull, we now have multigenerational surveys of the bovine genotype, studied in conjunction with objective performance data. The determined breeder can chase any genetic trait that 64

•

BREED

can be measured, and as the industry becomes increasingly sophisticated at screening out genetic defects, even the risks of inbreeding are receding. The livestock show, the breed fiesta that dominates the pages of breed histories (their turgid lists of prizewinners across the course of the centuries make them one of the world’s least satisfying literary genres), still exists. The websites of breed societies are full of pictures of beaming farmers next to their impeccably groomed animals. But those pictures of uniformed farmers and beasts with rosettes seem a quaint nod to another era. Even the dress codes for the farmers look to the past. How many people these days milk in a bowler hat and tie? But the numerical metrics of big data in cattle breeding disguise the hidden value judgments within the selection of data. Above all, the proliferation of indices of genetic merit for cattle demonstrate that all of this information is only meaningful when understood within very precisely defined farming systems. They constitute something of an alphabet soup: the USDA indexes for lifetime net merit (NM$), grazing merit (GM$), cheese merit (CM$), and fluid merit (FM$) each evaluate types of fitness, but they do so only in the context of distinct approaches to farming. Different systems require different indexes, and the values are purely contextual. In this case, “cheese merit” is something of a misnomer; it does not measure how good an animal is for cheesemaking, but rather how well it will do under a component-based pricing contract. The cow will not necessarily make great cheese, but it will make the farmer selling milk for cheese a bit more money. It is no surprise that an enterprising breed society, like that for the Montbéliardes, should try to monitor better traits for cheesemaking and promote its breed as the ideal cheese solution. Along similar lines, within the United States, Jerseys, the popular breed with the highest levels of fat and protein in their milk, are held up as the ideal cheese cow: with their rich yellow milk, they are the opposite of white water factories. For someone with a background in wine, it all feels rather familiar. The dairy industry is in the process of creating the bovine equivalents of “noble grapes.” In their monumental study Wine Grapes (Allen Lane, 2012), Jancis Robinson, Julia Harding, and Dr. José Vouillamoz described the phenomenon: “In the 1980s and early 1990s . . . it sometimes seemed as though learning the names of a handful of the most popular grape varieties—Cabernet Sauvignon, Merlot, Syrah/Shiraz, Chardonnay and Sauvignon Blanc—was enough to unlock the entire world of wine, so much did they dominate vineyards and therefore labels.” BREED

•

65

The concept of noble grapes created a world where it was easy to “improve” the quality of a region’s wines. Simply rip out whatever obscure local variety might already be planted and replace it with an international favorite like Chardonnay or Cabernet Sauvignon. The consequence was the creation of a so-called international style, wines that had lost their connection to a specific place or culture. There is no more challenging exercise than blind tasting a glass of oaked Chardonnay and trying to guess its origin. In wine, interest has now moved on to the chase for uniqueness, the obscure, and the native. It is almost as if the opinion formers of the wine world had been reading through a guide to cattle breeds written in 1834: “The grand secret of breeding is to suit the breed to the soil and climate. It is because this has not been studied, that those breeds which have been invaluable in certain districts, have proved altogether profitless, and unworthy of culture in others.”47 During an era in which, for the average farmer, breed meant not so much a pedigree as simply the character of the animals from a particular area (almost all of the classic European breeds carry geographical names of a place of origin), the point of breeding was to arrive at an animal that best suited the environment and system in which it lived. Indeed, modern breeders with all of the tools of genotyping and statistical analysis are not operating so differently. What has changed is the steady erosion of the variety of environments in which the cattle live, as, almost by definition, intensive systems operate by divorcing the animals from the surrounding natural world. At each level, it is optimization for the system, rather than a set of innate attributes, that matters most. In the Auvergne, farmers take this even further, referring to the “Holstein system” and the “Salers system” as proxies for ways of farming rather than as definitive descriptions of the breeds involved. The Holstein system, entailing free stalls and feed containing extra concentrates, might readily be used for Montbéliarde or even Salers cows, while the harsh, pasture-only Salers system could equally well be used to raise Montbéliardes or the odd, unfortunate Holstein. A typical Salers, fed on grass, will give around 3,500 liters of milk each year. As we discovered when visiting the experimental farm run by Dr. Marie-Christine Montel’s colleagues, under the same conditions, a (normally massively higher-yielding) Holstein gives about the same amount of milk. The difference is that while the poor Holstein would be starved half to death, the Salers would be in peak condition. The value of the Salers over the Holstein only comes within this system. And the only economic value of this system is that the cheese made with the Salers milk is more heavily rewarded than that made from generic 66

•

BREED

milk. This is the market’s inefficiency, the failure of the collection of indexes and breeding statistics. The economic data used in the index for cheese is entirely taken from commodity market bulk Cheddar. There is no room in the model to price the advantage of earning more money by making a more compelling cheese. This concentration on the headline production numbers is why multipurpose breeds like the American Milking Devon have gone out of fashion. Breeding quite literally gives flesh to abstract values, and the values of the modern dairy world are driven by production statistics. That is the great paradox. Multipurpose animals are not heroic supercows. They do not give extraordinary milk that will dramatically boost yields and cheese quality. The level of solids in Salers milk, for example, is entirely unexceptional. Of course, high solids alone do not necessarily make great milk for all types of cheesemaking: the rich Jersey milk is rarely happy as hard cheese, as the large yellow fat globules are easily damaged and quality suffers. What makes rugged animals like the American Milking Devon special is that they offer the opportunity to thrive in low-input systems in interesting places. Of course, that immediately raises the question: What is an interesting place?

BREED

•

67

FIVE

Feed

The game show Jeopardy! is an American institution and pop culture staple, with over seven thousand episodes broadcast since its debut in 1964. In the 1990s, Will Ferrell satirized it on a long-running Saturday Night Live sketch. When IBM tackled the challenge of building a computer system that could deduce the answer to questions posed in natural language, the company’s goal was to win Jeopardy! (Their system, Watson, took home the first prize of $1 million in 2011.) But even as Jeopardy! provides fodder for late-night comedians and computer scientists, its format also allows us to understand the world of cheese. Let us present the Jeopardy! theory of terroir. “Terroir” is perhaps the most misused word in the world of food and drink, more often than not coupled with an overblown French accent and no plausible mechanism of action. Terroir is the sense of “placeness.” In its strictest definition, it refers to the soil and climate of a particular place and the marks these local conditions leave on the foods grown there. Within the dairy industry, a claim to express terroir is too often simply a case of flashing some quick pictures of happy cows in a pasture, presenting a brief marketing spiel, and moving swiftly along. And this is where our game show comes in. Jeopardy! has a simple twist: the host presents the answer, and it is up to the players to deduce the question. For a contestant’s response to count in the game, he or she must start with the phrase, “What is . . . ?” Exactly the same is true of a cheese. The cheese is always the answer to the question of terroir: it is the sum of its parts. What we have to figure out is: What is the question?

68

A N AG R I C U LT U R A L G E S TA LT

Once we start to think about cheesemaking in the context of solutions to farming problems, we inevitably expand our working definition of terroir from the simple physical circumstances of the cheese to include its social reality as well: we consider not just the environment’s impact on humanity but also humanity’s response.1 Cheesemaking exists within a larger system. Indeed, the daily grind of life as a dairy farmer—the animals must be fed and milked every day, come what may—underlines that cheese exists as part of a holistic system. There are discrete moments during the day, from working with the animals to milking and cheesemaking. But none of these moments has the same impact as the decisions that farmers make at the systems level: the philosophy with which they approach their business and through which they choose to interpret every detail. Cheese gives us a unique opportunity to taste the totality of a farming system, to experience an agricultural gestalt within a single product. In the case of growing cereals, fruit, or vegetables, the consumer can only taste the single end product of the farmer’s endeavors. When animals are kept for meat, the approach to raising them for slaughter might reflect the biodiversity of the farmer’s stewardship of the land, but any sense of our farming at the microbial level is lost at the instant of butchery. It is only cheese that gives us the potential to taste every systematic decision made by the farmer, from macro to micro, encompassing flora and fauna. So what set of problems is answered by each cheese? The original puzzle solved by cheese was straightforward. Some land is unsuitable for growing crops like grains and vegetables. This could be due to poor soil, high altitude, or extremes of water, or it could be that the land is simply too steep or stony to be plowed. But just because land isn’t arable does not mean that it is worthless for agriculture. In fact, land unsuitable for growing crops makes up well over half of the world’s agricultural area, and this is where livestock-based agriculture arose and continues to thrive.2 Pastoralists harvest the energy from sunlight through grass to feed their livestock and then recycle the by-products of production (manure) back into the land. As time went on, exclusively pastoral farming evolved into mixed farming systems. As well as producing cheese and butter, these small farms might have raised whey-fed pigs and poultry and produced beef from their male calves. The fields and pastures were shaped by human activities, but in addition to providing nutrition for livestock, they supported a wide range of FEED

•

69

plants, insects, birds, and other wildlife. Very little was bought in to these extensive systems, and no more nutrients left the farms than could be safely replenished. While today they seem the stuff of rural idyll, such farms were alive and well through the early twentieth century; it was this style of integrated dairying that Bronwen’s great-grandfather Fred brought with him from Switzerland to Southern California. The details depended on the place. Farming practices were molded by all sorts of local factors, including the climate, the accessibility of basic resources like wood and water, and even the routes of goods to market. Craggy mountains presented different problems from marshy lowlands, and cheeses evolved as natural products of their ecosystems, the inevitable answers to blindingly self-evident questions. Transhumance, the practice of pastoralists following their animals to rich high-altitude grazing in the summer while extra feed is grown in sheltered valleys at lower altitude to sustain the herd through the winter, is still practiced on over four million hectares (almost ten million acres) throughout Europe.3 Pastoralists making cheese at the top of a mountain in high summer can rely on feed quality; the cows decamp expressly to reach a diet of superb and plentiful grass. The challenge is the remote location and sparse manpower. In these environments, cheeses need to be made quickly so the herders can get back out and look after their animals, and they need to be stable and long-lived enough to make it back down the mountain later in the season. Large mountain-style cheeses like Gruyère, Beaufort, and Comté were developed to fit exactly these criteria. Lowland areas like western England and the Netherlands presented different challenges. The animals in these places stayed closer to home, and while the lush fields afforded plentiful grass in the summer, making consistently good and plentiful hay (dried grass for winter feed) was a challenge in these cooler, wetter regions. A modern analysis of weather patterns over a twentyfive-year period reveals just how much harder northern Europeans have to work to put up good hay than their sun-drenched counterparts.4 These lowland farmers quickly learned to augment their animals’ grass- and hay-based diets with roots such as potatoes and parsnips. There is a problem with this tack, however: spoilage bacteria present in soil and poor-quality fodder are capable of producing gas within the curd of cheeses, causing them to “heave,” or blow up, during maturation. By the late eighteenth century, dairymaids had begun to add saltpeter (potassium nitrate) to rennet and cheese to prevent this defect.5 But through trial and error, British cheesemakers eventually 70

•

FEED

found different solutions to this climatic challenge: waiting for more acidity to develop during the make and mixing the curds with salt before molding were effective ways to avoid “hoven” cheeses. These techniques led to the use of saltpeter gradually fading away in the United Kingdom. The entire panoply of British territorial styles, from Cheddar to Stilton, is a response to the questions posed by a maritime climate. In contrast, Gouda remains a cheese of both relatively low acidity and salt content, and “late blowing” continues to plague Gouda makers to this day. As a result, almost all makers of Dutch cheese still use potassium nitrate (or other additives, such as the enzyme lysozyme) in their cheeses.6 The one similarity between all original cheesemaking ecosystems was that they made use of marginal land that was useless for growing crops. Yet there is a significant drawback to pastoral farming on marginal land: the yields are marginal too. Over the course of the nineteenth century, integrated farming began to give way to specialization as the mentality of industry was applied to agriculture. Governments and development agencies invested huge sums in helping farmers boost their productivity and optimize every part of their systems, from the grass to the cows to the cheese, for higher yields. Using new tools and knowledge, dairy farms have emerged in places—like the California high desert—so inclement that productive agriculture of any kind there was previously unimaginable. High-desert dairy cows don’t eat the local tumbleweeds but rather forage grown hundreds of miles away, which is then mixed with the by-products of other industries into a ration that delivers optimum energy at minimum cost. This progressive spirit has shaped the practices of farmers in longstanding dairying regions as well. Every farmer in the developed world has access to tools and technology that will help him coax the maximum yield from his land. Where the inputs have become standardized, the outputs have followed. The hidden price of yield has been identity. These modern, “improved” systems have a completely different relationship with the land than that of premodern dairies. Historian Barbara Orland has shown how, for early modern Alpine farmers, the landscape and the use to which it was put were inseparable.7 The vision of the land—what it looked like and how it was managed—was inherently part of the identity of the cheese that was made there. So perhaps it is no wonder that so many cheeses have lost their deep connection with place at the same time that farming has. No longer is the question, “What cheese does my farm and my milk tell me to make?” Now, the question new cheesemakers ask first is, “What cheese does the market want?” Bronwen has spent most of her career as a cheese FEED

•

71

buyer thinking of the latter question as second nature, as common sense. If the market wants high-moisture blue cheeses and sweet Cheddars and technology gives us the tools to make them anywhere, surely that is the very definition of opportunity and success? But before pronouncing the old limitations irrelevant and charging further down that path, we might ask: What are the implications of breaking that old link between landscape, farming systems, and flavor?

“ D O N OT I N S U LT T H E R U M E N ”

To understand the connection between place, system, and cheese, we must start from the ground up, with the relationship between the land and the animal. Superficially, ruminant digestion looks a lot like that of any other vertebrate: food is taken in, nutrients are absorbed, and shit comes out. But ruminants are built to run on a very different sort of fuel than the rest of us. If a human were to shred this book and eat it, he or she would derive little nutritional value from its pages beyond several weeks’ allowance of dietary fiber. For a cow, on the other hand, there is potentially sufficient energy in the cellulose in this book to support the production of around 1.5 liters of whole milk, enough to make 150 grams (a third of a pound) of Cheddar cheese.8 Farming ruminants is our only way of harnessing the energy locked within the grass, channeling it into milk and meat of great nutritional and economic value. We often associate ruminants with “chewing the cud,” the process in which they regurgitate and re-chew their grassy meals repeatedly over the course of many hours. While chewing the cud breaks down the plant matter into small pieces, that process is not what releases the energy. Rather, rumination is a process of microbial fermentation, the by-products of which power the animal. Grass doesn’t feed ruminants; microbes do. The first two chambers of a ruminant’s four-chambered stomach, the “reticulo-rumen,” are a vast microbial culture chamber with a volume of up to two hundred liters in an adult cow. Inside is a fizzy, olive-green brew of saliva and plant material with a vibrant community of bacteria, protozoa, and fungi. As the name implies, rumination is a long, slow process involving up to fi ft y cycles of regurgitation, salivation, chewing, swallowing, bubbling, and belching. The cellulose-eating bacteria in the rumen have evocative names like Ruminococcus and Fibrobacter, and they break down the fibrous material in grasses into sugars, which are then further processed into volatile 72

•

FEED

fatty acids that are absorbed into the bloodstream to power the cow. These cellulose digesters are delicate creatures: unlike the lactic acid bacteria used for cheesemaking, they are highly sensitive to acidity and will shut down if the rumen becomes even moderately acidic. In a classic pasture-based system, the only grain that a ruminant would ever eat would be the occasional seed head from grass that has gone to seed. But in addition to cellulose, the rumen can also metabolize starch. A different set of microorganisms, including lactic acid bacteria, do this job, and in the process they acidify the rumen liquor, just as if it were a jar of sauerkraut or a vat of milk for cheese. If they produce too much acidity, the cellulose digesters switch off. As a result, too much starchy feed quickly disrupts proper rumination; cows evolved to eat grass, not grain. The high-yielding dairy cow is a metabolic contradiction. The new generation of high-performance cows that we met in chapter 4 produce astonishing levels of milk compared to their lower-yielding ancestors. These cows are the Michael Phelpses of the livestock world, elite athletes whose energy requirements make the Olympian’s prodigious consumption of fried egg sandwiches and chocolate chip pancakes look like small potatoes. The cow holding the current world record for milk production, who yielded an average of over ninety liters a day over the course of a year, must have been consuming the equivalent of well over 120,000 calories per day.9 With food intake the most important determinant of potential productivity, every bite needs to count, and on a cellulose-based diet, there simply isn’t time to pack that much energy in. Compared to the unimproved stock of several hundred years ago, even today’s average dairy cows would waste away on an exclusively grassbased diet. Dairy farmers in the United Kingdom sometimes refer to a productive animal losing her reserves as “milking off her back.” A patch for this problem comes in the form of starch, which provides quick-release energy with no lengthy rumination required. But feed a cow too much starch, and the rumen becomes acidic, blocking the digestion of cellulose. Why not just accept that cellulose will be wasted in a nonruminating ruminant and allow the animal to get all its nutrients from energy-rich grains and starchy foods instead? It might sound sensible, but the reality is horrifying. Rumen stasis—the shutdown of rumination as a result of excess starch consumption—is a condition that can occur when beef cattle and sheep are moved from grass-based diets onto feedlots for finishing on grain. Rumen stasis caused by acidosis does not make for happy cows. The rising acidity in the rumen begins to kill the normal population of sensitive FEED

•

73

cellulose digesters and protozoa. Meanwhile, prompted by an influx of readily digestible food, the starch digesters multiply rapidly and lactic acid becomes the dominant metabolic end product in the rumen. The starcheating bacteria also produce viscous slime, which captures gas in the rumen, causing what is called “feedlot bloat,” which can lead to rapid death if animals are not properly acclimatized to the new diet. Even with a gentle introduction, in the presence of excessive starch, the rumen becomes irritated by the high levels of acidity over time and ulcers develop. The rumen bacteria migrate through these perforations into the animal’s bloodstream and liver, causing abscesses and eventually tissue inflammation and infections throughout the body. For ruminants, long-term starch-based diets, despite their high energy, are a direct route to total system failure.10 Nobody keeps dairy animals on the intensive starch diets used on feedlots, but as the productivity and nutritional requirements of dairy animals increase, there is more pressure for them to rely on starch to supply energy. While an “unproductive” cow yielding 5,000 liters of milk per year can glean the majority of its energy from forage (grass-based feed) and thus stay far from the ruminal “danger zone,” a high-yielding cow producing 10,000 liters per year requires a much larger proportion of its energy from concentrates.11 Meanwhile, a Salers cow of the Auvergne producing a very modest 3,500 liters per year thrives on 100 percent forage, consuming no starchy concentrates whatsoever.12 Cows can be picky eaters; comfort, availability, and deliciousness are all important to them, and with the delicate balance of the rumen at stake, many dairy farmers have switched to premixed complete feeds, known within the industry as total mixed rations, which contain an optimal ratio of cellulose to starch, protein, and vitamins. Acidosis can still be a problem, so there are other tools available to give the carefully calculated high-energy rations a further helping hand. Feed companies offer a variety of buffers based on seaweed, ground oyster shells, or sodium bicarbonate for use as feed additives. Often, the grains themselves are pretreated with caustic soda, which is alkaline and thus provides a neutralizing effect on the acid produced by their digestion in the rumen. Antibiotics have also been co-opted to influence the rumen microbiome and tip the balance in the right direction. Ionophores are a class of antibiotics that are selectively active against acidifying rumen inhabitants, like the lactic acid bacterium Streptococcus bovis. By knocking out this and other sensitive members of the rumen community, farmers can increase the level of starchy concentrates their animals consume while keeping them ruminating. 74

•

FEED

All of these interventions, if not perfectly honed, have the potential to take a heavy toll on the animals. Throwing a cow’s digestive system out of equilibrium can easily end up poisoning or starving the animal. An old textbook of cattle nutrition sums up the dictum best: above all else, “do not insult the rumen.”13

FA S T E R , H I G H E R , G R E E N E R . . .

Employing ruminants to harvest energy from grass sounds straightforward enough, but like every system that requires equilibrium, livestock farming is an exercise in careful control. There are many ways that things can go wrong. For example, if undergrazed, pastures immediately begin to revert to forest, while overgrazing leads to compacted soil, weed encroachment, and erosion. The following year’s food security hinges on protecting the health of the environment and the balance of the system, and farmers have a vested interest in never taking out more resources than can be reinvested in their land. In societies where pastures are publicly owned, rules for land use are rigorously enforced by the entire community to ensure that the assets of the group are protected.14 In cold continental climates, cattle can graze through snow in the winter and remain outside year round, but in wet areas, cows outside in winter will rip up the dormant pastures as if they are treading in chocolate cake. The European tradition, which was also transplanted early on to the dairying regions of the Americas, entailed feeding animals on grass during the summer while hay was conserved for the colder months. Over the winter, the animals were moved indoors and fed the dry food set aside from the summer surplus, and because they were often not milked during the winter, lessenergetic food was required during this rest period. In the spring, the next generation arrived, and the cycle began afresh. Making good hay was not a given. A beautiful account of haymaking in Wensleydale, in the county of Yorkshire in England, written in 1811, captures the energy and resourcefulness the residents of that wet, stormy area employed to outsmart nature year after year: “A desire of winning their hay crops well, as they have very little coin, long since incited the farmers to pursue various methods, and to make comparative trials, till at last they decisively and justly concluded the present method eligible and preferable; and by the unvarying practice of which they acquired their acknowledged and merited celebrity in haymaking.”15 They scythed at lightning speed. If the FEED

•

75

weather threatened to turn, the families took to the fields to mound the hay into small, smooth heaps, known as haycocks, which were carefully constructed to repel dampness and wind. The hay might be gathered and restrewn across the field many times before it was dry enough to achieve the desired quality. A livestock farmer from Maine wrote in 1832 about how the wet weather and inadequate drying period had made his hay “verry [sic] black with a sour, mouldy, musty smell.”16 Such spoiled hay, fi lled with potentially toxic mold spores, was worthless as food and dangerous to farmer and animal. Making good hay in the face of climatic uncertainties required knowledge, tools, and skills as specialized as those for cheesemaking. Accounts of preindustrial haymaking are populated with objects of strange emotional resonance such as scythes, haycocks, and horse-drawn sledges. Like fairy tales, they are at once both familiar and remote, but the fate of the animals and people over the impending winter once hung on those tools and the farmers’ skills. In early nineteenth-century England and the northeastern United States, hessian or cotton caps were sometimes spread over the haycocks to protect them from damp, and there are accounts of salt being mixed with partially dried hay to help it cure.17 When hay was being made, skilled labor was at a premium; the cost of potential failure was even higher. Since the ultimate challenge was to get the hay sufficiently dry, the system favored a late harvest, which was carried out after the grasses and flowers in the meadows had already begun to shed their seeds and lose their lushness. Many of those seeds dropped onto the ground as the hay was repeatedly raked, turned, and stacked, and were propagated the following year. During the winter, muck from the animals’ housing was spread back onto the fields, fertilizing them. In this way, what was taken out of the system was put back in. It is easy to idealize these old farming systems, which worked in concert with the natural rhythms of nature and were almost entirely self-sufficient. But the truth is, it was not an easy life. There was a reason these subsistence farmers had “very little coin.” Yields were extremely low by today’s standards. There were periods of extreme physical labor, particularly before the introduction of mechanization, when all the work was done by hand and horse. Without electricity, animals were milked by hand, and small amounts of cheese had to be made every single day during the season because there was no other way to preserve the milk. Woven into the fabric of those systems was plenty to reject as backward, difficult, and entirely inefficient. 76

•

FEED

It is in the context of this world that we must understand the two great innovations that, operating hand in hand, revolutionized livestock farming: nitrogen fertilizers and silage. Nitrogen fertilizers have the striking capacity to increase plants’ growth rate and yields, easily more than doubling the yield from an unimproved hay meadow.18 The nitrogen cycle is fundamental to all life on Earth. It is also perhaps the single most significant instance of a twentieth-century technical innovation reordering our relationship with the natural world. Proteins cannot be made without nitrogen: each of us needs to eat all ten essential amino acids in order to synthesize the proteins we require to grow and maintain our bodies. Whether we eat plants directly or the products of animals that have eaten those plants, we depend on the cycling of nitrogen. Before the twentieth century, this cycling was achieved through biofi xation by Rhizobium bacteria symbiotic to legumes and by cyanobacteria, through deposition from the atmosphere, and through recycling in the form of manure and crop residues.19 The practice of farming, with its crop rotations, muck spreading, and structured control of grazing is, more than anything else, an exercise in nitrogen management. But as farming systems applied the logic of industrial production and became more intensive, it was access to nitrogen that was the limiting factor in their productivity. This was well known by the late nineteenth century, and other sources of nitrogen were sought: bird guano has a higher nitrogen content than bovine manure but was strictly limited in supply; the same was true of mined Chilean saltpeter (naturally occurring mineral sodium nitrate).20 The solution came on July 2, 1909, in Karlsruhe, southwestern Germany, with the first trial demonstration of what would eventually become known as the Haber-Bosch process. Using metal catalysts and high pressures and temperatures, the process converts atmospheric nitrogen to ammonia through reaction with hydrogen. It was the defining—if largely unknown—event of the twentieth century, and chemists Fritz Haber and Carl Bosch would both be awarded separate Nobel Prizes, the former for the invention and the latter for its implementation. In the words of Professor Vaclav Smil, the HaberBosch process was the “detonator of the population explosion” that took the world’s population from 1.6 billion in 1900 to over 7 billion today.21 It was the literal detonator of many other explosions too, as it was a key source of the nitrogen required for explosives in the twentieth century’s bloody wars. Using nitrogen to promote growth was so instantly rewarding, and the results so visible, that its allure was hard to resist. Fertilizing with nitrogen FEED

•

77

rather than—or in addition to—farmyard muck has become standard mainstream agricultural practice around the world. Heavily fertilized fields allow farms to sustain many more animals than the land would have supported previously.22 Anyone who has ever dissolved a lurid blue spoonful of Miracle Gro to feed their garden is familiar with the concept of nitrogen fertilizers, but silage is a more remote and mysterious institution. Silage is a blanket term for fermented fodder, which can take many forms: grass silage is a chopped-up and fermented analog of hay—its relationship with fresh grass is the same as that between sauerkraut and fresh cabbage—but silage can also be made from cereal crops like corn, barley, oats, and wheat. When these grasses are harvested for human consumption, we are only interested in their mature seeds, but when they’re turned into silage to feed cows, the entire plant is harvested and cut up, and the resulting vegetal confetti is allowed to ferment. Silage fermentation is anaerobic—oxygen impedes the process and permits the growth of spoilage molds—and so the process is carried out in an enclosed vessel—typically either a silo or a large concrete enclosure with a weighted cover—or as individually wrapped bales. Silage has been an essential part of the diets of dairy cows in the United States since the late 1880s, when a slew of silos appeared in the Northeast and Midwest, mostly for the ensilage of corn. The effect of corn silage on dairying practices was profound; whereas farmers had previously been forced to sell much of their livestock in the autumn because of the prohibitive expense of feeding them through the winter, the introduction of silage made it affordable to feed more cows in the lean months from the same amount of land. In Wisconsin in 1866, the conversion of the masses was already well underway: “Nearly every owner of a silo is visited almost daily; farmers come five, ten and twenty miles, some singly and some in couples and sometimes in platoons; in short, farmers are flocking to the silo like pilgrims to the Holy Land.”23 Making grass silage is attractive for essentially the same reasons: it allows for higher yields at lower cost. Because silage does not rely for preservation on being dry in the same way as hay does, grass can be cut when it is younger, greener, lower in fiber, and higher in protein and energy. By cutting young grass, two or even three successive cuts of silage can be harvested where once it was only possible to collect one crop of hay and perhaps offer animals a little grazing of fresh growth (the “aftermath”) in the months that followed. For sheer efficiency in getting energy from a field of grass, silage can’t be beat: it remains the agricultural scientist’s preferred method to enhance yield. 78

•

FEED

Tellingly, the first great spikes in silage production in the United Kingdom came during the two world wars, as the government regarded farming yields as a strategic resource.24 As an added benefit, due to its higher moisture content, silage is much less weather dependent for its success than hay. Where hay requires constant attention and laborious physical manipulation to dry it down to a stable moisture content of around 20 percent, grass for silage can be successfully baled with more than twice as much moisture in the leaves; after just a brief wilt, it is ready to be carted away and packed up for fermentation and storage.25 With all of these clear advantages for wet and rainy climates, the fact that silage took so long to be popularized in the United Kingdom is surprising. More than a century after silos first became common in America’s dairyland, silage still had not really caught on across the Atlantic, aside from the brief flirtations during wartime; in fact, the level of total hay production in the United Kingdom peaked in 1971. Then, in the early 1970s, amid high labor costs and widespread mechanization, silage production in the United Kingdom suddenly took off, to the extent that today it is utterly ubiquitous.26 Silage has taken on something of an air of timelessness, sometimes even tinged with local pride. Visit a British dairy farm and you will quite likely be invited to take a look at the farmer’s “chopper,” the farmer reveling in the schoolboy smut (“Oh yes, mine’s a big one!”) of the double entendre. The methods of one’s grandfather are those one regards as definitively traditional, and British farmers now in their sixties have known no other farming system. How could any other way of feeding cows be possible in such a wet place as the United Kingdom?

FERMENTING A PROBLEM?

Silage allows farmers to get maximum productivity from their grass and to efficiently and reliably harvest and store the excess during times of peak production for use over the winter. Admittedly, there were initially obvious disadvantages. Early guides conceded that manual silage making without proper equipment was almost impossibly difficult: all that extra moisture more than doubled the weight of the product, which farmers first had to cart in from the fields and then, after fermentation, carry to the cows. There were also concerns that cows had trouble eating as much dry matter in the form of silage as they could in the form of hay.27 But silage, particularly coupled with FEED

•

79

the use of nitrogen fertilizers, has also brought with it disadvantages that have taken longer to reveal themselves. Because grass for silage is cut much younger than hay, the plants have not yet gone to seed when they are harvested. Meadows that have only ever been cut for hay are home to a wide range of plants, not just grasses but also flowers and herbs; in some, over seventy species cohabitate.28 Many of these are annuals with no bank of dormant seeds waiting in the soil; they must flower and drop seeds every year to maintain their place in the community.29 The act of cutting early and often quickly extinguishes these species. Fertilizer intensifies the pressure on marginal species further, as it allows leafy grasses like perennial ryegrass to muscle their way to dominance at the expense of the flowers and herbs. Over time, new grasses have been developed that are even more vigorous and nutritious. With their lush green leaves, they are so wet that making hay from them is impossible; they are grasses bred specifically for silage.30 Just a few seasons of fertilizer application, silage making, and seeding with selected varieties of these new grasses are all it takes for fields to be “improved” to the point where they support only a few shades of green. This loss of plant diversity has secondary effects on other species. The wide range of flowers in unimproved pastures provides a paradise for insects including butterflies, bees, and grasshoppers. Those insects and the seeds produced by the flowers and grasses provide food for birds, while the meadows themselves provide shelter for ground-nesting birds, rodents, hares, stoats, shrews, moles, and hedgehogs. Bees and butterflies depend on flowers; they have little use for perennial ryegrass. And while a person scything by hand can spare a nest, a large forage harvester will not. The destruction of their native habitat has driven many species of ground-nesting field birds, such as the corncrake in the United Kingdom and the eastern meadowlark in the northeastern United States, to near extinction.31 Silage brought the answers to some problems, but it opened up a new set of its own. These problems include serious practical limits when it comes to making cheese, particularly low-acid and soft varieties. Brothers Mateo and Andy Kehler founded Jasper Hill Farm in Greensboro, northern Vermont, with the recognition that making cheese held the key to producing a product of great value from severely marginal land. The value of the grass of Vermont’s Northeast Kingdom is the founding principle of their entire operation. As their business grew, they looked to expand, and with that came the prospect of adapting the farming systems of some of their neighboring farms. 80

•

FEED

It was a challenge. With clear-eyed logic, Mateo described the process of buying nearby Andersonville Farm and converting it from producing liquid milk to raw-milk cheese. The microbiological test results from the raw milk were, at first glance, brilliant, no different from the milk at their own small farm, which feeds exclusively hay. But longer-term analysis of the milk filters (the most sensitive practical form of milk testing) showed intermittent presence of Listeria species—which can grow in pockets within silage—at minute levels in the milk. Given that cheesemaking is a process of exponential microbial growth, the brothers decided that making raw-milk soft cheese with milk from silage-fed animals was too great a risk. But with their eyes on European best practice, rather than accepting the inevitability of pasteurization or switching to producing a style of cheese adapted to the challenges of milk from silage-fed cows, the Kehlers embraced a technological solution to their problem: the hay dryer. Their six-thousand-square-foot barn dryer uses technology developed in the Reggio Emilia region of Italy, the home of Parmigiano-Reggiano, to dry hay in thirty-six to forty-eight hours of sunshine rather than the five days it would take in the field. Its energy requirements are more than offset by the farm’s solar array. The hay dryer is the first of its kind in the United States. Back when herd sizes were small and stocking densities low, producing hay the old-fashioned way in places like Vermont was just about manageable. But by the late nineteenth century, methods for artificially drying hay were already being adopted by enterprising farmers.32 For the Kehlers, the hay dryer was a game changer, allowing them to feed their herd of cows at scale in a way that would otherwise be impossible given their climate. Mateo told us with pride: “We’re feeding two hundred milking cows; there’s nobody else in the state or probably in the Northeast feeding that many dairy cows a dry ration.” The hay dryer cost almost a million dollars to build, but given the cost of buying in good hay, they expect it to pay for itself in just five years. “We can produce dry hay with the hay dryer for a fraction of the price,” Mateo explained, “but the real payback is in quality. We’re able to put up this feed that’s amazing. Bright green hay, with 90 to 95 percent dry matter. It’s the best and most efficient way of feeding a cow.”

D R . R E B LO C H O N

Diverse and species-rich grasslands are in themselves a public good, just from an environmental perspective. This is the impetus that propels most approaches FEED

•

81

to their conservation. But instead of consigning these landscapes to national heritage conservancies, can they provide a real farming opportunity? To put it simply, does increased biodiversity provide a difference we can taste? To address this question, we look once again to an INRA facility in the Auvergne, this time in Theix, a suburb of Clermont-Ferrand, about two hours’ drive north from Dr. Montel’s lab in Aurillac. We have come to visit Dr. Bruno Martin, an agronomic scientist at INRA’s Herbivore Research Unit. Martin’s PhD research linking feeding systems to cheese characteristics in Savoie earned him the nickname “Dr. Reblochon,” which he wears with quietly bemused pride. His work was also swiftly adopted by the Reblochon producers’ group as part of their Appellation d’Origine Protegée (AOP) regulations. France was the first country to establish a system of quality marks that linked products to places; the integrity of the AOP and its pan-European equivalents depends on a demonstrable link between flavor and place. Martin’s work over the past twenty-five years has explored the forms that link takes in milk and cheese and how farming choices influence the flavor and texture of cheese before the milk even leaves the animal. A passionate native of Lozère—home of the Aubrac cow—Martin is as excited about the region’s gastronomic traditions as he is about his work. At dinner, he shyly serves us the excellent saucisson sec that he made with his young son. Too much an empiricist to be dubbed a guru, Martin and his colleagues are steadily, through long-running and rigorous controlled experiments, making the case that extensive farming systems produce demonstrably the most interesting cheese. His work has also made him an instrument of French soft power on the international stage: when we visit, he has just returned from a trip to Bhutan, where he advised the government on the creation of their own AOP equivalents. Now officially the cheesemonger to the Queen of Bhutan, he brought home a kilo of yak butter to share with his colleagues. When we visit Martin’s lab, our first stop is at the experimental farm nearby, where his team has been testing new ways to reduce the amount of methane produced by ruminating cows. He shows us their methane-capture chambers, the only ones of their kind in France, designed to measure variations in methane emissions linked to different diets. Having just enjoyed a lunch of haricots blancs at the canteen, we glance nervously at each other, but fortunately the object of the tour is not to capture methane from guests. Instead, we move on to look at a head-mounted camera designed to reveal the 82

•

FEED

plant preferences of grazing sheep (watching the videos and cataloging the animals’ choices is a task for long-suffering graduate students). Almost all dairy farmers have a story of when their animals ate something pungent and the milk the next day told the tale. The sulfur compounds in cabbage and wild garlic pass through the bloodstream and into the milk of dairy animals; a similar process occurs in our bodies, which is why you can taste raw garlic or onions in a meal for many hours after eating. But looking beyond volatile sulfur compounds from the odd rogue plant, Martin asked: How does the diet of the animals affect the character of cheese? A series of controlled experiments, where all factors but the feed were held constant, turned up clear patterns.33 You can tell a lot about the milk that went into a cheese by its color, at least for cow’s milk cheeses. That is, after all, the whole point of cheesemakers dying their cheeses orange with annatto. Grass contains high levels of the reddish-orange carotenoid beta-carotene, which is absorbed into the bloodstream as the grass is digested. The digestive systems of goats and sheep rapidly convert this pigment into colorless vitamin A, making their milk snowywhite regardless of the amount of grass in their diet.34 Cows lack the mechanisms to do this, and the color of their milk is directly linked to the levels of beta-carotene in their food. Ultraviolet light degrades beta-carotene, so milk and cheese from fresh grass and from grass silage, which does not need to dry in the sun for long, are the deepest yellow in color. Hay that has been dried in the sun for days is left with a lower level of pigment. Corn contains barely any beta-carotene, and so milk from cows fed on corn silage and grain make the whitest cheeses. The feed that animals eat also has a powerful effect on the fatty acid composition of their milk. While the total fat content of milk is proportional to the level of fiber in the diet of the animals (an odd concept for a human dieter), the degree of saturation and the length of the fatty acid tails are linked to which plants they are eating. Just as with grass-fed beef, the fat in milk that has come from cows eating only fresh grass is higher in polyunsaturated fatty acids and conjugated linoleic acid than milk from cows fed preserved grass feeds like silage; diets based on corn or containing more grain produce milk with a higher proportion of saturated fat. Because unsaturated fats have a lower melting point, milk from grass has a softer, more giving texture. Butter made from grass-fed cows is also more spreadable when cool. Of course, cheesemaking plays a huge role in cheese texture: Camembert made from the milk of cows consuming corn silage and grain will always be FEED

•

83

softer than Cheddar from the milk of cows eating grass. But all else being equal, the grassier the diet, the mellower the texture of the resulting cheese. Flavor, more than color or texture, defines the quality of cheese for most consumers. Here as well, the differences are there, and they are statistically robust. The impact of feed on cheese flavor and aroma has been explored on two levels. The simplest comparison measured the contrast between cheeses from cows fed grass versus cows given non-grass-based feed. Further experiments then examined the impact of plant species diversity on the flavor and aroma of cheeses from exclusively grass-fed cows. On the experimental farm, a herd of Montbéliarde cows was split into three groups. One group lived indoors and ate hay and concentrates containing corn, barley, and soybeans. The other two groups lived on pastures managed in two different ways: one group was put into an intensive grazing system with a high stocking rate (the cows were presented with a fresh strip of grass each day, which they grazed all the way down before moving to the next), while the other group was let loose in a large, unfertilized, highly diverse field and allowed to graze freely. Scientists from the cheese division made identical cheeses from the three milks during the experiment. When presented with the cheeses, the blind-tasting panel immediately identified the cheeses made from the milk of the indoor cows, even at a young age. In addition to having a paler color and firmer texture than the two types of cheeses from grass-fed milk, the indoor milk cheeses also had less intense and less sour aromas and flavors. On the other hand, the panel members could not tell the cheeses from the two grazing systems apart when they were tasted at twelve weeks of age. As the grass-fed milk cheeses aged further, however, they became more and more distinct. By twenty-four weeks, the sensory panel had more success in differentiating the cheeses from the two different grazing styles, to the extent that the difference became statistically significant, albeit not as pronounced as the difference between both grazing methods and the indoor system.35 The reasons for the flavor differences (as opposed to those of color and texture) due to feed are the hardest to pinpoint, but Martin and his team believe several factors might be at play. First, fatty acid breakdown products are a key source of cheese flavor, so it follows that the diet’s effect on those precursors determines the way those flavors develop during ripening. A second theory involves molecules called terpenes. Terpenes are a family of aromatic molecules produced by herbs and flowers found in species-rich pastures, including yarrow, chamomile, fennel, and many others. When 84

•

FEED

extracted directly, terpenes give many natural essential oils their scent, and when livestock eat plants containing them, they pass into the bloodstream and through to the milk, just as beta-carotene does. However, while the terpenes can be detected in milk, research groups have failed to detect their flavors in cheeses. But evidence is building to support a more intriguing possibility: that terpenes—many of which have antimicrobial properties—may work indirectly by influencing the growth of bacteria in the cheese.36 That’s right: the flowers that an animal eats in the field may well turn out to affect flavor by controlling the behavior of the microbes within cheeses made from the animal’s milk.

THE HILLS ARE ALIVE

The next morning, we see theory in practice. We set out with Martin to visit his friend Patrice Chassard, whose family has been making Saint-Nectaire cheese at their Ferme du Bois Joli for almost sixty years. The farm is located in the middle of the Auvergne Volcanoes Regional Park, and the domes and craters stand out against the bright blue sky. The Ferme du Bois Joli is a poster child for environmentally sensitive pasture management in the service of delicious cheese, and Chassard has collaborated with many INRA scientists, including Martin, over the years. Their system is extensive, with ninety milking cows and seventy heifers on their 240 hectares of permanent pasture. More productive systems might look to keep up to three-and-a-half times that number of animals on the same amount of land. The cows eat a diet of 85 percent hay and 15 percent non-GMO cereals to keep them in condition. Even the storied Montbéliardes cannot survive on grass alone, unlike the rugged Salers. After a cursory glance at the creamery—by midmorning, the cheesemaking is almost completed, though they’ll repeat the process in the evening with fresh milk from the afternoon milking—we visit the adjacent barn, where the animals are housed over the winter. An incredible sight greets us there: the hangar-sized barn is packed from floor to ceiling with hundreds of massive round bales of hay. Chassard digs into one with his hand and liberates a fistful of different grasses, dried seed heads, and withered flowers. All of his hay is dried in the sun, and its color is slightly bleached compared to the brilliant green of the artificially dried hay we have seen at other farms. But its aroma is vivid, sweet, floral, and sappy. FEED

•

85

There are no cows here; this food has been set aside for their winter ration. The cows are at their summer resort, a swath of high-altitude pasture about twenty minutes’ drive up the mountain. In the winter, this is a popular ski destination, Super Besse, but on this summer day, it is practically deserted, and we park in front of a vacant chalet and wander over to greet the cows, who are spread out across the slope, lolling on the grass and soaking up the summer sun. They are milked here in the field morning and evening with a mobile milking unit. On this summer day, it seems an easy life. But Martin’s attention is on the grass beneath our feet. He picks out different species for us to see, rub between our fingers, and sniff. The game is afoot, and he strides ahead, stopping for us to catch up and smell the samples he has found. As we examine the plants, he talks about the work of his group and his colleagues in the international mountain-cheese network. Sniffing one sample, he tells us about a group working at a nearby experimental farm mapping the “odorscape” of different fields containing these diverse species, looking only at the volatile compounds they release into the surrounding air and discovering their relationship with the insect species found in the same area; they plan to go on to experiment with milk.37 Biodiversity is smellable. With a chuckle, Martin twirls in the mountain pasture like a Gallic Julie Andrews: the hills are alive with the smell of terpenes. Chassard manages the Ferme du Bois Joli to maximize the diversity of the pastures, with the knowledge that doing so will produce better and more interesting cheese and benefit the environment. But even in the cheese Narnia that is the Auvergne, the same forces and pressures that plague milk producers elsewhere are lurking at the gate. Saint-Nectaire is one of the most successful AOPs in France: an astonishing 210 farmhouse cheesemakers produce over seven thousand tons of the raw-milk cheese each year. But the same AOP also makes allowance for large-scale factory production of pasteurized Saint-Nectaire, and four factories produce as much of the cheese annually as all the farmhouse producers combined. As the sitting president of the SaintNectaire producers’ organization, Chassard oversaw an amendment to the requirements in 2007 that tightened them considerably. The most contentious change was a ban on silage for all members—farmhouse and factory— effective from 2017. The proposal was heavily contested, and the vote was tied, so as president, Chassard made the decision to go forward with the amendment. When the amendment goes into force, any Saint-Nectaire made with silage will lose its AOP status and become generic tomme de montagne (mountain tomme), which has a much lower value. As the date gets nearer, 86

•

FEED

the producers’ group is under significant pressure and political strain. In theory, the ban was meant to take place within ten years, but the deadline has already been extended to 2020. Chassard remains steadfast in his conviction that the difficult choice was the right one: “Only by protecting and increasing the integrity of our cheese will we secure our future.”

THE VIRTUOUS CIRCLE

Restoring the link between dairy farming and place is possible, but the caveat is that turning away from more than a century of agricultural innovation means letting go of the measures of success that we have grown to prize. For decades, higher productivity really did mean higher profitability, but oversupply and the vagaries of the market have led to an inevitable conclusion: being an artisan-scale—or even a medium-sized—producer of a commodity like milk is no way to make a living. Small dairy farmers’ budgets are stretched to their limits, and margins are low; a significant drop in yield under such circumstances signals oblivion. But what if we change the rules of the game? Extensive systems are also lower input: they require less bought-in feed, fewer concentrates, and less fertilizer. Grass-based cattle farmer and author Julius Ruechel puts it well: “It isn’t enough to eke out a living from our maximum productivity. To maximize our profit, we need to adjust our priorities toward minimizing expenses and addressing logjams in our production systems before we focus on boosting production.”38 Even when remaining within a commodity system, a reconfiguration toward a more extensive approach can lead to a more profitable balance. There are drawbacks. First, you can’t do it everywhere; the high desert of California, for example, is not marginal land, it’s downright inhospitable. Without bringing in feed and water from afar, such desert dairies are an impossibility. Extensive dairy farming is also unlikely to stack up economically in areas of rich arable land, whose natural attributes give it potential to make more profit in other ways. Unfortunately, you cannot just flip a switch, stop applying nitrogen fertilizer to a field, and return the following year to find an ecologically diverse hay meadow in its place. Using nitrogen fertilizer disrupts the soil’s microbial community and the natural cycling of nutrients within the system. Most naturally occurring nitrogen in soil is found within organic matter, which requires microorganisms to break it down before the plants can access it. FEED

•

87

When those microbial communities have been disrupted and fertilizer is withdrawn, there is no way for the plants to access the nutrients in the natural organic matter all around them. Restoring the system takes time. A study carried out on a large estate in the process of converting from conventional to organic agriculture showed that the first phase after nitrogen stopped being applied was a significant decrease in grass yield without much change in species richness, but between years six and eighteen of the study, the number of species present more than tripled. Simply stopping the nitrogen was sufficient to restore a degree of diversity, assuming farmers could afford to wait ten to fi fteen years for it.39 Ecologists tasked with restoring the old hay meadows of the Yorkshire Dales are finding ways of helping the process along. The results are best if the soil fertility of the fields to be restored is already low, as the higher the richness, the more the competitive grasses push the others aside. The ecologists modify leaf blowers to work in reverse, collecting mixtures of seed heads to propagate elsewhere, and organize “hay transplants” between donor and recipient fields. Even after taking these extra measures, the ecologists estimate that it takes between six and fourteen years of careful resowing and cultivation for a field to regain a passable similarity to a northern hay meadow, and between twenty and forty years to achieve an exceptional level of diversity.40 One thing they’ve learned: livestock are integral to the process. Grazing in the spring and autumn aerates the soil surface, and the animals fertilize the fields with organic nitrogen from their urine and manure. Without the presence of the animals in the right amount and at the right times, certain species are lost. Teams of ecologists are working to understand the patterns of livestock management that best support the ultimate objective of healthy, diverse meadows.41 In contrast to Bruno Martin, Patrick Holden is a guru. He has farmed organically at Bwlchwernen Fawr, a 130-acre hill farm in west Wales, since 1973. It is the longest-standing registered organic farm in Wales, and it is now home to Hafod cheese.42 Holden was the founding chairman of British Organic Farmers in 1982, and in 1988 he joined the Soil Association, the charity that is the leading British voice of the organic movement and a key certifier of organic standards. He served as director of the organization from 1995 to 2010. Now, as founding director of the Sustainable Food Trust, he is at once a farmer, an activist, and a public intellectual specialized in extensive farming: “The type of farming we’re trying to practice has been rejected in favor of modern industrial farming methods. . . . The system itself is confined 88

•

FEED

to the margins, not mainstream at all.” He believes that, given the problems of resource depletion, climate change, and the “devastating” environmental impact that industrial farming has had, “there’s a need for all farming to change to become more sustainable than it has been for the last sixty years.” His conversations with other farmers and policymakers and his work as head of the Sustainable Food Trust have led him to question the tenets of the economic system that has given rise to intensive agriculture: “It’s been my growing conviction that there would be a case for doing more of what we’re doing if the economics weren’t so artificially skewed in ways that destroy the business case. . . . Emissions, pollution, damage to public health, damage to biodiversity, natural capital depletion—these have not been costed, and they give an illusory cheapness to mainstream products.” Even if we are not paying the real price for unsustainable food-production practices in the meantime, the theory goes, as a society, we will foot those bills eventually, whether through health care, environmental remediation schemes, or the decreasing fertility of our soil. Holden is on a mission to change the system so that it rewards the choices that are good for the environment rather than the shortterm patches, moving sustainable agriculture out of the fringe and into the mainstream. He believes that sensitive use of marginal land is essential for minimizing the environmental toll of food production. Others, such as journalist George Monbiot, believe that such land would be better left to revert to nature. He cites overgrazing, soil compaction, and erosion as inevitable consequences of this style of grassland management. While it is true that abuse of marginal land through overstocking does have these effects, it is telling that Monbiot cannot imagine a world where livestock farmers look beyond the short-term, or where grassland and meadow ecosystems have as much ecological merit as forest does. More than anything else, Monbiot refuses to contemplate a world where the economic return that farmers receive is entirely dependent on the tasteable consequences of their farming choices or consider that there could be anything other than a straightforward linear relationship between yield and profitability.43 British rural geographer Dr. Henry Buller was dissatisfied by the widespread belief among people like Monbiot that pasture-based animal husbandry on marginal moorlands and grasslands is an economic charity case. A Francophile familiar with the work of Bruno Martin and his French colleagues, Buller set out to explore the theory that the flavors of biodiversity confer extra economic value to products. He measured the “floristic richness” of both extensive farms and intensive control farms and then subjected samples of their lamb, beef, or FEED

•

89

cheese to a battery of chemical and sensory tests. The results showed a clear relationship between farming system and flavor for the meats, but less so for the cheeses. In some ways, this finding is not surprising, as the way the cheeses were made was not controlled for in his experiment. Martin’s group has already reported that processes such as pasteurization and microfiltration disrupt the connection between the milk and the cheese flavor; they level the playing field between interesting milk and commodity milk.44 The use of strong starters— let alone aggressive molds or ripening strains—has a similar effect. Whereas meat needs no transformation before it reaches the consumer, it is sadly all too easy to negate the influence of even the most interesting milk through brutal or sloppy cheesemaking. In his report, Buller writes that “biodiverse rich grazing systems are not a universal panacea and there are structural limitations but we do think that they offer a new potential for producers to create additional value in the food chain, particularly in those areas where the alternative is often ecologically damaging forms of intensification or abandon.”45 He proposes what he calls the “Eating Biodiversity Chain,” which shows how biodiversity can play the role of both output and input within a virtuous circle, where the higher economic return obtained by dint of unique and delicious flavor funds an investment in the continued richness of the system (see figure 6). The most marginal land is the most diverse, and for cheese, it holds the most potential. When cheese is made in a way that allows us to taste the distinctiveness of sensitively farmed raw materials, it once again becomes the answer to the unique set of challenges posed by each and every farm. Only cheese can turn those challenges into flavors that make good farming sustainable. This is something that the most perspicacious cheesemakers have already intuited. The owner of Uplands Dairy in Dodgeville, southern Wisconsin, cheesemaker Andy Hatch speaks passionately about the link between his farming system and his cheese. The son of a Milwaukee lawyer with a passion for the wines of Burgundy, Hatch has thought deeply about the connection between the land and his cheese. Thanks to the work of the founding owners, Mike Gingrich and Dan Patenaude, Uplands Dairy and its cheese Pleasant Ridge Reserve are—unlike most American dairies and cheeses—named after specific geographical phenomena rather than given a personal or fantasy name. At Uplands Dairy, place matters. Pleasant Ridge Reserve is also the only cheese ever to win overall champion at the American Cheese Society’s annual competition three times, including a win in 2001 in its first year of production. 90

•

FEED

Intrinsic natural value

Farm management practices

Pasture species diversity Farm microbial diversity

Animal feed composition Milk microbial composition

Economic viability

Cheese quality Cheese taste

Explicit market value

FIGURE 6. The cheese biodiversity chain. Adapted from Henry Buller and Carol Morris, “Eating Biodiversity: An Investigation of the Links between Quality Food Production and Biodiversity Protection; Full Research Report; ESRC End of Award Report, RES-224–25–0041” (Swindon: Economic and Social Research Council, 2007).

“We’ve never planted anything especially exotic here,” says Hatch, “mostly because our chemical and mechanical management of the pastures is pretty minimal, so the species composition will probably always drift toward what’s best suited for the soils and climate. We focus instead on maximizing diversity and especially on the stage of growth at which the cows harvest the grass.” Hatch and his business partner, Scott Mericka, bought Uplands Dairy from Gingrich and Patenaude in 2014 after serving long apprenticeships, and they aim to achieve a balanced mixture of grasses, nitrogen-fi xing legumes, and wildflowers in their pastures. “The wildflowers appear on their own and we don’t discourage them, but we don’t sow them, and I couldn’t really trace any certain aromas or flavor to them, although when we make hay in a field with a lot of flowers, the hay aroma is noticeably different.” Hatch now makes two cheeses, each produced at a different point in the year. In the summer, when the herd is exclusively on grass, he makes Pleasant Ridge, his hard cheese; the second cheese, Rush Creek, is an oozy, barkwrapped Vacherin style made in the autumn, when the cows are eating conserved fodder. “With the Pleasant Ridge,” he explains, “we acidify the milk, dry it down, get out of the way, and wait to see how good it’s going to get.” When he was developing Rush Creek, he took a different tack, recognizing that that milk had less inherent complexity but also more fat: “We looked to FEED

•

91

that [autumn] milk as a canvas to which we could apply flavors generated more through cultures, the rind, and the spruce band. . . . So technically, making that cheese was completely different from the hard cheese, and philosophically, or at least mentally, it was also completely different.” While the sensuous Rush Creek sets social media afire each November, Hatch’s heart belongs to the Pleasant Ridge Reserve and its ability to communicate the quality of his pastures: “If you look at our advantage, we have a special piece of ground, a special herd of cows. . . . Our advantage is in producing distinctive milk. I don’t know that my skills as a cheesemaker or affineur are so much better than other people’s. Our ability to produce special milk is more unusual, and so that’s where I see our advantage.”

92

•

FEED

SIX

Microbes

In the middle of the bustling creamery, we stand entranced by the curd: the milk has been transformed into a glassy gel. Set in small, wheeled basins, it quivers as the cheesemaking team pushes past us. It is sexy. An assistant cheesemaker slides in the harp-like curd knife and makes two quick cuts at ninety degrees to each other. This is the first great intervention to encourage drainage; suddenly, the silky slices of curd are suspended in clear whey. We blink, and the next basin is brought forward, the cutting progressing with hypnotic regularity. This is a precisely choreographed production line, but each operation is performed with sensual delicacy. Beside us, David Aubree beams with delight. He is the production manager at Fromagerie Graindorge, located in Normandy in northwestern France. Aubree oversees the production of all four of the region’s Appellation d’Origine Protégée cheeses across multiple facilities, but the Camembert de Normandie here at Domaine de Saint Loup is the jewel in his crown. We are visiting the creamery, in the village of Saint-Loup-de-Fribois, with Dr. Marina Cretenet, a microbiologist from the University of Caen, and she has already described Aubree as the “poet” of Norman cheesemakers. In middle age, he moves with the absolute confidence of a man who knows that he is the very best at what he does. Each of his muttered comments to the team on site is met with a swift click of the heels: Aubree is master and commander. We move from room to room of the creamery, following the path of the cheese from liquid milk through to dispatch. Each room is climate controlled for the appropriate stage in the cheese’s maturation, and Aubree nods approvingly as we identify some of the puff y molds growing on the adolescent cheeses. We have a brief discussion about the choice of ripening molds with which he inoculates, but the reason we are here, and the thing that gives 93

Aubree the greatest pride, is raw milk. He slaps the milk tank as we walk past, “It is all raw milk, all from [the local breed of] Normande cows.” It says something about the culture of Camembert that, within the region, Domaine de Saint Loup is tiny. With a daily production of thirteen thousand cheeses, it dwarfs any farmhouse operation, but its total production is a mere rounding error in the three hundred and sixty million nonappellation Camembert cheeses made annually in Normandy. To qualify for the protected name, Camembert de Normandie must be made with raw milk; the term Camembert itself is, like Brie or Cheddar, regarded as too heavily genericized to deserve legal protection. Unsurprisingly, there is a profound tension between AOP Camembert de Normandie and its pasteurized cousin, Camembert fabriqué en Normandie. In 2007, against this backdrop of simmering rivalry, Lactalis and the Isigny Sainte-Mère cooperative—two producers that together represented 80 percent of Camembert de Normandie production at the time—announced that they would drop out of the appellation altogether if they were not allowed to heat-treat their milk, inspiring an extended bout of national navelgazing.1 Ultimately, the requirement remained that the protected cheese be made from raw milk, but as a result, Lactalis and Isigny Sainte-Mère withdrew their cheeses from the appellation, and by 2011, AOP Camembert de Normandie made up just 4.2 percent of the national production of cheeses called Camembert, down from 10 to 15 percent just three years previously. Eventually, the conflict erupted into a long-running and ultimately abortive lawsuit initiated by the remaining Camembert de Normandie producers, who accused the larger companies of masquerading as the real thing, or as they put it, “usurpation de notoriété.”2 With Camembert at the heart of French gastronomic identity, these perpetual legal tussles over the future of the cheese are a part of the national conversation, debated across the pages of the news media as much for what they say about the state of the nation as for what they say about the cheeses themselves. Pitting big business against tradition and commerce against the plucky artisan, the Camembert story is a familiar narrative.3 There is just one problem, and that is what brings Dr. Cretenet out to Saint-Loup-de-Fribois. Making a distinction between raw and pasteurized milk is all very good, and it is easy to educate consumers to look for the term au lait cru on a label. The call to the barricades—aux armes citoyens—to defend the incomparable gastronomic heritage of France fits perfectly with the country’s mission to civilize the world and how it chooses to see itself. But 94

•

MICROBES

the project that Cretenet is working on at the University of Caen, spearheaded by her colleague Professor Nathalie Desmasures, suggests that the story of raw-milk cheese owes more to classical tragedy than to the mythos of resistance and revolution. Milk has changed. The past thirty years have seen a holocaust of raw-milk microbes. This is not a term that we use lightly: the quest for control has caused the catastrophic destruction of the microbial communities on which cheesemakers rely to make their raw-milk cheeses distinctive and unique. In chasing perfect hygiene and absolute control, has that which cheesemakers most desire also become their tragic flaw?

D I S COV E R I N G M I L K M I C R O B E S

The modern era of milk microbe analysis began in the lab of Dr. Jean Richard, who was working for France’s INRA in the leafy Parisian suburb of Jouy-enJosas. In the early 1980s, he set about putting together the first taxonomically accurate and statistically rigorous analysis of the bacteria in raw cow’s milk. Those were not the carefree days of modern microbiology, where highthroughput sequencers churn out tens of millions of reads from a single sample in a matter of hours. Richard’s papers serve as a reminder of how painstaking the collection of even the most basic microbial data was little more than thirty years ago. Over the course of a year, he collected 231 milk samples, plated out and isolated over six thousand individual colonies, and then performed between ten and sixteen tests on the basic properties of each one just to identify them at a general level. Richard chose for his samples raw milks that were très pollués, or highly contaminated, which he defined as having more than one million bacteria per milliliter. His goal was to gather data about the relationships and potential competition between the different bacterial groups within them. He was not impressed by the picture that emerged: the lactic acid bacteria essential to cheesemaking were always in the minority, and the milk communities were dominated by the types of spoilage bacteria associated with poorly cleaned milking equipment and utensils. At the beginning of the experiment, he had wondered if high levels of lactic acid bacteria might have an inhibitory effect on spoilage bacteria. Alas, this was not the case. Richard found that both could coexist at high levels simultaneously, and even then, the lactic acid bacteria did not check the growth of their unsavory companions. He concluded: MICROBES

•

95

“It follows that if we are to reduce the detrimental bacteria to a level where they have no impact on the quality of cheese, improved hygiene on the farm (correct cleaning and adequate cooling of milk) is required.”4 This was the defining moment of modern dairying. This conclusion—that when it comes to raw-milk microbes, the fewer the better—is the basis of the milk-quality payments that now dominate the market for liquid milk. The concept sits easily. In a society where microbes are widely conflated with disease, it is simple to explain and train, and it corresponds with expectations and behavior in other parts of our lives. The total number of bacteria in a milk sample is now also easy to measure. In Jean Richard’s time, the only way to count the total number of bacteria in a milk sample was to make a serial dilution, spread it on a series of sterile agar plates, wait for several days for colonies to grow, and then count them using a handheld clicking tally counter. At the same time as his study was taking place, however, a new tool was being developed that would transform milkproduction practices: the Bactoscan test. This test counts the bacteria in a sample of milk by dyeing them with a fluorescent marker and then using a laser sensor to count them. It is a totally automated process that gives the total bacterial count of a sample in just a few minutes; top-of-the-line machines can grind their way through up to two hundred samples per hour, and centralized milk-testing laboratories now test thousands of samples per day. With such an easy and cheap way of measuring bacterial counts, it is little surprise that over 90 percent of milk in Europe is paid for according to Bactoscan results: the lower the number of bacteria in the sample, the higher the premium paid to the producer.5 The Bactoscan test and its more time-consuming manual counterpart, the culture-based “total bacterial count” or “standard plate count,” give an indication of exactly that: the total sum of all bacteria present in a sample. They lump all bacteria together, whatever the type. For producers of liquid milk, this makes sense as a measure of quality, as any sort of microbial growth in fresh drinking milk equates to spoilage. Left under the proper conditions, lactic acid bacteria—the cheesemaker’s friends—will quickly sour fresh milk, causing it to curdle. And while pasteurization helps, it is designed to eliminate only the organisms that might make people sick, not to kill all of the bacteria. Even pasteurized milk left in an unopened refrigerated container begins to clot and turn after a few weeks. The hardy bacteria that remain after pasteurization are the source of this spoilage, and the lower the numbers in the milk at the source, the better the keeping qualities of the pasteurized milk will be. 96

•

MICROBES

Producers of milk for cheese have adopted a similar approach to quality. This is not surprising since so many of them once sold, or continue to sell, liquid milk for processing. The depletion of the useful bacteria has gone unnoticed, since purified strains of starter cultures are used to reinoculate milk with just what it needs to be transformed into cheese. Milk with a low number of total bacteria is regarded as the optimal raw material, a blank canvas on which a selected set of added cultures goes to work to manufacture a consistent cheese. It is a philosophy that we have heard espoused by cheesemakers on multiple continents: “If I want a microbe in my cheese, I’ll put it there myself.” At every scale, from macro to micro, selective environments beget change. Thirty years after Richard’s study, with the widespread adoption of payment premiums and more efficient tools for analysis, the average total bacterial count of raw milk has fallen precipitously. Whereas, in the 1970s and 1980s, milk samples taken at the farm often showed total bacterial counts in the millions, today, the vast majority of bulk raw-milk samples contain less than thirty thousand per milliliter.6 Eager to demonstrate their commitment to ever-higher levels of safety and quality, many raw-milk cheesemakers wear their low counts as a mark of pride. A 2010 study in Vermont showed that the 86 percent of milk samples destined to be turned into raw-milk cheese contained less than ten thousand bacteria per milliliter, and 42 percent contained less than one thousand per milliliter.7

Q UA N T I T Y V E R S U S Q UA L I T Y

The dairy industry has succeeded in vastly reducing the total number of bacteria in raw milk, a case study in changing widespread behaviors in a short period of time. But hubris leads to nemesis. Total bacterial count is a perfect metric for milk quality when the milk is destined for pasteurization because it is a good proxy for the product’s ultimate longevity as fresh milk. As a measurement of quality for raw milk destined to be made into cheese, however, it completely misses the point. The safety of raw-milk cheese relies on milk that is free from pathogens, and therefore, lowering the total number of bacteria simply doesn’t work: pathogenic bacteria still turn up from time to time in the “cleanest” operations and in milk with extremely low total levels of bacteria. The whole point of making cheese is to create a perfect environment for bacterial growth, so MICROBES

•

97

even a tiny number of a pathogen such as Listeria monocytogenes can pose a problem if they are present in raw milk. Less than ten of these bacteria per milliliter can grow to harmful levels in certain types of cheese over the course of production and maturation. And in milk devoid of beneficial and useful species, pathogens and spoilage bacteria make up a larger proportion of the total starting population when they do occur. Cheesemaking deploys the microbes that live within milk to preserve it and produce delicious flavors, and farms are natural reservoirs of microbial biodiversity. Just as they have in our gut or in the soil, microbial populations evolve to fit perfectly into the ecological niches available to them. These communities are diverse, with many different species and strains performing the same ecological roles side by side. This functional overlap makes bacterial communities very robust. As the environment changes across the seasons or as different farming practices are adopted, different microbes adapt more or less quickly, but the underlying operation of the community remains undisturbed. In 1998, Nathalie Desmasures and her team set out to study the diversity of Lactococcus lactis, one of the most significant lactic acid bacteria for cheese, on six farms in Normandy. They found an extraordinary level of diversity, isolating and identifying 357 unique strains over the course of the study. Some single milk samples contained up to 60 different strains. Some of these strains dominated in winter, and others in summer; some were found at almost all the farms, while others were unique to a single farm. The scientists were also astonished to find that many of the isolates were quite different from the reference strains held by the government reference labs. During the study, they identified Lactococcus lactis strains belonging to ten different genetic subgroups. The reference labs, the most complete microbial libraries available at the time, held examples from just two of these subgroups.8 Eliminating these naturally occurring microbial communities denies the cheesemaker a critical tool. In Caen, Desmasures and her colleagues have coined a term for raw milk that has been stripped of its microbial character and useful properties: they call it “dead milk.” Only now are we beginning to appreciate the collateral damage caused by the out-and-out war on milk bacteria over the past thirty years: a decrease in raw-milk biodiversity and a worsening of the balance between useful bacteria and those associated with spoilage. Desmasures noted that between 1983 and 1997, as the total bacterial count of milk became lower, the average ratio of lactic acid bacteria to total bacteria also decreased, from around 30 percent to close to 10 percent.9 In 98

•

MICROBES

other words, the sanitary measures being introduced during that time were more effective at targeting lactic acid bacteria than other bacterial families. Milk with depleted microbial biodiversity and a low total bacterial count, made up of a large proportion of spoilage bacteria and a minute number of pathogens, will perform brilliantly on analyses designed to measure suitability for pasteurization and potential shelf life. For a raw-milk cheesemaker, however, this milk would be the first ingredient in a recipe for disaster. In 2000, bracing themselves for a battle with regulators over accepting milk with high microbial loads, the producers’ group for Comté, in eastern France, commissioned a series of experiments to test the impact of both the total number and the type of raw-milk microbes on the quality of cheese. Two milk producers volunteered to participate in the experiment, and their herds’ raw milk was put through a microscopic fi lter to remove all bacteria. Microfiltration is commonly used as an alternative to pasteurization for the production of extended-shelf-life milks and some cheeses—it is one of the most controversial interventions in the world of Camembert—and in those cases, the concentrated bacterial soup left on the outside of the filter is disposed of as a waste product. In this experiment, however, that microbial slurry was used to reinoculate sterile milk with known levels of bacteria ranging from 5,000 to over 250,000 per milliliter (the legal limit for use in raw-milk cheese within the European Union is 100,000 per milliliter). Experimental cheeses made from these inoculated milk samples were presented to sensory panels after five months of aging. The tasters failed to detect any difference in the flavor of the cheeses that came from the same farm but were made with milk that had started out with the different initial levels of microbes. But they were readily able to distinguish one farm’s cheeses from the other’s, regardless of the initial total bacterial count of the milk used to make them.10 This result is actually not so surprising: cheesemaking in itself is a process that involves the exponential growth of microorganisms. Even milk with a low starting total bacterial count provides everything necessary for successful cheesemaking as long as that initial population encompasses the right microbial diversity and balance. It’s diversity, not quantity, that counts. Now, scientists like Desmasures are asking: What does optimal raw milk for cheese look like on a practical level? Is there any way to reclaim the microbial biodiversity that we have spent the last thirty years trying to wipe out? It is important to note that this is not a call for the revalorization of fi lth, for an anything-goes microbial philosophy where a bit more shit in the milk just adds to the flavor. It is exactly the opposite: an entreaty to reexamine the MICROBES

•

99

current policy of blanket eradication and adopt holistic practices that contribute to conscious selection of the organisms that contribute to robust, healthy, interesting milk for cheese. The tools are already there. Simple farming and milking practices hold incredible power to unleash the useful microbes on dairy farms and change the microbial composition of raw milk. Who are these characters that, for better or for worse, inhabit milk? As we have seen, it is all too easy to think of them as an undifferentiated microbial chorus, the levels of which cast a moral commentary on the travails of farmers and cheesemakers. But before we can think about influencing the microbial inhabitants of milk, we have to understand where they come from, their role in the natural world, and the characteristics that might make them beneficial, harmful, or even both at the same time. Dramatis Microbia I: Lactic Acid Bacteria

Lactic acid bacteria are the fickle prima donnas of the milk microbiology world. While they are always present in raw milk, their numbers are often low, and they are more sensitive than the other families of milk bacteria to the effects of chemicals and low temperatures. In the natural world, they are associated with the surfaces of plants, where they can be found at extremely low numbers, often in a viable but not directly cultivable state.11 When plant cells are damaged and nutrients are released in the escaping fluid, the population of lactic acid bacteria can increase to very high numbers in a short period of time; this is what happens in the process of making sauerkraut and kimchi. Lactic acid bacteria dominate microbial communities by producing and tolerating high levels of acidity as they grow, making them worthy opponents of other bacteria in the environment. Their acid production also facilitates the physical process of cheesemaking, aiding the contraction of the casein proteins in the milk and the expulsion of moisture from the curd. In addition to acid production, lactic acid bacteria have evolved an impressive arsenal of antibacterial compounds called bacteriocins, and other natural antibioticlike molecules, that help them survive at the expense of other bacteria.12 Lactococcus is the genus that has most successfully specialized itself as a milk bacterium, and it includes a subspecies, Lactococcus lactis ssp. cremoris, that is exclusively found in dairy environments, on udders and in raw milk and milking machines. Outside of the dairy, Lactococcus species are present in a wide variety of plant material, from corn silks to clover to cantaloupe, as well as in wild grasses and plants. While it is more acid tolerant than many 100

•

MICROBES

other types of bacteria, Lactococcus is also fairly delicate, unable, for example, to survive the harsh acidic conditions of the human stomach. Unlike Lactococcus, with its relatively narrow set of natural habitats, the genus Lactobacillus is widespread in the environment and within the bodies of humans and animals. Lactobacillus is not only distributed widely on plant surfaces but is also found in association with decaying plant matter in the soil, as well as in the mammalian gut and (as a result) in feces and sewage. In humans, lactobacilli are the dominant microbes within the vaginal tract, as well as a minor component of the bacterial community of the mouth. Unlike the lactococci, lactobacilli ingested in food are capable of surviving the acidic conditions of the stomach and passing as viable cells through to the lower gut, where they are found in variable proportions depending on the species and the individual. Lactobacilli have received a lot of attention for their potential to benefit human health as probiotics, and there is evidence that as they form stable populations within the intestine, they change the environment in ways that may select against potentially pathogenic bacteria. However, there is much that remains to be understood about how and when these organisms establish themselves within the gut. The mechanics of their interactions with the other members of that community, and the complex interrelationship between diet, population dynamics in the gut, and ultimately health, are subjects of intense interest within the field of medicine.13 With their numerous uses in food technology and their role as members of the human microbiome, the vast majority of lactic acid bacteria have been granted GRAS (“generally regarded as safe”) status by the United States Food and Drug Administration, making them one of the few groups of bacteria with which the mainstream of society has—at least philosophically, if not always in practice—made peace. This is not to say that they always behave in human-friendly ways or that the line between beneficial and pathogenic is any more clear-cut in this group of bacteria than in others. The same Lactobacillus species so important in cheesemaking are also well-recognized pathogens and can cause lethal infections if they make their way to and multiply in parts of the body that are meant to remain sterile, such as the bloodstream and the lining of the heart valves (this type of septicemia has nothing to do with eating cheese, and it usually occurs in immunocompromised patients with severe illnesses such as cancer and diabetes). Another lactic acid bacterium, Lactococcus garviae, causes mastitis in cows and is a potent pathogen of farmed fish, where it is responsible for lactococcosis, a deadly disease characterized by protrusion of MICROBES

•

1 01

the eyeballs and hemorrhagic enteritis. Lactococcus garviae is not regarded as a risk to humans, however; it is widely found in raw milk and cheese and has even been shown to inhibit the growth of the human pathogen Staphylococcus aureus in milk, prompting some cheese technologists to suggest that it might be useful as a component of commercial starter cultures.14 Dramatis Microbia II: Other Gram-Positive Bacteria

Scientists have developed many different ways of classifying the bacteria within the microbial tree of life, but one of the most enduring and broadest distinctions is between the so-called Gram-positive and Gram-negative bacteria. This definition simply reflects the structure of their cell walls. In Grampositive bacteria, the rigid structural component of the cell walls (called peptidoglycan) forms the outer layer, while in Gram-negative bacteria, these peptidoglycan walls are covered by an outer membrane. During the Gramstaining test, bacteria are exposed to a dye that is absorbed by exposed peptidoglycan. The exposed walls of Gram-positive bacteria can be stained by the dye; the surface membranes of the Gram-negative bacteria, however, do not retain the stain. This distinction gave rise to a classification system. Lactic acid bacteria are Gram-positive, but not all Gram-positive bacteria are lactic acid bacteria. The wider world of Gram-positives includes a vast range of organisms associated with environments including soil, decaying vegetation, and skin and—as a result—with milk. One of the most important groups of Gram-positive bacteria in raw milk is the Actinobacteria, which are well known for their importance in soil ecosystems, where they are key recyclers of organic matter and crucial for the health and nutrition of growing plants. When most people are asked about where antibiotics come from, their immediate response is fungi, particularly the famous Penicillium. However, Actinobacteria, especially a genus called Streptomyces, are the source of far more antibiotics than are fungi. (Any antibiotic ending in -mycin, and many others, are a product of these bacteria, which developed them to defend themselves from other bacteria in the soil.) Beyond antibacterial agents, Streptomyces are also a potentially important resource for other bioactive compounds, including antifungal agents and tumor inhibitors.15 There is every reason to protect and encourage the biodiversity of this group of bacteria—in milk and elsewhere—as they may hold the key to novel therapies for disease. 1 02

•

MICROBES

The Actinobacteria also include genera that are old friends to cheesemakers, including the famous Brevibacterium, Corynebacterium, Microbacterium, and Arthrobacter. Most Actinos (as they are affectionately known by microbiologists) are aerobic, and these bacteria can often be found dominating the surfaces of cheeses as they mature. Many of them are closely associated with the pungent flavors of sticky, red, washed-rind cheeses, though they make up a significant component of the communities on the drier natural rinds of hard cheeses as well. The Actinobacteria also encompass the genus Mycobacterium, which includes the species that cause tuberculosis and leprosy. As we will see in the next chapter, Mycobacterium species present in raw milk from diseased animals were a significant source of foodborne illness in the late nineteenth and early twentieth centuries. The widespread adoption of pasteurization for drinking milk, along with more stringent approaches to assuring animal health, led to the eradication of milk-borne human tuberculosis infections by the latter half of the twentieth century. Interestingly, Mycobacteria, as well as other Actinobacteria, are some of the more heat-resistant microbes in milk. The time and temperature combination that was originally chosen for the pasteurization process was that required to kill Mycobacterium tuberculosis, since all the other milk-borne pathogens known at the time were susceptible to lesser heat treatments.16 Some, but not all, Actinobacteria have higher heat resistance and can survive pasteurization,17 and these organisms—which under other circumstances are valuable allies in the ripening of soft cheeses—are also the organisms whose presence often limits the shelf life of pasteurized fresh milk. In addition to Actinobacteria, staphylococci are also common Grampositive inhabitants of raw milk, and they too include species that play an important role in cheese ripening, as well as a few opportunistic pathogens. Staphylococcus species are often found residing on healthy human and animal skin. A few, such as Staphylococcus aureus, can either be benign commensal organisms or cause disease if they set up shop in the wrong place. They are a common cause of mastitis in dairy cows, and infected cows will shed bacteria directly into their milk. Listeria monocytogenes is perhaps the most notorious Gram-positive bacterium; its very name strikes fear into the hearts of cheesemakers. Its natural reservoir is soil, where it lives on decomposing plant material, much like the Actinobacteria, and it can also be found in poorly made silage and in cool, wet environments. It is possible to find Listeria species in milk when farming MICROBES

•

1 03

and milking systems are not well controlled, and it can also cause infections of the udder, which can be particularly problematic for the raw-milk cheesemaker. It is important to note that, as a primarily environmental bacterium, Listeria monocytogenes can make its way into cheese through various routes, not just via raw milk; pasteurized cheeses may also encounter problems with Listeria contamination if the bacteria are introduced to the system at any point during the make or maturation. Dramatis Microbia III: Gram-Negative Bacteria

The third and final major category of raw-milk bacteria are Gram-negative. Unlike Gram-positive bacteria, which play a well-known role in cheesemaking, Gram-negative bacteria have had a bad rap, and their presence is often blamed on poor hygiene in the milking parlor or during cheesemaking. But recent discoveries are leading researchers and cheesemakers to reconsider their role. For one thing, these bacteria are much more widely distributed than previously imagined, even among pasteurized cheeses. When the Dutton lab at Harvard University surveyed the microbial communities on the rinds of 137 cheeses, they found that Gram-negative bacteria dominated the rinds of the bloomyand washed-rind cheeses and were detectable on almost all the cheeses surveyed.18 Gram-negative bacteria are members of balanced communities too, and in cheese, they play a critical role in terms of flavor development.19 As with Actinobacteria, their metabolic pathways are equipped to break down sulfurcontaining amino acids in the cheese paste into volatile aromatic compounds.20 Working with these bacteria during her stint in the lab, Bronwen was struck by their vivid aromas; the colonies growing in culture emitted odors ranging from linseed oil to chicken broth to musky zoo to sweet rotten fruit. Without Gram-negative bacteria, Livarot and Munster would have none of their funk. A family within the Gram-negative category, the Enterobacteriaceae, has the socially awkward distinction of being largely—though not exclusively— associated with the intestinal environment, and to make matters worse, it contains a number of famous pathogens, including Salmonella, toxigenic E. coli species, and Shigella. Because their presence is sometimes used by commercial labs as an indicator of fecal contamination, these Enteros are generally reviled as fi lthy contaminants, but just as with so many other bacterial groups, the bigger picture is more complex. Some Enterobacteriaceae even serve as useful tools for the cheesemaker. Hafnia alvei is closely related to E. coli and is available to cheesemakers as a commercial adjunct culture. Not 104

•

MICROBES

only is it often slipped in among other ripening cultures to give Brie-style cheeses a barnyardy boost, but microbiologists are also exploring its potential to inhibit the growth of toxin-producing strains of E. coli.21 Dramatis Microbia IV: Yeasts and Molds

A tour of the microbial inhabitants of milk would not be complete without a brief survey of naturally occurring milk fungi. These include yeasts and molds from the environment, originating mainly in animal feed and bedding. They can also be found within the air of the milking parlor. Wild yeasts play an important role in the early stages of ripening, where they are among the first colonizers of the fresh cheese, lowering the acidity of the surface and making way for the procession of molds and bacteria to follow. Unfortunately for those who covet rinds as pure as driven snow, wild-type strains of molds like Penicillium are brightly colored greenish blue, and many other environmental fungi that readily set up shop on cheese rinds are grey, yellow, brown, black, or even red. The relationship between cheesemakers and these molds is one of the key drivers that shaped the evolution of Camembert, as we’ll see later in this chapter.

M I C R O B E FA R M I N G

But first, we must step back and remember our place. We humans have a tendency to construct a worldview that is all about us. It is deceptively easy to cast microbes as characters in our own grand drama, portraying them either as minions whose only will is to do our bidding or as archnemeses who are quietly plotting our destruction. Make no mistake: the farmer-cheesemaker is the architect of the system, with godlike power to change the climate, drop in foreign species from above, and call forth floods of toxic chemicals. But the microbes in milk have no inkling of the type of cheese they are about to make, of which of their activities we depend on for help, or which ones will ruin our day. To them, our human designs are incidental, our world many orders of magnitude removed from their own. In their names for the various groups of bacteria, cheesemakers and even cheese scientists fall under the spell of this functional view of microbes, to the point that we often refer to them as “acidifiers,” “ripening strains,” “spoilage organisms,” and “pathogens.” But even a pathogen’s reason for being is MICROBES

•

105

not to cause disease. Why would a simple soil bacterium like Listeria monocytogenes have any interest in hurting a human? Scientists are constantly looking for model systems to help them study their phenomena of choice as directly and inexpensively as possible, and in the case of many human pathogens, microscopic protozoa fit the bill. This is no strange coincidence but rather the unlikely conclusion to a story that began many millions of years ago. Soil bacteria don’t simply sit alone in the dirt, munching away at the smorgasbord of decomposing plant material laid out before them. They are under constant threat from predators, such as protozoa, for whom a bacterium like Listeria constitutes a tasty breakfast. To survive possible onslaughts, Listeria has developed the tools to invade and replicate within protozoa; it has learned to fight back. Protozoa, despite being microscopic, single-celled organisms, are not bacteria; like humans, they belong to another taxonomic domain, Eukaryota. Due to a shared lineage that stretches back an almost-unimaginable distance in time, the structure of protozoa and our own white blood cells (known as macrophages) are not so different from one another. And in the human body, it is in macrophages that Listeria replicates and spreads.22 This phenomenon is not limited to Listeria. Many other microbes that we know as human pathogens, including Mycobacterium tuberculosis, Pseudomonas aeruginosa (a common cause of skin and lung infections), Vibrio cholera, and Legionella pneumophila (the causative agent of Legionnaires’ disease), depend on anti-eukaryote warfare systems for survival in the wild.23 They just happen to be as effective against human eukaryotes as they are against microscopic ones. Nature is not benign, as some would have us believe, but it’s not out to get us either. Although humans are incidental to a microbe’s worldview, the invisible communities around, on, and within us have the power to affect the outcome of our macroscopic pursuits, for better or for worse. On farms and in dairies, microbes settle into communities encompassing thousands of different members, all of which play overlapping roles. These unique communities are farm specific, and they are shaped by the choices of the farmer. What the animals are fed, how often their bedding is changed, how they are cleaned—all of these decisions have the power to make the difference between dirty milk, dead milk, and milk with the potential to express the unique character of that farm and no other. Just as we saw with the work done to clean up milk over the past thirty years, if the goal is to change something, the first step is to be able to 106

•

MICROBES

measure it. Today’s commercial food laboratories are very good at counting total bacteria. Ask them to tell you if a sample has Listeria or E. coli in it, and they will spring into action. But full-population-level analysis of milk samples is still the domain of well-funded research labs. The act of commissioning a comprehensive survey would quickly bankrupt a farmhouse cheesemaker. Desmasures and her colleagues use a more practical method of measuring milk quality called the “relative index,” which was designed for use by cheesemakers. Instead of using DNA analysis to identify all the organisms in a sample, they use old-fashioned culture-based methods to quantify members of each of the broad groups of microbes—the lactic acid bacteria, the other Gram-positive bacteria that can play a role in cheese ripening, and the Gramnegative bacteria—as well as the total yeasts and molds. This doesn’t reveal the names of the individual species, but at the farming level, such names are little more than trivia. Through relative index testing, even a farmer with just ten cows can gain access to information about the microbial community within his or her milk. Even so, these relative index tests are at best a general indicator of what is going on in the milk. Problems such as finicky bacteria refusing to grow on plates in the lab as they would in their native environment and the shift in the balance of species that occurs as samples are transported to the lab are sufficient to cast doubt on the relevance of the results. Even the eye-wateringly expensive molecular tests that can screen a complete bacterial community with precision give just a snapshot of a sample at one point in time. The microbial world, with its unfathomable complexity and constant state of flux, still confounds us in our attempts to pin it down. With this in mind, the success of the original farmhouse cheesemakers in manipulating organisms whose existence they were not even aware of is astonishing. Their cheesemaking methods evolved over the course of centuries through trial and error, and while bad cheese has always existed, those early cheesemakers were, by and large, able to harness the native microbes in their milk to turn it into cheese. Those practices worked. But the continuous links in the chain of farmhouse cheesemaking practice have been broken, and piecing them back together again is a formidable challenge. Studies of the effects of various practices on the diversity and balance of microbial communities in milk are in their very early stages, but they have already shown that farming choices have a strong influence on the microbes that are carried through to milk and cheese. MICROBES

•

1 07

ADDRESSING THE UDDER

The quality of animals’ feed and pasture, the bacterial load of their bedding, and even the air quality in the barn and milking parlor can affect the microbial profile of raw milk, but these environments are hard to understand and even harder to control. For dairy farmers trying to decrease total bacterial counts in their milk, the surface of their animals’ teats is an easy target. Unfortunately, the bacterial community native to the teats is one of the most important natural sources of the microbes that are most useful to cheesemakers. These days, dairy farmers are obsessed with putting chemicals on teats. There is a cornucopia of products to choose from, including iodine-based disinfectants, bleach solutions, and newer antimicrobial compounds such as benzalkonium chloride and chlorhexidine. Some of these are designed to be applied before milking as part of the teat-cleaning process, and others are meant to be used afterward. Some are sprayed on; others take the form of foaming dips; even medicated wipes are available. Many of these products are selected not only for the speed and efficiency of the initial kill but also for their persistence, their ability to continue to suppress the growth of bacteria for many hours after they are applied. This chemical blast certainly influences the passage of teat-associated microbes into the milk, but not necessarily for the better. The problem is that the surface of the teats hosts a community of microbes in which the beneficial lactic acid and ripening-associated bacteria outnumber the undesirable spoilage bacteria by a ratio of a hundred to one.24 A team working with Comté producers in eastern France found that easing up on teat cleaning before milking significantly increased the proportion of lactic acid bacteria relative to the total bacterial count of raw milk. The milk from farms that did more precleaning of teats and udders carried significantly higher proportions of bacteria associated with milk and cheese spoilage.25 Let’s not pretend that cows always come into the parlor with pristine udders, and sucking dirt and manure into the milking pipeline is not an option. So how is it possible to clean off the teats without water or chemicals in a way that preserves the native skin communities? Newly promoted by the French dairy microbiology community, an old practice is being rediscovered by cheesemakers: the use of fine, soft wood shavings, the original nonscratch scourer. Very similar in appearance and texture to straw, wood wool (often called “excelsior” in the United States) is grippy enough to clear the teat skin of debris while being soft and absorbent enough to avoid scratching or chapping. And unlike paper towels or cloths, once used, it can be left on the floor 108

•

MICROBES

and sluiced out of the milking parlor, where it naturally degrades back into the soil. Wood wool is cheap, simple to use, and effective. Teat dips are sold on multiple promises: they claim not just to disrupt the transfer of microbes into milk but also to prevent udder infections. After milking, the teat orifice remains open for several minutes, and without the application of chemicals, the conventional thinking goes, bacteria may invade the teat canal and cause infection. A slash-and-burn approach to teat microbial communities is the industry standard: dead microbes can’t cause disease. However, a sterile teat surface is not something that even the strongest teat dips can deliver in practice, and the risk becomes one of disturbing healthy communities, which makes way for imbalance and infection. In this new theory of disease, microbes are still the agents of infections, but their ability to cause those infections is not dictated by their presence but rather by the balance of the entire community. Healthy teat microbial communities are composed primarily of organisms that are harmless or beneficial to both the animal and the cheesemaker. Managing disease risk is not just a matter of splashing around some sanitizers. Disinfectants may not even be effective at preventing mastitis. Preliminary evidence gathered by Dr. Marie-Christine Montel and her team over the course of two years showed that herds milked with and without teat post-dipping had statistically indistinguishable rates of udder infection.26 Even more powerful evidence comes from the population-level data collection from Montbéliarde cows. Over the past fi ft y years, the Montbéliarde society has been breeding cows for resistance to mastitis infection. Montbéliardes have lower rates of mastitis and lower somatic cell counts (a test for infection) in their milk relative to other breeds of cow, irrespective of teat-dipping practices. Although the default solution within the dairy industry is to reach for stronger chemicals, there are more inexpensive, effective, and elegant approaches to keeping milk safe and animals healthy. We just have to learn how to use them. In the popular imagination, the process of milking cows is one of dull repetition: cows march in, udders are prepared, clusters suck, milk flows. Rinse and repeat. In reality, however, milking is a skill that requires patience and empathy, keen observation, on-the-spot-problem-solving, prodigious multitasking, and speedy reflexes, as one must always be ready to dodge an abrupt kick. The consequences of tiny shortcuts add up; conversely, so do small extra cares and attentions. How does one get the best from each cow, work with her, and keep her happy and comfortable throughout the process? MICROBES

•

109

Is it pouring rain outside or dry and dusty? Are the cows coming into the parlor clean, or are they caked in mud or manure? Are incipient issues spotted before they flare up into acute problems? Is the time taken to do it right, or is it a headlong rush where any problems are left to be sorted out downstream? This is the art of milking, and it is as much a part of cheesemaking as anything that goes on in the vat or the affineur’s caves. As such, milking for raw-milk cheesemaking and milking for pasteurized liquid milk are two completely different disciplines. The extent of the distance between the two worlds is something that we were properly able to appreciate only when we visited Albert Straus at his Straus Family Creamery, located just outside the town of Marshall in Marin County, California. The northeastern coast of Tomales Bay has been dairy-farming country since the 1850s, but the consolidation of the Californian dairy industry in the 1970s and the associated growth of huge dairies was destroying the local family farms. By the early 1990s, faced with a milk price from the regional processor that did not even cover the costs of production, Straus decided to convert his entire operation to organic and sell directly to consumers under his own label. And so in 1994, Albert launched Straus Family Creamery—at the time, the only fully certified-organic creamery in the USA—and a muchloved Bay Area brand began. As well as liquid milk, Straus makes and sells ice cream, yogurt, butter, cream, and sour cream, and his clients include many of Northern California’s most obsessive sourcers of produce, like Alice Waters’s Chez Panisse restaurant in Berkeley. A visit with Straus and Pankaj Uttarwar, his director of research and development and quality assurance, reveals that the two are shrewd technicians with a thoroughly justified pride in their product. Their plain Greek yogurt is easily the best example we have tasted of a vat-set strained yogurt. But it is clear when we talk about milk and milking that their concerns are very different from those of a farmer milking for raw-milk cheese. The abiding concern for Straus, just as for every producer of pasteurized liquid milk, is to make sure that his milk stays fresh and sweet until its use-by date, something that he addresses with great skill. This requires a dedicated approach to producing milk destined for pasteurization, which includes monitoring the total bacterial count of both the raw and pasteurized milk as well as how it tastes over the course of its shelf life. Questions about promoting the presence of native lactic acid bacteria and ripening microorganisms in raw milk are inherently irrelevant to the market Straus is in. To the contrary, milk that sours rapidly of its own accord, or milk containing significant numbers of 110

•

MICROBES

heat-insensitive ripening bacteria, would be highly problematic for a purveyor of liquid milk. There are cheeses made from the milk of Straus’s herd and the eight other farms from which Straus Family Creamery sources its milk, but they are made with pasteurized milk. Raw-milk cheesemaking requires raw-milk dairy farming.

THE FINAL HURDLE

The microbial balance and diversity in raw milk determines its flavor potential and its safety. Farming and milking practices have the power to shift those initial populations, for better or for worse. But the bacteria in milk face perhaps their most significant hurdle at the very last stage of production. Refrigeration, the essential tool of fresh milk production, has the capacity to disrupt within a matter of hours the microbial potential that the farmer and milker have worked so hard to create. As warm milk leaves the animal and passes through the milking machine, it is imbued with its optimal microbial potential. Some milks have greater diversity and balance than others, but at that moment, they are the best versions of themselves. But the microbes within the milk do not have cheesemaking on their agenda. They are primed to respond to whatever environment they are presented with, ready to stake their claim according to their relative abilities. At the moment the milk enters the world, a microbial race has begun. Most bacteria—even many of those strains optimized for life in the soil— grow and divide faster at warmer temperatures, hence the attraction of a rapid and sustained cold snap as a way to preserve liquid milk in an unspoiled state for as long as possible. But within that general rule, individual species are more or less adapted for a chilly lifestyle. Lactic acid bacteria useful for cheesemaking have an unequivocal attitude toward refrigeration: their growth rate grinds almost to a halt in the cold. At a normal refrigeration temperature of 39°F (4°C), Lactococcus takes thirty hours to muster the resources to divide just once.27 For this fundamental class of cheesemaking bacteria, refrigeration essentially serves as an off switch. It won’t kill them, but it knocks them down for the count. Other Gram-positive bacteria show more variable attitudes to cold, depending on the species. The growth of the famous Staph aureus is effectively checked by refrigeration temperature, hence the food-safety regulator’s rationale for requiring fresh milk that isn’t immediately processed to be MICROBES

•

111

refrigerated. Many others, like species that play a role in ripening soft cheeses, are simply slowed down a bit. And a few, such as Listeria, are adapted to cold temperatures and positively thrive under refrigeration. Many Gram-negative bacteria also find ice-cold raw milk an inviting place to grow. Members of the genus Pseudomonas are among the most dreaded of the cheese spoilage bacteria. They cause the cheese rinds on which they grow to become slimy and taste of bile, and many strains have the disconcerting ability to glow in the dark. Many of the reviled Enterobacteriaceae grow quite happily at refrigeration temperatures as well. A study of raw-milk samples taken from farms and sampled twice, first immediately and then again after twenty-four hours of refrigeration, showed significant shifts in their bacterial makeup even over that short time. Storage at low temperatures amplified some species that had been at subdetection levels when the milk was fresh; three milk samples in which Listeria had not initially been detected showed its presence after being stored cold for twentyfour hours. Refrigerated storage also depressed initially significant populations of more cold-sensitive bacteria within twenty-four hours. Lactococcus was one of the biggest losers.28 Milk, raw or pasteurized, warm or cold, is a moving target. It is constantly evolving, its bacterial populations and balance a reflection of its life history up to that point. Before the invention of the refrigerated bulk tank, souring began as soon as milking was completed. The two classic cheesemaking solutions to this problem are either to make cheese after every milking—this is the approach used for making many Alpine cheeses—or to take the British route and leave the milk from the evening milking to begin to sour and ripen at a cool room temperature overnight before mixing it with still-warm morning milk and immediately commencing cheesemaking the next day. Even with all the plate chillers and icy bulk-milk tanks in the world, milk for cheese can’t just be put away and returned to when it is convenient, at least not if cheesemakers expect to find the same milk they left behind when they switched off the lights the night before. Chilling milk stacks the deck against the partners the cheesemaker needs the most.

A FA R M E R A M I D T H E FAC TO R I E S

Back in Normandy, we are restless to get out into the fields and see some cows. The technical skill in the cheesemaking we have seen has been 112

•

MICROBES

supremely high, but all of the conversations we have had about the importance of raw milk and its optimal microbiology have been conducted on the purely abstract level, discussions of numbers and testing systems as we stand next to huge factory milk tanks. We want to see what this looks like in practice and what it means for the farmers who supply the milk for the cheese. And so we find ourselves, together with Marina Cretenet and Nathalie Desmasures, carefully navigating the winding country lanes through the bocage on our way to visit Patrick Mercier and his wife, Francine, at their Ferme du Champsecret. Patrick is an intense man with the air of an evangelist. The Merciers are one of only two remaining farmhouse producers of AOP Camembert de Normandie. If the amount of cheese made by Domaine de Saint Loup could be considered a rounding error in the total French production of Camembert, the production of the Merciers, with their ninety Normande cows, would be nothing more than a statistical irrelevance. As Patrick sits with us at a table in his small creamery, it is clear that his farming decisions are moral choices: he is firm in his convictions. La Ferme du Champsecret is the only producer of Camembert that is triple certified: his farming is organic and his cheese is fermier (as opposed to the industriel operations that dominate Normandy) and Appellation d’Origine Protégée. Each of these choices brings Patrick pride. He also exclusively farms Normande cows, the sturdy brown-andwhite, dual-purpose beasts that are associated with the region. With their broad barrels and chunky frames, the Normandes taste as good as they milk, and their meat was a constant feature of our meals in Normandy. The position of Champsecret as an outlier in the appellation causes us initially to get off on the wrong foot with Patrick. When he mentions that some other people have told him that the “appellation is not for people like him” but rather for the large producers, we mutter that they might have a point and ask him if he has thought about following the route taken by some wine producers who feel that their production is better than their appellation and declassifying it completely. “Oh no!” he insists, “I could never do that.” What we have not properly understood is that Patrick is also the president of the organization responsible for the management of all of the AOPs of BasseNormandie. He is the battered public face of the failed lawsuit against the industrial producers, and when he talks about how his cheese is undermined by industrial Camembert, he becomes animated. “People do not understand our cheese!” he says. At Champsecret, “we make a Camembert from the 1950s, and our cheese is not what they expect.” MICROBES

•

1 13

The connection between cheese, milk, and land is infinitely more palpable as we stride through the verdant meadows, but it soon becomes clear that Patrick is far more comfortable talking about his farming than his cheese. Perhaps we should not be surprised: he has been farming since 1980 but only started to make cheese on the farm in 2000. The Merciers employ a cheesemaker, and we get the sense that he must struggle with Patrick’s combination of absolute certainty about what he wants to achieve and lack of experience with the practical side of cheesemaking. Patrick understands all too well that he is in the very rare position as a Camembert producer of actually being able to make cheese that tastes of his own land: this is his mission. Enthused by Patrick’s passion for organic farming, and enchanted by the magnificent hay (the product of his own hay dryer) that he delights in showing us, we wonder what we will find in the creamery. How will it be different from creameries that make cheese from the pooled milk of hundreds of individual farms? As Mercier leads us through his little creamery, the most striking thing is not the contrast, but how similar everything is to the factories we have visited. The cheesemaking lacks a little of David Aubree’s technical polish—the curd does not shimmer quite so sexily, and the whey is cloudy when the curd is cut—but that is to be expected from a farmhouse operation, where it is not possible to standardize each and every variable. But as we look at Patrick’s cheese growing its snowy white rind in the maturing room, it is clear that at La Ferme du Champsecret, the aspiration is to make a Camembert de Normandie that looks as pure and pristine as its factory-made counterparts. It is open to question whether Camembert is at heart a cheese designed for large-scale production. For all of its iconic status as a symbol of French gastronomy, in historical terms, Camembert is extraordinarily young. The origin story that receives the most airtime firmly places its mythic origins within the context of the French Revolution. According to this story, in 1791, Marie Harel, a farmer’s wife from the Auge, hid a recusant priest who happened to watch her making the local cheese. As he followed her cheesemaking, he made suggestions taken from his experience of Brie, the cheese of his home region. Thus, Camembert was born, a synthesis of the older Norman cheese Livarot and Brie, from the Ile-de-France. Whether or not the legend is true, it is clear that Camembert was commercially developed in the early part of the nineteenth century.29 With this comparatively late start, Camembert addressed a set of problems quite different from those of more ancient cheeses. As a dominant form 114

•

MICROBES

of agriculture, dairying came late to Normandy. Before the disruption to the textile industry of the Industrial Revolution, it was flax that was the principal agricultural commodity of the region, and fields were extensively planted with cereal crops. Our modern image of Normandy, of cows grazing amid apple orchards, is entirely a product of the mid-nineteenth century, reflecting the industries that took over once the land devoted to wheat and other grains was left to go fallow.30 The production of Camembert—and in fact all modern pastoral farming in Normandy—is not an attempt to eke value out of marginal land. Rather, Camembert is a cheese designed to access new markets, a cheese defined by the coming of the railway to Normandy during the 1850s, which, suddenly, made it possible for a small, soft-ripened cheese to conquer the market in Paris and beyond.31 Even the distinctive white bloomy rind is not an inherent facet of the raw milk’s microbiology, nor even something that a nineteenth-century Norman cheesemaker would recognize. The choice of the cocktail of ripening cultures with which the cheese is inoculated is a vital aspect of modern Camembert production, so much so that when we asked the plant manager at one factory for some more details about his cultures, he replied with a studious and conspiratorial “no comment.” These cultures are the closest things that the industry has to secret ingredients. But as a consequence, much of the energy of modern Camembert makers is directed toward waging a war with wild molds. By inoculating the cheeses with high levels of known strains of bacteria and providing them with perfect conditions to dominate the rind, producers aim to push out any natural strains that might be native to the milk or dairy. Such an approach requires constant vigilance and paranoia, but even under the best conditions, it is not always successful. As we toured Normandy, all of the cheeses we saw and tasted were dominated by the mushroom flavors of snowy white Penicillium candidum. In a style of cheese where the rind is responsible for much of the flavor, it is a huge point of convergence. However, if we look back beyond the 1950s, the ubiquitous white rind was almost unknown. When British cheese authority James Long visited Normandy in the 1890s to research “Continental fancy cheeses,” he found that it was “the blue mould which is responsible for so much of the work in the process of ripening.”32 The cheeses that Long describes were not inoculated with any ripening cultures: their rinds were the product of the interaction of the milk’s own microbes and the environment in which the cheeses were kept. White rinds as we know them now only became common between the 1920s and 1930s. As Pierre Boisard notes in his history of Camembert: “It was only MICROBES

•

115

in the latter part of the century, sometime after 1975, that the cheese began to flaunt the pale hue that had once been viewed with a certain suspicion.”33 This change was driven by the work of three men, all closely associated with the Pasteur Institute in Paris: during the course of the 1890s and early 1900s, Georges Roger, Émile Louïse, and Professor Pierre Mazé sought to bring scientific control to the fickle rinds and fermentations of Brie and— eventually—Camembert. Roger, the retired owner of a factory producing millstones in the Brie region, devoted his dotage to the study of microbiology, and in 1896, he was asked by the producers of Brie to find a method to ensure greater consistency and stability in their cheeses. After consultation with the Pasteur Institute, he began a program of the careful selective culturing of samples scraped from cheese rinds. By 1897, he settled on snowy white Penicillium candidum as the best option for controlling and avoiding unwanted mold growth. He then proceeded to sell the pure cultures to cheesemakers. Roger advised them against working with native molds and instead recommended that creameries be thoroughly cleaned with an antiseptic solution to make sure that there could be no competition against his pure cultures. His laboratory became the first private laboratory to “deliver to cheese manufacturers . . . culture media made up of several types of microbes acknowledged to be the best.”34 With its synesthetic association of purity and hygiene—not to mention its enhanced shelf stability when compared with other molds—the near monoculture of a Penicillium candidum rind now defines the cheese. In effect, Camembert is a physical embodiment of the quest for absolute control. As we talk with Patrick, we become increasingly curious to taste his Camembert for ourselves. Unfortunately for us, while we have been sitting and chatting, locals have been dropping by the farm shop. One by one, they help themselves to the Camemberts in the fridge, leaving money in the honesty box. Once it is time for us to leave, no cheese remains. We ask Patrick if it would be possible to purchase one of the cheeses that we have seen maturing on his shelves and once again discover that he is a stalwart supporter of the appellation rules. He tells us that it is forbidden to sell Camembert de Normandie from the farm at less than twenty-one days old; the only cheese he has is twenty days old, so it is not available for sale. With a glimmer of stubborn pride, he gives us a list of local shops that might have some in stock, and we are on our way. In the car, classical tragedy is rapidly followed by farce. We hurtle down the autoroute surrounding Caen in search of Patrick’s cheese and find a suc116

•

MICROBES

cession of disappointing organic co-ops, none with any in stock. Desmasures and Cretenet become ever more apologetic, and we retreat to the car after each failure, giggling in exasperation. Ultimately, we find some of Patrick’s Camembert in an anonymous organic supermarket, hidden in the refrigerated section amid the health drinks and probiotics. The line at the checkout counter consists of elderly pensioners concerned above all with securing their month’s worth of dietary fiber. As we wait behind the bulk purchasers of Metamucil, we reflect on the price differential between the Camembert du Champsecret and its industrial counterparts: at retail, the Champsecret is just one euro more expensive than the Camembert de Normandie from industriel producers in the local hypermarket. Not only does it look like cheese made on a larger scale, it is distributed and priced in relation to industrial production. But still, we have not tasted it. Our opportunity to compare the cheeses comes the next day. On our way home, we are invited to lunch with friends in Paris, and in their chic apartment in the sixth arrondissement, we serve the cheeses that we collected in Normandy. Our fellow guests are intrigued as they taste their way through the selection. Unanimously, they prefer the cheese from Domaine de Saint Loup. Quite frankly, we agree. It is a technical marvel of modern Camembert, harmonious, rich, and at ease with itself. There is a delicious whiff of truffles and cabbage. It tastes, says one of our friends, “more like Camembert!” In contrast, the Camembert du Champsecret is caught in an awkward no man’s land, neither different enough from nor similar enough to the cheeses made on a larger scale. Partly, this must be an excess of youth—the cheese is at a painfully young stage of development—but the cheese is also more lactic in character, with higher moisture and spiky, angular flavors behind the mushroom notes of the rind. From our visit with Patrick, we cannot but wish that he would be willing to pursue a different path. His herd is wonderful, and we applaud his commitment to grazing, hay, and sustainable agriculture. We did not have the chance to witness him milking, but we can conjecture from his collaboration with Cretenet and her team that his raw milk has excellent biodiversity for cheesemaking, at least as an expression of the summation of his farming practices. Each little decision he has made helps make his milk more potentially interesting for cheese. But his subsequent cheesemaking decisions result in a cheese that will inevitably be judged within the context of the Platonic ideal for industrial Camembert. If raw-milk microbes are important, if they are something that is worth campaigning for and saving, then the best possible way for MICROBES

•

1 17

Patrick to showcase them would be by making a Camembert inspired by the 1890s, not the 1950s. Camembert de Normandie with a natural rind, most probably dominated by blue mold, would make the case for the vital importance of raw-milk microbes. It would not taste like Camembert as we know it, but that would be the point. Just as we have consistently seen in pastoral farming, to help a rare breed survive, you have to eat it.

118

•

MICROBES

SEVEN

Risk

We nearly kill someone before we even see a cheese on our visit to HauteSavoie. Our friend Jean-Pierre Missillier, a local affineur, has invited us to visit cheesemakers in his picture-perfect corner of the French Alps, and on this bright February day, the resort town of Le Grand-Bornand is packed with skiers, who fill its toytown streets. Missillier’s affinage facility is near the center of town, but in order to get there, we must drive across the municipal cross-country ski track. The curve of the hill creates a lethal blind spot, and as we are about to pull out, a skier suddenly emerges: we slam on the brakes and come to a skidding halt, and he skims across in front of us, hurtling along in blissful ignorance of his narrow brush with catastrophe. We sit, hearts racing, and then gingerly continue across the track. When we make it to the facility, Missillier, jovial and ruddy faced, greets us and excitedly starts to show us around. We are here to see one of France’s cheese icons. Reblochon is a one-pound pancake of a cheese, with a glossy paste and a thin pink rind. Often misclassified as a stinky washed-rind, Reblochon is neither smelly nor washed. When it is normally eaten, at around four to six weeks of age, it tastes surprisingly gentle, like milk with a hint of sour fruit and warm yeasted dough. It is a study in quiet balance and understated complexity, at its best intensely expressive of the milk from which it is made. Despite its mellow character, Reblochon is a cheese of defiance. Its foundation myth dates back over seven hundred years, to an era when farmers paid for their animals’ grazing rights based on the amount of milk they produced. Eager to reduce the amount they had to pay, the farmers hatched a plan to milk their cows out only halfway on the days when the milk volume was measured. After the landlords went on their way, the farmers returned 119

to the barn to “re-pinch” (reblocher, in the local dialect) the teats to extract the rest of the milk and turn it into cheese. Reblochon has continued to find itself at odds with rule makers in more recent years. Even within tolerant France, where raw-milk cheese is vehemently defended, Reblochon is considered a high-risk cheese, and the services vétérinaires (health inspectors) have come down hard on farmhouse cheesemakers over their rustic infrastructure and reluctance to modernize.1 One can imagine the rationale. If raw milk has been reviled by some health officials as “liquid death,” then Reblochon, with its meltingly soft texture and almost complete lack of protective acidification, is a perfect petri dish. Reblochon is liquid death amplified. Several months before our visit, Missillier had visited London with a group of his fellow Reblochon producers and maturers. Bronwen and her colleagues showed them around the maturing facilities at Neal’s Yard Dairy, hoping to impress them with their innovative approaches to climate control and the wide variety of farmhouse cheeses produced in the United Kingdom. Instead, the group seemed most impressed—and more than a little bemused—by the Anglo-Saxon fi xation with donning hairnets, overcoats, and shoe covers and by their hosts’ obsession with handwashing. Missillier offers to take us to see one of his favorite producers, Madeleine Daviet, whose winter cheese dairy is just a few miles up the road, occupying the ground floor of her chalet-style house. When we pull into their driveway and step out of the car, there is nobody around; the gentle rustling of the cows within and rough-woven cheesecloths air-drying in the winter sunlight are the only signs of habitation. After a few minutes, Daviet appears, pulling her young son on a plastic sled behind her across the snowy terrain. She greets us enthusiastically, and we stride directly into the dairy, pausing briefly to say hello to the forty Abondance cows lolling in their stalls just steps from the open door to the cheese room. Built to thrive on the rough mountain slopes on a diet of grass, the cows stand there like great contented tanks; they look up with mild curiosity from their hay to regard us. “They sure do know how to eat!” says Daviet, rolling her eyes; providing highquality hay through the winter, which is critical to the quality of the cheese, is the largest cost to the business. What is first striking about the dairy is its stripped-down minimalism. There is no refrigerated milk tank, no hulking pasteurizer, no trolley stacked with tools. It is just a spacious room with a tiny vat and a table, on which seventy fresh cheeses, only a few hours old, sit plump and glistening in their 120

•

RISK

molds. Daviet, her hands clean but unwashed, pops one out of its mold and gives it a squeeze to show us its pliant texture; within a few hours, these rubbery discs will move onward to the brine bath, making room on the table for the evening cheeses following fast on their heels. The second thing that is striking is the dairy’s spotlessness. Sunlight streams through the windows, filling the room with light. The smell is lactic and wholesome, and there is not a speck of curd out of place. In an adjoining room, several hundred young cheeses sit perfectly aligned on wood planks, spaced as if by a ruler, exuding thick, glistening serum. Again, Daviet picks a cheese up, revealing an identically shaped imprint on the plank; over the years, the whey and calcium deposits given off by the cheeses have etched a circular pattern onto the wood, the ghostly footprints of batches past. Glancing nervously at the boards—after all, this is why we have come here—we ask about her cleaning protocols. She is quick to reply: “Oh, we just wash with water. We’ve never used a disinfectant in here since my father started making cheese over twenty years ago. Not on the molds, not on the boards. Actually, the only chemical we use is on the milking machine, and then only after the morning milking. In the evening, we just rinse it with water.” Both Daviet and Missillier smile sweetly, enjoying our incredulity a little too much for comfort. Daviet beckons us down a corridor and past an alcove containing a collection of prizes and trophies arranged into two neat shrines, one for cheesemaking and the other for skiing. Everyone in the family wins awards: her brother Benjamin is on the French Paralympic cross-country skiing team and won a bronze medal in Sochi. We progress through into a bunker-like chamber at the very heart of the building. In stark contrast to the make and draining rooms, the ripening room is dark, with thick cement walls and dirt floors, and it smells of clean earth and minerals. The air is still and moist, and the walls are wet with condensation and dark with black mold. Looking up, we are greeted by an extraordinary sight: the ceiling is covered with caviar-like beads of condensation, ranging in color from vermillion to lemon yellow to black. Large plastic boards sit poised on the top shelves to deflect inevitable drips. On the wooden shelves lining the walls sit hundreds of pristine young cheeses, each with a flawless rind resembling the dusting of fuzz on a just-ripe peach. Here, amid mold-covered walls, dirt floors, visitors in street clothes, and a phantasmagorical array of wild microbes practically pouring from the ceiling, are cheeses more visually pure than anything we have encountered in any plastic-clad, positive-pressure, hospital-inspired maturation facility. Perfect rinds are maddeningly difficult to achieve, but looks are skin-deep, and, like RISK

•

121

a mushroom, the beauty of a cheese has no bearing on its safety. As if anticipating our next question, Missillier digs out a sheaf of recent test certificates. Despite their relaxed attitude toward the native microbial residents of their dairies, Reblochon producers spend a fortune on microbiological testing to verify for themselves and for their customers that the residents of their finished cheeses are indeed benign and flavorful rather than dangerous. A quick glance through the results confirms that the cheeses are made from raw milk, but page after page attests that they contain no pathogens. The dairy smells clean, it feels clean, the test results are fine, and the cheese tastes amazing. What supernatural force or Alpine god has been invoked by the Reblochon makers to descend from the hills and protect their cheese?

DISCIPLINING MILK

The regulation of cheese, deciding what is and what isn’t safe to eat, is caught up in the fraught discussion of milk, hygiene, and safety. Cheese sits awkwardly in this debate, on occasions a belligerent, but often simply a collateral casualty. The important truth, seemingly self-evident but frequently forgotten, is this: cheese is not liquid milk. We all too easily forget that the widespread consumption of fresh milk is not rooted in historical practice; before the era of refrigeration and efficient transport, the products of dairy farms were cheese, butter, beef, and whey-fed pork, not fresh milk. At the World Expo in Paris in 1900, one of the exhibits that caused the biggest stir was a display of sweet milk and cream from American dairy farms.2 Today, the feat of being able to consume fresh milk thousands of miles from the place where it was produced is something we take for granted, but it has only been made possible through modern technology. On early dairy farms, fresh milk was made directly into cheese, and the natural souring of the milk played an essential part in that process. Dirty milk had consequences, and the cheese would tell the tale, either immediately, in the form of gas-producing microbes that inflated the curd, or later on, as off-flavors. There was also recourse: the person who milked the cows was right next door, and the decreased resale value of the product, or simply having to eat bad cheese, was a penalty that the whole team would take together. Everyone had skin in the game. But when farmers began to sell their milk at a distance, the consequences were out of sight, out of mind. If anything, liquid milk was a more sensitive 122

•

RISK

material than cheese, and the most common complaint was that it would sour before it reached its customers. Any pathogens that were present in the milk had the opportunity to multiply over the course of hours, causing stomach upsets or worse. But milk often changed hands several times before it was consumed, and this made meting out punishment for a substandard product more difficult. Until the appearance of the railroads, urban milk came from city dairies, where cows were often kept in cramped conditions and fed on industrial byproducts, such as brewer’s grains. The dirt, hair, and straw that were often found in the churns and the sediment at the bottom of glasses of milk were testament to the questionable (or at least variable) standards of production. The squalid conditions made production of safe milk a challenge, and as early as the 1840s, social reformers in New York were already raising awareness of the threat posed—particularly to infants—by such “swill milk.”3 While the advent of the railroads meant that milk for cities could be produced under better conditions in the countryside, practices like decanting milk from many farms out in the open on railroad platforms provided even more opportunity for contamination. The quality of the lot was reduced to that of the least scrupulous producer. Early attempts to improve the quality of commercial milk focused on reducing the level of visible contaminants entering the milk supply, but as the science of bacteriology took off in the late nineteenth century, the attention shifted toward invisible agents of disease. Pathogens could be introduced and propagated through careless handling, but in 1882, microbiologist Robert Koch identified the bacterium responsible for tuberculosis, leading to the rapid realization that milk from dairy cows—which were known to carry the disease—might be a significant vector for a pathogen that at the time killed one out of every seven people.4 Public health officials sprang into action. Farms began to be subjected to veterinary inspections, and animals with signs of disease were condemned; farmers who knowingly sold milk from diseased cows were prosecuted.5 These were the early days of microbiology, but tests of the milk itself (including direct microscopy, bacterial plate counts, and injecting guinea pigs to see if they developed signs of disease) were a start. Unfortunately, the tests did not work within a practical timeframe for removing contaminated milk from the food chain.6 Veterinary inspections of farms and certification schemes designed to ensure higher standards of production made slow progress, but even these were not very effective at eliminating tuberculosis in drinking RISK

•

123

milk. A study carried out in the early 1930s in the United Kingdom revealed that over 5 percent of samples of Grade A milk fed to school children tested positive for the presence of Mycobacterium bovis, the organism responsible for bovine tuberculosis.7 There were also brucellosis and mastitis bacteria such as Staphylococcus aureus (which produces a toxin that causes rapid-onset food poisoning) and Streptococcus pyogenes (the organism behind scarlet fever) to contend with. In short, raw milk was a public health nightmare. Pasteurization, the heat treatment of milk to kill potential pathogens, represented a downstream control capable of breaking the cycle of infection, and it was adopted in fits and starts over the course of the early to mid-twentieth century as various jurisdictions became aware of the dangers posed by raw-milk consumption.8 Pasteurization brought additional benefits beyond safety: it also kept milk from souring and increased its shelf life. Fears over pasteurization depleting the nutritional value of milk were also gradually laid to rest.9 For those who don’t live on dairy farms but want access to fresh and safe liquid milk, pasteurization is an ideal technology. And that is the point. Pasteurization is an ideal control for liquid milk, a product that will not be further handled until it reaches the consumer. Cheese is different. Cheesemaking is another technology for preserving milk, which it accomplishes through controlled microbial activity: the whole point of making cheese is to select for and encourage the growth of specific microorganisms. As might be expected, cheese brings with it very different risks from raw milk. To take one example, Campylobacter jejuni, responsible for diarrhea, stomach cramps, and fever, is one of the most common causes of illness associated with the consumption of raw milk. Indeed, it was responsible for some 54 percent of all the illness associated with the consumption of raw-milk dairy products in the US Centers for Disease Control and Prevention’s survey of the United States from 1993 until 2006.10 But C. jejuni is astonishingly fragile and vulnerable to just about every aspect of the cheesemaking process: it is sensitive to acidification, curd heating, the normal levels of salting, and even the ripening temperature. Properly made cheese does not present a health risk from Campylobacter.11 We are not attempting to suggest that raw-milk cheese is in some way magical or that the act of cheesemaking conveys a shamanistic protection from all pathogens. To the contrary, we are deeply suspicious of any cheesemaker who tries to make this case. Let us be very clear: cheese can carry 1 24

•

RISK

pathogens that can result in very serious illness. Invasive Listeria monocytogenes infections have a fatality rate of over 20 percent, and highly pathogenic serotypes of Shiga-toxin-producing Escherichia coli can lead to acute renal failure, particularly in young children.12 These risks cannot be shrugged off or taken lightly. But cheese, with its processes of fermentation, drainage, salting, and aging—as well as the extra time its maturation period provides for running tests and monitoring systems—can be managed and controlled. And what of the distinction between raw and pasteurized cheeses? Pasteurization eliminates any pathogens that may be present in the milk, but the process of cheesemaking offers plenty of opportunities for contamination further down the line. These risks are the same regardless of whether a cheese is made from raw or pasteurized milk. Indeed, speaking off the record, one microbiologist described a fully ripe pasteurized-milk Camembert as “the environment you would design from first principles to incubate Listeria monocytogenes.” A high-risk cheese is simply a cheese that is a good petri dish for whatever makes its way into or onto it. A low-risk cheese is a barren desert in which no pathogen will survive for long. Epidemiological evidence bears this out; the simplistic division between “risky” raw-milk cheese and “safe” pasteurized cheese is flat-out wrong.13 We wince when we observe pregnant women refusing (dry, aged, acidic) raw-milk Cheddar but tucking into (highmoisture, low-acid) pasteurized Gorgonzola. Another approach is needed.

S PAC E AG E F O O D S A F E T Y

Sixty years ago, there was no industry-standard approach to food safety. Each company approached quality control by starting from first principles, and there were a wide range of strategies. Against this backdrop, the role of regulators was to inspect each facility for evidence of health code violations: things like missing window screens, leaky gaskets, poorly cleaned equipment, staff failing to wash their hands, or euphemistically termed “evidence of pest activity.” By and large, this style of quality control worked all right; most food was safe most of the time, as most food tends to be. But the early 1960s were the era of the space race, with the Soviet and American governments mobilizing their collective resources to demonstrate dominion over the heavens. NASA was all too aware that space capsules were not the only things that might experience explosive decompression at zero gravity. An astronaut with RISK

•

125

food poisoning on a Gemini or Apollo spacecraft would jeopardize not only the lives of everyone on board, but the entire world order. To ensure that the “vomit comet” was only ever part of training, a team was assembled at the US Army Natick Laboratories in Massachusetts and given the task of producing zero-risk food for the astronauts. Their rigorous tests of raw ingredients showed that viruses and bacteria were omnipresent, and the risks were heightened by the fact that astronauts under stress would likely have compromised immune systems to begin with. The original plan— to test samples of every batch of food for every conceivable pathogen—was prohibitively expensive, and much of the food was destroyed in the very process of testing it. In charge of solving this problem was Dr. Howard Bauman, a microbiologist recruited from Pillsbury, which had signed a contract to help develop and produce some of the space food. He and his team started at the beginning, by listing all the known hazards—physical, chemical, microbiological— and in doing so, they began to conduct the first rigorous food-risk assessment. The protocol that they came up with required their suppliers not only to pass stringent requirements of sanitation (much of the space food was manufactured in positive-pressure “clean rooms”) but also, and more importantly, to map out the risks within each process and the control points in place to negate them. It brought the same logic of engineering the safety of complex systems that was used for the rockets themselves and applied it to the food for the astronauts. Each and every point had to be signed off on for every batch, and careful records were kept to ensure this. The Hazard Analysis and Critical Control Point (HACCP) system was born, and the Gemini and Apollo mission astronauts were the first to consume its fruit. Early space food produced using HACCP-based quality systems included midcentury favorites such as butterscotch pudding, shrimp cocktail, and chicken with vegetables.14 Bauman and his colleagues at Pillsbury became practiced in the efficient application of HACCP in the factory, though the special method was used only for the items the company manufactured for NASA. But when shards of glass were found by a customer in Pillsbury’s farina baby food in 1971, Bauman decided that the most effective way to prevent another food scare would be to apply a HACCP-based approach across the board for all Pillsbury products. That same summer, botulism in canned vichyssoise soup from a different manufacturer killed one person and sickened others in New York, and public concerns over the safety of canned food forced the National Canners Association to take action to save their tarnished reputation. Low-acid can126

•

RISK

ning suited the HACCP approach to a tee, and the Canners Association gave their members sixty days to adopt it. The US Food and Drug Administration (FDA), under pressure to show that it was policing the industry, signed on to the new requirements, and the use of HACCP soon became a legal requirement in the United States for all low-acid canned food producers.15 Since then, and largely in response to specific food-safety crises—such as the Jack in the Box E. coli O157 outbreak in 1993—the use of HACCP within food-safety systems has become more and more widespread, and it is specifically required for juice and meat production within the United States. Also in 1993, the Codex Alimentarius (a set of international food-safety recommendations) incorporated guidelines for the implementation of HACCP systems, and since 2004, HACCP has been required for all food-business operators within the European Union.16 Using HACCP does not mean that food producers never need to be inspected by the authorities, but it does fundamentally change the nature of such visits. While inspectors always spend time on the production floor observing working and sanitation practices, the emphasis of inspections within HACCP-based systems is on examining how the risks specific to a company’s own processes are being managed. Some may complain that HACCP is “all about paperwork,” but this is a fundamental misunderstanding of the approach. HACCP is about identifying real risks and documenting that the critical steps to control them have been taken for each and every batch. It doesn’t excuse people from keeping their premises clean or keeping rodents out; these elements are also a part of every quality system. But using HACCP means that food producers—and inspectors—have to look beyond the dust on the windowsill and prioritize the real sources of risk at the heart of their processes. HACCP principles work perfectly for foods that are designed to go through at least one “kill step” that obliterates all pathogens, such as heatprocessed canned green beans or pasteurized milk. But raw-milk soft cheeses are different beasts. The whole point when making these cheeses is to provide a clement environment for the microorganisms in the raw material to grow exponentially. Surely, under such a system with no true critical control points, HACCP becomes null and void? Far from it. In practice, HACCP is a process not of risk elimination but of risk reduction, and proper use of HACCP principles means identifying risks and putting systems in place to lower them, not to zero, but to acceptable levels. In this, raw-milk cheese is not an exception. These cheeses are RISK

•

1 27

actually very similar to all other fresh, ready-to-eat produce, which cannot undergo a heat-treatment step: just try pasteurizing some salad leaves for a delicious light lunch. The best and most robust quality systems put numerous controls in place at all points in a process to reduce overall risk to a low level. It’s a very different approach than relying on a kill step to make things that are out of control suddenly safe again. As one cheesemaker graphically put it, “If I took a dump in a glass of milk and then pasteurized it, would you be happy for me to make cheese with that?” The new US Food Safety Modernization Act, signed into law by President Obama in 2011, is a forward-thinking piece of legislation that recognizes this fundamental distinction and makes it explicit. Rather than mandating the use of HACCP for all food businesses, it stipulates a related approach called Hazard Analysis and Risk-Based Preventive Controls (HARPC) for industries other than canned food, meat, and juice, which already have HACCP requirements in place. The new food law recognizes that complete microbiological safety for all fresh, ready-to-eat foods—not just cheese, but also things like fruits and vegetables—is impossible to guarantee unless we plan to irradiate them immediately before consumption. But the same hazard-analysis principles are also highly effective at bringing the risk to the customer down to a very low level. However, for HACCP and HARPC systems to operate effectively, they require a profound reorientation of the relationship between producers and regulators, one that many jurisdictions—most notably the United States— have failed to embrace. HACCP makes that sitcom plot staple, the traumatic visit from the health inspector, largely obsolete. Its focus on systems and verification positively encourages collaborative work: inspector and producer are working together to make the safest food possible rather than simply surviving a confrontational one-time inspection. Under a good HACCP system, there would be no sitcom plot. Basil Fawlty, his employees, and the inspector would instead sit down and discuss the appropriateness of their testing regime over a cup of tea. This testing is the key. HACCP and HARPC necessitate sensible, targeted, and diligent testing on the behalf of the producer. Just as with the space program, this does not demand the destructive end product testing of each batch but rather calls for an astute assessment of the individual potential risks, along with systematic testing to a schedule that will verify that the systems in place are operating effectively. Absolutely verifying the absence of pathogens is a physical and philosophical impossibility, but a well-designed 128

•

RISK

regime can demonstrate that the system is running according to plan. It is the construction of the system that keeps risk down to an acceptable level. The problem comes when there is a legal and regulatory incentive not to test. In a jurisdiction like the United States, where the presence of a foodborne pathogen at a processing facility means that the producer is in violation of federal law, there is every incentive for him or her deliberately to look the other way. The consequence of a positive test is simply too great. It interferes with research and the acquisition of knowledge too. When Professor Catherine W. Donnelly of the University of Vermont, the foremost US authority on Listeria in cheese, was conducting her survey of Listeria monocytogenes in creameries, she first made sure that the president of the university gave her a guarantee that he was prepared to go to prison rather than let the federal government subpoena her results to look for the positive results.17 Test results have to be morally neutral tools for ensuring safe food rather than a cue for the threat of prison. At a time when the medical profession is exploring how to use mortality and morbidity meetings (the debriefings between physicians after a patient has suffered a negative outcome) to systematically improve care, it is time for public health infrastructures similarly to embrace the essential utility of a safe space to discuss what went wrong and how to learn and improve from the experience.18 All of this brings to mind an old rabbinical story of a shipwreck. A particularly devout sailor is clinging to the wreckage and struggling to stay afloat, when along comes a rescue helicopter. “Ah no,” he says, “I have prayed to God, and he will rescue me.” The helicopter flies away. Sometime later, a merchant vessel encounters the sailor, and again he insists, “I have prayed to God, and he will rescue me.” It sails on. Eventually, a lifeboat reaches the sailor, but yet again, he insists that God will save him, and it sails away. Finally, he can cling on no longer, and he drowns. As he encounters God in heaven, the perplexed sailor asks why he was not saved, to which God replies, “But I sent a helicopter, a merchant ship, and a lifeboat. What more do you want?” So it is for positive tests for pathogens within a well-run HACCP or HARPC system. We are not waiting for some mysterious intervention from the Almighty to make our cheese safe. Testing is our lifeboat. Indeed, occasional positive test results are part of a system at work. They allow producers to detect and quarantine any products that are not up to standard, but just as importantly, they allow them to correct the weaknesses in their systems that allowed the problems to occur, continuously improving their level of control. RISK

•

129

T H E S H O C K O F T H E R AT I O N A L

Back in Haute-Savoie, after our visit to Madeleine Daviet, we drive down the snowy valley to the village of Thônes, where we meet Bruno Mathieu, the head of research for the Syndicat Interprofessionnel du Reblochon. The syndicate is the organization that represents all the members of the Reblochon making and selling community, including the 130 farmhouse producers, maturers and distributors like our friend Jean-Pierre Missillier, and larger factories. Mathieu has spent over thirty years within the industry. His job is to conjure up the resources of the scientific community and employ them to understand and defend traditional practice, helping producers work smarter and better in the process. With his shock of white hair and calm and eloquent manner, he is the man called in to bat when Reblochon faces its periodic existential crises. One of the most profound of these was over the use of wooden boards in the dairies. The maturation of hard cheeses on wooden boards is widespread—and for some cheeses, like Comté, it is positively required—but the maturation of soft cheeses on wood, particularly during the early stages of their production, is much less common; the use of wooden boards for Reblochon is one of its salient and unusual characteristics. When new hygiene regulations were put in place in 2006, maturation on wooden boards was one of the first practices to be challenged. As Mathieu tells us, he and his colleagues approached the problem directly: We analyzed the different ecosystems on the boards and showed that the most numerous organisms were the ones that were technologically interesting. Geotrichum, lactic acid bacteria, white molds, and ripening bacteria were all installed squarely in the wood. It’s one way in which each farmhouse cheese dairy has its own ecosystem, unlike the industrial producers who have taken those reservoirs away and then have to add the ripening cultures to their milk each day. But those who opposed the boards said, “That’s fine, you’ve shown that there are beneficial microbes on the boards. But what about when there are pathogens present?”

So in the lab, the scientists proceeded to inoculate Listeria onto two sets of boards, one that had been cleaned in the traditional way, with cold water and no chemicals, and one that had been heat-treated to kill the microbial biofi lm living on its surface. They then seeded a moderately high level (one thousand organisms per square centimeter) of Listeria onto the boards, popped them into an incubator at a comfortable growth temperature, and 13 0

•

RISK

waited to see what happened next. After twelve days in the incubator, the living biofi lms had significantly inhibited the growth of the Listeria; viable Listeria could still be detected, but the levels had dropped by a factor of five or more. The story was very different on the boards whose native biofilms had been burnt off. With no competition for resources or nutrients, the contaminating pathogen grew unchecked, its levels rising to over 10 million per square centimeter. While the protective effects of the living biofi lm would not “fi x” a contaminated cheese, it is a powerful ally, helping to ensure that foreign colonizers—be they pathogens or ugly spoilage molds—don’t take up residence where they don’t belong.19 Mathieu went on: Of course, we also know that the boards have to be used carefully and cleaned well. People have had problems with Listeria from time to time, it’s true. And systematically, you have to check the boards. I’ve heard from producers who say, “I’m having problems with my results from the boards. I’m going to chuck them on the fire.” And I ask them, “How long have you been using those boards? They might be from your grandparents, used for thirty years without problems. According to Pasteur, there’s no such thing as spontaneous generation. If there’s Listeria on your boards, you put it there.” So I ask them how they wash their boards, and it turns out they take them outside and spray them with a power washer. They pulverize the ground and distribute Listeria from the dirt over their boards. It’s not the boards that are the problem, it’s the way that they’re handled. If you treat them well, they’ll do great service. Now, some people are using metal or plastic shelves instead, but if they are handled the same way, they’ll have exactly the same issues.

Finding a middle ground between the demands of the health inspectors and the practices involved in making farmhouse Reblochon has not always been easy, and despite the efforts of Mathieu and his colleagues, concessions have had to be made that have spelled the end for some producers. When Mathieu began working for the syndicate, there were still farmhouse producers making cheese entirely without starter cultures; the changes in hygiene and sanitation standards over time have meant that the old wooden cheese rooms up in the alpages, where the milk was warmed over open fires and the floors were bare earth, could no longer continue. “What a Reblochon! It was lost.” He pauses, deep in his thoughts for a moment. On one occasion, he was called in to help a cheesemaker upgrade his dairy, and together they installed a new brick wall in the nursery, where the young cheeses spend their first days: RISK

•

13 1

The brick wall was perfect for maintaining the humidity. If the room got too dry, you could rinse it with water and it would maintain the ambience. The local inspector from the vet service was a hygienist like no other! He said to me, “What if you get Listeria on there, how are you going to get rid of it?” And I replied, “I’m not going to ripen my cheese on the walls!” We don’t normally have those kinds of problems anymore; they have accepted the use of porous materials. But at the farmhouse level, the disappearance of wood, copper vats, beaten earth and stones—it has disturbed the ecosystems. Back in 1980, when I had just started at the syndicate upon leaving school, I visited a producer with a maturing room made of wood. The walls were black! But there was not a trace of grey mold on the surface of the cheeses. I visited that cheesemaker again years later, after they modernized their ripening room. Within six months, they had to call in the technicians to help them. The room was spotless, but the cheese rinds were entirely rotten with black mold. There are still compromises that we’ve made; it’s an evolution. At the level of working with the inspectors, if we make an assertion, we need to prove it. It’s difficult and expensive. But at least, if there’s a disagreement, they listen to us. We have the power of being a group.

One visiting foreign health official once expressed astonishment that in France, the homeland of the great Louis Pasteur, such effort was spent trying to fight pasteurization. Mathieu’s response was nimble: “We aren’t arguing with Pasteur at all; to the contrary, it is his work that has set us on the path to understanding and managing our microbial ecosystems effectively!” This brings him to another key point: It is great that the American government is beginning to adopt HACCP more widely, but the manner in which they’re interpreting it is very different from ours. There, it’s a matter of, “We can work carelessly as long as at the end of the process, we have a treatment that reduces the risk to zero.” Here, in France, we apply HACCP principles from the start. We know that we have to work carefully and obsessively right from the beginning because there is no treatment later on. That’s it: the power of raw milk! Beyond delivering a good taste, it’s that everyone along the line of production respects the importance of controlling the hygiene at each and every step.

As we leave Mathieu and head back toward the airport, we are delighted and just a little giddy. He is, to our minds, the ideal figure to work with producers on cheese safety: his approach uses a genuinely evidence-based framework to apply sensible controls. Mathieu’s work with his 130 small farmhouse producers also underlines to us one of the key limitations of the new Food Safety Modernization Act in the United States: the exception that 132

•

RISK

is made for “very small” businesses, those with sales of less than $500,000 per year.20 The degree of risk to the individual consumer and the effectiveness and practicality of risk assessment–based approaches like HACCP or HARPC have absolutely nothing to do with the size of a business. These systems work at every scale. For the sake of the industry, every commercial producer must take responsibility for understanding, controlling, and monitoring their control over the risks in their own process. Even hobbyists making cheese at home have the capacity to understand and apply hazard-analysis principles; doing so will help them understand their process and may well improve their cheese at the same time. While Mathieu is not the regulating authority—indeed, he bears the scars on his back from tussles with inflexible officialdom—he is an advocate for rational pragmatism: he is aware of the risks involved in making cheese, and his work has been part of a collective effort to discover the best methods to keep them under control. His is a world in which every advocate for smart regulation, from Professor Catherine W. Donnelly in Vermont to worried dairy technicians throughout France, would love to live.

RUNNING THE NUMBERS

So, exactly how risky is “high-risk” cheese, and how does this risk compare to that of other foods? We can examine the prevalence of pathogens in animal and food sources by testing them, but as the CDC’s own team acknowledges, without the capacity to cross-reference this data with outbreak surveillance, it is useless for attributing disease to an individual food. We might know that a potential pathogen is out there, but we have no idea if it is actually making anyone sick. The use of modern genetic techniques, including whole genome sequencing, offers an extremely promising avenue of enquiry for connecting pathogen and source.21 Approaching the question from the opposite direction, the challenge of obtaining accurate statistics for outbreaks also makes it difficult to arrive at a coherent statistical picture. To put it bluntly, we do not like talking about our shit. The confessional privacy of the water closet makes people reluctant to talk in public about the intricacies of their gastrointestinal experience. It is only when patients both present themselves to a physician and have their samples tested that an absolute diagnosis can be recorded. Under these RISK

•

13 3

conditions, the number of laboratory-confirmed cases of illness is sensibly assumed to be the tip of the pyramid: at each stage, some cases fail to progress to a point where they will eventually be recorded in outbreak data.22 To bridge the data gap between laboratory-diagnosed cases and the plausible suggestion that many people probably never bring their symptoms to a doctor, the data on the prevalence of outbreaks takes the form of surveys and “expert elicitation,” essentially a structured form of best guess. We do, however, have good evidence—perhaps reassuringly—that the two symptoms most heavily associated with seeking medical attention and submitting the all-important stool sample are experiencing diarrhea for more than three days and suffering from bloody diarrhea.23 This does at least suggest that the outbreak surveillance data that we do have most probably overrepresents the most severe cases. The potential pathogens in cheese are most certainly not the bacteria that originally caused the introduction of pasteurization in the dairy. Although pasteurization played an essential role in bringing food-transmitted cases of tuberculosis to an end, a simultaneous campaign was being fought by veterinarians and farmers to eliminate these diseases from their herds altogether. Today, animal-borne diseases are strictly controlled at the farm level, making the risk of infectious raw milk reaching humans extremely low.24 Now, it is the more recently discovered organisms related to environmental contamination—such as Listeria monocytogenes and pathogenic strains of E. coli—that milk and cheese producers worry about. Armed with this knowledge, our problem becomes one of weighing relative risk, something at which humans are singularly poor. Outbreak and surveillance data can easily be crunched and presented as alarming statistics of the relative threat of raw versus pasteurized milk cheese. In their effort to communicate these risks, many public health agencies attempt to humanize the story and argue through affect, recognizing that bringing statistics to an emotional fight is never a winning strategy. The CDC is typical of such agencies, at least within the Anglo-Saxon world, and its web page to answer queries about raw milk features videos of plaintive mothers ruing the day that they ever decided to feed their children raw-milk products.25 The impact of these emotive stories is compounded by the persistence with which regulatory agencies present the risks of raw-milk cheese—that is, when they take the time to disaggregate it from liquid raw milk—in terms of the risk relative to pasteurized cheese. CDC researchers, in their study of outbreaks associated with raw-milk dairy products in the United States from 13 4

•

RISK

1993 to 2006, indulged in some back-of-the-envelope calculations in an effort to demonstrate the danger of unpasteurized milk.26 Estimating total US dairy production at 2.7 trillion pounds of milk during this period, they then came up with a best-guess figure that 1 percent of this was not pasteurized. Plugging this into their outbreak data, they arrived at the disturbing headline figure that, per unit of production, the incidence of outbreaks of foodborne illness is approximately 150 times greater with raw-milk products than with those made from pasteurized milk. Such a figure is immediately scary. No wonder they want to ban raw-milk cheese. However, we can do some back-of-the-envelope calculations of our own. We know from psychologist Paul Slovic that risk perception is heightened for activities that are unusual or alien.27 So let us take the CDC data for risks associated with raw-milk cheese and see how it measures up relative to that most culturally normative of American activities, driving a car. We can use the CDC researchers’ approximation that 27 billion pounds of raw milk were consumed between 1993 and 2006. They offer no suggestion of the relative balance between liquid raw milk and cheese and other dairy products, so for our initial calculation, we will take the most extreme numbers possible and say that all 27 billion pounds were consumed as cheese. This is clearly not true, but we can adjust our figure later. With standard cheese yields, 27 billion pounds of milk will give approximately 2.7 billion pounds of cheese. In the CDC’s data set, there were two deaths associated with the consumption of raw-milk cheese during the period of study, so we arrive at a figure of 1.35 billion pounds of cheese per fatality. Now let’s analyze the road accident data. In 2006, the last year of the CDC’s data, there were 42,708 traffic deaths in the United States. During that same year, there were 3.014 trillion vehicle miles traveled.28 We can now calculate the relationship between pounds of cheese per death and vehicle miles traveled per death. Using these figures, we determine that an American consumer would need to eat 19.13 pounds of raw-milk cheese to have the same risk of dying as they would have from driving a single mile in the United States in 2006. To give the figure a little more context, in 2006, average per capita annual cheese consumption was a shade over 32 pounds, while average per capita vehicle miles traveled was 10,101.29 If one of those average Americans decided to eat only raw-milk cheese rather than pasteurized for the entire year, he or she would still be six thousand times more likely to die in a motor vehicle accident than from eating the cheese. Suddenly, raw-milk cheese does not look so dangerous. RISK

•

13 5

Of course, this is an overstatement of the safety of raw-milk cheese, simply because not all of the raw milk produced was made into cheese. We have also not taken into account negative outcomes other than death, but for the purposes of our model, let us assume that road traffic accidents injure and maim at an at least comparable rate to the nonlethal consequences of foodborne illness. Even if only a tenth of the raw milk produced was turned into unpasteurized cheese, it would still equate eating 1.9 pounds of cheese with the risk of death from driving one mile. Expressed this way, we are reassured in the creamery, if somewhat scared on the freeway. Nonetheless, such weightings of relative risk are the key measures used to form policy at the national level. A recent joint report by the FDA and Health Canada, Quantitative Assessment of the Risk of Listeriosis from SoftRipened Cheese Consumption, is especially instructive.30 Their statistics, which model the risk of invasive listeriosis from the consumption of a ripened Camembert-like cheese, express risk in a predicted number of servings required to generate one case of illness. Again, the relative numbers appear to be a huge cause for concern. For the general US population, the model predicts that a raw-milk cheese would increase the risk 157 times over that of cheese made from pasteurized milk. (In this relative degree of risk, it is consistent with the CDC’s earlier study of outbreaks between 1993 and 2006.) But the assessment also lets us see the predicted absolute risk. Staying with the US general population, the model predicts that 55 million servings of raw-milk cheese would be required to generate one new case of invasive listeriosis. This makes cheese from raw milk more dangerous than that made from pasteurized milk, but the absolute risk is still vanishingly rare. To give an idea of the scale, in 2013, there were only just over 20 million 250-gram raw-milk AOP Camemberts de Normandie produced for the entire world.31 Now, the FDA would no doubt say that the average serving size of a softripened cheese like Camembert de Normandie is considerably smaller than the entire cheese; their modeling suggests that the median amount consumed is less than (or equal to) 40 grams.32 But even if we allow for a 40-gram serving size, it still means that the US general population would need to eat half of all the global production of AOP Camembert de Normandie each year just to give rise to one single case of invasive listeriosis. The assessment is even more revealing when it breaks down some of the demographics of the US population to arrive at a more nuanced and specific idea of risk. Pregnant women immediately stand out: whereas for the general public, it is predicted that there will be one case of invasive listeriosis per 55 13 6

•

RISK

million servings, for pregnant women, there is a risk of one case per 570,000 servings. This is a striking contrast, one that helps us understand why the identification of Listeria monocytogenes as a foodborne pathogen was made extraordinarily recently: it was so rare for any otherwise healthy person to contract listeriosis that nobody imagined that the illness might be caused by an organism present in food. The original study that made the connection was only published in 1983, and making that leap required a perfect storm of circumstances.33 Unlike the other pathogens we have discussed, L. monocytogenes does not present in humans with the symptoms of enteritis. According to the original 1983 study, in adults, it was “an uncommon cause of bacterial meningitis and a rare cause of sepsis, endocarditis, peritonitis, or focal abscess,” while it was the third most common cause of bacterial meningitis in newborns.34 It took a major outbreak at the maternity hospital in Halifax, Nova Scotia, in the summer of 1980 for a prolonged and systematic investigation to arrive at a foodborne route of transmission. The culprit? Coleslaw made from cabbage that had been fertilized with the manure of sheep that had died from listeriosis. Looking back, the pre-1983 understanding of L. monocytogenes makes perfect sense in light of the FDA’s risk assessment. While the risk is extremely low for the general population, it is two orders of magnitude greater for pregnant women. Thinking back in terms of how it was understood before 1983, invasive listeriosis is not a risk to the general population; it is a risk associated with a specific condition: pregnancy. This matters, because as a society, we are quite prepared for food to be sold that has the potential to result in death if the wrong people happen to ingest it. The risk of death from acute anaphylaxis, for example, is all too real for those with severe peanut allergies. At the level of mere gastrointestinal distress, we are happy to allow milk sales even though many people cannot digest lactose. It is up to those with allergies or intolerances to understand and manage their own conditions.

B E YO N D T H E P R E C AU T I O N A RY P R I N C I P L E

That the same is not true for raw-milk cheese is a regulatory consequence of the precautionary principle. In legal scholar Cass R. Sunstein’s definition: “Avoid steps that will create a risk of harm. Until safety is established, be cautious; do not require unambiguous evidence. In a catch phrase: Better safe RISK

•

137

than sorry.”35 This is the logic of food regulation, where any risk at all is considered too high. Unfortunately, as we have seen, for all fresh foods that lack a definitive final kill step, we have to deal with the concept of acceptable risk rather than zero risk. This is as true for salad leaves as it is for cheese, and it is the inevitable consequence of eating a diet beyond canned food and MREs. But it also requires a sophisticated public conversation about what exactly constitutes an acceptable level of absolute risk and how that can be calculated. In his book Laws of Fear: Beyond the Precautionary Principle (2005), Sunstein launches a sustained assault on the intellectual edifice of the precautionary principle. In an instructive critique for those of us concerned with food safety, Sunstein suggests that the greatest logical flaw of the principle is its implicit failure to calculate the negative costs of retaining the status quo. If we operate within a model where there are no costs or consequences of inaction, then any change looks potentially terrifying. While Sunstein directs his attack at the European predilection for aggressive environmental regulation, his model is equally applicable to food safety on both sides of the Atlantic.36 Sunstein’s tool for overcoming the limitations of the precautionary principle is cost-benefit analysis, and this is officially part of the methodology of risk assessments in most jurisdictions.37 The problem comes in assigning values to both sides of the equation. While there is a thriving literature within the world of public health that seeks to provide quantitative estimates of the economic burden of foodborne disease, there is nothing of the sort that tries to estimate the value of ostensibly riskier alternative behavior. In 2003, the US National Research Council condoned milk pasteurization as eminently economically feasible simply because “virtually all fluid milk processors that ship milk products via interstate commerce have invested in equipment for pasteurizing their product.”38 There was no discussion of lost potential value in the cheese. Unsurprisingly, for regulators who find that there is no benefit to be gained from distinctive flavor, any extra risk—however small in absolute terms—is unacceptable. And so the challenge becomes: What is the benefit of raw-milk cheese? Let’s turn our thoughts back to AOP Camembert de Normandie. By law, this must be a raw-milk cheese, and cheeses made within the appellation attract a moderately healthy price premium. Just to keep our calculations simple, we can say that this premium is €1 (around $1.10) per cheese at the creamery door. In 2013, there were 5,112 metric tons of cheese produced, 13 8

•

RISK

which represents 20,448,000 individual 250-gram cheeses. So, with our estimate of the raw-milk price premium, we have already arrived at in excess of €20 million ($22 million) in extra value from raw-milk production, just within one appellation. For context, the US Environmental Protection Agency set the statistical value of a life at $7.4 million in 2006.39 This is not to say that the extra value generated from raw-milk AOP Camembert de Normandie should permit it to be responsible for almost three deaths a year to achieve a statistical break-even, but it does begin to chip away at the conviction that raw-milk cheese is all risk and no economic reward. A skeptic might point to the criticisms we made of cheesemaking in Normandy in chapter 6. If most of the character of a cheese comes from the mold and other ripening cultures with which the milk is inoculated, it is not dependent on its own microbes for anything other than a good story; in this case, surely the elimination of any extra health risk that can be achieved through pasteurization of the milk makes absolute sense? We are inclined to agree. Where a cheese is raw in name only and derives no substantive value from its native microbes, pasteurization is the only sensible course of action. But this is why AOP Reblochon de Savoie, with its vigorous farmhouse tradition and milk-expressive make, is such a good flagship for raw-milk production: it is a cheese that finds every bit of value in its milk. There is one additional revelation that we can find in a cost-benefit analysis. The dairy industry as a whole is remarkably safe and does an extremely good job of not making its customers sick. However, farming does reap a bloody harvest of the farmers themselves, largely by their own hands. Th is is a global problem, but US data is illustrative.40 According to CDC statistics from seventeen states in the United States in 2012, males employed in farming, forestry, and fishing (occupations that were grouped under a single National Standard Occupational Classification code for the purposes of the statistics) had the highest suicide rate of any occupation. This devastatingly high rate of 90.5 suicides per 100,000 was six times higher than that for men in the occupational group with the lowest rate, those working in education, training, and libraries.41 This is clearly a multidimensional problem, with causes ranging from the difficulty of accessing mental health support in rural areas to the ready availability of means of great lethality to the impact of chronic exposure to pesticides. However, the vicissitudes of the commodity market and the great capacity for financial loss in farming are also factors acknowledged by the CDC researchers. The pressures of small-scale commodity production put many farmers in an impossible position. In this RISK

•

13 9

context, if public health agencies have a mission to reduce preventable deaths, we strongly advocate that these organizations embrace schemes to teach and incubate farmhouse raw-milk cheese production as a structural solution for farmers confronting the challenges of the commodity market. Raw-milk cheese can save lives.

C R A F T I N G A D I A LO G U E

Cass R. Sunstein’s solution to the problems of the precautionary principle is simple. Taking into account the pathologies of public assessments of risk and their apparently near-inevitable cascade into moral panic, he suggests that regulatory authority should be delegated to an independent cabal of expert agencies: “If the public demand for regulation is likely to be distorted by unjustified fear, a major role should be given to more insulated officials who are in a better position to judge if risks are real.”42 For the food lover, this is very interesting because it exactly describes the existing situation. The drafting and implementation of food-safety legislation are complicated and highly technical; in short, they are well suited to technocratic elites. Th is should be where dull number crunching and probabilistic models of costs and benefits rule. The move within the food industry to embrace HACCP and HARPC is a welcome trend. In the United Kingdom, the Assured Code of Practice, drawn up by the Technical Committee of the Specialist Cheesemakers’ Association, has the force of law, codifying industry standards and providing a point of reference and appeal in disputes with individual local authorities.43 And yet, despite its excellent track record for safety, raw-milk cheese is not globally accepted. From North America to Europe to Australasia, raw-milk cheesemakers find themselves on the back foot, forever facing the next existential threat, and as concerned about regulators as they are about the microbes in their cheese. At a private conference on food safety in cheese organized in Providence, Rhode Island, in 2015 by Dr. Dennis D’Amico of the University of Connecticut, cheesemakers, retailers, and wholesalers spoke openly about their experiences with surprise inspections that just happened to follow on the heels of their making public statements that were critical of the aggressive, day-to-day implementation of regulation. When one considers these stories, it is unsurprising that, in black moments and private conversations, some cheesemakers assign hidden motives to the 14 0

•

RISK

actions of their regulators. It is the thrust of a vague and emotionally charged conspiracy theory: through the revolving door linking corporate big dairy with government agencies, regulatory capture has been achieved, resulting in a hidden agenda of dismantling raw milk and farmhouse dairying. The recent history of food regulation does not make such a possibility entirely unprecedented.44 It had been with a mixture of dazed panic and disbelief that the global cheese industry awoke in June 2014 to news that the FDA had apparently banned the use of wooden boards for ripening cheese. The ban ultimately proved little more than a farce. Amid a frenzy on social media, congressmen waded into the debate, and the FDA rapidly backed down, insisting that they had never actually banned wooden boards in the first place. As an exercise in galvanizing the will of American cheesemakers and giving them a winnable battle with their regulator, the wooden board fiasco was in fact useful for the farmhouse dairy industry in the long run. But as commentators celebrated the victory and indulged in rigorous Kremlinology to interpret the FDA’s pronouncements, the real and fundamental cultural issue driving the affair went unnoticed. In the original letter that communicated the ban to the New York State Agriculture Department, Monica Metz, the chief of the dairy and egg branch of the New York Office of Food Safety, noted that wood boards “absorb and retain bacteria, therefore bacteria generally colonize not only the surface but also the inside layers of wood.”45 This is undisputed, but it was the impossibility of killing all of these bacteria with chemical sanitizers that made her regard wood as an unhygienic surface for aging cheese. Metz’s vision of hygiene through sterility is typical of modern industrial food science. There were dark mutterings from some quarters about Metz’s previous employment with Leprino Foods, an industrial producer of mozzarella, but the worldview that she expressed—what anthropologist Heather Paxson has called a “Pasteurian perspective”—would have been shared by all of her fellow graduates in the master’s program in Food Science and Technology at the University of Nebraska-Lincoln.46 More than that, it was likely shared by almost all graduates of food science programs anywhere in the world. According to the Pasteurians, because pathogenic microbes make us ill, the best route to safety is to annihilate the entire microbial community. In the case of wooden boards, as with so many other examples of microbial communities in action, we have seen how this is demonstrably not the case. RISK

•

141

However, it is a fascinating statement in an incipient culture war, one that ultimately offers us the potential tools to heal the rifts between cheesemakers and their regulators. It opens the door to considering the problem of foodrisk assessment as a case of cultural cognition. Within democracies, the capacity to discuss complex risks in public is vital to the health of society. The defining issues of our age, from climate change to social inequality, all involve the interpretation of factual evidence. However, the political reality of these debates is that they suddenly descend into a culture war over empirical data. It is not the scientific complexity of the data that causes the problem: the sides of the battle over, for example, climate change do not segregate according to education. Rather, the cultural cognition approach, championed by Dan M. Kahan, a professor of both law and psychology at Yale Law School, is to explore “the influence of group values—ones relating to equality and authority, individualism and community—on risk perceptions and beliefs.”47 The work of Kahan and his colleagues demonstrates that “people find it disconcerting to believe that behavior that they find noble is nevertheless detrimental to society, and behavior that they find base is nevertheless beneficial to it.” The cultural values being advocated or attacked, rather than the essential aspects of the evidence itself, determine the taking of sides, progressive polarization, and the manner in which individuals critically evaluate “experts” as sources of unbiased opinion. One set of experiments used the case of human papillomavirus vaccination for schoolgirls (the vaccine is recommended by the CDC but mired in controversy and moral panic). Kahan constructed arguments for and against mandatory vaccination and assigned them to two fictional male experts whose appearance (one wore a suit and had grey hair and the other wore a denim shirt and had a beard) and list of publications were crafted to emphasize a particular cultural perspective. When opinions were voiced by an expert who appeared to share their cultural values, the experimental subjects tended to adopt those positions more easily than when they were presented by someone the subjects regarded as being from a different group. Presenting challenging information through the guise of a figure who appears to share the same cultural values is an effective way of changing opinions.48 Kahan makes the point that the aggressive cultural contest over scientific data is ultimately self-defeating. An individual can support his or her “team,” but unlike in a sporting contest, the aim of debating climate change is not to win a trophy but to make sure that the world is kept safe for future genera142

•

RISK

tions. The same principle is immediately applicable to food safety. The goal of absolutely everybody concerned with food production and regulation is to have safe food. With this insight in mind, we can turn to the world of rawmilk cheese regulation. To examine the dairy industry, we will adapt the cultural cognition model introduced above. In her anthropological study of American cheese, Heather Paxson highlights the binary choice between raw milk and pasteurized milk, “one that I as an anthropologist cannot help but see as a Pasteurian dualism between ‘nature’ (not us) and ‘culture’ (us).”49 We encounter the same distinction between nature and culture in the regulation of the dairy industry, although not as a binary choice but as a continuum. At one end of this continuum, we have Paxson’s Pasteurians, the modernists who are delighted to control nature. Some of them may even use raw milk, but their emphasis is on the power of technological progress. At the other end of the continuum, we have the dedicated traditionalists, who represent Paxson’s “nature” and are deeply skeptical of the modernist project. In conversation, they shift the topic to the loss of traditional practice: cheese is never as good today as it used to be in the past. While we will discuss in subsequent chapters the important tensions within cheesemaking between the celebration of individual and collective identities, the biggest debate in cheese regulation is not along the individualist-communitarian axis. But if we replace that axis with our new continuum of traditional to modernist, we have a cheese-adapted version of Kahan’s cultural cognition model.50 In the bottom left quadrant of figure 7, we have the (libertarian-traditional) people who constitute the base of the modern raw-milk movement. It is notable that Carlos Yescas, program director at the US lobbying organization for raw-milk cheese, has recently changed the name of the organization from the Cheese of Choice Coalition, which emphasized the libertarianism of its values, to the Oldways Cheese Coalition, which relocates its values firmly toward the traditionalist.51 In this quadrant, we also find the Raw Milk Freedom Riders, who transport raw milk across state lines as an act of civil disobedience.52 To the right, the libertarian-modernist position is exemplified by the work of the Kehler brothers at the Cellars at Jasper Hill Farm in Greensboro, Vermont. With its own on-site laboratory and a full-time microbiologist on staff, their entire organization is driven by the promise of technological developments. At the same time, with the hugely diverse range of cheeses that they produce and mature, the Kehlers self-consciously do not restrict RISK

•

143

Authoritarian

Appellation d’origine protégée (authoritarian traditional)

FDA (authoritarian modernist)

Libertarian

Oldways/Cheese of Choice Coalition (libertarian traditional)

Cellars at Jasper Hill (libertarian modernist)

Traditional

Modernist

FIGURE 7. The cultural cognition model applied to cheese.

themselves to a rules-based territorial framework of the sort that is found in Europe. This is in direct contrast to the top left quadrant of the figure, the authoritarian-traditionalists, where we find the logic of all rules-based, topdown classifications, such as those found in the Protected Designation of Origin system within the European Union. Finally, in the top right, we have the regulatory agencies responsible for public health. One example is the FDA, but it could just as easily be any one of similar authorities in Europe, the Americas, or Australasia. In the cheese world, these authoritarian-modernists are typified by people like Monica Metz, those who are comfortable with top-down rules and regulations and enthusiastic about the modernist project to control nature. Here is where we find Paxson’s most dedicated Pasteurians. Most importantly, this is not an exercise in putting people into cultural boxes just for the sake of it. The sociologist Joseph Gusfield describes the process of “status conflicts,” the symbolic political acts through which cultural groups legislate to glorify their values and demean those of others. While his principal historical case study was the temperance movement, we see exactly the same phenomenon in operation in the legislation governing our food.53 Just consider what would happen if actors associated with the wider food movement were suddenly appointed to key legislative and admin14 4

•

RISK

istrative roles. The established order would be swift ly overturned, and the regulatory agenda would ignore small-scale farmhouse producers and instead direct its fire at GMOs and the role of big business. Whether this would make the world a safer place is moot; it would legislate a different set of risks. Despite the imminent prospect of more bitter culture war, considering our typology in light of Kahan’s research offers a tantalizing glimpse of a strategy that might solve raw-milk cheese’s communication problem. At present, those in the libertarian-traditional base who are the most enthusiastic advocates for raw-milk products are concerned that the government wants to take away their right to enjoy traditional products. It is an argument with tremendous emotional resonance to those who are already converts, but it completely cedes the science to regulators who find the prospect of raw-milk products culturally and emotionally distasteful. Instead, we need to craft a message that has two components, each of which will resonate with both regulators and raw-milk advocates. The first is that anybody enthusiastic about the prospect of raw-milk cheese has to be enthusiastic about the necessity of regulation, not just in its existing sense, as a method to ensure safe products—although acknowledging this is a prerequisite of responsible production—but also as the only way to achieve fairness in the market by guaranteeing authenticity. Regulation in this sense refers not to demanding a protected name for each cheese but rather to recognizing that in a world dominated by an ideology of farm-to-fork authenticity, state regulators have a vital role to play to prevent widespread fraud. The groundbreaking investigative journalism of Laura Reiley at the Tampa Bay Times has demonstrated the prevalence of barefaced fraud on restaurant menus, and there is no reason to think that Florida is an outlier in this regard.54 Good, honest cheesemakers and their customers need regulators to keep the market fair. We even have the advantage that much of the research in France and Italy on identifying unique markers within the cheese for particular agricultural practices has been conceived of as a means of fraud prevention. Regulation is good, should be the message to the authorities: we need you to keep the market safe and fair. The second pillar of our communications strategy is that cheesemakers need to engage with science rather than ceding technical discussions of microbiology to white-coated regulators. The work of Dr. Marie-Christine Montel with Salers stands out in this respect, but perhaps the definitive case here is the example of wooden boards: to the instinctively Pasteurian modernist, these might seem archaic and dirty, but they actually represent the RISK

•

145

most sophisticated possible material on which to age a cheese. But the conversation between cheesemaker and regulator must not be, “Oh, we age on wooden boards because we always have and it’s traditional,” but rather, “We have executed a diligent survey of the scientific literature, and this choice of materials is best for us to make safe and delicious cheese. I’m sure that you have read the papers about it in the learned journals?” Our beacon for the future comes from France. Dr. Valérie Michel is a dairy microbiologist who works for ACTALIA, the self-proclaimed French “Center of Expertise for the Food Industry.” Michel has a residual affection for the Anglo-Saxon world. Although she trained as a microbiologist in France, she learned the nuts and bolts of the cheese industry during two and a half postdoctoral years in New Zealand. “I learned my technical words for cheese in English,” she says, “ ‘curd’ and ‘cutting’ rather than caillé; I prefer that.” We speak with her shortly after she addressed the American Cheese Society on the issues surrounding the use of wooden boards, and she is still smarting from her first encounter with a regulator from the US FDA. That said, her message is one of hope: Twenty years ago, we had this in France, but now progressively we have a bit more confidence and some exchanges and dialogues with our food safety authority. But we made a lot of efforts trying to have scientific data and showing them that we aren’t blind, that we don’t want to see and have no problems. If you present that face, they won’t believe you. I’ll take the example of Listeria—it’s an old one. In the 1980s, we had trouble with Listeria in certain cheeses, and the first thing was to deny it: we didn’t look at it, so we didn’t find it. But doing that, the risk is that you have an outbreak and it will be a disaster. One producer had an outbreak and had a problem with a pregnant woman, and the food safety authority came and imposed really tough rules because they weren’t confident in the way they were doing things. Part of the cheese industry said “We can’t have this attitude: we need to look, analyze, see what we find, and if we’ve got a problem, face it.” Try to find a solution, improve yourself, show that you are progressive, and they will help you. It contributes to establishing a feeling of trust between the authority and the cheese industry. This is the global overview of the situation. On the ground, you could have tricky people. But it’s the best way to do things.

14 6

•

RISK

EIGHT

Cultures

In his writing, eighteenth-century British cheese merchant Josiah Twamley does not go out of the way to create a likeable persona for himself.1 This is something that has cost him in the eyes of posterity.2 But behind his blunt self-satisfaction, he offers a tantalizing glimpse into the English cheese industry at the close of the eighteenth century. Above all, as a London cheese factor, Twamley vividly dramatizes the collaborations and tensions between producer and wholesaler, although as equal part Enlightenment improver and professional curmudgeon, there is a sense that the world would be a much better place if everyone would just do as he tells them. He is, perhaps, our greatest inspiration for this book. For all his bluster, Twamley was a shrewd observer who had the priceless advantage that he daily witnessed cheesemaking at different farms. Cheesemaking, more perhaps than any other profession, is vulnerable to tunnel vision: even now, cheesemakers very rarely see each other make cheese. To make cheese was and is a daily commitment, all the more so in an era without refrigeration. Twamley himself was aware of the advantage of his position. His skill, as he saw it, was to identify best practice. And so, as we visit producers ourselves and peer into their make rooms, it is tempting to ask: WWJD? That is, what would Josiah do? Fond of withering put-downs and untroubled by self-doubt, Twamley would rather enjoy the twenty-first century: his is a voice built for social media. And if we were to take him into a modern creamery, he would recognize almost everything. While the exact process of cheesemaking has experienced some minor changes in the past two hundred years, its broad contours have remained the same. Initially, it would be dairy farming that he would find the most alien. He would be intrigued by and most probably 147

incredulous at the average size of twenty-first-century herds and the volume of milk given by each animal. In contrast, modern conveniences like clean running water, electric light, temperature control, and stainless steel would impress him because he would be able to readily imagine their utility. Heat, cooling, and cleanliness were central to Twamley’s vision of good cheesemaking; to be able to achieve them with the flick of a switch or through an ingenious new material would seem the natural march of progress.3 However, one stage of modern cheesemaking would appear completely alien to a visitor from the eighteenth century: modern cheesemakers add mysterious containers of liquid—or even more mysterious sachets of powder—to their milk. These are starter cultures, the purified strains of lactic acid bacteria that cheesemakers use to ferment their milk. In modern cheesemaking, their use is ubiquitous at every level of cheese production, from giant factories to the most boutique farmhouse operation. They are the single tool without which the modern industry would not exist: without starter cultures, pasteurized milk cheeses would simply be impossible to make. And yet, while they are a topic of constant debate and discussion within the industry itself, they are almost completely outside the frame of reference of the consumer. They have no place in the public conversation about cheese. Commentators have not helped. When Professor Paul S. Kindstedt published his history of cheese and cheesemaking for a popular audience in 2012, he chose the title Cheese and Culture.4 At first sight, this is an admirable pun for a study with the ambition to address the place of a fermented food in Western civilization. But starter cultures, their discovery, their use, and their evolution are completely absent from the book. It is a glaring omission, akin to writing a history of the Titanic without mentioning any icebergs. It is all the more surprising given the author. Kindstedt actually wrote the book on starter cultures—at least, he contributed a chapter on starters titled “The Heart of Cheesemaking” to American Farmstead Cheese, the industry textbook that he edited in 2005, in which he clearly outlines his opinion of their importance: “The identity, quality, and safety of almost all cheeses are profoundly shaped by the starter culture.”5 The adoption of starter cultures over the course of the past 125 years is the cheese world’s secret history. The dairymaids of the late eighteenth century were unaware of the existence of microbes. The methods they inherited from their mothers and grandmothers were the result of centuries of trial and error, during which practices 14 8

•

C U LT U R E S

that worked were retained and refined. Now, in retrospect, we can see how those methods unwittingly increased the population of the right bacteria in the milk and then built a house for those bacteria to make their own. As we saw in chapter 6, fresh raw milk has an extremely low population of lactic acid bacteria. Even in the early days of cheesemaking, when milk was produced without the emphasis on low total bacterial counts that dominates the mentality of mainstream dairying today, the bacteria in milk needed an extra boost to ensure successful cheese. High in the mountains of the Auvergne, the wooden gerles inoculated fresh warm milk with high levels of useful bacteria within seconds. Further east, in the Alps, mountain cheesemakers used whey as the medium for extracting rennet from calves’ stomachs each day. Carefully incubated at warm temperatures, the freshly prepared rennet extract was teeming with lactic acid bacteria that started the fermentation. In England, cheesemakers took a different tack. While in a few areas, cheese was made twice a day with milk fresh from the cow, as in France, for the most part, evening milk was collected and stored in the dairy. Without refrigeration, the cool ambient temperature was perfect for the lactic acid bacteria to multiply overnight, and with the addition of warm milk and extra heat the next morning, acidification was on its way. In the world before starters, with few cleaning chemicals and even less in the way of temperature control, the problem experienced by most cheesemakers was not that their milk wouldn’t sour but rather that it soured too fast. Dairy instructors advised their students that the acidity of the preripened milk could be judged quite accurately just by measuring its temperature the following morning.6 Warmer milk meant more activity and no time for delay. Far from today’s concerns about slow makes and stuck fermentations, for farmhouse cheesemakers in a prestarter world, getting milk to sour was not a problem.

A DA N I S H S TO RY

The story of commercial starter cultures begins in late nineteenth-century Denmark, not with cheese but with butter. Their introduction was heavily contested at every stage and thoroughly dependent on unrelated technological developments and market pressures. It is a Danish national story too, one of parochial concerns fought out in countless conversations between farmers, technicians, and scientists. And amid all this debate, it is the unlikely block C U LT U R E S

•

149

of Lurpak butter, the Danish Dairy Board’s collective brand, that can claim to be the first modern cheese.7 We easily forget in today’s world of sweet cream that butter was once exclusively a fermented dairy product. Butter production at the farm started with a process of spontaneous souring. Milk was left undisturbed until the cream rose to the surface, and this cream was skimmed and reserved for several days until there was enough to make churning it worthwhile. By this time, the same natural lactic acid bacteria so important in cheesemaking had naturally soured the cream. With a much higher level of moisture than cheese, the naturally risen cream required the utmost cleanliness and care to ensure the quality of the resulting butter. In cheese, a few “impure odors” might be acceptable, but in butter, there was nowhere to hide.8 Like cheesemakers, farmhouse butter makers developed methods for encouraging proper fermentations in their cream, such as adding buttermilk or soured cream to fresh cream to get it started on the right path. Even so, butter was a variable product. Late nineteenth-century Denmark was a nation at a competitive disadvantage. Having been defeated by the Austro-Prussian coalition in the Second Schleswig War in 1864, Denmark lacked the coal, iron ore, or mineral deposits that would normally kick-start economic growth. The country’s surface area was just under half that of Ireland, and its key resource was its agriculture.9 To compete and survive, the Danes had to be smart. Denmark might lack mineral wealth, but it was blessed with an abundant coastline and ready access across the North Sea to the United Kingdom, which was an economic powerhouse and also an open economy, ready and willing to accept imports. In particular, demand for butter was skyrocketing in the United Kingdom. Up until the mid-nineteenth century, the Danish dairy industry had been dominated by butter makers using time-honored methods, but this changed rapidly in the face of demand from its neighbor to the west. Dairy farmers organized themselves into cooperatives and ramped up production, taking butter making off the farm and into purposebuilt factories starting in the early 1880s. Over the period between 1850 and 1914, Danish agricultural exports more than quadrupled.10 Factory production of butter on a large scale was first made possible by the invention of the centrifugal separator in 1878.11 On the farm, cream had been skimmed by hand, but with the help of the separator, milk could be collected daily from farms and separated efficiently at the factory. Skimmed milk had many uses back at the farm, such as feeding the calves and the family, so the mixed skimmed milk was pumped back into the milk cans and returned to the 150

•

C U LT U R E S

farms. There was only one problem with this brilliant set up: the skimmed milk was the perfect vector for the spread of disease, particularly tuberculosis.12 Drawing on the full resources of the Danish veterinary and agricultural research institutes, help for the dairy farmers was on the horizon. Around 1885, N. J. Fjord, head of the new Danish Agricultural Research Laboratory, developed a machine that ran on steam and could be used to pasteurize milk.13 The cooperative creameries scrambled to put this new technology into place to protect the health of their members and their herds, and the proportion of creameries using pasteurizers increased from 11 percent in 1892 to 97.5 percent in 1897.14 Beyond the threat of giving all their members tuberculosis, the cooperatives were also struggling to address a second problem: inconsistency. On the farm, butter production had depended on the conscientiousness of the milker and skill of the butter maker. The quality of the blended cream at the factory, however, was only as good as the cream of the least-careful member of the cooperative, and a satisfactory fermentation was hardly a given, even with rigorous policing of the farmers and the addition of soured cream or buttermilk to encourage the process along. The problem was so great that in 1878, the Scandinavian Preserved Butter Company offered a prize of 1000 Danish kroner for a thesis offering “a thorough investigation of the conditions which influence the souring of cream, for it must be recognised that the way in which this is usually carried out in the dairies makes it extremely difficult to deliver a uniform product.”15 (1000 kr. was a significant sum in 1878: it is equivalent in economic status value to around $93,600 in 2015.16 This was a problem that the industry was desperate to solve.) Several theses were submitted, but none of them were considered adequate, and the prize went unclaimed. Eventually, technology offered a solution. One of Fjord’s colleagues at the Agricultural Research Laboratory, C. O. Jensen, demonstrated in 1891 that pasteurization could be used to eliminate the spoilage bacteria in cream. His experiments on cream pasteurization dovetailed conveniently with work being done at the time by another member of the same institute, a chemical engineer named Vilhelm Storch. For several years, Storch had been researching the sources of off-flavors in butter, and his work had culminated in the isolation of several strains of lactic acid bacteria. Together, Jensen and Storch set up an experiment in which they demonstrated that Storch’s lactic acid bacteria were capable of souring Jensen’s pasteurized cream.17 With the equipment to pasteurize skimmed milk already in place, it was not a tremendous leap for the cooperative dairies to start pasteurizing their C U LT U R E S

•

151

cream as well. However, stripped of its native lactic acid bacteria, the pasteurized cream would not sour on its own, so something had to be added to start it fermenting. Many creameries initially chose sour cream or buttermilk for the job, as even Storch himself admitted that the butter produced using the isolated strains “did not give such a fine-flavoured butter as that produced after ripening with buttermilk from a dairy . . . supplying particularly good butter.”18 But even the naturally soured buttermilk and soured cream could be inconsistent in their fermentations. Since the batches were large, failure was expensive, and the isolated strains offered the most dependability and consistency, even if there was a trade-off in flavor. Pasteurizing the cream and adding starters offered other advantages. The reduction in waste and increased potential for scale meant that the butter could be sold at a competitive price, and the greater microbiological purity meant the butter stayed fresh for longer as well. A low-priced, mild-flavored, consistent butter with a long shelf life was perfect for distance selling, and the growth of the Danish cooperatives was largely driven by voracious demand from the United Kingdom, whose urbanizing northwest and mining districts were in the middle of massive industrial expansion. Denmark’s mild, cheap butter went down just fine. As we will see in chapter 9, which discusses the rise of American factory cheesemaking, the bland taste preferences and parsimonious approach to value of these British consumers has defined the modern dairy industry.19 By 1891, just one year after Storch’s research was published, four different Danish companies had already brought commercial butter starters to market. A report from the Danish Research Laboratory written in 1895 predicted that “the time is not far distant when [commercial starters] will practically supersede the older means of souring.”20 In the course of less than twenty years, the Danish butter industry had been transformed. It was a transformation that reverberated all around the world. In 1894, as far away as Queensland, Australia, the Brisbane Courier lauded the “scientific methods in butter-making” of the “modern system in Denmark.”21 But North American butter makers were not ready for a scientific revolution. In a neat reversal of every European food snob’s ideas about Americans and the industrialization of taste, the preference in the United States was for more flavorful butter. At that time, most American and Canadian farmers were still skimming their milk by hand at the farm and bulking the cream over the course of several days before selling it to a local factory, just as the

1 52

•

C U LT U R E S

Danish farmers had been doing twenty years before. The result was a strong and variable butter, with a short shelf life to match.22 Bland pasteurized-cream butter made with starters was simply not to the taste of the American consumers, who “demand[ed] a higher, stronger flavor than that produced by the Danish process.”23 Unlike in Denmark or the United Kingdom, sniffed one American commentator in 1898, “in this country this delicate but extremely evanescent quality as to flavor controls the market to a much greater extent than is true in other dairy countries.”24 Ambitious North American businessmen were busy developing transatlantic trade routes for butter and cheese, but the butter consignments were often thoroughly spoiled by the time they arrived in Europe, and a lot of off-flavored American butter ended up being used as axle grease in Liverpool.25 The Danish butter industry had no need to worry about losing their precious British customers to the Americans. Although it undermined their export ambitions, Americans were not interested in butter that was bland and tasteless, no matter how “scientific” its production. Raw cream was set to remain the standard for American butter, but sometimes it could be excessively characterful, even for the Americans. In an attempt to decrease this variation, factory managers gradually began to adopt cultures, both homemade and commercial. By 1904, almost ten years after the introduction of commercial cultures to the American market, a growing minority of butter producers were using them, although almost exclusively as a steadying influence on raw cream rather than as the sole acidifier in pasteurized cream, as recommended on the label.26

A C U LT U R A L S H I F T

The original Danish Heaters, as Fjord’s pasteurization machines were called, pasteurized milk and cream at 180°F (82°C) rather than 161°F (72°C), which is used today. While fine for making butter, this high temperature leached the minerals out of the milk proteins, resulting in a sloppy, poorly formed curd that was useless for making cheese. While pasteurized butter made with starters was widespread in Denmark by the turn of the twentieth century, no pasteurized cheese was being made anywhere at all. However, just as starters were being co-opted for raw butter making in America, it was also possible to use them to make raw-milk cheese, and it was

C U LT U R E S

•

153

not long before American factory cheesemakers (who were facing similar challenges with pooled milk of variable quality) were exploring the use of commercial starters in raw-milk cheese. The results were mixed. The transcript of a 1907 meeting of the Wisconsin Cheesemakers Association conveys the heated spirit of the debate on the utility of starters. A representative from the Dairy School in Madison, Gottlieb Marty, had been invited to present the results of his research on using starters for a Wisconsin specialty, Brick cheese. His attempts to convince the audience of the superiority of starters met with resistance from cheesemakers working with older methods, such as whey-based rennet extracts, for encouraging acidification: mr. marty. Every up-to-date cheese maker knows that there is no other one factor which influences the quality of the product to a greater extent than the use of a good commercial starter. . . . Every cheese maker, no matter whether he makes American or Brick cheese, will turn out a more uniform product when a good starter is used. mr. parkin. I would like to ask . . . if in using starter in this brick cheese it does not develop a somewhat cheddar cheese flavor? mr. luchsinger. The cheesemakers of Dodge county, Wisconsin, have the reputation of making the best brick cheese, and I would like . . . to find out how in practice they find it has compared with the methods described by Mr. Marty. mr. westphal. Gentlemen, we have not been using any starter so far but are beginning to use home made rennet.27 Of course by using a [commercial] starter in brick cheese you will lose a little of the original flavor and get some of the cheddar flavor. In our section they are using the home made rennet more and more of late years. mr. marty. [A cheesemaker] must . . . understand the use of an acidity test and the preparation of starters for if he does not understand these fundamental points, he will get himself into all kinds of trouble when he commences to try to use a starter. mr. westphal. Our makers down there have not the right hang of [using commercial starter]. . . . We have not studied it and that is where we are falling down on it. In the end, Marty backed down and offered a more equivocal endorsement:

154

•

C U LT U R E S

mr. marty. Of course brick cheese is clean, sweet curd cheese and if it has too much cheddar flavor I would prefer a cheese that has not so much of that flavor; but if you have a tainted milk I would rather have that cheddar flavor than a blowy or overworked cheese.28 It was to be a recurring theme, just as it had been with the butter: starters turned out cheeses that tasted different. These cheeses weren’t as good as the best naturally soured cheeses, and the starters brought unfamiliar flavors with them, but at least they lifted the worst to a salable standard. The Wisconsin cheesemakers also raised a critical point: getting the hang of using commercial cultures was no easy task. Early liquid cultures could not travel far from the lab where they were produced and lost their strength after a day or two, while powdered cultures were weakened considerably by the drying process. Both needed to be cultured several times in sterile media before they were ready to use.29 Techniques that seemed straightforward to the progressive dairy instructor from Madison took cheesemakers in rural Dodge County well beyond the edge of their comfort zone. Once again, the key attraction of starter cultures was not that they made milk sour—it did that on its own—but that they reined in wayward milk and made life easier and more predictable for the cheesemaker. There is no doubt that much of the cheese made before the advent of starters was faulty, particularly in factories or on farms where milk production was anything less than obsessively fastidious. Second-rate cheeses that stank of the barnyard or were rife with gassy holes were an expensive waste of valuable resources. Starters were a convenient patch. Soon after the turn of the century, commercial starters adapted for cheese were beginning to appear on the market, and dairy schools played an important role in distributing and popularizing them. The schools held collections of starter cultures that had been gathered from successful dairies, each a lively community of up to a hundred different strains that naturally worked together. Nobody quite knew exactly what was in these “undefined” mixedstrain cultures, but they were robust and fairly flavorful. Each day, a bit of leftover starter could be inoculated into sterilized milk to create a fresh batch of culture for the following morning, making them economical as well. The most skilled cheesemakers did not embrace starters with open arms. In England, a Cheddar instructor remarked in 1917, “It was the misuse of starter when it was first introduced that brought it in bad repute with first class cheesemakers. The starter in the hands of careless people, who expected

C U LT U R E S

•

1 55

it to cover any dirt, was used in such large quantities that hard dry cheese was the result.”30 Even in the 1930s, a Scottish guide to cheesemaking was also cautionary: “If it has not the essential quality of purity, starter will always be a source of danger in cheese-making. A coarse, ‘sharp’ starter is seldom pure, and is a common cause of sour cheese.”31 As more production moved from farm to factory, where the milk being used for cheese contained the products of many farms, a conservative approach became more and more justified. As batch sizes increased, the financial penalty for making faulty cheeses was greater than the reward for turning out superlative ones. With the refinement of pasteurization technology, by the 1920s, it became possible to pasteurize milk at a lower temperature that did not destroy the structure of the curd.32 Finally, milk could be cleaned up and starters added to do their work alone, just as they had been designed to do for butter decades earlier. Cheese-starter technology has continued to evolve rapidly during the past hundred years, making cultures more convenient and more specialized. The mixed strain, undefined starters popularized up through the 1920s were well suited to making cheese at the farm level, but their complexity proved to be their Achilles heel. A by-product of some strains of lactic acid bacteria (including those found in many undefined starters) is carbon dioxide gas produced during the fermentation process. In a farmhouse cheese, this is unlikely to cause much of a problem; it may create some fissures in the paste of the cheese, but this is made up for by the flavor, and the gas produced escapes through the natural rind. When cheeses are waxed or sealed in impermeable materials, however, gas production is a bigger issue. By the 1930s, factory cheesemakers in New Zealand were tired of dealing with gas production in their Cheddar, so they revisited Storch’s single-strain approach. They isolated one strain that only produced acid and made cheese with that. It worked perfectly, but there was a catch: the strain conked out without warning after just a couple of batches. The New Zealanders soon realized that these much simpler “defined” starters were susceptible to attacks by viruses called bacteriophages, or phages. Not all bacteria are vulnerable to the same phages, but when there is a match, a phage can debilitate a starter, leaving in its wake a “dead vat” of curds that refuse to acidify. In a diverse natural fermentation or when using a mixed-strain starter, even if phage knocks out a few strains, the others carry on with the fermentation. The fermentation may slow slightly, but the cheesemaker may not even 156

•

C U LT U R E S

notice a difference. Defined cultures demand a different approach. The New Zealand team solved the problem by identifying which strains were sensitive to different phages, making separate starters from each of them, and then adopting a strict rotation system. This way, if a phage got into a susceptible vat on one day and started to build up in the environment, it would lose its host when an unrelated starter was used the next day. Any residual phage would be cleansed from the dairy before the susceptible starter was used again. Simple enough. But phages are tiny and omnipresent, and trying to keep them out of dairies required a new level of obsessive sanitation on the part of the cheesemakers. Keeping defined strain starters vital for more than a few days demands hospital-like conditions: pasteurized milk (raw milk is a notorious source of phages), positive-pressure airflow systems, aseptic technique for starter production, and closed cheesemaking vats for extra security. In the cheesemakers’ quest for complete control over the bacteria that went into their cheeses, defined cultures created a new beast to contend with. These are the unintended consequences of monoculture on a microbial scale, and they are as much a concern as problems arising from monocultures of any cereal crop, fruit, or vegetable.

T H E C U LT U R E H O U S E

With their headquarters in Hørsholm, Denmark, just outside Copenhagen, Chr. Hansen is one of the world’s largest culture producers. It is by no means solely, or even mainly, a producer of microbial cultures for the dairy industry. Rather, Chr. Hansen styles itself as a bioscience company, manufacturing enzymes, probiotics, and natural colors used in an astonishing array of foods. Though not a household name, its products are everywhere: no matter where you are on the planet, if you pick up a cheese at random, there is a 50 percent chance that Chr. Hansen has produced at least one of its ingredients. The company’s founder, Christian Hansen, was a student of chemistry at the University of Copenhagen during the 1870s, where his research on digestive enzymes led him to launch a business making purified rennet for cheesemakers in 1874. At the time, preparing homemade rennet from the dried stomachs of veal calves was still the darkest of dairy arts and a major source of frustration. The concentration of the homemade rennet varied considerably from batch to batch, and poorly cured or maggot-infested dried stomachs were sometimes all that were available. Pure and reliable, Hansen’s rennet C U LT U R E S

•

1 57

extract was a wild overnight success, and in the space of a generation, cheesemakers almost universally embraced its dependable consistency. Tellingly, unlike the contested and controversial reception of starter cultures two decades later, commercial rennet extract was greeted with a near-universal sigh of relief across the industry. Appropriately enough, it was through rennet that we first encountered a representative of the company. Trish Dawson is a senior scientist at Chr. Hansen’s Milwaukee-based US operation, and her role is as mediator between white-coated researchers and practical cheesemakers on the ground. At the 2015 American Cheese Society conference in Providence, Rhode Island, she led a session on coagulants for cheesemakers, and we watched as she fielded the earnest questions of the audience. Her accent still flecked with her Australian origins, Dawson has near-infinite patience in the face of elementary queries. For many of the smaller-scale cheesemakers in the room, this was their annual opportunity to engage with a technical expert, and Dawson deftly addressed their concerns about their cheese. Equal parts saleswoman, sounding board, and emergency service, Dawson is the point of contact for cheesemakers and the recipient of a thousand requests. We were, quite frankly, impressed. And so, a month later, we arrive at the Chr. Hansen global corporate headquarters in Denmark. It is a comfortable campus, modern and just a little bland: the future as a triumph of business casual. While awaiting our hosts in an atrium filled with natural light, we browse through current issues of Nature and Science. Before our meeting, we dine in the company canteen on a diverse spread of vegetable salads, smoked fish, and dark rye bread, surrounded by disturbingly attractive scientists and technologists conversing in multiple languages at communal tables. It is a distinctively Scandinavian vision: efficient, progressive, responsive, and egalitarian. But at the back of our minds, we can’t help wondering whether behind the alluring facade, this might be the equivalent of the Bond villain’s lair, where nefarious schemes for world culture domination are hatched and carried out. Our hosts are two scientists from the Cultures and Enzymes Division, Dr. Eric Johansen, the company’s associate vice president for innovation, and Dr. Jannik Vindeløv, the senior scientist responsible for the development of the cultures program. Together, these two soft-spoken men select the microbial winners that drive the production of a large proportion of the world’s cheeses. When it comes to starter cultures, commercial selective pressure is the factor that drives evolution. “It’s important to realize that we serve mostly the big dairies,” Vindeløv reminds us. “They are maybe not making the most 158

•

C U LT U R E S

exciting products, but they need the acidification to work every time, day in and day out. And we’ve been very good at providing robust acidification without off-flavors.” The evolution of the cultures is a direct reflection of changing commercial demands. “The cheesemakers’ wishes have changed over time,” he continues. “In the old days, they had lots of time, but as time goes on, they’re more pressed, and faster acidification is becoming more and more in demand.” As with everything, there is a trade-off: the greater the acidification potential of the culture, the less the strains that produce more rounded, buttery flavors contribute. Trish Dawson had voiced similar sentiments when we met with her in Providence: “The majority of the people we work with are the very large cheesemakers: they are looking for high-throughput, consistency, robustness, yield, and low costs. A huge part of their drive is to maximize yield and have a consistent product day-in and day out.” These customers’ demands differ substantially from those of the farmhouse cheesemaker, Dawson told us: “ ‘I want to maximize yield, I want my whey-stream to be clean, my casein proteins to be in my cheese, my whey proteins in my whey, nothing that negatively affects the value of my whey.’ We have to look at the whey industry now more than we’ve ever had to before.” The very elements that make starters suitable for factory production— such as turbocharged acidification—present technical challenges for cheesemakers working on a different scale. We think immediately of one small producer we spoke with who struggled with a persistent problem: his cheeses had a gritty, powdery texture, a classic symptom of acidification racing ahead of drainage. A review of his method revealed that he was using starter cultures of the so-called RA series. He was shocked to hear that RA stands for “rapid acidification”; one of the sources of his texture problem was itself a culture designer’s triumph. Package sizes are an even more mundane—but no less critical—challenge. Cheesemakers working with just a few hundred liters of milk might need only a couple of grams of freeze-dried culture for each batch. But the market for piddling five-gram sachets is so tiny as to be little more than an annoyance for the big culture houses. Vindeløv tells us, “We actually try to work in the other direction, to produce bigger bags. Our most popular size is our one kilo bag.” But subdividing larger packets—precisely breaking down their contents so that the fine powders and coarser flakes are distributed evenly, without contaminating them in the process—is a frustrating and messy challenge for cheesemakers. Many just skip this step and work from large open packets C U LT U R E S

•

1 59

for weeks at a time, sprinkling small amounts of starter directly into their milk. In doing so, they unwittingly add yet another layer of uncontrolled variation each day. If there isn’t an economic case for the major culture houses to offer their starter cultures in smaller packets, there is even less incentive for them to develop new cultures honed to meet the more specialized requirements of small-scale cheesemakers, such as those that lend themselves to lower doses and slower acidification speeds. The economics are simple, Johansen tells us. Research and development costs aside, he explains, “when we produce a fermentation tank of starter cultures, we make several tons of cells. They have a shelf life, and it’s not worth it to make tons if someone’s only going to buy a kilo or two.” Even for their largest customers, their approach is necessarily modular, drawing on blends formulated from universal single-strain building blocks. “Ultimately, we want to develop products that are useful globally,” he says. “Part of our business model is economy of scale.” Of course, there is far more to the Chr. Hansen culture-development agenda than identifying fast acidifiers and turning them into giant batches of cultured, freeze-dried cells. Vindeløv describes his role as that of “culture architect.” “The culture program has evolved over time, and certain key themes have made [our cultures] successful,” he tells us. “We are trying to find past patterns and mold them into new ways of solving problems.” He and his team spend much of their time studying how the different bacteria in their cultures interact and work together. By looking at each strain’s genetic makeup and behavior in isolation, they identify patterns and features that have the potential to be complementary and then test how those theoretical synergies play out in practice. These are intriguing scientific questions, but for obvious reasons, their approach treats cheesemaking as a rigid process to which microbes are subjected rather than as a dynamic conversation between the physical manipulation of the curds and the biology going on within them: “As a culture developer, it’s more important that I know about combinatorial science and physics and predictive mathematical models than that I have my hands in the cheese.” Even so, a more sophisticated understanding of how microbes interact with each other in the context of cheesemaking is a tantalizing prospect. Could the research that Vindeløv is doing be laying the groundwork for the future development of complementary cultures designed to work alongside native raw-milk microbes? He is intrigued by the idea, then pragmatic: “I think it could be a possibility. It could be done, but I don’t know whether it 160

•

C U LT U R E S

would be done. We would rather go for a pasteurized version and then add in what would make it simulate raw milk.” Johansen and Vindeløv are intrigued and a bit bemused by the farmhouse cheeses we have brought them as gifts, and they ask questions, turning the cheeses over carefully to examine them as if they are artifacts from an ancient civilization. Despite our very different views on cheese, as we speak with the two men, we are impressed by how much we like them. They are open and curious. Here are talented scientists using imaginative approaches to answer fundamental questions about how microbes behave. While our ultimate ends are different, their goal, just like ours, is to understand how better to harness microbes for practical purposes, and it is inspiring to hear about the creative ways in which they are generating new knowledge. When we take our leave and head back to Copenhagen, it is the market that we are disappointed in. The talent pool within a company like Chr. Hansen is extraordinarily rich, and just a little of their energy and resources devoted to the questions of the farmhouse cheese producer could bring transformative results. We are still pondering this point the next day, when we visit a research lab in a shipping container—dubbed the Science Bunker—at the restaurant noma on the Copenhagen quayside. The atmosphere could not be more different from that of Chr. Hansen: where Hansen’s headquarters is quiet and contemplative, the laboratory at noma is infused by the manic improvisational energy of a restaurant kitchen. Our friend Dr. Arielle Johnson is in charge of the research, and she leads us through the various climate-controlled rooms full of experiments and production fermentations for the restaurant. From her office overlooking the harbor, Johnson runs a pirate crew. Chef René Redzepi’s team has been experimenting with variations on garum, the ancient Roman fermented fish sauce, and the internal guide to fermentation that Johnson and Lars Williams, noma’s research and development chef, have authored contains the injunction that you should “man the fuck up” if you cannot abide the smell of fermenting fish and seafood.33 While Johnson lacks the expensive laboratory equipment she enjoyed during her PhD on flavor chemistry at the University of California at Davis, she has near-infinite human resources. At a globally famous restaurant like noma, there is a stagiaire for all occasions. Above all, the entire operation is driven by a relentless sense of the necessity of innovation. At noma, it is not a question of trying to make anything marginally more efficient or consistently reliable. Customers pay to experience flavor and novelty, and that is the selective environment in C U LT U R E S

•

161

which the research program moves. Both Chr. Hansen and noma sit in entirely rational positions within their own ecologies. Ultimately, and unfortunately for us, the market for all the starters used by artisan cheesemakers combined is of negligible consequence to the financial success of a company like Chr. Hansen. A large factory produces as much cheese in a few hours as a respectably sized farmhouse cheesemaker might make in a year. Inevitably, the requirements of the financially significant customers shape the research and development process, and farmhouse cheesemakers are left to use a set of tools crafted for somebody else.

OUR DINNER WITH ANDRE

There has been much toing and froing over email to set things in motion, but we have finally arranged the chance to sit down for an informal meeting with “Andre” (a pseudonym). Responsible for cheese in one of their country’s major supermarkets, Andre is a well-known figure within the international dairy community: they are a cheerleader for the category and an enthusiast to whom major players within the industry listen attentively. Among those producers they most support, planning for Andre’s eventual retirement is a necessary strategic precaution. The portfolio of cheeses with which Andre works ranges from farmhouse to factory. Across their supermarket, they shift enough cheese that they can make or break a supplier, although their reputation is one of earnest enthusiasm rather than corporate aggression. Appropriately enough, this will be a meal of cheese. The relative status of Bronwen and Andre is complicated—volume versus prestige?—and our initial greetings have a sense of wariness about them as each party appraises the other and tries to get a feel for the situation. From a professional perspective, Andre is consistently on the same side as Bronwen, except when they are not, and on both accounts, there is a desire to be taken seriously. We exchange some awkward small talk, but things begin to warm once the topic switches to industry gossip. Having taken to heart our suggestion that we wanted to talk about the state of the cheese market, Andre has brought a huge bundle of market data and trend reports. As we tuck into the cheese, these are rapidly forgotten, but they inadvertently structure our conversation, and the encounter plays out as if written by a playwright more concerned with a drama of ideas than of character development. Both sides have a role: we are the idealistic dreamers, advocating for the experimental 1 62

•

C U LT U R E S

pushing of boundaries, while Andre is down to earth, insisting on ideas grounded in practical day-to-day realities. It is from these perspectives that we taste each cheese. Rapidly, it becomes clear where the differences in outlook and market position play themselves out most clearly: while we talk about farming and site, Andre argues for a different route to uniqueness. We agree on the importance of a unique and exclusive product as a commercial necessity, but for a buyer of Andre’s scale, it becomes a question of commissioning dedicated recipes: “In the past, [cheese would] be specially matured for us—[but] everybody can do that, it’s [like the] emperor’s new clothes. We’ve moved toward dedicated recipes, which we have a big input into.” With the limited number of ingredients in cheese (milk, cultures, coagulant, and salt), a new recipe will almost always be a result of changing starter cultures. Andre’s role is as a mediator between consumers who do not know exactly what they want (invoking Steve Jobs, Andre points out that a consumer focus group would never have come up with the iPhone) and producers who simply want to sell more cheese. In this environment, the embrace of adjunct cultures selected to produce sweeter styles of cheese has become a commercial powerhouse: You have to take feedback in the round. With our [flagship] Cheddar, we sent that to sensory analysis. It came out third on the day. It was a flagship cheese, but the feedback was, “It wasn’t sweet enough.” At that time, it was very savory and a very successful brand. Sales were going up—it wasn’t broken. The feedback was that they preferred somebody else’s, and we tasted theirs, and it was newer on the market, and . . . they’d put more [Lb.] helveticus in. I know artisan cheesemakers hate it, but in mainstream it works very well. But often it’s done to extremes . . . [in some cheeses] which are just pure sugar off the scale. So we refined our recipe slightly, took it back to the panel: top. We have more savory backbone through our cheese than a lot of the competitors. For me, that’s very important, because I hate these sweet cheeses; I don’t like it overly sweet. Comté and Gruyère have been going up and up, because of the sweetness, nuttiness, complexity.

This is the essence of Andre’s argument. Within the supermarket environment, which requires selling a large volume of cheese to consumers with minimal engagement, making cheeses that achieve their character entirely from their starter cultures is not so much a betrayal of artisan ideals as it is the only route to commercial survival. The sheer scale of the operations involved is a different world from the farmhouse producers with whom we C U LT U R E S

•

163

spend our time. One small, in the global scheme of things, experiment that Andre instigated with the addition of adjunct cultures to an established cheese style resulted in a cheese that now sells comfortably more by volume than the entire British farmhouse Cheddar industry. In this market, it is after their conversation with Andre that cheesemakers—or more likely the technical directors of cheese plants—will be banging on the door of people like Trish Dawson and her colleagues at Chr. Hansen. The buyer chases the market; the producer follows the buyer; the culture house just lets both achieve what they want. Our dinner with Andre has been amiable and convivial, but the conversation has never quite become a genuine dialogue. We each deliver soliloquies and politely affect to listen to the other’s perspective without paying it too much attention. At an emotional level, we suspect Andre feels judged for not embracing the avant-garde, just as we feel marginalized by the mainstream discourse. As a final roll of the dice, we approach the last pair of cheeses, a piece of Bleu de Termignon from Savoie in the French Alps, not too far from the home of Reblochon, and a piece of Cambozola, from Bavaria in the south of Germany. In theory, the cheeses represent both sides of the debate at their apotheosis. The Bleu de Termignon is a raw-milk blue cheese made on a tiny scale in a remote part of the French Alps. It is made entirely without inoculation of starters or ripening agents: all of the microbes that sour the milk and the fungi that cause the bluing in the finished cheese are indigenous to the milk and its environment. From a purely ideological perspective, this should be our perfect cheese, one that lets the native microbes free. In our heart of hearts, this is what we would want all cheese to be. The Cambozola is quite the opposite. The definitive blue Brie, it is a cheese made on an immense scale from the pasteurized and bulked milk of, we have to assume, hundreds of producers. Our request to visit the production facility in Germany was repeatedly turned down: Käserei Champignon, the producer, is not friendly toward requests from the press. The cheese is entirely the product of its inoculations. The thick, snowy-white rind is a consequence of sprayed-on Penicillium candidum, and the streaks of blue mold that traverse the paste are a product of inoculation with Penicillium roqueforti. Pitting Bleu de Termignon against Cambozola represents nature versus culture in the most visceral way possible. We taste. The Cambozola has a paste that is bland and rubbery, even though it appears fully broken down. Its thick white rind does not feel inte164

•

C U LT U R E S

gral to the cheese as a whole: it has a papery texture and peels off easily. The blue has not developed beyond the channels that mark where the cheese has been stabbed with the piercing needles. Amid the flaccid paste, the nodules of grainy blue mold are a bizarre punctuation. In comparison, the Bleu de Termignon is wild and feral and feels savagely out of control. It is not a cheese that is trying to be pleasant, preferring aggressive volume of flavor, but the ammoniated aromatics give it too much of the character of a latrine. After an evening of advocating for our worldview, this is not the flagship that we want. Andre remarks on the commercial success of the Cambozola. We are intensely disappointed: neither cheese is appealing. With the injunction that we should get in touch if we come across any interesting cheese recipes, Andre departs, and we are left staring disconsolately at the Bleu de Termignon. Is this the inevitable consequence of working with indigenous microbes?

W H I T E L I S T I N G I N AU S T R A L I A

With its untamed flavors and flirtation with barnyard aromas, Bleu de Termignon is a cheese precariously balanced between commercial success— Parisian affineurs grumble to us that it is the most expensive cheese that they buy—and the steady decline of the number of producers prepared to embrace the toil of twice-daily cheesemaking. Only three producers of the cheese remain, and each sign of ill health or threat of retirement is cause for concern. However, the same wildness at the heart of its appeal is also what makes it the raw-milk skeptic’s idea of what happens when a cheesemaker embraces native milk microbes. Time and again, when we encounter dairy scientists from Anglo-Saxon institutions, they repeat the same mantra: raw milk inevitably leads to unpredictable variations in quality and the ever-looming threat of tainted, dirty flavors. Why not pasteurize in the spirit of cleanliness, control, and progress? Dr. Mark Johnson, the assistant director of the Center for Dairy Research at the University of Wisconsin-Madison, expresses the sentiment most clearly: “Pasteurization levels the playing field.” Dispirited at the showing of our Bleu de Termignon against the Cambozola, it is a tempting thought. But restricting our choice of microbes to those available in a culture company’s catalog goes against the spirit of chasing unique and distinctive characteristics. Without the scale to approach a company like Chr. Hansen for a proprietary cocktail of cultures, what is the small cheesemaker to do? Is it possible to tame the beast? What if one were C U LT U R E S

•

1 65

to level the playing field and then subsequently make it a little less level, to reintroduce the milk’s own microbial communities in some way that allows greater control but can still make a claim to the unique character of the farm? The pursuit of native starters is not new. Indeed, cultures are often named to reflect their origins. One of the world’s most successful undefined multistrain starters, common in everything from crème fraiche to Brie, is known by the name Flora Danica, “Danish flora.” The mellow, buttery flavors that it imparts are delicious, but its ubiquity reflects the power of Storch’s original innovation: we all make Danish cheese now. In France too it is possible to buy special culture blends tailored to the cheeses of specific regions. Browse through the catalog of a boutique dairy ingredients and equipment distributor like Coquard, based just north of Lyon, and you are greeted by a comprehensive collection of different cultures, each boasting of their rustic typicity for a different individual cheese. Local regional practice is taken into account too, with the starters offering cheesemakers the tempting prospect of a discreet microbial helping hand, even when making some of the least technologized cheeses. The cultures designed for Salers are fully equipped to make the cheese by themselves, but the instructions assiduously note that they are only for use “in addition of natural flora, especially in wooden vats.”34 However, this sort of diversity is at best a choice between regional blends. It offers cheesemakers the prospect of conforming to collective ideals but does nothing to magnify the individual character of their own farms. It is in this context that we can conceptualize the holy grail of starter cultures: farmspecific cultures, each with sufficient diversity to capture the essence of that farm’s milk but with all the nasty bugs removed. The most enthusiastic proponents of this method are convinced that it offers the potential to eliminate raw-milk production altogether. Variability and risk would be gone forever in this brave new world of absolute control. Unsurprisingly, it is those countries with the most restrictive laws governing raw-milk cheese production that have shown the most interest in innovating with starter cultures, and in this case, it is Australia that has taken the lead, although the sentiments behind the Australian cultures trial of 2014/15 mirror work contemplated everywhere from Europe to North America. Certainly, as a collaboration between the Australian Specialist Cheesemakers’ Association (ASCA) and Dairy Innovation Australia Limited (DIAL), the cultures trial resisted the normal logic of the starter culture market: here were farmhouse cheesemakers dictating the problem to be solved by scientists, a 166

•

C U LT U R E S

problem expressed as something other than just the requirement for ever faster and more reliable acidification of their milk. Three farmhouse cheesemakers in the state of Victoria participated in the trial by providing fresh raw-milk samples in October 2014. Holy Goat Cheese, l’Artisan Cheese, and Prom Country Cheese represent goat’s, cow’s, and ewe’s milk, respectively, and their samples were analyzed by a team of scientists at DIAL. The raw-milk microbes were screened: colonies were isolated and identified, and selections were made for those with no record of pathogenicity. The investigators then prepared four culture blends: one specific to each of the three farms and the fourth a “super blend” of cultures from all the farms put together. Most notably, there was no attempt to select bacteria for their role as acid producers during fermentation; the cultures that the project produced were designed to inoculate pasteurized milk alongside conventional commercial acidifying strains. The cheesemakers then made cheese using these adjunct cultures, and the results were compared with control batches made without any addition of farm-specific strains. At the final sensory evaluation in November 2015, a blind-panel tasting compared three cheeses from each producer: the control, cheese made with the specific adjunct cultures of each farm’s indigenous milk microbes, and cheese made with the “super blend” from all three farms. The three types of cheese involved in the trial showcased a diversity of styles, ranging from a semihard goat’s milk tomme to a Reblochon-style washed rind to a ewe’s milk blue, and the initial results of the blind tasting were statistically noisy. With just one iteration of the experiment, finding definitive answers amid the batch-to-batch variation of small-scale farmhouse production was a challenge: the tasting panel’s preference was for one cheese made with the indigenous cultures, one with the “super blend” of cultures, and one made without adjuncts as a control.35 But that ambiguous finding is not the point. Although the project was widely reported as an exercise in selecting beneficial bacteria and promoting microbial terroir, that was not the real agenda.36 Dr. Ian Powell, the lead scientist behind the project, was frank in acknowledging that the adjunct cultures—each made up of five to seven species, representing up to ten strains, and inoculated into the milk at levels far higher than would occur in nature—do not capture the full microbial diversity of raw milk. Within DIAL, there was no pretension otherwise. The process of whitelisting, by which microbes were selected for inclusion in the culture blends, was by definition one of simplification, undertaken with a C U LT U R E S

•

1 67

conservative perspective on risk. Instead of including everything, barring known pathogens, in the cultures, the project adopted the opposite strategy: microbes were only included if they could be positively identified and had a record of nonpathogenic appearances in food. Where there was a lack of information about a bacterium or its association with foods, it was avoided. Rather than capturing the essence of each farm’s unique microbial ecosystem, the Australian project can be understood as a feasibility study, an exercise in testing if it is commercially viable for farmhouse producers to commission their own work from microbiologists. When we spoke with him, Powell was upfront about his opinions on the subject: “Cutting the on-going production cost whilst retaining culture composition and safety is a bit of a challenge, especially because these are (by definition) not going to be mass-produced cultures. The estimates are rubbery, but you’d need about A$20,000 [over $15,000] to get started.” That is a huge capital outlay for a small farmhouse cheesemaker, but it must be considered in the context of an Australian industry that feels starved of access to cultures appropriate for small-scale producers. Alison Lansley, the secretary of ASCA, is acutely aware of the problem: “We now have just a few cheese cultures available to Australian artisan and farmhouse cheesemakers . . . all [designed to meet] just the needs of Australia’s big dairy producers. And the list is shrinking all the time.” Faced with this existential threat to their cheesemaking community, the cultures trial was just one strand of ASCA’s response; the organization has also started a project to import a selection of dairy cultures from France and distribute them among its members. Despite its reception in the media, the project was not so much about the lure of combining consistency with uniqueness as it was about creating structures through which raw milk can be studied as something other than a source of potential pathogens. In an environment where regulators treat raw milk with suspicion, ASCA has subtly moved the goalposts. DIAL can fret that “further development of this concept should involve regulators and food safety experts to assist in defining acceptable microbial content for commercial cultures,”37 but that involvement is exactly the process that will make raw-milk cheese seem less threatening. The more raw milk is studied and its microbial populations are named and identified, the more it becomes part of the normal conversation among cheesemakers and their regulators. To know is to understand and to remove fear. The cultures trial and her work on enabling Australian raw-milk production are for Lansley inseparable, two parts of the same thought. They are both simply “cultures-related projects.” 168

•

C U LT U R E S

A H O U S E O R A C A N VA S ?

Farmers and cheesemakers are usually practical people with a pragmatic regard for solutions that work. Questions of philosophy and worldview rarely intrude into the creamery, but the prospect of farm-specific, bespoke starter cultures brings to the surface perhaps the most profound philosophical tension in the cheese world today. What is the relationship between the modern cheesemaker, their milk, and their cheese? Must cheesemakers nurture and steer their milk and then build a house to host its microbes? Or is the milk instead a blank canvas on which cheesemakers paint with their choice of bacterial cultures? The latter is certainly the impression that you would get from a culture catalog. Some factories still produce thousands of gallons of thick, sour starter cultures each day under aseptic conditions to seed their fermentations. But in the 1970s, a new innovation appeared that has rapidly grown to dominate the cultures market: Direct Vat Set (DVS) technology. DVS starters are highly concentrated, standardized blends of bacteria in suspended animation. Unlike weakened powdered cultures that need several days of intensive care to revitalize them before use, or the undefined liquid cultures that require daily propagation to keep them active and healthy, DVS starters can be added straight to the vat, with no incubator, sterile media, or special skills required.38 Is ripening too slow? Then just add CR-500 to bring “power and balance” to low-fat continental cheeses in as little as nine weeks. Alternatively, the CR-Full Flavor range allows the cheesemaker to choose from a portfolio of flavor profi les, from “Farmhouse & Savory” to “Fresh Buttery & Malt Chocolate.”39 DuPont’s Danisco cultures are sold by one of their US agents with copy taken straight from a fashion catalog: “The MM series makes a great base culture for customizing with adjunct cultures to make specific cheeses: Add TA50 series to make stabilized soft cheese or semi-hard Tommes. Add MD88/MD89 to double up on the gas production and butter notes. Add LM57 to get the buttery aromas and creamy mouthfeel without adding more gas and eyes, combine with Aroma B to make Swiss, Edam, or Gouda, the possibilities are endless.”40 Mariano Gonzalez, the cheesemaker at Fiscalini Farms in Modesto, California, the producers of one of the most highly regarded raw-milk Cheddars in the United States, carefully emphasizes this point, telling the website myFarmLife.com, “John [Fiscalini]’s milk is a blank canvas. . . . I can push John’s milk in any direction with the right culture.”41 C U LT U R E S

•

1 69

In the United Kingdom, DuPont cultures are distributed by a food technology firm whose other operations include the production of cake mixes, the sort where you just add an egg to make the batter.42 In discrete moments and after a few drinks, this is all too good to be true for cynical wiseacres within the cheese industry: the cheeses made with these cultures, they say, are “Betty Crocker cheeses.” All the cheesemaker needs to do is open the packet and add milk. To what extent is this true? Can you really create a cheese using a paintby-numbers approach with the right selections from a catalog? Back at Harvard, in 2013, Rachel Dutton and Ben Wolfe put this theory to the test in a simple and clever way.43 As they analyzed cheese rinds, they identified the types of bacteria and fungi that turned up most often and then made a mixture of representatives from the six most common bacterial and five most common fungal species across all of the different cheese types. Next, they took this initial “universal” mixture of microbes, inoculated it into sterile, curd-like agar plates in the lab, and then treated these model cheeses in different ways. One group was bathed in a salt solution to simulate rind washing; another was kept in a drier environment, as a hard, natural-rind cheese might be. A third group of cheeses got an extra dose of fi ft y times more Geotrichum, to simulate what might happen if a cheesemaker added ripening cultures to their milk before making cheese. A set of controls received no special treatment whatsoever. After four weeks of “maturation,” the microbial communities that had grown from the initial mixture were sampled again, and just like in real cheeses, their balance had changed in response to the different treatments. When subjected to different conditions, the generic mix of microbes rapidly organized themselves into the characteristic communities found on cheese rinds in the real world. The model cheeses that had been washed were dominated by bacteria similar to those found on real washed-rind cheeses, while the drier, natural-rind treatment led to communities enriched in bacteria and fungi found on harder cheeses. Most interestingly, the massive dose of Geotrichum, which was meant to produce a bloomy-style cheese, had no effect. Although the Geotrichum-enriched cheeses had been matured under the same conditions as the cheeses in the control group, the two microbial communities, which were initially different, evolved and settled into exactly the same balance. Artificially raising the number of one species—the equivalent of inoculating with a commercial culture—made no difference to the final result. The environment determines all: build it, and they will come. 170

•

C U LT U R E S

The same principle sheds light on the results of the Dutton lab’s survey of rind communities on real cheeses. The microbial composition of the rinds of cheeses was similar for the same styles of cheeses, regardless of where in the world they were made. It was moisture level that was most strongly correlated with what grew on the rind. Acidity had less of an impact, and salt level none at all. The finding that something as mundane as moisture trumps terroir left some members of the artisan cheese community a bit chagrined. The tacit hope had been that Dutton and Wolfe’s high-tech tools would demonstrate the uniqueness of each cheesy snowflake. Instead, the experiment showed that similar and predictable communities assemble on cheeses of the same style, whether they are in California, Somerset, or the Pyrénées. Having thus killed the concept of microbial terroir, Wolfe—now an assistant professor of microbiology at Tufts, with his own lab—is in the process of reviving it. For, although the underlying conclusion of the initial survey holds true, that survey cataloged microbes according to the relatively broad taxonomical category of genus rather than by species or even strain. As a result, communities that looked similar at the survey level have turned out to be made up of different species playing similar roles. Wolfe and his team have been constructing model communities in which one strain is switched for another and testing the knock-on effects that these minor substitutions have on how the communities perform. A particular strain—which might be unique to a single farm—might be more competitive in a given context and grow to dominate a community where a similar organism from the same species might have played only a minor role. Likewise, specific strains might have unique abilities, such as churning out volatile sulfur compounds, thus completely changing the aromatic profi le of a cheese. A study done in 2002 on Gubbeen, an Irish washed-rind cheese, called into question the principle of inoculating with commercial ripening cultures in the first place.44 The premise of the experiment was simple: make a batch of cheese, split it into two lots, inoculate one with a commercial strain of the orange-pigmented, salt-loving bacterium Brevibacterium linens (lauded for its critical role on washed cheese rinds), leave the other lot culture free, and then treat the two lots normally, washing them on a regular basis with brine solution to ripen them. The result? Despite rubbing the surface of the inoculated batch with a culture containing many billions of purified B. linens cells and keeping those cheeses in an environment that should have encouraged that species to prosper, not a single viable sample of B. linens was detected on the experimental cheese over the course of the study. Instead, the team found C U LT U R E S

•

17 1

different species of orange-hued bacteria (corynebacteria and some staphylococci) that had the ability to inhibit the growth of the commercial B. linens. The inoculated strain was annihilated by the native inhabitants of the creamery—it never stood a chance. Among the native microbes, the research team found a new species on the rind of the Gubbeen, which they dubbed Microbacterium gubbeenense in tribute. Then they took it one step further, making it into a purified culture that the cheesemakers now add to each batch of cheese. It is an exciting story of the triumph of the native. But given that M. gubbeenense was growing just fine on the surface of the cheeses already, it’s worth asking whether adding more of it makes any difference. How will the domesticated strain fare as the Gubbeen creamery and rind community continue to evolve naturally over time? Will it eventually cede its home turf to a fitter, better-adapted strain, becoming another snake oil culture like the commercial B. linens before it? Like the Gubbeen study, the Dutton survey turned up a wide range of microbes in cheese-rind communities that no cheesemaker had added intentionally, ranging from ocean bacteria to soil fungi. No company commercializes these as adjunct cultures—they made their way onto the cheese the oldfashioned way, either through the milk or by chance encounter during the make or in the ripening rooms. Contamination does not have to be dramatic to be effective. As little as one spore of Geotrichum per kilogram of curd can lead to levels as high as one hundred thousand per gram after twenty-one days if the conditions are right.45 It is not the microbes that make the cheese, but the cheese that picks the microbes—and then tells them how to behave. Cheese made and stored in a given way becomes a selective environment that encourages specific microbes to grow. What is striking is that this is not a new idea. Back in 1912, at the very dawn of the use of starter cultures within cheesemaking, a team of researchers from the US Department of Agriculture wrote a guide called The Bacteriology of Cheddar Cheese.46 Over a century later, their perspective on microbial ecology sounds strikingly modern, a view that might be expressed by microbiologists working with the most sophisticated new equipment. These are words to live by for all cheesemakers, wisdom that the last century all but lost: “As has been stated, the microorganisms necessary for the preparation of most kinds of cheese are contained in the milk. Whether one form or the other is to develop in the cheese depends upon the environment established by the cheese maker. The manufacture of cheese is thus a problem in the ecology of microorganisms.”47 17 2

•

C U LT U R E S

T H E H O U S E T H AT J O S I A H B U I LT

In that case, what is the point of starters? They are obviously an absolute necessity for cheeses made from pasteurized milk, but do they have value for cheesemakers working with raw milk already rich in its own microbes? Where traditional methods for souring the milk work well, we are enthusiastic partisans for their use. These methods are not used only for the production of idiosyncratic cheeses like Salers, with its wooden gerles—they are thoroughly mainstream techniques. In particular, the use of continuous whey starters—simply adding a little of the previous day’s sour whey to the new milk—is a robust and effective method for making lactic and Alpine styles. It is not without good reason that the technique is ubiquitous among goat cheese producers in the Loire Valley in France and a positive legal requirement for cheeses like Comté, Gruyère, and Parmigiano-Reggiano. At the same time, we are deeply skeptical of inoculation with ripening (as opposed to acidifying) cultures. The research on Gubbeen and the work of the Dutton lab demonstrates that these are an unnecessary and expensive distraction. At worst, they lead cheesemakers toward snowy microbial monocultures, as we saw in the Cambozola. But what of other styles of cheese? What should a cheesemaker do with a cheese like Cheddar, where whey starters are markedly less successful? Addressing the question with the pragmatism of a dairy farmer means that we should not make a Manichean distinction between “good” cheeses made exclusively with native microbes and “cheats” made using commercial cultures. Between working absolutely without inoculation and dosing heavily with the most aggressive and muscular strains of starter bacteria, there is plenty of room for nuance. Indeed, the choice of which cultures to use and the level at which to dose them is just as significant as the choice of whether to use starter at all. The problem comes when the choice of starters and adjuncts dominates the character of the cheese. Our test when tasting is a simple one: if the flavors derived from the starter are clearly perceptible, then they are too much. In the same way, even a pasteurized cheese made with defined cultures under the cleanest possible conditions will only be as stable as it is well made. If the acidification stalls, or too much moisture is left in the curds, the result will be a bitter, drippy mess, covered in all of the wrong sorts of microbial “contaminants,” species that were there all along in minute quantities but suddenly found themselves at an advantage. The same principle works in reverse. Bleu de Termignon is a low-acid cheese whose sweet, wet curds are left bathing in whey before being squelched C U LT U R E S

•

173

through a grinder and stuffed, still soft and dripping, into molds. When it comes to microbes, such high-moisture, low-acid environments are the ultimate in permissive parenting. The bitter, yeasty flavors of the cheese are not the essential expression of wild microbes. They are flavors of process. Working with exactly the same milk and tools in the same cheese room but treating the curd differently—coaxing more moisture out earlier or adjusting the temperature to encourage a bit more acidification—would produce a much more polished cheese, but of course, it would no longer be Bleu de Termignon. Our greatest guide in reaching this understanding, the leaping-off point before we contemplated the detailed scientific research, was the wise words of Josiah Twamley. While he wrote before there was any understanding of the microbial origins of dairy fermentations,48 Twamley had a sound empirical appreciation of cheesemaking and how potential faults could be avoided. As we noted in our discussion of farming and cattle-feeding systems, the greatest concern to the eighteenth-century English dairymaid was a problem described as “hove cheese.”49 This is cheese that, as it matures, develops a yawning chasm within the paste, accompanied by putrid, rotten flavors. In modern terminology, this is the “late blowing defect,” the product of butyric acid fermentation, where lactate is fermented into hydrogen gas (the cause of the holes) and butyric acid (which tastes of vomit).50 By the 1780s, cheesemakers had discovered that adding saltpeter controlled the problem; this is the method that many modern Gouda producers still use to prevent the late blowing defect.51 But Twamley’s preferred solution, his “most powerful preventative to the heaving of Cheese,” was simply to encourage a full fermentation and make sure that the curd was well drained. Both of these operations make the selective environment within the cheese hostile to the Clostridium tyrobutyricum that causes late blowing.52 This was the driving force behind the evolution of British territorial cheese as a style, an evolution that was powered by a robust practical application of the principles of microbial ecology. Even in complete ignorance of their existence, Twamley was building a house for his favored microbes.

174

•

C U LT U R E S

NINE

Families and Factories

It is getting dark as we pull into the gravel drive at Beesley Farm. A cold November rain pours down as we stumble toward the dairy; we are thoroughly soaked by the time we get inside. It is wet here, up in the northwest of England. Preston, just down the road, is the wettest city in the country. The Pennines, the mountainous backbone of England, rise a few miles away at Beacon Fell, but to the west it is flat all the way to the sea. Standing there in our sodden clothes, it feels like we have been lashed by the North Atlantic. Seeing us at the door, Graham Kirkham flashes a smile and welcomes us in. We are here to taste his Lancashire cheese. Within the county of Lancashire, the Kirkham family stands proudly alone. In 1939, there were 202 farms making cheese in the county, but now they run the only farm making raw-milk cheese in Lancashire. Graham is an amiable personification of their enthusiasm. A sturdily built man in his early forties, he pads around with surprising delicacy while we hang up our wet coats and don aprons and white wellies. He searches around for his cheese iron, the long bayonet-like knife that allows him to remove thin core samples of cheese for us to taste. He calls it “the stabby thing.” Finding the iron, Graham draws the first sample, and we begin to taste. The cheeses are astonishing: they are unlike any other cheese being made today. Even many of the locals have forgotten the taste of this kind of Lancashire. Francis is the child of two Lancastrians—indeed, his parents’ wedding reception was in a pub a mile from the Kirkhams’ farm—but they are perplexed when they encounter Graham’s cheese: “It’s not crumbly; is it really Lancashire?” When Graham tastes finished cheeses, he looks for a combination of mellow creaminess and weightlessness. He slides in the long sampling iron, twists it with exaggerated care, and then pulls it back slowly 175

at an angle to make sure that he reveals a fluff y collection of curds. This is the reward for his time, a texture that is on the verge of complete extinction, sacrificed to the demands of speed and efficiency. The texture has a special name that perfectly captures its balance between richness and lightness: a “buttery crumble.” For Graham, the best cheeses are “fluff y monsters.” The flavors too are delicate and nuanced. Kirkham’s Lancashire was the only cheese we served at our own wedding, a decision that was met with some surprise from Bronwen’s family: “Is that really your favorite cheese? I was expecting something, well, stronger.” Bronwen can understand this reaction; when she started work as a cheesemonger, she too gravitated toward oozing washed rinds and piquant blues. When we tell him this story, Graham chuckles and recounts his experience at a local food show. Some people were going from stand to stand, looking to gorge themselves on as many free samples as possible. They scarfed down the samples of his Lancashire cheese and walked away. “It wasn’t until they had taken about ten steps that they suddenly turned around; I could see that they had really tasted the cheese,” he says. “That’s what Lancashire is like. It’s a slow release that gradually builds up layers of flavor.” “Slow” is the key word here. In its taste and texture, no cheese better illustrates how before sophisticated biotechnology, before even mechanization, the first great epochal shift in the world of dairy was the attempt to control cheesemaking in terms of the clock. This was not a change that came through great technical innovation. It did not require new machines or even immense facilities. No, for cheese, “industrial” was a mentality more than anything else. And with its gentle succulence—achieved through techniques that directly contradict the logic of factory production—Kirkham’s Lancashire is a vestige of another age. In that, Graham, so unassuming as he potters around his cheese store, is a great subversive. Confronted by the modern world, he is an inadvertent cosmologist, an accidental philosopher of time.

C H E E S E A N D T H E FAC TO R I E S

One can’t help but think about the Industrial Revolution on the drive to visit the Kirkhams. Their corner of the county of Lancashire bore witness to Britain’s industrial history. Preston was the birthplace of Richard Arkwright, the inventor of the spinning frame, and it was the water cascading down from the Pennines that powered the industrial transformation of the cotton indus176

•

FAMILIES AND FACTORIES

try. In the nineteenth century, Lancashire was the workshop of the world, and the local mill towns grew rich. That has all gone now, although the industrial past still scars the landscape. Beesley Farm is on Mill Lane, but the mill—which processed feed rather than cotton—that gives the road its name has long since closed and been converted into an expensive luxury house. The scars of industrialization persist on a deeper level too, and not just in Lancashire. The advent of factories and mechanization brought the working population of the United Kingdom suddenly under the discipline of the clock. Time in an agricultural society is inherently task oriented: dawn, dusk, and how long it takes to plow a field are more relevant than the precise hours on the dial. When the factories arrived, everything changed. Clocksmiths were the first precision engineers, and their skill contributed to the complex mechanical operations of the early factories. But more than intricate moving parts, machines required an external sense of time. In the textile industry, before Arkwright’s spinning frame, cotton was spun by women in their homes as piecework and then passed on to men for weaving; the day was malleable and could be adjusted around the tasks at hand. But once machines came into play, time was suddenly absolute: workers had to be on time if the mill was to be efficient. But this drive for mechanical and temporal efficiency did not suddenly plunge the county’s smallholders into a world of factory farming. Just as the immediate consequence of the mechanization of spinning was a rapid expansion in the demand for skilled handloom weavers, whose numbers dramatically increased before mechanization caught up with them, so feeding the new mill towns ushered in a boom time for the small tenant farmers clustered in the surrounding countryside.1 As late as 1895, the Preston Guardian could insist that “the peasant has held on so well that in this section of the country he is supreme, and the agriculture of the peasant is the prevailing style of farming.”2 At that time in the United Kingdom, the greatest concentration of small farms was to be found in the three northern industrial Pennine counties dominated by mining and the textile industry: Lancashire, Derbyshire, and the West Riding of Yorkshire. Smallholders they might have been, but ignorant bumpkins they were not. Their close proximity to towns meant that small farmers in these counties could sell directly to the consumer. In the words of historian Michael Winstanley, these smallholders should be understood “not as peasants, but as market-orientated entrepreneurs.”3 Given the damp climate and marginal FAMILIES AND FACTORIES

•

17 7

land they had to farm, which were deeply unsuited to the increasingly intensive arable farming popular with the era’s agricultural improvers, dairying was the natural choice. This is the context in which Lancashire cheese originally evolved, in which the lure of factory wages drove up the cost of agricultural labor to the point that in the county, “the capital of the small farmer is the labor of himself and his family.”4 By the time that the production of Lancashire was codified by Joseph Gornall in the 1890s, its niche was clear: it was a family cheese, made by the wives, mothers, and daughters of the local smallholders and sold directly to customers in the nearby towns. Those characteristics of industrialization that transformed the lives of the mill workers—the strict time discipline, the specialization, the disposable income from factory wages—allowed Lancashire cheese producers to thrive in a system that inverted the logic on which the factories were built. We should be careful not to romanticize these smallholders. They might well have eschewed specialization to produce and sell poultry, eggs, and perhaps some meat or vegetable crops as well as their cheese, but they were not living lives of bucolic rural idyll on the fringes of towns. Quite the opposite: contemporary commentators praised the smallholders for their sober industry and diligent toil.5 But the extent to which they thrived demonstrates that when it comes to cheese, Britain was not the first industrial nation. Even though Lancastrian farmers were engaging directly with factory workers every week, not one of them attempted to become the Richard Arkwright or Henry Ford of cheese. No, for the application of the industrial mentality to cheese, we need to look across the Atlantic.

A M E R I C A’ S G I F T

Discussions of the impact of the United States on cheese inevitably descend into a conversation about American cheese and the ubiquitous orange Kraft singles. But America’s industrial gift to the global cheese system predated James L. Kraft by a generation. The dialogue that transformed cheesemaking was a transatlantic affair, and its prophet and evangelist was Xerxes Addison Willard, a dairyman-turned-journalist in upstate New York. Willard might have started life as a farmer, but it was as a commentator and visionary that he defined modern dairying and its relationship with cheese. For all of Kraft’s subsequent innovations with processed cheese, he was merely playing out the logical conclusions of Willard’s work. 178

•

FAMILIES AND FACTORIES

The New York State Cheese Manufacturing Association was just one year old when Willard delivered the keynote address at a meeting in Utica in January 1865. Speaking nearly a decade before the birth of Kraft, Willard outlined his vision with an almost religious fervor, flitting between descriptions of the rewards that would flow from factory production and the inherent weaknesses that threatened to bring the entire social experiment—not to mention the audience’s financial security—crashing down. Jesse Williams, a local dairyman and the founder of the association, had built the first cheese factory in a neighboring town only fourteen years before, and in that short time, the number of centralized cheesemaking operations had exploded to almost five hundred. In the year leading up to Willard’s address, they had collectively churned out over fourteen million pounds of produce. While today, the word “factory” conjures images of conveyor belts and belching smokestacks, the cheese factories of nineteenth-century New York more closely resembled what we now call farmers’ cooperatives. Rather than relying on a third-party investor, most of these so-called associated dairies were joint-stock companies owned by dairy farmers within a tight catchment area. Without commercial refrigeration, distance was a crucial limiting factor; around four miles was the upper limit beyond which the trouble of hauling heavy churns of warm milk over rough dirt tracks twice a day outweighed the advantages of outsourcing the cheesemaking. For a factory to cover its overheads, it needed to produce at least fifty tons of cheese a year, but the larger dairies could turn out up to three times that much, processing the collective milk of up to a thousand cows.6 Without the ability to store the milk, factory employees worked seven days a week, just like the dairy farmers, but the larger rhythms still took their cues from nature. Just as it had been on the farms, factory cheese was only produced between April and November.7 Willard was known for his weekly reports on cheese-market activity and his commercial expertise. Intimate knowledge of market forces and inefficiencies was key to his very modern vision of success: maximizing profit margins while growing volume aggressively. Willard’s ambition was to see dairy farming elevated to a powerful business; no longer would farmers merely muck about with animals, struggling independently to little effect. By banding together to promote their mutual interests, dairy farmers would become a powerful force. Factory production was not booming just in New York. Dairy farmers further west in Ohio and Wisconsin were also building factories at great speed, and Willard was acutely aware of this: “We have let the whole world FAMILIES AND FACTORIES

•

179

into this factory system, and if low priced cheese is to be the rule, we of the old dairy districts, will not be able to compete with those of the new on low priced lands.”8 Instead of scuffling with the upstarts to the west, Willard laid out the more ambitious goal of exporting to Europe: “Without a market in Europe the supply, it is evident, will be so great as to glut the home trade, and render cheese dairying unprofitable. . . . We can compete with the dairymen of the old world as to prices, and when we shall be able to out-do them in quality a market for our goods is secured for all coming time.”9 Quality was the problem. When the first American cheeses found their way to London in the early 1840s, a local commentator sniffed, “In some matters the Americans have adopted modes and customs different from ours . . . and by no means superior. . . . As a consequence, they produce a quality of cheese decidedly inferior to our own.”10 He had a point. The first American cheeses to reach Europe were spongy, bitter, and reeked of the barnyard, and they quickly acquired a reputation as second-class goods. The gleaming metropolis of London, capital of the most powerful nation on Earth, was a sophisticated market where price was no object. British cheesemakers were hardly perturbed by the American imports that began to trickle into the market. “Let Americans make cheese for the lower classes,” said one British cheesemaker, “but let it be the aim of the home farmers to make cheese for the higher classes, which always commands a better price.”11 Cheddar was uniquely suited for both factory production and export to Great Britain. First, there was a voracious demand for the cheese, particularly in London, where, apart from the rare and precious Stilton, Cheddar had recently taken pride of place as the best and most fashionable cheese. Willard told his audience in New York, “It is unquestionably the highest grade of cheese produced, or at least it best satisfies the English taste, and as England is our largest customer, it is important that we pay attention to and humor that taste.”12 Indeed, a description of the English farmhouse Cheddar of the time evokes a glorious cheese: “close and firm in texture, yet mellow in character or quality; it is rich, with tendency to melt in the mouth; the flavor full and fine, approaching to that of hazle nut [sic].”13 The cheese was also immensely flexible. It could be sold at just thirty days of age, but it could also be kept for a year or more, during which time—provided the cheese was well made—its flavor only became more profound. And it was robust enough to travel. Ironically, the boat ride across the Atlantic was the least of the physical trials that an American Cheddar was likely to face. Unlike the temperate United Kingdom, New England’s climate posed great 180

•

FAMILIES AND FACTORIES

challenges to the cheesemaker. New England factory “dry houses” (maturation stores) often experienced extreme fluctuations between heat and cold, and dampness and dryness, and these took their toll on the young cheeses, splitting them open and making way for cheese mites and spoilage molds. American cheesemakers devised a novel solution: binding their cheeses in cloth to protect them. It is a feat of collective forgetting that today we think of Somerset Cheddar as the quintessential clothbound cheese when, in the 1860s, Cheddars with cloth bindings were regarded by the English market with suspicious derision. Willard recounted of his grand tour of the United Kingdom: “I have been at hotels where American cheese is always purchased in preference to English, and I have been amused to hear Englishmen contend that no such cheese could be produced in America, and nowhere else except in the best dairies of England, but who were forced to give way on pointing out to them the bandage, which is an indisputable proof of American manufacture.”14 Competition from the best English farmhouse Cheddars was an impediment to the growth of the American industry, but the English Cheddar makers were also a tantalizing source of know-how that could be used to make American factory production better. The New York State Cheese Manufacturing Association hatched a plan to send Willard to England to collect the technical knowledge necessary to give the American factory cheesemakers the upper hand.

THE SOMERSET EMPIRICIST

The man to whom Willard turned to understand the intricacies of the Cheddar make was Somerset cheesemaker Joseph Harding. Born in 1805 to a family of cheesemakers, Harding witnessed the arrival of time discipline in rural England. His youth was infused with the natural rhythms of the rural community. Local parish church clocks were set according to the position of the sun, their chimes providing little more than sonic punctuation to village life. Personal watches were luxuries far beyond the reach of the peasant laborer, but more than their cost, the time they displayed was irrelevant to the pacing of the day. Harding was already in his early thirties by the time Brunel’s Great Western Railway cut a path from London to Bristol, the main port for trade with America, passing directly through Cheddar country on its way. Village-specific variations of time were incompatible with the rigor of FAMILIES AND FACTORIES

•

181

timetables and railway schedules, and in 1840, the Great Western Railway was the first of Britain’s rail companies to synchronize its local schedules with the time in London, 120 miles to the east. Rural communities that had operated on a loosely acknowledged local time were jolted into a new era of time discipline just as the new transportation network began to bring them into contact with a global market. It was not long before American factory Cheddar appeared at local village stores throughout the United Kingdom. Harding embraced the spirit of improvement through control and empirical inquiry. He deplored the inefficiency of the methods used to make farmhouse Cheddar and the enormous waste that resulted when batches turned out badly: “The process of manufacture was unsystematic and irregular . . . [and] consequently the cheese was of unequal quality—some good, some bad—from causes unknown to the dairy-women.”15 He also recognized the immense toll of laborious methods that had not changed for centuries: “The curd, when put into the vat, was broken into small pieces by the hand, so laborious a work that I have seen dairy-women whose finger-joints were grown large and stiff in consequence.”16 However, he didn’t envision a factory solution, as Willard did. Instead, he set out to devise a better way to make Cheddar on the farm. His experiments were extraordinarily successful. The new methods he devised for breaking and scalding the curd increased yields by almost a quarter, locking butterfat that had been lost in the whey within the cheese to make it richer and creamier. He also introduced a grinding machine—or “curd mill”—that ripped the curd apart much more quickly and efficiently than an aching pair of hands. Cheesemakers were spared repetitive stress injuries and the pressure to cut corners at the end of a long day. Harding’s work inspired an era of farmhouse cheesemaking innovation, and a cascade of inventions flowed forth. Some were elegantly practical, others hilariously overengineered. Progressive cheesemakers traded in their wooden tools for tin, fitted their make rooms with boilers that provided hot and cold running water, and tried out the latest technology. The names of the new cheesemaking inventions were as evocatively Victorian as the etchings that depict them: Keevil’s Curd Press, McAdam’s Telescopic Whey Separator, and the revolutionary Cockey’s Cheese-Making Apparatus, the first temperature-controlled milk and cheese vat. Eccentric inventions helped, but Harding was convinced that far more than any labor-saving device, it was “the thermometer and the clock [that] have greatly assisted in reducing cheesemaking to a regular system.”17 He also 182

•

FAMILIES AND FACTORIES

recognized the limitations in the understanding of the process: “[We] have acquired, so far, some knowledge of what we are doing, [but] we have not yet arrived at perfection. Cheesemaking, as a science, is not understood.”18 While Harding himself knew that more remained to be discovered for “cheese [to] be made (as it ought to be) upon principles scientific and, consequently, unerring,”19 there was no doubt that his improved Cheddar method got results. Cheeses made according to his system took top honors at cheese shows throughout the British Isles and were “regarded by Englishmen as the finest cheese . . . made anywhere.”20 An idealistic social activist, Harding was keen to share his method with as many people as possible. He traveled to Scotland to lecture and teach, advocated the publication of scientific findings in journal articles, and advised that dairy schools be established to spread skills and knowledge efficiently. He also hosted guests from across Europe, as well as the ever-curious Willard from America. For his part, Willard returned from his European tour somewhat bemused. He intuitively grasped the achievements of Joseph Harding: “Laying all prejudice aside I must, in truth, say that we have not yet been able to surpass in excellence the fine specimens of English Cheddar. It is a very high standard of cheese, and is deserving of all the encomiums which it has received.”21 However, the rest of British cheeses were another matter, and he considered them to be little more than backward old-world curiosities. Of Wiltshire cheesemaking, he said, “It is a matter of surprise that the people of this district are so bound up in old practices as to waste their time and substance in manufacturing cheese of this character,”22 and he dismissed Cheshire as “somewhat peculiar . . . to an American, [it] would be called decidedly antiquated.”23 The relationship between Willard, Harding, and Cheddar, the cheese that they championed and transformed, highlights the same intriguing paradox we encountered in nineteenth-century Lancashire. Within our modern cheese world, we readily draw a distinction between artisan and factory, between European tradition and New World artifice. And yet, both sides of all these distinctions have their origins in the same moment, their inspiration in the work of the same men. Lancashire cheese, its logic so antithetical to the factory, owes its existence to industrialization. The same tension exists within Cheddar. When the three remaining raw-milk Somerset farmhouse Cheddar producers gathered together in 2003 to write a definition of their cheese for a Slow Food Presidium in the Ark of Taste, they made great play of the cloth binding.24 The cheese must be bandaged with lard-soaked muslin; this, in FAMILIES AND FACTORIES

•

183

opposition to shrink-wrapping in plastic or coating in wax, is one of its defining features as a traditional cheese. And yet, as we have seen, marketing cheese bound in cloth was an American innovation of the nineteenth century. All modern Cheddar makers are careful to genuflect before Harding, acknowledging his impact on British farmhouse cheese. Without his work, there would be no Cheddar as we now know it. But at the same time, the use to which Willard put Harding’s innovations was instrumental in the industrial transformation of cheese.

FAC TO R I E S A N D T H E C H E E S E

Willard described the advent of the cheese factory as “one of the most remarkable steps in the history of progressive farming.”25 Transferring production from the farm to a communal dairy was, in one sense, entirely superficial: the same process was used to make the same cheese, just a few miles away. But the shift from farmhouse to factory production represented an irreversible change in mentality. There was no going back, and Willard recognized that, even as early as January 1865: “The [factory] system is a progressive step; and all history teaches that when that is taken it is difficult to retrace the step.”26 Before the advent of the factory, cheese was one of many products of the family farm, an economic unit that wove together the diverse skills of the entire family and put them to use turning the resources at hand into a viable living. Making cheese was not a farm’s reason for being; rather, it was just one important part of a diverse system. Family members augmented their work on the farm with seasonal labor in industries as diverse as spinning and fishing. A side benefit of this mix-and-match approach to making a living was self-determination. While people were often engaged in work-related activities from early morning to late at night, they were the masters of their own time, adjusting their own schedules to fit the tasks at hand. There was little distinction between working and “passing the time of day.”27 By necessity, farmhouse cheesemakers were also champion multitaskers. Being a great cheesemaker didn’t mean neglecting other duties; it just meant that the cheesemaker managed the process in a way that allowed her successfully to tuck up the curd and turn to more pressing matters until the time was found to pick it up again. Even the ministrations of Joseph Harding and his systematization of Cheddar did not succeed in getting every farm to produce 184

•

FAMILIES AND FACTORIES

great cheese. Some dairymaids or farmers’ wives were just better cheesemakers than others; they had a more intuitive “feel” for the process or simply cared more about getting it right. For these reasons, Willard dismissed Harding’s approach to improving the general standard of cheese at the farm level: “Whatever progress can be made towards improvement will naturally develop itself more rapidly [in the factory] than among persons scattered over a broad extent of country, and who are so occupied with a variety of work as to have little time to spend in the improvement of any one particular branch.”28 The factory was safely isolated from the domestic world; it was a space devoid of escaped livestock, crying babies, or supper to prepare. Factory cheesemakers were professionals, and they were specialists, something Willard readily acknowledged: “The managers employed must exhibit high skill in manufacturing, and they make cheese-making a study, and adopt it as a profession. We pay high wages for skill, and this induces manufacturers to great exertions for success.”29 Under their watch, cheesemaking was transformed from a flexible and adaptable method into a set of rigid rules, measurements, and timings. Women making cheese on the farm could be taught better methods and given better tools, but it was only when cheesemaking was removed from the domestic sphere that it was finally brought under the discipline of the clock. The spread of the factory system was not always smooth. Willard recognized that the requirement to transport and pool milk was an Achilles heel: “There is nothing, perhaps, which indicates the progress and skill of our manufacturers more than the fact that they are able to take imperfect milk from the hands of patrons, manipulate it among the fetid odors of whey slops and decomposed milk, and yet turn out cheese that will compete with the great bulk of English make. But these conditions will not and cannot produce the fine, delicate flavor of the best Cheddar.”30 Combining milk from many different farms meant that the overall quality of the cheese was dependent on the lowest common denominator, the farmer who couldn’t be bothered to wash his churns properly or who milked his sick cows in with the rest. Working collectively, even on a small scale and as part of a jointly owned concern, required discipline. While it saved the farmers’ wives the trouble of cheesemaking, hauling milk to the factory twice a day was a drag. On joining an associated dairy, many farmers complained that the cheese could be “made by the family with but little more trouble and labor than that of carting the milk.”31 For many, relieving family members of the cheesemaking hardly seemed worth the FAMILIES AND FACTORIES

•

185

effort: “The cheese being made either by the farmer’s wife or one or more of his daughters, he considers that the labor costs him nothing; that the work is a duty in the former case, and a wholesome discipline in the latter.”32 The factories waited for no one. The fiercest tension between the factories and the dairy farmers was over “the necessity of keeping regular hours for [milk] delivery under all circumstances of weather, &c., since no delay can be made at the factory for the milk of a single dairy without hazarding the acidity of a large quantity . . . besides deranging in some degree the regular factory-work.”33 The natural and inherently flexible rhythm of work on the farm had to be brought into line with the clock at the factory down the road. When deliveries were late, milk soured prematurely, vats were lost, and penalties were inflicted. The factory apparatus possessed a social power that reached far beyond the building where the cheese was made. The efficiency of centralized production, the comparatively cheap price of American land, and inexpensive transatlantic shipping meant that by the mid-1860s, American factory cheeses were appearing on shop counters in England at almost half the price of the locally produced farmhouse cheeses. By the admission of all involved, factory Cheddar did not match the quality of the very best farmhouse cheese, but as it got better and stayed cheaper, the objections became progressively quieter. Crucially, when given the choice, customers on both sides of the Atlantic bought factory cheese, finding better value for money in the consistency of the cheap, mass-produced product than the vicissitudes of the farmhouse version. In America starting in the 1850s, farmhouse cheesemaking operations had been shoved to the margins in the clamor to join and build associated dairies. The dominance of factory cheese in foreign markets followed soon after. In 1874, 25 percent of the cheese consumed in England was imported factory cheese; by 1924, this figure rose to over 75 percent.34 Willard’s prophecy had come to pass. The cheese factory reigned supreme.

T H E C U R D V E R S U S T H E C LO C K

Harding’s improved farmhouse Cheddar required a full day to make. If the weather was cool or the milk’s natural acidification sluggish, cheesemakers often had to wait late into the night for the curd to be ready to mold.35 After the curds were in the molds, they still had to be pressed for several days before the cheese was ready to carry to the ripening room. Despite the economies of 186

•

FAMILIES AND FACTORIES

labor and scale, factory production in the late nineteenth century was based on the same general principles and was almost as cumbersome. Factory cheesemakers recognized that increasing efficiency was key to their continued success. Working in conjunction with agricultural research stations and the US Bureau of Animal Industry, they explored ways to speed the process and boost their yields. Adding artificial acids to speed milk ripening, boosting moisture content by soaking the curds in water before salting, adding extra rennet enzymes to hasten the ripening process—these techniques and more were tried and tested. Perhaps not surprisingly, they produced some terrible results: abnormal gassy fermentations, bulging cheeses, and sometimes, cheeses that remained rock hard and refused to ripen at all.36 One set of experiments was remarkable for its success, however. Two agricultural chemists, Stephen Babcock and Lucius van Slyke, based at the US Department of Agriculture’s experimental stations in Wisconsin and New York, respectively, demonstrated in a series of experiments carried out in 1902/3 that simply keeping cheese cold as it matured could reduce weight loss by up to two-thirds.37 Sealing the rind in paraffin wax decreased losses even further, boosting profits both by preventing moisture loss and by blocking mold growth and attack by mites. When properly made from excellent milk, cheese responds well to aging at cool ambient temperatures. Accordingly, dairy farms in Somerset kept their cheeses not in underground cellars but in well-ventilated attics. But these conditions will also accentuate the slightest imbalance, sending a wobbly-flavored cheese straight off the rails. By keeping cheese much colder, Babcock and Van Slyke found that they could delay the development of any off-flavors and extend the potential selling window, helping to soothe the inevitable disjunctions between supply and demand. It was a safety-first approach, but it suited the market well enough and built on the logic of Willard’s drive for consistency above all else. Midtwentieth-century descriptions of the cheese market dwelled on the tastes of housewives, who preferred “a mild flavor and economical spreadable texture in place of the firm-bodied drier cheese of pungent taste.”38 Whether those housewives preferred the smooth, dull cheese because of its flavor or because it was cheap is unclear. Was it simply a convenient coincidence that, as cheese technologist and historian Val Cheke noted in 1959, “the physical properties of so much current cheese . . . largely correlate with modern methods of manufacture, subsequent storage at low temperatures, and . . . marketing when immature”?39 FAMILIES AND FACTORIES

•

1 87

No cheese better exemplifies this change in mentality than Lancashire. These days, two different cheeses bear the name of the county, but they might as well be from different planets. The first style, the one that dominates the market, is known as Crumbly Lancashire. Two respected Lancashire creameries, Greenfields and Singleton’s, both claim to have invented this style in the 1950s. In Singleton’s telling of the creation myth, the new crumbly cheese was a commercial miracle that ensured the postwar survival of the dairy. Alan Riding, who in the early 1950s had only recently taken control of the company, was struggling with his cash flow. Milk had to be paid for on the sixteenth of every following month, but traditional Lancashire cheese took a good two months to mature. The business was hemorrhaging money until a cheese factor asked for an acidic, crumbly cheese to be sold in East Lancashire. It only required a couple of weeks to mature before it could be sold, so suddenly, money was coming in before the bills had to be paid. Cash flow and cost controls, rather than taste or tradition, have been the defining feature of Crumbly Lancashire from the very beginning. It is a style of cheese that is loathed throughout the industry. In 1982, a cheese grader at Singleton’s, speaking within living memory of Crumbly Lancashire’s creation, confessed: “It’s a bastard cheese; they call it sawdust in the South; I don’t like to see any on my shelves after two weeks.”40 He was not forthcoming about why his own company proudly claimed to be the originators of the style.

P R O PAG A N DA O F T H E C H E E S E

Back at Beesley Farm, we have finished tasting, and Graham is making tea. As it brews, he proffers some of his mother’s famous fruitcake. The cake is dense and moist, ideal for a wet Lancastrian afternoon, and we sit and gossip. Before Graham built the new creamery nine years ago, all of this conversation would have taken place at the farmhouse kitchen table, and the creamery break room where we now sit retains some of the sense of cozy warmth despite its utilitarian design. That Graham can afford to take a leisurely break is no fluke: it is an inherent characteristic of the cheese that he makes and underlines how his cheese subverts the modern world. The British Marxist historian E. P. Thompson wrote, in the context of the experience of industrialization, “In mature capitalist society all time must be consumed, marketed, put to use; it is offensive for the labor force merely to ‘pass the time.’ ”41 As an exercise in time and motion, the sitting and chatting appear grossly ineffi188

•

FAMILIES AND FACTORIES

cient—surely there is productive work to be done! But the Kirkhams are conducting their own rebellion. If it is not quite propaganda by the deed, then it is at least propaganda by the cheese. For most cheesemakers, of whatever scale, the clock looms large. Clocks structure the day, defining a typical eight-hour shift and letting people know when they can go home. On a more granular level, many cheesemakers choose to wear a stopwatch around their neck as they work with their cheese, recording in laborious detail the exact timings of each interaction with the curd. When cheesemakers are troubleshooting problems or being inspected by public health officials, these timings always receive close scrutiny. Not so the Kirkhams. Where other cheesemakers obey a stopwatch and precisely time each of their actions, Graham takes his feedback from the curd. There is a clock in the make room, but it sits propped against the window ledge; in his nine years in the new creamery, Graham has still not got around to putting it up on the wall. Of all modern Lancashire producers, Graham alone makes raw-milk cheese from the curd from two separate days. The entire process, from milk to bandaged cheese, takes three days. The result is a cheese that retains the fluff y texture of the milled curds but also breaks down into mellow buttery softness. It makes for a tender and compelling cheese, but it takes time. You can’t hurry real Lancashire. The Kirkhams do not make a crumbly cheese. They never have and never will. Graham adds less than a hundredth of the dose of lactic acid bacteria that is used in industrial cheeses, such a tiny amount that he struggles to meet his supplier’s minimum order requirement. The acidity development is barely measurable for the first eight hours or so of the make, by which time the team have had the chance to cut the curd, drain it, stack it in the vat, transfer it to the cooling table, press it and break it up several times in succession, and have a long and leisurely lunch in between. Being family helps. When we visit, a longtime and stalwart employee has just left after over a decade with the Kirkhams—Graham is still having some surprises in the process of discovering just how much of the business side of the operation she quietly kept on an even keel—but on balance, he thinks that it is good for the cheese. With an employee, there is always the pressure to get the make done on time for the end of the shift, whether the cheese is ready or not. Now, he can be more leisurely, and he already feels the benefit; he has pushed milling back to the afternoons, and it seems to be working well. The cheesemaking takes a long time—in total, the working day of the farm runs from 5:30 a.m. until 9:30 p.m.—but it is not all spent slaving in front of FAMILIES AND FACTORIES

•

189

the cheese. Unlike most conventional paid jobs, which are regulated by the clock, there is flexibility for family life. Graham remembers his mother, Ruth, adding the rennet to the milk before bundling him and his sisters into the car for the school run; the curd was ready to cut when she returned. Ruth herself said: “It’s a job where you can set your own pace, that lets you be with your children when they’re growing up, and which you can stop and go back to.” That in itself perfectly illustrates the difference in mentality between the Kirkhams’ way of doing things and any factory operation. But although large factory cheesemakers might rely on extensive automation, their Cheddarmaster machines (capable of producing twelve tons of cheese an hour; in six hours, one machine can match the Kirkhams’ annual production), and their robots to make their production efficient, we have encountered exactly the same time-driven mentality within systems operating with nothing more sophisticated than a curd ladle. The curd of French AOP Camembert de Normandie must be ladled into molds in order to satisfy the appellation rules, and the best cheeses are ladled by hand. With Gallic pragmatism, at the Gillot factory this is achieved by an army of men and women who glide across the production floor pushing tubs of curd on wheels. Each ladler makes a circuit of her table, scoop clicking like a metronome against the edges of the molds. Round and round they go in perfect synchrony; after five passes, the tubs are empty, and it is time for a five-minute toilet break before the next tubs of curd are cued up, waiting for no one. The hierarchy of the factory runs from the clock down to the curd—which is carefully managed with inputs geared toward producing consistent, predictable outputs— and then down to the individual employees. With ruthless discipline, the day of each ladler is controlled down to the minute. As we take our leave of Beesley Farm and drive down the lane past the signs advertising the county of Lancashire’s industrial heritage to tourists, we cannot help but think about the Kirkhams and how their cheese embodies an alternative model. It is true that Graham has not avoided specialization. His parents dabbled with poultry farming before deciding to concentrate on the cheese, but Graham, although he has experimented with raising some of his bull calves for rose veal, remains very much the dairy farmer. With the exception of a short break at Christmas, all of his milk goes into his cheese. He is also every bit as market orientated as his nineteenth-century predecessors, it is just that his market is spread over more of the world: Graham is the only farmer in his little village of Goosnargh whose name is emblazoned across cheese counters from San Francisco to Hong Kong. Now his sons are 190

•

FAMILIES AND FACTORIES

learning the business too, and things are looking up. But it is the cheese that is key. Try and push Lancashire, bully it into the convenience of an eighthour shift or use it as the dumping ground for indifferent milk, and the end product is corrupted. There is simply nowhere to hide. This gentle, delicate cheese that evolved to feed the world’s first industrial workers cannot itself be industrialized without being changed beyond recognition. Turning to join the main road, we realize that it is the cheese, not Graham, that is subversive of modern life.

FAMILIES AND FACTORIES

•

191

TEN

Expertise

The Loire Valley belongs in a fairy tale. It is a region of picturesque medievalism, marked by aristocratic châteaux and the flying buttresses of the cathedrals at Chartres and Le Mans. This sense of timeless tradition immediately comes to mind when we visit Vincent Peltier and his wife, Mary Agnès, at their farm in Bossay-sur-Claise. Striding at the head of his herd of 130 milkywhite Saanen goats, Vincent is the very image of the grizzled French peasant. With his bushy beard, he has a touch of swords-and-sorcery epic about him: squint hard enough and it is not difficult to imagine a leather jerkin and coat of chain mail. Here, in the western center of France, we are in goat-dairying territory. Indeed, this is the home of some of the most widely imitated styles of cheese in the world. With their bright citrus flavors, the lactic goat cheeses of the region define what most of the world thinks of as goat’s milk cheese, to the extent that they are often marketed simply as chèvre. The Peltiers make ten different cheeses, but almost all of them are derived from the same supple lactic curd. Most of the names are simple statements of the different shapes of the cheeses: there is le Touré, le Pavé, and le Button. But taking pride of place is the AOP Sainte-Maure de Touraine, the definitive ashed goat’s log. From a technical perspective, the cheeses are an exercise in spectacular virtuosity. Their production involves a long lactic coagulation that makes use of a continuous whey starter, and a delicate dusting of Geotrichum candidum on the rinds. As the young cheeses are turned out of their molds, the whole process seems effortless. Surely, this must be the result of countless generations’ accumulation of knowledge? Joan of Arc led the French army that drove the English from the Loire in the fi fteenth century, and as we watch the Peltiers work, we joke that we would not be at all surprised if they were 1 92

to point to a relation who had served as personal cheesemaker to the Maid of Orléans. Remembering what became of her, the other English members of our party shift awkwardly and nervously remark that, in that case, they hope that the Peltiers do not hold a grudge. Eventually, we pluck up the courage to ask Vincent. “Ah, no,” he laughs. “I used to be a physical education teacher at the local high school, but we fancied a change of pace.” The Peltiers, it transpires, have only been making cheese since 1999.

A C A L I F O R N I A N CO N T R A S T

The Peltiers are very much on our mind as we visit Soyoung Scanlan at her Andante Dairy, just outside Petaluma in Sonoma County, California, five and a half thousand miles away from Bossay-sur-Claise. The landscape could not be more unlike that of the Loire. In place of the lush greens of France, the parched Northern California earth is yellow and cracked, baked under the August sun. Each farmer we have visited on this trip has spoken at length about the challenges of the drought and their hopes for a wet El Niño winter. Like the Peltiers, Scanlan specializes in lactic goat cheese, and she has been making cheese for about the same length of time. She is, however, definitely not a rustic ex-gym teacher. From an academic family in South Korea, Scanlan found her way to cheesemaking via the biological sciences and a love of music. After completing a master’s degree in biochemistry in South Korea, she came to the United States in 1993 to start a PhD program. Within a fortnight, she had met her future husband, James Scanlan, at an Itzhak Perlman concert at Boston Symphony Hall. She married and moved to the San Francisco Bay Area, leaving her doctoral work, but not her scientific curiosity, behind. A combination of inspiration from European vacations and food magazines brought cheese to her attention, and she learned that California Polytechnic University in San Luis Obispo offered short cheesemaking courses. Never one to believe in half measures, she immediately signed up for the university’s master’s program in dairy technology. Two years later, strained by the distance between San Luis Obispo and the Bay Area, she left her postgraduate studies and took the first tentative steps toward her own cheese production. Diminutive and self-possessed, Scanlan shows us around her creamery as her young daughter assiduously wraps cheeses for a big order. Scanlan watches the girl’s careful work proudly: “I raised Jamie at the dairy. . . . Everybody EXPERTISE

•

1 93

wants to write about me, but I’m a boring person, just ask her!” She has a fierce pride for the dairy, and it is this same spirit and self-belief that saw her through the fi fteen inspections necessary to get all of the required permits for cheese production. As a trained scientist, Scanlan takes a broad view of the discipline, looking more toward Aristotle than Pasteur for inspiration: “There is so much pseudoscience going on, but I don’t say anything. I can test variables, but I am more in that ancient Greek category of scientist: they were philosophers, musicians, scientists all at the same time. I think that our problem is [intellectual] segregation.” She brings out a cheese to taste. She is a talented musician—there is a piano in the creamery’s office—and all of her cheeses take their names from music. There is Acapella, Nocturne, Picolo, and Pianoforte. All cheesemakers have a close relationship with their cheeses, but for Scanlan it is especially close: they are a public projection of her personality and sense of self, of her own passions and interests. She acknowledges that this can seem strange to outsiders: “Cheesemakers are very headstrong; it’s hard work, they are a little bit crazy.” As we work our way through her various lactic goat cheeses—the cheese wrapped in fig leaves has an astonishing depth of flavor—we ask about her experience of cheese education in the United States. It is a difficult question, and one that Scanlan immediately expands to include the rest of the Anglo-Saxon world: “There is a great passion for amateurism, maybe from your [British] side. . . . There is a romantic myth of it.” That people benefit financially from this myth is something that troubles her deeply: “Because of that, there is not much of an intellectual background, [it’s] more of a beauty industry. . . . [There is] a tradition of lack of self-evaluation [and instead a focus on] ways to package it beautifully.” For Scanlan, all of the years spent on her education are no match for the knowledge that comes from hands-on experience. She works hundred-hour weeks and restricts her wholesale customers to those she feels will best look after her cheese: 60 percent of her production goes to chef Thomas Keller of the French Laundry and his eponymous restaurant group, which includes his three-Michelin-starred temples of haute gastronomy in the Napa Valley and New York City. For Scanlan, cheesemaking is above all else the act of communing with the milk, something that transcends science: “You smell it, and you can touch it. I can actually touch, taste, smell [the milk, and it] will tell me that the pH meter can be wrong and that my senses are more acute. . . . The science is the tool. Scientific rigor is a myth, and people follow it because they are intellectually lazy.” 194

•

EXPERTISE

The contrast between Scanlan and the Peltiers says everything about their respective routes to cheesemaking. Through a course at the local technical college and the support of local consultants, the Peltiers were able to achieve a degree of technical proficiency that allows them to perform brilliant cheesemaking with nonchalance. Quite frankly, they make cheesemaking look easy. The Peltiers deserve their success—they are highly skilled cheesemakers—but the ease with which they made the transition from neophytes to artisans highlights an important truth about expertise: structures matter. When we tell the story of a cheese, we are readily seduced by the Great Man theory of history. It is true that for every nascent region, there is the rugged and heroic band of early adopters, the outsiders willing to invest their money and time in a dream. These are compelling stories, but they are narratives of struggle and separation, of the sheer bloody-mindedness necessary to succeed against all odds. By definition, they require atypical individuals; the path is difficult. We see Scanlan’s ferocious self-belief hidden underneath her gentle demeanor. Our ready willingness—even our desire—to believe that the Peltiers are channeling generations of folk wisdom reflects a similar romantic impulse. Amid the medieval heritage of the Loire Valley, we want the cheesemakers to be dairying Joans of Arc, the recipients of direct instructions from God, performing heroic rites of cheesemaking as revealed truth. But just as wars are won by big battalions and not by individual teenaged girls, so we need to strip divine inspiration and tests of character out of the route to expertise. How do normal people learn to make great cheese?

E CO LO G I E S O F E V E RY T H I N G

Ecology is the universal dairy metaphor. Just as we describe the interactions of dairy animals, their pastures, and the microbiology of their raw milk as a coherent ecological system, so anthropologist Professor Harry West has characterized the relationship between cheesemaker, milk, and curd as an organic dialogue. In language that combines anthropology with poetry, he describes how the skilled practitioner resonates with their environment, how the artisan enters into a conversation with their surroundings. Working with the Belontes, a family of Saint-Nectaire producers in the Auvergne, West makes the case for the inadequacy of the linguistic vocabulary for communicating the cheesemaking process. Terms like “dry,” “fragile,” or “hard” have little utility for the apprentice cheesemaker. Instead, it is the cheese and its behavior—how the EXPERTISE

•

195

curd drains, what it looks and feels like—that provide instruction. Effectively, the cheese teaches its own production.1 That is a poetic ideal of artisanship and mastery to which most cheesemakers aspire, but our own intentions are more prosaic. We want to understand how and why some regions are so successful at communicating and fostering cheesemaking skills. After all, all cheesemakers have regular access to milk, curd, and cheese, yet the cheese seems to speak more clearly to some than to others. Why is it easy for the Peltiers or the Belontes to learn to make cheese? We could put this all down to the individual character of the cheesemaker, to their disposition and God-given skill. But charming hosts that they are, the Peltiers are no more virtuous than any number of other producers. No, if we are to attempt to explain why they prosper, we need to flesh out the ecology of knowledge beyond the cheese vat, beyond even the institutional infrastructure—the schools and formal training—through which knowledge is reproduced. We need to examine the wider evolution of the ecological system in which knowledge about cheese and cheesemaking interacts with both the structure of the industry and the cheese itself. Above all, the system is dynamic: change one dimension, and the whole is transformed. The arrival of cheap American factory cheese had a dramatic impact on the nineteenth-century British dairy industry. In the face of this competition, in 1891, the Bath and West and Southern Counties Society—a confederation of agriculturalists that survives today as the Royal Bath and West of England Society—initiated the first great attempt to bring British cheese into a scientific domain of knowledge. Elementary technical proficiency was the secret behind the New World cheeses, and so the progressive response was to bring in a scientist. At thirty-eight years old, Dr. F. J. Lloyd was a consultant chemist to the Royal Agricultural Society of London, part of a small office of technical troubleshooters applying the discipline of academic science to improve British agriculture. His mission for the next five years was to “formulat[e] a complete scheme of investigation of the science . . . which underlies the existing practice of the best cheese-makers.”2 In other words, he was there to “solve” Cheddar. In order to provide practical context for his experimentation, he was invited to work with the school for Cheddar makers run by the Bath and West and Southern Counties Society. Each season, the school took up residence at a different farm in the region, and the cheesemaking there was given over to a dairy instructor. The course ran seven days a week over six months, from April to October. Students—generally the wives, sons, or daughters of 196

•

EXPERTISE

dairy farmers—could sign up for a complete four-week residential course or dip in for a week or even a day. Teaching at the school was Edith Cannon, a twenty-two-year-old prodigy whose Cheddar had been named supreme champion at Somerset’s premier agricultural show when she was just nineteen. It was an era of transition. Joseph Harding’s innovations were working their way down to the farmhouse level, and competing methodologies jostled formally to be approved as the “correct” method for Cheddar making. Among these, the society deemed the Cannon method, developed by Cannon’s father, Henry, also a teacher of cheesemaking, the most promising both for further study and for adoption by the Somerset cheesemakers attending the school. Rival approaches, such as the Candy method and the Scotch method, offer a tantalizing glimpse of the lost potential for diversity within the county of Somerset, where each village or even farm had its own techniques. We live now with the technical bottlenecking that is a consequence of this nineteenth-century codification. Shepherding milk through the process to become a proper Cheddar curd required cheesemakers largely to use sensory inputs—appearance, touch, taste, and smell—to inform their decision making. These Cheddar makers were artisans to delight a twenty-first century anthropologist, but even they backed up their intuition by measuring temperature and performing a few crude tests using everyday materials that could be found on a dairy farm. It was hardly cutting-edge technology. The “rennet test” for acidity required cheesemakers to time the action of a large dose of rennet on a cup full of milk, while the “hot iron test” involved pressing a bit of curd onto a heated metal bar and looking at how it stretched as it melted. Far from dismissing her skills, Lloyd showered praise on Cannon’s exquisite expertise and the astonishing sensitivity of her palate. Her ability to produce first-class cheese day in and day out simply by responding to the information provided by her senses was akin to the prodigal skill of a genius like Mozart. If they had been based in the Loire Valley, Lloyd would no doubt have referenced Joan of Arc and divine intervention. But what hope was there for the cheesemaker who lacked such superhuman virtues? Lloyd hypothesized: “Evidently to overcome this natural inaptitude of estimating the various stages in the progress of the curd, it would be necessary to substitute some means of determining them, which would not depend on individual capacity.” As his guide, Lloyd planned to use Cannon’s “remarkable ability at every stage” and “her exceptionally keen senses of taste and touch EXPERTISE

•

1 97

and smell,” revealing through his measurements the chemical conditions within the curd that she alone was capable of deciphering with her senses.3 Lloyd brought with him from London the equipment of a professional chemist, particularly an acidimeter, a titration apparatus already widely employed by chemists of the day. In rural Somerset, however, it was foreign, exotic, sophisticated, and fascinating. The instructress and her students alike enjoyed learning to use it and witnessing the development of the pink color of the indicator in the milk and whey samples. “Given the necessary solutions and the apparatus, the determination of acidity is very simple,” Lloyd wrote, and by the second year of the study, he had acquired an acidimeter for the use of the school, “such as I hope may subsequently be found capable of introduction into cheese dairies generally.”4 Diligently setting out to reveal the quantitative basis of successful Cheddar making, Lloyd began by measuring every conceivable variable. Over the first two years, he generated reams of records. The minute details of hundreds of batches—including the names of the fields where the cows grazed, the volumes of milk and whey, and the temperatures, acidities, timings, and yields— were recorded in fi ft y-six different columns in his record book. Within those first few months, which Lloyd spent simply measuring, patterns began to emerge that showed—as he had expected—that the progression of acidity development was important to the behavior of the curd and that Cannon was intuitively steering the process through a series of fairly consistent measurable milestones. By September of the first year, Lloyd had gathered enough information that he felt ready to progress to his first and only experiment that year: “I determined to make a cheese myself without touching, tasting, or smelling the curd from commencement to end, and to be guided at these stages by the determinations of acidity alone. . . . The mechanical operations, which of course require skill and experience that I do not possess, were kindly performed for me by Miss Cannon; but she offered no opinion during the making, so that I might not be in the least biased.” Lo and behold, when it came time to sell the cheeses, the experimental batch was pronounced excellent. Lloyd reported that his cheese, made on September 10, 1891, was “probably the first cheese ever made with scientific standards whereby to estimate the critical stages in its manufacture.”5 Lloyd’s ambition was to develop a foolproof method, to unravel the feminine mystery of the curd and lay it bare in the form of cold, hard numbers. But by dismissing soft data gathered through the senses as dependent on 198

•

EXPERTISE

extraordinary talents, he negated the possibility that those sensory judgments might be teachable. Lloyd did not acknowledge the fact that Cannon, after all, had had plenty of careful guidance from her father and years of daily practice, and what might have appeared to an external observer to be sheer natural genius was the result of driven parents, serious determination, and systematic repetition. Practice makes perfect. No doubt Cannon’s sense of smell was fairly acute, but her ability to make sense of and act on those sensory inputs was entirely the product of her vast experience. Nonetheless, as a way to teach the unskilled to make a passable cheese with minimal fuss, Lloyd’s approach had a certain charm. A world where acidity held all the answers was much simpler to manage. Unfortunately, understanding Cheddar entirely through acidification was not so clear-cut as it first seemed. The society had decided that in order to dispel common notions that some lands could not produce good cheese, the school would be hosted by a different farm each year—and the more problematic the farm, the better. As the school moved from farm to farm, Lloyd began to realize that the readings of the acidimeter did not tell the whole story. “Evidently what held good in 1891 did not hold good in 1892,” he reported, perplexed, at the end of the second year.6 While the acidity readings were hardly comparable between farms, Cannon succeeded in producing a season’s worth of top-quality cheese at each, against all odds. Lloyd could only surmise that other factors were affecting the outcome: perhaps the richness of the milk, or maybe the amount of calcium in the soil. “One thing seems certain,” he wrote, “that the quality of the milk affects most seriously the amount of acidity which may with advantage be obtained in the curd before grinding. . . . While I state fully the facts that have been obtained . . . I do not feel in a position to draw conclusive and practical deductions therefrom.”7 But Lloyd was soon diverted from the ambiguity of the acidity measurements by a shiny new preoccupation: bacteriology. This new field, resonant with discoveries about disease and the universal problems of mankind, was far more intriguing than the results of a simple acidity meter. Tracing the causes of common off-flavors, or “taints,” to bacterial contaminants promised to improve the quality of cheese as much as any map of acidification. Lloyd’s further finding that cheese made from milk prone to off-flavors could be improved by focusing on achieving high levels of acidity during the make was enough to reinforce his foregone conclusion. As far as Lloyd was concerned, the key to good Cheddar making was consistent development of acidity. His EXPERTISE

•

199

urge to develop a more nuanced theory to explain the difference in the acidification profi les of successful cheeses made at different farms dissipated and was eventually forgotten. Even at the end of his six years of study, Lloyd had not satisfactorily solved the puzzle, and he wrote it off as a footnote. The next era for research—and the preoccupation of the industry—was to be biological rather than chemical. Scientific supply companies began to commercialize acidimeters for cheesemaking as early as 1894, no doubt as a result of students from the school returning home with an urge to apply this up-to-date scientific knowledge themselves.8 And while it’s known today in most dairies simply as “checking titratable acidity,” Lloyd’s Caustic Soda Test is still a standard method—a cheap, simple test that provides cheesemakers with a straightforward indication of the progression of acidity development during their makes.

REVENGE OF THE NIRD

The industry that received Lloyd’s research was in steep decline. British dairying was undergoing a profound structural change in the response to cheap imported Cheddar and the ready domestic market for liquid milk.9 The 1908 Census of Production revealed that cheese production on British farms exceeded that of factories by a factor of almost seven to one. Less than thirty years later, in 1935, that ratio was almost exactly reversed. At the same time, cheese as a proportion of English dairy output had dropped from 40 percent in 1860 down to a mere 5 percent by 1930.10 Dairy farming, and particularly the production of liquid milk, had become a huge industry and was the only form of agriculture that actually achieved growth during the long agricultural depression that lasted from 1870 to 1940.11 English farmhouse cheese was quickly going extinct, but the liquid milk market was thriving, and its spirit was progressive. In this spirit, the Board of Agriculture made another call to science. The National Institute for Research in Dairying, founded in 1912 in Reading, both was shaped by and helped to institutionalize Lloyd’s thinking. Indeed, in 1950, the then director, H. D. Kay, self-consciously claimed Lloyd as a forbear, describing him as one of the “very few experimentalists who concerned themselves with dairy problems” before the founding of the institute.12 Universally known by its acronym, NIRD, the institute was not much interested in cheesemaking. From the beginning, it was plunged almost immediately into 200

•

EXPERTISE

World War I, where it was at the forefront of British agriculture’s response to the challenges of feeding the nation in wartime. As such, NIRD devoted its resources to figuring out how to increase the efficiency of production and distribution, as well as to convincing the industry of the “fact that clean milk production turned largely upon the . . . almost surgical sterility of everything coming in contact with the milk.”13 With the 1930s came a series of experiments on the nutritional effects of various methods of processing—pasteurization, spray-drying, and the like— with the result that “no member of the Institute staff subsequently extolled raw milk to the detriment of milk rendered safe by pasteurization.”14 The pressures placed on the food system during World War II resulted in NIRD turning its attention to the development of synthetic hormones to boost milk production and even to beget artificial lactations from cows without pregnancy.15 NIRD epitomized its era’s vision of progress: it was clean, efficient, modern, and scientific. The research conducted there shaped the conventions of the modern industry while farmhouse cheese production gradually went extinct. Val Cheke, a lecturer and instructor at the British Dairy Institute, also in Reading, was born in 1902 and was herself a product of that era of change. Her first teaching job was in 1923, but she commented with approval on the students who entered the industry after 1945: “From the colleges emerged a new type of agricultural personnel . . . a budding scientist. . . . Scientific cheese-making was a suitable outlet for many of the college [graduates] who found themselves in farm or factory. . . . There ensued record trade orders for acidimeters, centrifuges, hygrometers, thermometers and other equipment for chemical and microbiological testing.”16 In this brave new world, the highly credentialed technical manager of a creamery would not be caught dead with the tools of a previous generation— such as an old rennet cup or a hot iron—in his or her dairy. For these graduate food scientists, the acidimeter was the instrument that represented progressive practice and robust scientific knowledge. The entire ecology of cheese production had changed, from the economic structure of the industry to the educational institutions to the form of knowledge that was regarded as most accurate and appealing. Acidity measured in degrees Dornic was the route to understanding cheese, not the intuition of a Cheddar instructress and her homely tests. For British territorial cheeses, indeed for Cheddars everywhere, this is the world we now live in. We eat F. J. Lloyd’s Cheddar, not Edith Cannon’s. EXPERTISE

•

2 01

T H E A R T O F WO M E N A N D T H E B U S I N E S S O F M E N ?

The working relationship between Cannon and Lloyd was almost too good to be true: it makes the perfect parable for the relationship between art and science, intuition and reason, women and men. The young cheese prodigy and the older scientist working long hours together in the creamery is the stuff of romantic fiction, although there is absolutely no evidence that their relationship was anything but professional. What is clear is that by 1896, things had soured somewhat. In place of Lloyd’s eff usive praise for Cannon’s palate and cheesemaking skill, his final report barely makes mention of his longtime assistant and Cheddar whisperer. Whatever might have happened between them, the Lloyd-Cannon collaboration taps into an established discourse on women’s work within the dairy industry and its transformation over the course of the nineteenth century. It was an international debate. We know that in Canada and the United States, where factory cheesemaking was swiftly adopted after 1860, the transformation of the industry rapidly removed women from cheesemaking; industrialization was even something that women themselves promoted as a solution to the excessive physical demands of cheesemaking. In the United States, as early as the 1850s, complaints from women about the excessive labor of dairying were appearing in the agricultural press: educated farm daughters increasingly found less physically arduous, more prestigious work as teachers within the rapidly developing public school system.17 The Danish creameries of the 1870s that gave rise to the development of starter cultures in butter and cheesemaking were managed by men; the women who had previously dominated the industry were relegated to subordinate positions and suffered a decline in wages.18 Within modern British cheesemaking, it is certainly true that making Cheddar has now become heavily masculinized. Cheddar is no longer the preserve of daughters and wives: the collective shorthand for the tiny farmhouse Cheddar community is the “Cheddar boys,” and the muscular demands of manipulating the curd from a four-thousand-liter vat has shifted cheesemaking to a narrative of athletic struggle. The modern Cheddar make involves brutally hard work—it is a rush to the finish, where speed and industry are paramount skills. Slight women in neatly starched linen aprons have been replaced by burly men in sleeveless T-shirts. But was this hypermasculine world ushered in by F. J. Lloyd’s appropriation of cheesemaking as an arena for the application of scientific knowledge? 2 02

•

EXPERTISE

Professor Deborah Valenze has contended that this shift was complete well before Lloyd was even born. She believes that it was the commercialization of English dairying in the century after 1740 that saw the erosion of women’s role as directors of cheesemaking. In place of the tradition-bound dairymaid, she argues, it was scientifically informed men who shaped the industry from the 1850s onward.19 Male dairy commentators in the late eighteenth century like William Marshall and our friend Josiah Twamley applied Enlightenment principles to improve the manufacture of cheese and butter for commercial sale. According to Valenze, in place of a small-scale household economy came the market-orientated adoption of new techniques. The irascible Twamley, intent on bringing market discipline to remote rural England, is particularly emblematic of this thesis. It is certainly how he would like to think of himself. (In his defense, we should note that Twamley dedicated the second edition of his book, published in 1787, to the very dairywomen he so chastised.)20 However, for all of Twamley’s self-importance, this thesis does not tally with the actual gender divisions of nineteenth-century British cheesemaking, as demonstrated in the relationship between Lloyd, Cannon, and their cheese. Before she even took part in Lloyd’s experiments, Cannon was not just a prize-winning Cheddar prodigy in her own right but also a respected teacher of cheesemaking. This was key. One of the most significant responses to the glut of cheap American cheese flooding into the English market was the establishment during the 1880s of an institutionalized framework of dairy education. The Bath and West and Southern Counties Society that sponsored Lloyd’s research was one part of this network, which included local county councils and organizations like the Royal Agricultural Society of England and the British Dairy Farmers Association.21 The expertise to make cheese was actively taught; in Somerset alone, approximately one thousand women received dairy education from 1908 to 1911.22 This was the system on which Lloyd’s experiments piggybacked. Moreover, in 1896, even before NIRD existed as a research institution, the British Dairy Institute was founded in Reading as a top-tier national center for advanced studies. A formal qualification, the National Diploma in Dairying, was awarded by examination, and from 1896 to 1939, women formed 65 percent of successful candidates at the English examination center.23 Even now, where modern British territorial cheeses can claim a continuity of tradition, they do so through powerful matriarchs of a generation old enough to have been educated in this system: Lucy Appleby of Appleby’s EXPERTISE

•

2 03

Cheshire and Elizabeth Montgomery of Montgomery’s Cheddar both inhabited Edith Cannon’s world. When Val Cheke, herself a lecturer and instructor at the institute, wrote about Cannon in 1959 under Cannon’s married name, Mrs. Sage, it was as a still-living trusted former colleague and inspiration.24 The paradigm shift from Cannon to Lloyd did not happen overnight. For all of its modern ubiquity, the supposedly scientific discipline of Lloyd’s testing did not immediately come to dominate the industry. Well into the 1930s, pupils continued to be taught using the older tests that Lloyd had ignored, using the acidimeter only for occasional backup confirmation. It was not that the chemist’s tool was too conceptually complex or too fiddly to use—far from it. The old, seemingly crude techniques were simply better at giving the cheesemaker relevant information, even if it was not expressed in a recognized scientific unit. Despite its trappings of scientific universality, the acidimeter had—and continues to have—serious blind spots when it comes to gathering information from the depths of the vat. In British styles of cheese, the amount of acid present when rennet is added to the milk is critical. Very slight differences in acidity will have a massive impact on the progress of the make later on, which in many cases cannot be solved by compensatory adjustments by the cheesemaker. In 1917, another Somerset Cheddar instructress, Dora Saker, cautioned her students that “during the time before renneting the number of bacteria present may be increased many thousandfold, but the acidity registered [using an acidimeter] remains stationary.” The old rennet test, on the other hand, was “simple and inexpensive . . . [and] small differences in acidity can readily be detected by it.”25 Instructresses like Saker were confident enough that they knew better than Lloyd that they could happily teach the deficiencies of titrating to measure acidity. As a means to acquire information about the milk in the vat, Saker’s preferred “rennet test” is remarkable for its elegance and simplicity. It works on the basis that chymosin, the enzyme in rennet that causes milk to clot, is much more active in an even slightly more acidic environment. The equipment needed is no more complicated than a small cup and a stopwatch. The tester simply puts milk in the cup and adds a very large known dose of rennet at the moment that she starts the stopwatch, then stirs until the milk suddenly seizes up. The number of seconds that this coagulation takes is inversely proportional to—and exquisitely sensitive to—the extent of acidification of the milk. This test was made even more simple with the development of the 204

•

EXPERTISE

“rennet cup,” nothing more than an enameled tin cup with a hole drilled in the bottom and graduations marked along the side. Rennet is added to the full cup, and for as long as the milk is still liquid, it runs out of the hole in the cup; once it has coagulated, this stops. The time that this coagulation takes— and by deduction the degree of acidity developed in the milk—is readily visible from how far the level of the milk in the cup has gone down. In 1932, Renwick Leitch, a professor of dairying at the West of Scotland College of Agriculture, agreed with Saker about the test’s merits: “Experimental evidence shows that the rennet test is a more correct index of the progress of ripening the cheese milk than the acidimeter test. . . . Frequently no increase in acidity . . . may be recorded after a normal [milk] ripening period, and yet the milk may be ready for renneting. . . . Cheese-makers rely almost wholly on the information provided by the rennet test, though some profess to determine the ripeness of the milk by the sense of smell.”26 Instructresses were not consigned to a ghetto of female institutional wisdom or folk knowledge: they were teaching at the cutting edge of scientific cheesemaking. In the interwar years, dairying was a suitable and potentially lucrative career for a bright young woman, with Farmer and Stockbreeder in 1926 reporting that it was possible for women to “secure good posts at substantial salaries if they have received a sound, practical and scientific training.”27 The irony here is that within the modern cheese industry, there is a tacit acknowledgment that the absolute values obtained for percentage lactic acid through titration have little meaning when transferred from one creamery to the other. Variations in the execution of the test make the numbers as sitespecific as any information from a rennet test. From our own perspective, the frustration is that a test as magnificently useful and sensitive as the rennet test could readily have been adopted as an ISO standard. The strength of coagulant is easily measured in International Milk Clotting Units, and the dimensions of the cup and protocol would have been straightforward to standardize. We have attempted our own improvised domestic attempts at constructing rennet cups, but if they were to be commercialized on a grander scale, it would be a great advance for the world of cheesemaking: compared even to a slick modern pH meter, they are substantially more sensitive and easier to use. While the rennet test measured the acidity in the milk as it fermented, the hot iron test measured acidity development within the dry pieces of curd as they were manipulated and drained. It too was used by cheesemakers well into the twentieth century and also offered significant benefits over titration. Not EXPERTISE

•

205

only is it difficult to squeeze the ten milliliters of whey needed to check the titratable acidity from well-drained curd, but even then, the whey that can be coaxed out does not necessarily reflect the curd’s true condition. In contrast, the hot iron test works directly on the curd itself, as Saker explained: “Acidity has the effect of partially digesting the casein of the curd so that it will ‘draw’ when heated, and the length and fineness of the threads before they break from the curd are reliable indications of acidity. . . . It is viewed with great contempt by some chemists as there seems no adequate reason for it, but cheese of good uniform make is turned out where it is the only test used.”28 This also illustrates the profound problem of Lloyd’s single-variable interpretation of the world of Cheddar. In a British territorial cheese, the art lies in synchronizing the twin processes of fermentation (which develops acidity) and drainage. Lloyd considered drainage something that was impossible to measure; instead, he turned a two-variable problem into one of a single dimension. While this simplified his attempt to deploy quantitative methods, his model is inherently deceptive. To make cheese using Lloyd’s approach is akin to trying to land an airplane with indicated airspeed as your only information: you know that the process will be over once speed reaches zero, but you can only hope that you are on the ground at the same time. In contrast, the hot iron test not only gives the cheesemaker information about the acidity in the curd but also provides a sense of its physical condition: more than anything else, it measures the retained calcium in the curd, returning cheesemaking to a multidimensional proposition. These tests—and the women who taught and used them—were not immediately displaced by white-coated scientists. Indeed, the understanding displayed in these tests and expressed in a fully developed technical vocabulary illustrates that these women were better, more specialized scientists than the gadfly Lloyd. The question becomes not why science displaced artisanship or why men replaced women but rather why a less sophisticated strand of scientific knowledge came to dominate the industry after the 1930s. To answer this, we again have to look no further than the wider changes taking place across British dairying. As we have seen, the enemy of farmhouse cheese production was not the rise of British cheese factories. It was the switch by farmers to the sale of liquid milk that rendered all of the highly specialist knowledge of cheese science suddenly arcane. In this, dairy farmers were simply making a rational business decision. As early as 1894, Sir Gabriel Godney told the Royal Commission into the Depressed Condition of the Agricultural Interest: “I have tried on one of my farms (in North Wiltshire) 206

•

EXPERTISE

adopting the very best process to make Cheddar cheese; we succeeded very well, but after trying it for a couple of years I found I could not make it so profitable, taking the labour into account, as selling milk.”29 This was the conclusion in one of the key centers for Cheddar production, although it was admittedly a county well served by the railways for the transport of milk.30 With the foundation of the Milk Marketing Board in 1933, the shift was complete. The board held the rights to purchase and collect all milk produced on farms in England and Wales, providing a guaranteed price and market. In such an environment, with strong demand (average daily liquid milk consumption increased from 0.4 pints per head in 1933 to 0.69 pints per head in 1949) and a secure and profitable price, it is no wonder that farmers removed themselves from the risk, toil, and expense of cheesemaking.31 Farmhouse cheese production suffered a precipitous fall: by 1956, there remained just 140 farmhouse cheesemakers in all of Great Britain.32 As the ecology that supported their skills disappeared, the cheese instructresses ceased to be relevant. Where cheesemaking was but one of many parts of the dairy-processing industry, their knowledge was no longer a vital part of the curriculum. Skills like using the rennet cup and hot iron test and the accumulated expertise that came from decades of hands-on experience were suitable for only a single task: making cheese. They could not help the bright young things of British dairying design flavored yogurts or innovate novel forms of whey processing. Eventually, even NIRD itself became surplus to requirements. It was replaced in 1985 with two new institutions: the Grassland Research Institute in Hurley and the Food Research Institute in Reading. This is the paradox of the industrialization of British cheese: as roles—dairy farmer, processor, wholesaler, and retailer—became ever more clearly defined and specialized, it was a generic training in food science that became the gateway to employment in the industry. The ultimate triumph of Lloyd’s acidimeter was simply thanks to its broad range of applications; the hot iron test is irrelevant to the testing of concentrated fruit juice, and the rennet cup has nothing to say about the safe operation of a canning plant. And those strapping modern cheesemakers in their sleeveless T-shirts? They are simply a function of scale. As herd size increased over the course of the twentieth century, so did the size of the cheese vat. Gradually, with each generation, vats have crept from a manageable 250 liters to the modern standard of 4,000 liters. With the sheer quantity of curd that needs to be moved around, it is no surprise that cheesemakers today are prized for their muscles as much as for their deft skill. In this, there is even a strand of historical EXPERTISE

•

2 07

continuity. In the late eighteenth century, Josiah Twamley noted that “where a large quantity of cheese is made, [a] man is employ’d as an assistant, the weight of a large Cheshire cheese being too great to be wrought by a Woman.”33

T H E LO S T C R I B S H E E T

Today, if we were to ask Cheddar producers on either side of the Atlantic to describe their process, the overwhelming consensus would be that Cheddar is a cheese of rapid acidification, with drainage achieved through heating (“scalding”) the curd and aggressive mechanical action. “Cheddaring” itself refers to the method of flipping and stacking blocks of stiff, rubbery, matted curd so that they stretch and knit together as they acidify and expel whey. Even on the farmhouse scale, making Cheddar is physically punishing. The pace must be maintained to stay ahead of the acidity, which is constantly nipping at the cheesemaker’s heels throughout the make. The best farmhouse Cheddars have a muscular granularity and savor; roast beef and horseradish are typical tasting notes. Behind the savory hit lies a streak of fresh acidity. In this context, it is disorienting to read the descriptions of late Victorian Cheddar, all of which wax lyrical about the cheese’s mellow nuttiness. Modern British farmhouse Cheddars are delicious, but they bear little resemblance to Edith Cannon’s finest. The testimony of Dora Saker exposes the extent to which the cheese has changed over the past hundred years. Written in 1917, Saker’s Practical Cheddar Cheese-Making is as good as its name, a no-nonsense guide for farmhouse Cheddar makers looking to make cheese “of the first quality.” Unlike Lloyd’s reports, which contain lovingly composed pictures of test tubes but not a glimpse of curd or cheese, Saker’s book contains photos of the process itself, taken in the creamery. To the intellectually curious cheesemakers who passed the book around, the pictures were puzzling. Saker’s Cheddar curds looked nothing like the rubbery blocks to which every Cheddar maker was accustomed. Instead, they resembled billowy blankets, draped over the side of the mill like chamois cloth (a contemporary recreation is shown in figure 8). What was going on? When Saker’s book resurfaced after many decades of obscurity, Jamie Montgomery, who makes archetypal Somerset farmhouse Cheddar, showed the strange photos to his mother, Elizabeth. She had spent many years in charge of production at their farm starting in the 1940s but was long since 208

•

EXPERTISE

FIGURE 8. Top, Cheddar curds from a typical “modern” make immediately before milling; bottom, curds from an experimental batch of Cheddar inspired by Dora Saker’s Practical Cheddar CheeseMaking (1917). Photos by Bronwen Percival.

retired. He recalls his shock when, on seeing the picture of the sensuously draped curds, she said, “Oh, of course that’s how they looked.” He shakes his head: “I nearly fell off my chair!” As the ecosystem changed, and as the work of Lloyd and the specialists on starter propagation at NIRD diffused throughout the industry, the remaining farmhouse cheeses quietly adapted and evolved. Nobody set out to change the texture of the curds or to transform a supple cheese with a rounded flavor into the Cheddar we know today. We saw with Lancashire the subtle but profound effect of transferring cheese production from the domestic sphere to the factory and the associated pressures toward greater efficiency and molding the cheesemaking to complement an eight-hour shift. But faster acidification had other advantages to recommend it beyond time discipline. Lloyd himself was delighted to find that more acid curds could cover, to a certain extent, a lack of cleanliness in the milk. In the summer of 1893, while working at a notoriously substandard farm about four miles from Glastonbury, Lloyd and Cannon noticed a change in the milk. There was “a faecal smell, which, though very faint at first, became quite strong before the curd was [molded], and was so unpleasant on some occasions that it made one feel quite sick to be near the curd for any length of time.”34 Lloyd observed that when the make did not achieve enough acidity, the fecal flavor came through in the cheese. On the other hand, if the acidity developed rapidly enough, the contaminating organism was controlled and good cheese could be made. If by some chance “too much acidity were produced, then the cheese, though it would not be a good one, was at least free from any taint.”35 Acidity-focused cheesemaking was conservative cheesemaking. Much later, in the 1960s, as the understanding of microbes and concerns about pathogens progressed, another sort of conservatism set in. Scientists studying Staphylococcus aureus in milk noted that they were poor competitors and that large numbers of other bacteria mitigated the risk of their growing to levels sufficient to produce toxins.36 Rather than lending weight to the important role of healthy raw-milk microbial communities or provoking a discussion about upstream controls on the management of diseased animals to effectively control the presence of staphs in the first place, this finding prompted a different conclusion. Large doses of starter bacteria could serve as competition against S. aureus, and even better, their growth was easily measurable through the development of acidity. Cheddar makes of five and a half to six hours, once completely normal, were redefined as “slow” and therefore dangerous, triggering expensive microbiological testing and other 210

•

EXPERTISE

corrective actions when they occasionally occurred. The focus on speed as a guarantor of safety created an abiding suspicion of slow makes, the depths of which we only discovered when we showed an experimental cheese made over the course of almost eight hours to a technician who had come up in the glory days of NIRD. The cheese was far from perfect; Dora Saker and Edith Cannon would not have been impressed. But this technician’s mind was elsewhere: “I hope they have tested it for staphs!” In order to increase the speed of the make, more starter needed to be added, and this was followed by a scramble to stay ahead of the acidity development as the cheese hurtled toward the finish line. Soon, cheesemakers were finding creative ways to move faster: hotter scalds to drive out moisture more efficiently, shorter cheddaring times, and new mills that rapidly sliced the curd rather than slowly grinding it. The pace of cheesemakers themselves often reflects this pressure. As the curd dries and tightens, it can be a struggle to squeeze out the drops of whey needed for the acidimeter test. There is almost always a furrowing of brows as the pink color develops under titration to show the extent to which the curd is “moving on.” Neither the cheesemaker nor the curd has the time to stretch and relax during the cheddaring process. Over the course of years, mellow bedsheets of curd morphed into springy blocks. Nobody set out to change Cheddar—after all, this is a cheese whose selfdefinition is firmly rooted in traditional practice—but tiny incremental changes in the selective environment, multiplied over decades, have turned a cheese that was once defined by its round, soft, milky flavor into something much drier and more acidic. Lloyd’s philosophy was accepted and carried to its logical conclusion. The Cheddar we know today is the incarnation of that way of understanding cheese. With the recent rediscovery of Saker’s book, the conversation about the nature of Cheddar is evolving too. Some cheesemakers, like Jamie Montgomery, wish to use this new knowledge not to go back to the past but rather to make the best possible versions of their cheeses. Others are experimenting with slower makes and older styles; Tom Calver at Westcombe Dairy—coincidentally next door to the site of Edith Cannon’s father’s farm—is even working on recreating a Cannon make. But while curds that look like bedsheets are simple enough to achieve, moving from a single-variable system based on acidimeter readings to a process reliant on integrating sensory data is a tall order. Learning to taste and react alongside Edith Cannon or the formidable Dora Saker—a structured environment in which EXPERTISE

•

211

to be spoon-fed the “tells” and guided carefully away from common mistakes—is a luxury that disappeared with the last generation of Cheddar instructresses. Faced with reacquiring that knowledge through trial and error, experimenting on vats full of expensive milk and paralyzed by the number of unknowns, one can forgive the modern British farmhouse cheesemaker yearning for a bit of divine intervention.

TERROIR AS CLUSTER

When we are shown around gleaming new INRA laboratories in France or witness the casual technical accomplishment of recently trained cheesemakers like the Peltiers, we cannot help but be jealous. The Britain of F. J. Lloyd, Dora Saker, and Edith Cannon and her father was a Britain committed to generating new cheesemaking knowledge, and it had the institutions and sheer density of cheesemakers to sustain a thriving ecology. The Britain of today is no longer that cheesemaker’s paradise. Whenever an Anglo-Saxon cheesemaker has to rely on the work of a French consultant or apply French government-sponsored research at second hand, it rubs salt in the wound. That infrastructure is something that we once possessed, even within living memory. But when we mentioned to a French affineur of our acquaintance that in 1939, there were 333 producers of farmhouse Cheddar in the southwest of England, his reaction was disbelief.37 To see what a still-vital environment for British territorial cheesemaking might look like, we turn our attention to the north and look beyond the dairy industry to the single greatest modern British territorial product: Scotch whisky. Here is an industry that not only survived its encounter with white-coated, mid-twentieth-century scientists but prospered from it. Even more importantly, it has—at least when compared with cheesemaking—successfully started to reconcile craft and science and to institutionalize the teaching of both. Brian Higgs has recently retired from one of the most significant positions within the entire Scotch whisky industry. A maltster and brewer by training, he came from the English Midlands and eventually became head of the Scotch whisky producing division for Diageo, the world’s largest producer of spirits. His time within the whisky industry perfectly coincides with the decline of British farmhouse cheese production—if he had been a cheesemaker, he would have been one of Val Cheke’s last pupils in Reading—and it 212

•

EXPERTISE

is fascinating to track the divergence of the two industries. Scale played a part. There are barely more than a hundred distilleries in Scotland, but the smallest Diageo distillery, Royal Lochnagar, still produces around half a million liters of whisky a year, making it vastly larger than a small dairy farm operating at barely more than a subsistence level. The products are also inherently different: once bottled, whisky is remarkably resilient, so it did not have to adapt to the new retail environment of the supermarkets as milk did. Even so, the people operating at every level in the industry needed to be trained to make whisky, and the mechanics of the spirit are no less mysterious than fermented curd. While coopers have their own four-year apprenticeship in the craft and blenders depend heavily on the sensitivity of their own sensory apparatus, distillery managers and operators have direct analogs within the cheese industry. At the management level, the modern prerequisite is a degree in chemical engineering or a related scientific discipline, followed by specialist professional training. The same would be true for the manager of a large industrial creamery or other milk-processing facility. It is in the training of the distillery operators that there has been the most change in the past generation. These men and women are the counterparts of cheesemakers at the farmhouse level. In his last quarter century at Diageo, Higgs witnessed a transformation of the distillery operators. What began as a group of people who knew “how to make whisky” from years of on-the-job experience (but were forever concerned that they would feel the wrath of the white-coated scientists from quality control) evolved into a team trained in both the fundamental science of distillation and the craft and soul of whisky. Within Diageo—which itself represents about a quarter of Scottish distilleries—the internal training program developed not only a syllabus of scientific and technical education but also a two-day course that put this knowledge in context. Every distillery operator, office manager, and warehouse person was taught to consider whisky as an aesthetic object, not just a collection of technical numbers. More than the cheese industry, the Scotch whisky industry has been acutely aware of the tension between the science necessary to increase the scale and the craft of the product and the appeal to tradition that creates value in it. This is something that Higgs explicitly articulates: “The years of 10 percent annual compound growth played into the hands of the ‘techies’ with the quick solutions to production increases, and the years coping with ‘whisky lochs’ and flat-lining sales figures helped the business understand the importance of craft, heritage, provenance, and differentiation in markets.” EXPERTISE

•

2 13

Distilling spirits is easy; making a whisky of unique character is hard. In particular, the scientists were tasked with finding out which parts of the process at each distillery are the “pinch points” that give the spirit its unique character. Higgs acknowledges that “in the 1970s, there was a tendency for the distilleries to be pushed as hard as possible to maximize volume output, and this led to some loss of differentiation.” And so, in a market where the value of the whisky comes from its unique attributes, the technical acumen of the company’s research scientists was deployed to identify those factors that made the spirit from each distillery unique, be they related to the fermentation, the still, or the practice of distillation itself. In contrast, after F. J. Lloyd’s initial work, the beleaguered British cheese industry never even thought to commission similar research into the differences between the cheese from different farms. The Scotch whisky industry is a classic example of a cluster, a group of businesses that operate in close proximity, produce similar goods, and compete for market share. Appropriate natural resources are prerequisites to cheesemaking success, but there are dramatic limits to what can be accomplished in isolation, even with every advantage of climate, animals, and soil. Harvard Business School’s Michael E. Porter has made a career out of studying the advantages of clusters for building and maintaining vibrant industries: “The physical proximity of world-class rivals is so common across nations as to hold important insights into the process [and benefits] of competition.”38 The density of technology companies, venture capital, and banking in Silicon Valley constitutes a classic cluster. Rather than harming businesses, Porter argues, competitive pressure provides a selective environment that drives and rewards rapid progress and innovation. Clustering provides other benefits beyond a sense of competitive urgency: it makes way for competition among businesses’ suppliers to provide better raw materials, and it brings with it a local customer base that knows their stuff and rewards the best. “Once a cluster forms,” Porter concludes, “the whole group of industries becomes mutually supporting. Benefits flow forwards, backwards, and horizontally.”39 The most dynamic cheese-producing regions we have visited are taking full advantage of this clustering phenomenon. The Reblochon syndicate’s funding for research on the use of wooden boards has overcome opposition from the local health authorities, and their trainings provide members with the tools and skills to promote the health and microbial diversity of their rinds. Meanwhile, over in the Auvergne, Saint-Nectaire producers participate in government-funded research on the link between field biodiversity 2 14

•

EXPERTISE

and the complex aromas of their cheeses. Not only do clusters provide an impetus for research and development, they also allow for government intervention that works at an ecological level rather than attempting to pick individual winners. Clusters work for cheese. But there is a catch: the cluster must be tight. If the Savoyard producers made many different styles of cheeses, from hard to soft to blue, it would have been impossible to secure a mandate to expend resources on defending the use of wooden boards for Reblochon. The lack of regional clustering in our post-terroir, contemporary, Anglo-Saxon cheesemaking landscape—a land of one thousand cheeses and rugged individuality—subverts our ability to deepen our expertise. American farmers source their know-how from a variety of places. Large producers often have an in-house research and development department, where proprietary methods are developed and tested. Companies that supply raw materials for processing, such as culture houses like Chr. Hansen, invest heavily in educators, like Trish Dawson, who are charged with helping their customers get the best results from their products. Another important source of expertise is land-grant agricultural colleges, which provide teaching, research, and agricultural extension services for farmers at the state level.40 The centers of knowledge and expertise perform the same role as France’s INRA, whose regional outposts employ many of the dairy scientists we have met on our travels.

A CALIFORNIA CLUSTER?

Located halfway between San Francisco and Los Angeles on California’s arid Central Coast, Cal Poly San Luis Obispo is home to the state’s premier dairy science program, the training ground for both Albert Straus and Soyoung Scanlan. The head of the Dairy Science Program, Professor Phil Tong, is a genial figure. Liberated by the prospect of his imminent retirement, he is in the process of working through his cheese bucket list, and we compare notes on our recent experiences at Slow Food’s Cheese festival in Brà, Italy. For a thirty-year veteran of the California scene, it was a walk on the wild side. Artisan cheesemakers are not his bread and butter. Most of his career has been spent addressing the needs of the larger-scale end of the cheese industry. About twenty years ago, Tong tells us, requests began to fi lter in from small-scale cheesemakers for a course designed specifically for them: “Our EXPERTISE

•

215

philosophy was, the basics of cheesemaking are the same, no matter whether you’re doing a million pounds a day or a hundred pounds a day. . . . We taught them composition control, consistency, yield from the point of view of the economic realities.” With time, the short courses have started to incorporate more considerations for specialty cheesemaking, but Tong believes that “a lot of that is experience rather than principles. It’s kinda hard to pass that on. We tell them, ‘That judgment will come as you gain more experience.’ ” It’s not that American land-grant universities lack the technical expertise of their counterparts in France. Regional approaches to grazing and grass management are the perfect example to the contrary: land-grant catchment areas are by definition climatic clusters, and detailed information is readily available to grass farmers about new varieties of tetraploid perennial ryegrasses suitable for the climate of Pennsylvania41 or the relative attractiveness of orchardgrass and bluegrass for cool-season pastures in Kentucky.42 It is the American artisan cheese industry’s lack of regional clusters that presents a challenge to the landgrant university framework’s ability to provide in-depth support in the way that INRA scientists and AOP producers’ groups do with such success. The exception, to some degree, is Wisconsin. According to Tong, “because of the way their industry is organized, if it’s not cheese, there’s not much else going on there. Cheese and whey.” Perhaps it is no surprise then that Pleasant Ridge Reserve, a great American artisan success story, was developed in collaboration with the Wisconsin Center for Dairy Research, part of the College of Agriculture and Life Sciences at the University of Wisconsin-Madison. Far from clusters specializing in a single variety of cheese, the modest number of producers making a vast number of different styles between them means that the only option is general education. In fact, Tong says, one would be lucky to find an academic outside of Wisconsin with a specialized focus on cheesemaking: In California, we have just as much of any dairy product as we do of cheese. Cheese is big, but butter and powder are just as big, fluid milk is big, everything is big. That’s always been my challenge in my career: if I have to pick a focus, what’s it going to be? For instance, I did some master’s work on ice cream and I’ve always had an interest [in that]. . . . I could have developed a full career. Even with three dedicated faculty with expertise in dairy foods, it still isn’t enough to cover the full range [of cheese].

After we take a tour of Cal Poly’s campus and their new, state-of-the-art laboratories and teaching facilities, Bronwen gives a departmental seminar 216

•

EXPERTISE

on the impact of farming and milking practices on raw-milk microbiology in cheesemaking. Due to declining enrollment in the Dairy Science Program, the decision has recently been taken to fold it into the Animal Science Department, and veterinary students and professors make up a significant part of the seminar’s audience. It is a fascinating and thoughtful group that brings together different backgrounds and expertise, and as we taste cheeses that we have brought with us from London alongside experimental Cheddars made in the department’s teaching creamery, ideas swirl through the room. Department head Dr. Jaymie Noland had been surprised by the information linking the cow’s environment with cheese quality. “We are coming around full circle in a way. When I was younger, I milked a Jersey for years and always thought even her mood made a difference!” A student asks whether we think it might someday be possible for people to work with cheese in restaurants in the same way sommeliers do with wine. The coming together of the Dairy Science and Animal Science departments at Cal Poly, while born of necessity, may have the potential to change the way that the students interact with milk and cheese, teaching them to look at these first and foremost as products of a farming system rather than a manufacturing process. As we are about to leave, Professor Rafael Jiménez-Flores makes an offhand comment about how the rugged Central Coast terrain and its Mediterranean climate are so much more naturally suited to small browsing ruminants than to the milky dairy cows that dominate the California academic and factory cheese scenes. The concept of unique regional foods as the products of close interaction between climate, historically informed approaches to production, and a discerning local—and international—consumer base is not foreign to California.43 Even in Cal Poly’s backyard, Paso Robles has secured a global reputation for its wines, which feature classic Mediterranean grape varieties. One winemaker on unproven land is a curiosity. But by dint of their collective identity—even as they compete for market share and argue about the character of the region—these winemakers gain access to all the advantages of a dynamic cluster. And through their vigorous conversation, they are unlocking the potential of natural resources they have only begun to understand and to tap.

EXPERTISE

•

2 17

ELEVEN

Markets

There is something peculiarly French about the career paths of great chefs. Each generation might make a show of renouncing such aspirations, but the pursuit of three stars from the Michelin Guide is the dramatic arc around which culinary careers are made and broken. The problem is, what comes next? Temples of haute gastronomy have fearsome fixed costs; they are monuments to ego rather than establishments of commercial good sense. And so, while chefs remain the arbiters of gastronomic authenticity, they also indulge in the side projects necessary to earn a living. In 2010, UNESCO included the French gastronomic meal on its list of the Intangible Cultural Heritage of Humanity. But it is each individual chef for him- or herself when it comes time to make money. A walk down the frozen food aisle of a French supermarket is a tour of the great names of French gastronomy, a practical demonstration of how far tradition and reputation can be pushed, one ready meal at a time. This tension is very much on our minds as we approach the Pavillon Ledoyen. Its busy neoclassical columns and moldings, capacious windows, and satisfyingly garish buttercup-and-bluebell paint job are familiar symbols of decadent Parisian high life. Ledoyen has occupied the gardens to the east of the Champs-Elysées for over two hundred years, making it one of the oldest restaurant sites in Paris. On this particular crisp and sunny January morning, Ledoyen and its three-Michelin-starred restaurant, Alléno, are playing host to Le Cheese Day, a “celebration of all the cheeses of France and the world, accompanied by the finest wines and spirits.”1 At the entrance, we pass a fleet of shiny black cars sporting Le Cheese Day livery parked in the drive. Inside, somber and purposeful white-jacketed commis scurry around us making final preparations for the impending cheese-themed lunch. Passing into the exhibition area, we find a sea of 218

booths. Looking slightly incongruous in the sunlight fi ltering through the eighteenth-century sash windows is a procession of big brands, interspersed with the odd regional producers’ group and a lone farmhouse cheesemaker. Armed with glasses of water and breadsticks, we venture forth into a realm of puzzling and disjointed signs and signifiers, a microcosm of the cheese market as it exists today. It is a semiotician’s wet dream. As we survey the room, we begin to see a pattern: the vast majority of the cheeses represented are subsidiary brands of a single company, Savencia, formerly known as Bongrain. A dairy company with a global reach, Savencia processes over four billion liters of milk a year. The company is the second-largest cheese producer in France, though many of its brand names—including Camembert Le Rustique and Alouette cream cheese—are just as familiar abroad. We approach a booth promoting Chaumes, a soft cow’s milk cheese painted with orange beta-carotene dye to resemble a smelly washed rind; in contrast to its appearance, its taste is bland and innocuous. Rather than dwell on the cheese itself, we are encouraged to appreciate its novel resealable carton, which invites us to cut and extract a serving without touching the cheese with our fingers. “People do not want to get their fingers sticky or have stinky hands,” the man behind the table tells us. “In this way, you can enjoy cheese without inconvenience.” At another booth, a young woman is busy representing Saint-Agur, a blue cheese with added cream made in the Auvergne. She moves fluently between English and French, working her way through her talking points. “Roquefort is a cheese for grandparents,” she says. “Our customers are looking to treat themselves to a luxurious and pleasurable indulgence.” Rather than chasing an aura of authenticity, Saint-Agur’s cartoonish, tongue-in-cheek advertisements parade an unsubtle sexuality. We think of the remote, mountainous region of Auvergne, which is also the homeland of monolithic Salers and subtle Saint-Nectaire. Saint-Agur is a cheese from the Auvergne, but not of it. Bemused, we wander over to a booth sponsored by the AOP producers of the Savoie, one of the few tables not featuring a brand name. Here, the cheeses are generic representations of the region’s AOPs, and we sample Reblochon, Abondance, and Tomme de Savoie. When we ask about their producers, we are told whether the cheeses are fermier (made on a farm) or laitier (made in a factory), but the trail ends there. Instead of being given more information about the producers, we are told about the rugged terrain and the high mountain pastures that characterize all the cheeses of the region. MARKETS

•

219

Leaving the booth, we are surprised to see an old friend across the room: Fromagerie Graindorge, whose Camembert de Normandie production we had visited just a few months earlier. Independent and family-owned, Graindorge specializes in making regional cheeses with protected names, such as Camembert de Normandie, Pont l’Eveque, Livarot, and Neufchâtel. In addition to being sold to supermarkets, many of its cheeses go on to cheese maturers, and they can be found on the shelves of numerous independent cheese shops. The team from Graindorge were pleased to be invited to participate in Le Cheese Day, but two hours into the day, they are having second thoughts. Designed to be universally appealing, Le Cheese Day succeeds only in projecting inauthenticity. The same could be said for the cheese market at large. Whether they sport a fantasy name or a generic appellation, cheeses are reduced to easy-to-consume, sanitized caricatures of themselves. At the most superficial level, the result is that the customer never has the opportunity to understand the origins of the cheese. But more fundamentally, the lack of a real conversation denies cheeses the opportunity to simultaneously express their multiple, complex, and sometimes-conflicting identities.

L I B E R T É , É G A L I T É , . . . FAC TO RY ?

As we stood surveying the room alongside the representatives from Graindorge at Le Cheese Day, little did any of us know that we were peering into a vision of their future. Just four months later, we receive a press release: Graindorge has been sold to a global dairy giant, Lactalis, a company with almost four times the annual turnover of Savencia. With the addition of Graindorge to its portfolio, Lactalis acquires a valuable brand and also becomes the largest producer of Camembert de Normandie. Over and over again, we have been awed by the power and resources that the French have mobilized through their producers’ groups and collective action, the public “good fight” they are waging to protect the gastronomic heritage of their country. And yet, despite the resources and creative intelligence standing behind farmhouse and artisan cheese production, the French cheese landscape is shifting. Graindorge is just one example of many: France is witnessing the decline of the independent cheesemaker. Make no mistake: French factory cheeses are technical masterpieces. The products of Anglo-Saxon factories have been distorted and reshaped by the 220

•

MARKETS

process of scaling up: nobody would ever confuse a plastic-wrapped block of mass-produced Cheddar with a natural-rind farmhouse cheese. French factory cheeses are a different story. They are sophisticated exercises in pooling the milk from hundreds of farms, correcting the component elements for optimum performance, and adding exactly what is needed to create cheeses that look identical to the farmhouse versions but sell at a fraction of the price. Many of them even use raw milk. And are they ever cheap. This is not just so for the Chaumes and the SaintAgurs; the same is true for many AOP classics. From Camembert de Normandie to Brie de Meaux to Epoisses de Bourgogne, over the course of decades, competition from factories with finesse has driven legions of farmers to abandon cheesemaking and become exclusively milk producers, or to leave dairying altogether, just as the march of progress did in the United Kingdom a generation before. Despite their collective groups, research budgets, technical prowess, and status as minor celebrities of the cheese world, French farmhouse cheesemakers—the Guy Chambons and Patrick Merciers—are becoming an endangered species. But what is the problem with a system that churns out millions of technically faultless, consistent, inexpensive cheeses? First, when cheese is made in a factory, milk becomes a commodity, with all of the social and economic implications that accompany that distinction. Farmers who sell milk for AOP cheeses may be paid more than those producing milk destined for UHT Tetra Paks, but in both cases, the milk is still a blank, an undifferentiated raw material. As the market fluctuates, so does the security of the farmers. They are trapped. In this system, quality is added in the factory, where the milk of many farms is blended. There, the components are corrected, and a collection of cultures is added to create a consistent product. These cultures are sophisticated, but compared to the natural functional ecosystem of a single farm’s milk, they are a simplification, a synthetic community with a limited range of expression. For a factory selling a cheese with a defined flavor profi le to a supermarket, where it will not be tasted before being taken home by a customer, this is a positive advantage. But it does not make the most interesting cheese. Consolidating production among a small number of large producers is convenient for large retailers who want to have simplified supply chains and carry the same cheese in every store. But when hundreds of producers of a cheese become one or six or twelve, the bottlenecking effect in itself MARKETS

•

221

fundamentally changes the style of that cheese. It becomes more constricted in its scope of expression; it no longer has the same dynamic range. Let’s imagine if there were one miraculous factory in Somerset churning out hundreds of tons of cheap, flawless Edith Cannon Cheddar. We would be no closer to realizing the distinctive potential of the small farms and unique sites that produced the milk, not to mention different expressions of Cheddar cheese itself—what about the Scotch and Candy methods, in this scenario still extinct and unimagined? With this addition to the Somerset landscape, we would just have . . . one more cheese. What gives shape to a style of cheese, what gives it depth and wholeness, is its diversity. The differences encompassed by all of its producers are as definitive as their similarities. Lose that diversity, and the cheeses lose their realness, their vitality, their soul. And that is what is happening in France.

T H E F O N D U E P OT

Few dishes are more inviting than fondue, the ultimate comfort food. At the bottom of every bubbling, sensuous pot waits la religieuse: a crisp, savory disk of caramelized cheese, a gold medal for those who push on past the first signs of satiety and emerge triumphantly overstuffed. There is no better place to enjoy Switzerland’s national dish than Le Gruyèrien, an Alpine-themed restaurant in Geneva. We are here with cheesemaker Marc-Henri Horner and his one-time apprentice Guy Arpin, who is now the export manager at Fromage Gruyère, a regional affineur that manages the maturation of many producers’ cheese, including Horner’s. Like most residents of the canton of Fribourg, Horner speaks French with a thick Swiss accent; Arpin has gallantly stepped in to facilitate communication. We have spent the day visiting Horner’s tiny creamery, shop, and maturing cellar in the tranquil Frenchspeaking town of Marsens, seventy or so miles to the northeast, and a dinner of fondue is a fitting conclusion. Sitting around the table in a satisfied postprandial stupor, it strikes us that there are few more credible endorsements of a cheese than its producers spending a rare night off enjoying large quantities of it themselves. Fondue as a dish was appropriated and popularized by the Swiss cheese industry during the early decades of the twentieth century. It was an inspired choice, both as a metaphor for Swiss society at large—itself a melting pot of French, German, and Italian identities—and as a way to sell loads of cheese. 222

•

MARKETS

Foundation myths aside, tonight’s fondues are resolutely francophone. Hundreds of kilos of Horner’s cheese are melted down into fondue at Le Gruyèrien each week. The white-wine-spiked Gruyère fondue is the most popular, but the standout for us is the one featuring Horner’s other specialty: young, semisoft Vacherin Fribourgeois. Made with water rather than wine, this second fondue highlights the deeply fruity aromas of the supple Vacherin. Horner’s Vacherin Fribourgeois is a very special cheese, and not just because it is rare and distinctive. It embodies a central dilemma of producers’ groups and collective identities: when hundreds of cheesemakers all produce the same cheese, what becomes of creative genius? The variety of cheeses made in Switzerland decreased precipitously during the twentieth century as a result of the policies of the Swiss Cheese Union (Schweizer Käseunion), a cartel that exerted control at every level of the cheese industry, from milk production to marketing. The union’s bosses decided early on that the most efficient route to growth was to focus on producing and marketing a few regional classics: Gruyère, Emmental, and Sbrinz. Producers who made those standard cheeses could tap into a system that used governmental support to subsidize aggressive growth. Lesserknown local specialties were denied the same export subsidies and consigned to local markets, low prices, and relative oblivion. When the extent of its anticompetitive practices and rampant corruption became clear in the late 1990s, the union was dissolved, and in 1999, the Swiss cheese industry entered a new era. Designations of origin were quickly secured—Emmental in 2000 and Gruyère in 2001—and the control over production volumes and quality that had been invested in the union was turned over to the legal authority of the producers’ groups. With structures no longer in place to push large volumes of cheese at high prices, strict quotas were implemented to match supply with demand. Some cheesemakers whose capacity exceeded their share of the quota began making unbranded knock-offs on the side. Production of the local specialty Vacherin Fribourgeois was encouraged in an attempt to stem the flow of cheap unbranded cheese from the dairies. “When they created the AOP [for Vacherin Fribourgeois] back in the early 2000s,” Arpin explains, “25,000 kilograms of milk quota was distributed to each dairy for making Vacherin. For most of the dairies, that might have been one week per year’s production, and that’s it. But it was still an incentive to make less of the copy cheese.” Horner began his career producing only Gruyère, Arpin tells us: “His idea at first was [to use Vacherin as] just a little extra sale, something to offer over MARKETS

•

223

the counter in his dairy shop.” But he became intrigued when he came across an old, black-and-white photo of a man selling Vacherin Fribourgeois at a nearby market in the 1950s. The cheese in the photo bore no resemblance to the gummy Vacherin of today: it had a crumbly, acidic core and a gooey breakdown underneath a dark and wrinkled rind. A few weeks after Horner first saw the picture, a roving antiques salesman offered the cheesemaker an old book containing a recipe for Vacherin, and an idea began to take shape: he would try to recreate an archaic version of the cheese. He began by making a few small experimental batches alongside his Gruyère, drawing on the photo and the recipe for inspiration. At first, he made just four or five wheels a month, but as word spread and the cheese developed a following, he increased his production. Now, he makes up to forty per day during peak season. The standard Vacherin Fribourgeois on the market is hardly a glamorous cheese. Over 70 percent is made expressly to be sold grated as fondue mix, and around half of all the cheese within the AOP is made by a single company. Risk-averse cheesemakers make it quickly by scalding the curds at a high temperature to drive off moisture and then pumping them into molds, an approach that yields rubbery cheeses with a mild flavor. Horner’s version is a different beast: its delicate curds require gentle stirring and must be molded by hand, as putting them through a pump would smash them to pieces. It is a softer cheese, with a furrowed and delicate rind. Not just for show, the double-cloth binding around the perimeter is essential for holding the shape of the cheese. Its flavor is deep, earthy, and funky, with a hint of freshness from the chalky core apparent in some of the wheels. It’s sublime in fondue, but it is also a serious cheese that holds its own on a cheeseboard. Membership in the AOP Vacherin Fribourgeois association means several things. In order to use the protected name, Horner pays a tariff of 80 cents per kilo of cheese, and each batch must be evaluated and approved before sale. When the graders visit, his unconventional cheeses receive low technical scores; as far as the industry standard goes, they are defective. Horner is philosophical. His low grades would diminish the price he would get from an affineur, but he doesn’t sell his cheese to affineurs. He matures it all himself before selling it directly to wholesale customers like Le Gruyèrien or across the counter in his little shop. His customers are happy to pay for the privilege: his wholesale price is 30 percent more than the market standard. “The only scores I need are the lines of customers outside of my shop,” he tells us. The practice of a cheesemaker maturing all his own cheese is almost unheard of in the region. “It can be very scary when I go to my cellar,” Horner 2 24

•

MARKETS

tells us. “Sometimes I look around and feel a sense of panic: ‘Whoa, I need to sell all of this!’ ” But his determination has paid off. Even with his high prices, the demand is strong, and as a result, the age of the cheese in his storeroom is almost a month younger than he’d wish. He simply doesn’t have enough cheese to go around. Under normal circumstances, a producer would increase production to match demand. But Arpin tells us, “He’s asking for more quota, but [the producers’ group] won’t give him any more. He isn’t allowed to make more, and he can’t buy the others’ quota because there is a rule that the quota has to stay within each village so that there are no empty vats.” He sighs: “Politics! In the longer term, the idea is to do less wholesale and to sell everything retail, locally, to maximize the value from what he can produce.” Why not just leave the group and sell the cheese by a different, proprietary name? Horner looks troubled. Arpin explains, “Making Vacherin Fribourgeois is a way of being loyal to an organization which gave him something. . . . It would be foolhardy to leave the group.” Horner believes he is being adequately rewarded for the cheese that he produces. Exceedingly humble, he doesn’t want to attract attention or, worse, be thought of as greedy. The only thing that bothers him is the constraint on the quantity he can produce. Why should he have to limit his production on the basis of quotas when customers are beating down his door asking for more? When it comes to Gruyère production, the rules are even more restrictive. Unlike with Vacherin, those who produce Gruyère are required to sell at least 96 percent of their total production to an affineur, who will age it and pay according to the scores it receives when it is graded. There is a limit to what producers are allowed to keep for direct sale to customers. As Horner’s retail demand grows, the amount he is permitted to sell himself is barely enough to satisfy it, and this problem is about to get worse. The week before our visit, the AOP Gruyère group announced that due to oversupply, the quota will be reduced by 10 percent in the coming year. The cuts are rolled out evenly across the board, so regardless of whether producers make amazing or inconsistent cheese, all of them will have to cut back production by the same amount. Arpin confides that despite its drastic nature, this intervention is unlikely to have the intended effect: “What do you do with the rest of the milk? You generally make a semihard [unbranded] cheese, and it kills everything.” It’s a classic story, the fountainhead of many a producer profi le and political campaign advertisement: the noble hero struggling under the weight of repressive bureaucracy. But when we return to London, we receive a phone MARKETS

•

225

call from a worried Arpin: Horner is concerned that he has inadvertently spoken against his colleagues. Who is he to talk down the appellation? He stresses that it was the income and security provided by his membership within the Gruyère AOP that gave him the resources to develop his Vacherin in the first place. And much of what the Gruyère syndicate requires is laudable, including raw milk, continuous whey starters, and a minimum standard of quality. Cheeses that don’t make the grade are not allowed to use the name, so when customers buy a cheese called Gruyère, they know that it has been tasted and judged worthy by experts. The cheeses at the bottom of the heap are policed and brought up to standard. For a cheesemaker like Horner, who must operate within a closely knit community, social cohesion is a necessity. But sitting back in London, we remain unconvinced. The same rules that guarantee a basic level of integrity simultaneously run roughshod over the capacity for individual cheesemakers to add value. A system designed to police a minimum standard represses those who excel, denying them a reward for the extra value they create. In an efficient market, producers like Horner would eventually find themselves making cheese of great renown, while the less talented would find another profession. But because the market for Gruyère and Vacherin Fribourgeois is not allowed to float, it can’t become stratified. And by continuing to produce Vacherin Fribourgeois rather than leaving the appellation and giving his cheese a fantasy name, Horner is unwittingly enacting a mountainside experiment in game theory. At what point does it become more compelling to compete than to cooperate? For Horner, the answer, it seems, is not quite yet.

M OV I N G T H E CO N V E R S AT I O N

Maison MonS is an affineur and distributor based near Roanne, in central France. Founded in the early 1960s by a couple from nearby Auvergne, the company has grown under the leadership of their sons, Laurent and Hervé Mons. On the day that we visit their maturation facility and warehouse, their cold store is lined with pallets of cheese ready to be sent to their nine shops in France and exported to the United States and the United Arab Emirates. Fanny Thivoyon, the company’s cheese buyer, is the first to admit that the dominant model for selection, maturing, and distribution in France has historically been at odds with the cheesemaker’s name ever reaching the customer: “At the beginning we didn’t keep the name of the producer [on the 226

•

MARKETS

cheese] because the company needed to make its own name.” Now that it is established and its reputation is secured, however, MonS is beginning to communicate more about the identity of producers and what makes their cheeses special. With MonS’s family links and proximity to the Auvergne, Saint-Nectaire is a particular specialty of the company, and it works with four different suppliers. Two of them farm Montbéliardes and Holsteins, while the other two farmers work with the temperamental and lower-yielding Salers breed. As the MonS team tasted the cheeses over time, it started to become clear to them that some Saint-Nectaires were better than others, and they recently decided to take the leap of highlighting the distinction between them. “The final products are completely different,” Thivoyon tells us. “Now that we are 100 percent sure of the [higher] quality of production with the Salers, we sell [the cheeses] by different names, different prices, different prestige.” In this more stratified market, both the cheesemaker and the affineur have the capacity to win. “The people with the Salers cows are now getting a higher price. . . . Hervé pushed them a lot to improve their quality and to put their value in terms of price.” In several cases, MonS has found that by breaking away from the AOP rules, it is possible to bring cheeses to a fuller expression of their potential, and occasionally they have decided to step away from the protected names entirely. By encouraging some of their suppliers of soft goat cheeses to dispatch their cheese earlier than is permitted by the decrees, they have gained more control over the ripening process while freeing up extra time for the producers to concentrate on the farming and cheesemaking itself. “We collect the cheese at four days old,” Thivoyon explains. “Now we do not sell any more Sainte-Maure, or Valençay, or Pouligny-Saint-Pierre. Each of them has a different [non-AOP] name, and they are better than before. If you have the final quality, you don’t need the designation. The cheese speaks by itself.” Thivoyon also shows us a few cheeses being developed by MonS entirely outside of the AOP system, the products of new and independent-minded cheesemakers who want to do something different from their peers. In the United Kingdom, development hell is a rite of passage for every good cheesemaker, and the reassuring if slightly depressing thing is that when deprived of a collective technical infrastructure, the French cheesemakers have just as many problems. Hervé Mons has spent enormous time and resources helping one young cheesemaker painstakingly develop and troubleshoot her new cheese. Only now, after several years, is it almost ready to launch in earnest. MARKETS

•

2 27

This process is a far cry from the Peltiers’ effortless sidestep into the world of Sainte-Maure production. Although letting go of the more restrictive requirements of the AOP has helped MonS increase its quality, it is a gentle reminder of the ambiguity of the situation: collective identities bring with them collective technical support, infrastructure, and experience as well. While MonS is experimenting with abandoning some of their use of collective names, another affineur, Laurent Dubois, is taking a different tack: he is keeping the collective names but adding more information on top of them. When we mention Dubois’s name to another affineur, the man whispers incredulously, “He puts the name of the cheesemaker on the label of all of his cheeses. He is giving away all of his competitive advantage!” We are intrigued. In contrast with rapidly growing and export-driven MonS, with its impressive collection of underground maturation caves and tunnels, the empire of Laurent Dubois consists of three tiny shops in bohemian districts of Paris. His reputation is that of someone who walks the line between charismatic visionary and turophile dictator. A friend who is an authority within the Parisian cheese scene shows some trepidation at the prospect of an audience with Dubois himself, even though she buys large quantities of cheese from his shop on a weekly basis. By the time we present ourselves for our appointment, we too are apprehensive. “Monsieur Dubois will see you now,” an assistant tells us gravely. And suddenly, he is before us, a severe éminence grise, sporting the bleu-blanc-rouge collar of the Meilleur Ouvrier de France on his black cheesemonger’s smock like a military decoration. He leads us to a small staff room at the back of the shop, and as we begin to chat, the mask comes off, and his public persona as austere ideologue melts. He shows a genuine geek’s enthusiasm as we discuss the extended aging of lactic cheeses and is intrigued to try a sample of a six-month-aged English goat cheese that Bronwen has brought from London. When we ask him about his decision to put the producers’ names on his cheeses, he launches into a passionate and articulate soliloquy. There is the sense that his conclusions are truths arrived at through hard-won experience, and we sit back and listen. We dare not interrupt him. Dubois believes that sharing the identity of each cheesemaker with whom he works, alongside the regional appellation of the cheese and his own identity as the affineur, has the power to change the market. Coming from the United States or Britain, this hardly seems revolutionary; within the AngloSaxon market, all cheeses are first and foremost brand names rather than regional styles produced by imagined anonymous peasants. But Dubois has 228

•

MARKETS

recognized that even within France, celebrating the producer is the only way to encourage and sustain the best of them: “We need to find people who make cheese with the same passion as we sell it. To make cheese, to take care of animals, it’s a true vocation. There’s no vacation. It’s not easy.” He tells us about a farmhouse cheesemaker in Périgord, Jean-Yves Pain, who supplies him with goat cheese. Pain milks his goats by hand and accompanies them into the fields. “He milks by hand because he sees the difference, even though it takes him three to four hours more each day,” Dubois explains. “When he is with his goats in the pasture, it’s almost as if he says to them, ‘Don’t eat that, eat that!’ ” For a moment, he is lost in reverie. Then he continues: The flavors of that cheese, the incredible aromas! If it’s to continue, there needs to be more money. . . . Money is important because it permits the small producers to buy new equipment, make a better life for their animals. It is a big problem with the price. We have arrived at a place where cheese, it is too much of a bargain! The inputs at every stage, when done with integrity, there is a big difference in quality. But the price difference between this cheese and something that’s made with many shortcuts is very small.

He points out that a wine made with care and sensitivity can command a price ten or even one hundred times that of one that is mass-produced. By contrast, the wholesale price of the world’s best Camembert is barely four euros. I am horrified when people say cheese is expensive. Each product lives in its ecosystem, it’s fragile, it’s complicated. . . . If we can sell something more expensive, we can encourage people even more, motivate them. We will have more choice. It is important to give people a helping hand. When there are quality problems, we split the cost. When the producer says, “I don’t have money to buy the food for my animals,” we put the price up a few euros so we can pay them what they need. The important thing is the taste and having a good product. If the thing is not so good, or just average, we cannot sell it.

An hour races by, and as the interview concludes, we make our way back through the shop, where we load up on cheese to take home with us. Dubois talks us through his wall of Comté, which features cheeses of different ages made by the same fruitière (producer), as well as cheeses of the same age made by different fruitières. As we chat, members of his team busy themselves turning and caring for the cheeses maturing in the tiny cellar, advising customers on cheese selections, and breaking down thirty-kilo cheeses for display with MARKETS

•

229

swift, precise movements. Despite the bustle of activity, the atmosphere is calm, and Dubois himself is sedate and unflustered. He is a charismatic leader who is prepared to trust his system.

CO M M O D I F Y T H E M O N G E R , N OT T H E C H E E S E

Dubois’s manner in his busy shop reminds us of a story about another famous Meilleur Ouvrier de France, chef Paul Bocuse. With over fifty years of experience, Bocuse has no qualms about announcing his mastery: hanging in his restaurant near Lyon is a reproduction of Da Vinci’s Last Supper with Bocuse himself in the role of Jesus. As the story goes, Bocuse ventured out into the dining room one night in his perfectly pressed chef ’s whites to greet his guests as they dined. One asked, “But Mr. Bocuse, if you are here in the dining room, who is doing the cooking?” Without missing a beat, he replied, “The same people who are doing the cooking when I am in the kitchen.” Both Dubois and Hervé Mons deserve credit for bringing the approach and rigor of haute cuisine to the world of professional cheesemongering. Both men hold the status of Meilleur Ouvrier de France, the highest professional accolade that a French craftsperson—of any kind, including chefs, aeronautical engineers, and cheese maturers—can receive. The award is one part of a system of training and professional qualifications that dignify the choice to work with cheese, a structure that both Mons and Dubois have helped to shape. “In a previous generation,” says Dubois, “ambition meant having a high-flying job, an expensive car, et cetera. Now we have a new generation, wanting to live life with choice: to work hard, to have a laugh, to exchange ideas. . . . Everyone wants a bit of money, but [this new generation of cheese professionals] is going out into the world to become business owners with high status, following a different path.” In 2000, Hervé Mons’s brother, Laurent, founded a training program for people within the French cheese industry and integrated it with the professional cheese associations. At the school, which is called Opus Caseus Concept, students can earn certificates and qualifications in affinage and service. Students are also trained through internships. Each year, Opus Caseus Concept trains between fi ft y and eighty students, who subsequently go out into the world as ambassadors for cheese through their work as journalists, maturers, cheesemongers (fi fteen of Dubois’s current employees have been trained through the program), or independent business owners. In 2012, 230

•

MARKETS

an English-language arm of the school, the Academie Opus Caseus, extended its reach to an international audience of cheese professionals. Both Dubois and Mons recognize that in order to have a healthy industry, you need to give people a career trajectory and reward and recognize expertise at every level, from producer through to salesperson. The skills the two men value are not just showy tricks for competitions; they are a prerequisite for selling a food that is notoriously delicate, variable, and tricky to handle. A century ago, such skills were something that every respectable Anglo-Saxon grocer or cheesemonger would be expected to have. In his study The Shopkeeper’s World, Michael Winstanley talks about the sale of cheese in a prewar grocer’s in the southern English town of Tonbridge. The shop owner personally selected his cheeses at agricultural shows before maturing them in his cellar for a year or more. With his cheese trier, he would taste them periodically and determine when they were ready to be cut to order in his shop.2 The selective environment that drove those highly skilled grocers out of business during the early part of the twentieth century is the same one that caused the simultaneous demise of farmhouse cheesemaking. Grocers capable of selecting, maturing, and properly selling farmhouse cheeses had the tools and knowledge to do the products justice. In contrast, the supermarket model thrives on consistency and low prices. With the tentative resurgence of farmhouse cheese in the Anglo-Saxon world, recognition is emerging— among even the large multiple retailers—of the need to regain some of that lost expertise. Without the proper knowledge and infrastructure, the attempt to sell these cheeses at any scale is expensive folly. In the United States, an attempt to institutionalize industry knowledge has taken shape as the Certified Cheese Professional (CCP) qualification. Jane Bauer, the American Cheese Society’s education and outreach manager, oversees the program, and going into its fi fth year, almost six hundred candidates have been awarded CCP status. “We have sixty-four different companies represented [among those who already hold the certification],” she tells us. “People are very marketable once they have the certification. . . . They’re very proud of it, and we’re proud of it.” Quite honestly, we are skeptical of the body of knowledge enshrined in the CCP’s 150-question multiple-choice test. There is no practical element or assessment of candidates’ ability to handle cheese as opposed to memorize facts about it. The current exam will be a fascinating document for scholars of the cheese world in a hundred years’ time: it is a perfect encapsulation of the industry’s values and limitations at this moment. The irony is that, despite MARKETS

•

231

its shortcomings, the exam is working. The act of professionalizing the role of the cheesemonger—in this case, simply deciding on a body of knowledge and training people—is in itself transformative. After an almost century-long hiatus, selling cheese is being recognized as a skilled profession once again. When we spoke with CCP candidates at the 2015 American Cheese Society Conference, the message was unanimous: studying for and taking the exam brought them a new respect for the craft of making and selling cheese. Samantha Newton, a cheesemonger from southern Florida, told us, “There is so much hard work that goes into [making cheese], and [the knowledge gained from the exam] helps me as a retailer show that to my customers. Making cheese is an art, and we need to bring our customers more awareness of that art.” Th is ability to have an informed conversation is crucial. Encouraging customers to ask questions about the farming and the methods of production, and linking that to the flavor of the cheese on the knife, holds the power to transform the market. Several months later, back in London, we run into Amanda Finco, another CCP-certified cheesemonger whom we’d met at the conference. Finco is on sabbatical learning to make Cheddar in Devon. Since we spoke the previous year, she has followed her partner to Lincoln, Nebraska, where she continues to work with cheese at Whole Foods. “In Virginia Beach, where I worked before, we didn’t have small cheese shops,” she tells us, “and the same is true in Lincoln.” However, her experience studying for the CCP and traveling in Europe has given her a new ambition: to open her own small cheese shop in Lincoln, across town from Whole Foods, in a neighborhood where, currently, “people have to drive twenty minutes to buy good cheese.” In starting their own small businesses, newly minted cheese professionals like Finco are building the infrastructure to bring real cheese to a wider audience. Several hundred people take the CCP exam each year. Not all of them are brilliant retailers, and many will move on to do something else within a few years of taking the test, leaving no lasting mark on the cheese industry. No matter. The same could be said of the Somerset Cheddar-making community in the 1890s: there were some cheesemakers who were downright not very good, many who were trained and competent, and a few bright stars, like Edith Cannon. No matter what the discipline, having the numbers at the base of the pyramid is a prerequisite to finding excellence at the top. And that is the point about cheese retail: there is no future for farmhouse cheese without a thriving community of people who know how to handle and sell it. If we are to decommodify the cheese, we must commodify the monger. 2 32

•

MARKETS

An efficient market is not one that makes cheese more expensive or that defines and canonizes the “best” cheeses, as if such a thing could even be said to exist. Rather, it is a dynamic equilibrium that rewards producers, sellers, and consumers in equal measure. An efficient market is one with room for diversity, where information flows and ideas are passed back and forth and refined along with the cheeses that are bought and sold. We think back to Dubois’s enthusiastic description of his favorite customers: “They are brilliant, so open and intelligent, ready to exchange ideas. This is why I have a boutique. This is enriching for us, too.”

MARKETS

•

233

T W E LV E

Reinventing the Wheel

While the lives of herdsmen in the nineteenth-century American West have given us an entire cinematic genre, there are precious few movies about cheese. The gentle pace of dairying is not conducive to dynamic fi lmmaking: the Man with No Name did not cut the curd. This makes those fi lms where cheese is central to the plot all the more fascinating: this is where cheese and cheesemaking leave the privacy of the creamery and enter mainstream culture. What does the encounter tell us? Wallace and Gromit are endearing symbols of Great Britain. The heroes of Nick Park’s Oscar-winning claymation fi lms, the absent-minded inventor and his smarter dog have a benign nobility of spirit. In the United Kingdom, the adventures of the two are a highlight of television schedules during the holiday season, the British equivalent of A Charlie Brown Christmas. Their gastronomic impact started early too. Their first short fi lm, 1989’s A Grand Day Out, features the pair building a rocket to visit the moon so that Wallace can indulge in his passion for cheese. The character has remained an archturophile ever since. His signature love of Wensleydale inspired enough popular attention to awaken merchandising potential for the cheese: the image of Wallace and Gromit can now be found peering out from supermarket shelves, emblazoned on special waxed mini truckles. But beyond corporate marketing synergies, Wallace and his beloved Wensleydale are themselves the embodiment of how far cheese has changed in the course of a single lifetime. Peter Sallis, the veteran actor who provided the original voice for Wallace, was born in 1921, and this brings us to an interesting question. Whereas a modern Wensleydale is a white, acidic, and crumbly hard cow’s milk cheese (the original template from which crumbly Lancashire was derived), cheese from Wensleydale was described as late as 234

1935 as a pungent, naturally blue cheese, softer than Stilton and spreadable with a knife.1 When Wallace rockets off to the moon in search of Wensleydale, which cheese is it that he is pursuing?

T H E A P O S TAT E

Andrew Hattan’s face shines with earnest intensity. A youthful forty-seven, he and his wife, Sally, have been working their holding at Low Riggs Farm in the Yorkshire Dales for eight years now. The farm is at an elevation of just one thousand feet above sea level, which is not particularly significant, but theirs is easily the most remote farm we have visited. With hard-won skepticism, he doubted that visitors from London would bring a car equipped to traverse the three miles of stone track, so he insisted that he pick us up from the nearest village in his four-by-four. Bumping along the rough terrain, he comments: “If we are ever going to sell cheese, it had better be good enough to make it worth getting it up this drive!” As we look out across his hay meadows while he talks about the thin soils and acid grassland of his hill farm, we feel that he is in his element. But as he cheerfully acknowledges, this was never the life that his professional training prepared him for: “I have been Holsteinized myself. . . . My background was a degree in agriculture, followed by a PhD in energy utilization in high-yielding dairy cows. At that point, I didn’t want to look at milking cows ever again.” Instead, he works with a handful of Northern Dairy Shorthorns. Small cows with mottled red-and-white coats, these hardy animals were once the mainstay of subsistence farming in the Yorkshire Dales. The breed society separated from that of mainstream Shorthorns in 1944, and even to the untrained eye, it is easy to see that they are vastly different beasts. They are much smaller, for a start, and they lack the huge milky udders of the improved breed. These are upland cows: they are dual purpose—as good for beef as for milk—and have a reputation for docile hardiness and a capacity to thrive in harsh grazing-based systems. This is England’s Alpine cow. Aside from a handful of enthusiasts—there are just sixteen registered herds in the United Kingdom—the breed died out after World War II, and along with it, farmhouse Wensleydale cheese production. Hattan shows us two herd books bound in green: one, from 1947, is thick and proud, but the other, from 1964, is not much more than a pamphlet. At Low Riggs, the Hattans get a paltry 1,800 liters per lactation. Andrew has come a long way from that early academic blooding in the way of the Holstein. REINVENTING THE WHEEL

•

235

In the time that they have been living and working at Low Riggs, the Hattans have focused on the environmental reinvigoration of the farm. Subsidies have allowed them to restore two kilometers of dry stone walls and begin to rehabilitate the farm’s hay meadows. As Hattan enthuses about the biodiversity target for the hay meadows of twenty-five different species per square meter, we sense that the environmental projects are closer to his heart than the cheesemaking. However, he is wary of a life entirely dependent on subsidy. This is where cheese comes in: making Wensleydale, even at the very modest scale on which the Hattans operate, might give the couple something to sell that would be valued enough by the market to make the entire farm viable with no subsidy at all. He tells us about this possibility: Last year, we got the opportunity to go into a ten-year agrienvironment scheme. . . . One of the options available was to receive a payment for keeping rare-breed cattle, and I thought it was an opportunity, it was beginning to fit together. I’d always had a pipe dream about whether we could keep Northern Dairy Shorthorns, whether I could ever think about milking cattle again. . . . Thinking that the subsidies aren’t going to be here forever, I ought to use the ten years of this scheme to do something that’s going to take us beyond those ten years, to make this sustainable.

It is early days, and the talk is of trial batches, but the Hattans have already started exploring the old tradition of seasonal milking and cheesemaking. This involves abandoning any pretense that the cheese will be the same throughout the milking season, but it is a method that a nineteenth-century farmer’s wife would recognize. “Make one cheese and make it well,” Hattan states boldly. “That original Wensleydale was a cheese made in summer from grazed pasture. That’s our primary aim, to make a cheese of that ilk produced from the milk made from the fifteenth of May to the thirtieth of September.” But he explains that the wildflower meadows that have taken so much of his effort offer another enticing prospect: “We could use that hay to produce a second cheese, a late-lactation cheese. The forage is based on the flower-rich meadow hay; that might be a completely different cheese that would also have a unique selling point.” Shyly, he produces some test cheeses and invites us to try them. The cheeses, made to an approximation of a pre-1930 Wensleydale recipe, are a revelation. No more than a couple of pounds in weight, each wheel has a greyish natural rind specked with blue. The paste is rich yellow in color, a testament to the pastures on which the cows have been grazing, and the tex236

•

REINVENTING THE WHEEL

ture is supple and giving, neither soft nor hard. Aromatically, the cheese is dazzling, all delicate milky richness with a dusting of pollen and fresh cut grass. With the trepidation of the cheesemaking novice, he asks: “Do you think that they might be saleable?” One of the fundamental rules of working with cheesemakers is that you should never show an excess of enthusiasm for a trial batch of cheese. Curiosity, yes. Mild enthusiasm and gentle encouragement, certainly. But praise-the-Lord, shrieking, wild-eyed, hysterical passion is a definite no. Too much enthusiasm too early in the process can cause cheesemakers to lose concentration as they attempt to scale from one-off experiment to daily production. We instead mutter something polite about the cheeses being a highly promising start, and we quickly feel a jolt of self-loathing. This is the cheese that we want to eat. Later, when we talk with Hattan about his journey through the world of agriculture, the journey that brought him from, as he puts it, the “throw the nitrogen on, sell vast quantities, and breed high-yielding cows” mentality of his university days to his present concerns for environmental stewardship and renewal and his fascination with letting his raw milk spontaneously sour, we see the thread that runs through all of our favorite cheeses and cheesemakers. At some level, each cheese we adore and all of the cheesemakers we most respect operate as closely as possible to the idea of a dairy permaculture. In their farming and in their cheesemaking, these producers nurture systems that do not require excessive external inputs. The farm, the pastures, and the animals operate as a holistic system. Good raw milk contains all of the microbes needed to sour and then to ripen it into interesting cheese. The Daviet family, with their Reblochon de Savoie, and Guy Chambon, with his Salers Tradition, take this system an intriguing step further: the whey from their cheesemaking provides the liquid to clean the creamery. From this realization comes the single simple question—although it is one that will produce nuanced and complex answers—the consumer can pose to all cheesemakers and dairy farmers: What do you bring onto your farm and into your creamery from outside? When we ask this question, it consistently causes farmers to pause and think; cheesemakers have a tendency to furrow their brows. The list of things brought in might be long and can include milk, animal feeds, seeds and fertilizers, and cheesemaking supplies like starter cultures and rennet. Then there is the question of cleaning chemicals. We do not presume to dictate REINVENTING THE WHEEL

•

2 37

what might constitute a correct answer to the question, and in many ways, starting the conversation is far more important than arriving at any single answer. Each farm has its own unique circumstances. But we do know that our own favorites, the cheeses we love above all others, are made with the minimum of external inputs.

W H AT F L AVO R M E A N S

Whenever we think about issues of taste, we always take the dynamics of our own marriage as a starting point. As an Anglo-American union, we constantly run up against the small differences of two cultures divided by a common language. Aside from mild quarrels over vocabulary and spelling, we notice this narcissism of small differences most sharply when Francis interacts with his in-laws. Quite frankly, Bronwen’s parents struggle with his jokes. It is a test of cross-cultural communication that Francis flunks every Thanksgiving. Supposed witticisms fall flat. Allegedly smart comments and sly references are greeted with blank incomprehension. And so when we think about the meaning of the flavor of cheese and about what qualities should be celebrated, we have the differences within our own family at the back of our minds. This is because quality is a cultural judgment; it must be learned. Beyond our craving for salt, sugar, and fat, we do not come out of the womb with a nuanced taste for cheese any more than we come primed to love late Beethoven string quartets.2 Indeed, sensory scientists are deeply reluctant to talk about the properties of anything in terms of such a vague and unquantifiable concept as quality. Instead, the battery of tests that they deploy includes discrimination testing (which proves if there is a perceptible difference between two samples), descriptive analysis (which uses trained panels to characterize those differences), and affective or hedonic tasting (in which end consumers are asked to rate how much they like a product).3 “Quality” is an abstract noun for which there is no laboratory test—a cultural Rorschach test for the taster. As we think about the struggle to bridge transatlantic and cross-generational humor, it is clear that there is a similar divide in approaches to the idea of quality in food. It is the difference between defining the quality of a cheese (or for that matter a wine, coffee, or anything else) in terms of its production versus defining its quality in terms of the intensity of the experience that it gives its end consumer. Like Francis and his in-laws, these two approaches to 238

•

REINVENTING THE WHEEL

quality are divided by the weight that they place on wider meaning and cultural fit. Are the best cheeses those that are the hardest to produce, that deliver unique flavors impossible to replicate elsewhere? Or are the best cheeses those with the biggest flavor, those that place primacy in the experience of the naive taster and make no demands on consumers to understand a wider web of relevance and meaning? In the modern world, there is a distinct fi xation on the experience of tasting. Even academic historians, inspired partly by historically minded reality television programing, have taken an affective turn in their approach to the past. The question of how things felt in the past—or, in our terms, how they tasted—has become a topic of inquiry that receives as much attention as the study of the structures and institutions of society.4 When this is applied to cheesemaking, the results are cheeses of awkward anachronism that follow the modern separation of the disciplines of dairy farming and cheesemaking while indulging in selected archaic techniques in the creamery. Such an approach treats all milk as equal, except perhaps as far as it conforms to a handful of quality marks, such as “raw,” “pasture fed,” or just “fresh.” One of the most amusing seminars we have ever attended featured a distinguished professor of food history who spent much of the session talking about his travails attempting to make cheese without starter cultures using microbially inert pasteurized milk sourced from his local supermarket. He took its refusal spontaneously to ferment as a sign of the dangerous deficiencies of modern commodity milk rather than as an indication that the pasteurizer was working properly. Any attempt to separate cheesemaking from the farming of its raw materials directly mimics the specialization of roles within industrial systems. And as the Hattans and their Wensleydale amply demonstrate, it is in the milk itself that value can actually be created. If the Hattans made the same cheesemaking decisions but used milk from lowland Holsteins farmed within a pasture-based system making use of dressings of inorganic nitrogen fertilizers, their cheese would have none of its character. Turning milk into cheese is just a question of baseline technical competence so as to avoid faults. One of the great untranslatable French words in the wine industry is the idea of the vigneron, a farmer who grows grapes and makes them into wine. There is no comparable English term, so commentators talk awkwardly about vineyard enology. As we gaze on the hay meadows at Low Riggs farm, it is abundantly clear that our language needs an analogous term for cheese. In this spirit, we introduce the idea of the cheesefarmer, a dairy farmer who raises their animals exclusively for cheesemaking and benefits from making every REINVENTING THE WHEEL

•

239

decision in terms of its impact on the cheese. To this end, when asked questions about recipes for making cheese at home, our own response is: “Well, first you need to get a cow . . .” Perhaps we should not be surprised that cheese appreciation has taken this affective turn. Asking consumers to enjoy the meaning and context of a cheese is difficult when, at the consumer level, the cheese world operates in a cultural vacuum. Where there is no public conversation about what constitutes good cheese and about the mechanics of cheesemaking, how can the consumer hope to know which cheese is best? In this world, anything can become a quality mark as long as the story is framed appropriately. Even the leading English-language consumer cheese magazine can select as one of its “75 Cheeses of the Year” a lactic goat cheese that is made from milk that arrives at the creamery as solid frozen blocks because “freezing the milk . . . keeps it extremely fresh.”5 The same magazine unquestioningly repeats as a positive selling point the “ ‘happy cow’ feed” of a cheesemaker who mixes grass with grains and industrial byproducts from the almond and citrus industries.6 Without an informed public sphere, nobody can blame consumers for their uncertainty. In the context of this confusion, it is unfortunate there is an inherent tension between the diversity and nuance of flavors achievable through farming and the massive one-dimensional hit that is possible in cheese where all of the flavors are a function of the process. Just think of the sweet butterscotch flavor of a pasteurized Cheddar made using Lactobacillus helveticus as an adjunct culture. If, as a consumer, what is important to you is the volume of flavor in your mouth, that sweet Cheddar would blow away the milky delicacy of a cheese like the Hattans’ Wensleydale. The Cheddar makes no demands of knowledge or cultural resonance on the part of the consumer. It is easy. By stripping the cheese of any cultural baggage, this attitude to quality is democratic and instantly accessible. In products like wine, about which there is a more developed conversation about quality, that democratic accessibility is central to the debate: American wine critic Robert M. Parker Jr. has dubbed people who advocate for wines of meaning over those of bombast the “anti-flavor wine elite.” His rhetorical slur is now a badge of honor worn with some pride by sections of the industry.7 However, the logical endpoint of a fascination with sheer volume of flavor is fi xation on those pharmaceutical properties of the cheese—the sweetness, the salt, the umami, and the fat—that offer lizard-brain appeal to the taster. It is an approach that has served the processed food industry well, an industry in 24 0

•

REINVENTING THE WHEEL

which snack food has been successfully stripped down to those properties that deliver the greatest instant hit.8 The logical progression of the approach is to abandon the pretense of serving food at all. If what you crave each time you put cheese in your mouth is an intense and not necessarily culturally meaningful experience, a sublethal dose of heroin will deliver pleasure at an even greater intensity. And like salty snacks, it is distinctly moreish.

A M OV I N G TA R G E T

Enjoying cheese for its capacity to deliver a unique experience is all well and good, but we also have to acknowledge that as contexts change—in particular as technology changes and adapts—the meaning of flavors changes too. When the quality of a cheese lies in its capacity to be unique, then quality itself is a moving target. Wallace and Gromit films demonstrate the extent to which Wensleydale has changed in the past seventy-five years, but to understand the impact and consequences of technological change, we can usefully look beyond cheese and to the art world. Just as painting reacted and interacted with the new technology of photography, so farmhouse cheesemaking now faces the challenge of ever more sophisticated dairy microbiology. Just as it was the most marginal—or perhaps the least able—farmers in Wensleydale who first abandoned farmhouse cheesemaking when there came the option to sell liquid milk, so it was the fringe players—the itinerant journeymen artists, painters of signs and quick portraits—whose working life was completely changed by photography.9 The mid-nineteenth-century career of Englishman William Atkinson, an artist of little renown, is typical: Atkinson started his life as an itinerant portrait painter and then opened a pair of photographic studios, finally abandoning painting altogether to concentrate on photography. His pragmatism matched that of any dairy farmer.10 However, among the artistic elite, photography was not as widely embraced, or at least it was used for a different purpose. Baudelaire considered photography the “refuge of failed painters of little talent”11; photographs were mere reproductions that cheapened the products of the beautiful. Once the capacity to produce likenesses had been successfully technologized, the role of the artist became to move beyond photographic reality. In the words of James McNeill Whistler, “If the man who paints only the tree, or flower, or other surface that he sees before him were an artist, the king of artists would be a photographer.”12 By the time Vincent van Gogh was painting, he realized REINVENTING THE WHEEL

•

241

that there was little point in chasing photographic resemblance. Instead, his technique and skill as a painter lay in representing what the camera could not see, in exposing and intensifying the character of his subjects.13 Our experience visiting Camembert producers in Normandy underlined to us that farmhouse cheesemakers should take their inspiration from Van Gogh. There is no value to be gained from cheese made on a small scale taking its aesthetic cues from the products of creeping industrialization: “farmhouse” needs to mean something, and to do that it needs to venerate a different—and potentially changing—range of flavors and textures. In this way, cheeses can avoid the charge that the emperor has no clothes, or in other words, that they are otherwise unremarkable products enhanced by the romantic resonance of a good story. As we have seen, there are objectively testable differences in the physical properties of cheese made from the milk of animals on biodiverse pastures and those of cheese made from commodity milk. Handled with scrupulous care at every stage in the process, the emperor is clothed, and if we cannot taste the difference, then it is our problem and not that of the cheese. If it is a story that we are buying, then it should be a story that we can taste. And if we value these environmental and farming decisions, these are the attributes that we need to value. This is the “best” taste for now. It is the moral dimension of flavor.

THE WHEEL REINVENTED

For producers like the Hattans, as for so many other cheesefarmers we have visited, the desire to discover the unique flavors of their farm has pushed them toward the techniques of their grandparents’ generation. Twentiethcentury innovations, from nitrogen fertilizers and black-and-white cows to pasteurization and inoculation with commercial starters, just get in the way. Cheese was made at Low Riggs farm until 1962, and to witness the Hattans and their experimental cheesemaking is to gaze back in time. Fade the image to sepia, add some starched linen aprons, and this could be a scene from the daily work of a Victorian farmer’s wife. But reinventing the wheel is hard. The tacit knowledge and quiet technical accomplishment of the skilled dairymaid died with the last generation of farmhouse cheesemakers. In the Yorkshire Dales, the course of the twentieth century witnessed the progressive extinction of these skills, and a frustratingly incomplete written record has been left behind: the last cheese factor 242

•

REINVENTING THE WHEEL

and grader who traveled to visit remote Dales farms and select cheese started work in 1933, and the last raw-milk farmhouse Wensleydale was made in 1957.14 What recipes and instructions that remain are sparse, written for the quick direction of people who fundamentally already knew how to make the cheese. The work of the Hattans and those like them is driven by trial and error rather than the easy replication of a historical formula. If ecological metaphors can be applied at every level of cheesemaking and dairy farming, they seem especially apt here. “Extinction” is exactly the term in this case. The farmhouse production of British territorial cheeses does not meet what ecologists refer to as the “minimum viable population.” Territorial cheeses like Cheshire, Lancashire, and Cheddar that as late as 1939 had hundreds of producers of raw-milk farmhouse cheese have now been reduced to individual cheesemakers.15 Somerset is lucky enough to retain three Cheddar producers, but even this represents a hundredfold reduction in numbers over the course of the past seventy years. In the grim statistical euphemism of ecological studies, the population is insufficient to survive “stochastic variation,” the random hand of fate: one car accident, one heart attack, or one new generation not enamored with the prospect of cheesefarming, and an entire style of cheese will disappear.16 From this perspective, institutions like Neal’s Yard Dairy have a role more akin to a zoo than a commercial enterprise. To pragmatists within the dairy industry, these endangered species have no value in the modern market. They sit within their gilded cages in the zoological garden that is the specialist cheese retail sector, but they have no commercial relevance. In ecological terms, the facts of natural selection should be acknowledged, and these precious panda bears should be allowed quietly to die off. From this perspective, individuals who experiment with archaic techniques are irrelevant antiquarians exploring evolutionary dead ends; they might as well be trying to initiate a breeding program for dodos. The Darwinian cut and thrust of the market has spoken, and these cheeses have been replaced by their better-adapted descendants. But this is a fundamental misreading of evolution. There is no teleology in natural selection, no intelligent design in the dairy industry. Evolution is not a straightforward march of progress in which each generation is fundamentally superior to those that came before. Rather, the immediate circumstances of the selective environment dictate reproductive success: change the environment, and different selections will be made. Compare humans with the original marine vertebrates from which we have evolved: we have many advantages in terms of our adaptability and intellectual capacity, but hold a REINVENTING THE WHEEL

•

24 3

modern human underwater for more than a few minutes, and it is our distant ancestors who look fitter and better adapted. Exactly the same phenomenon is true across the board, whether it is being applied to monitoring a microbial fermentation or deciding the fate of cheeses in a competitive marketplace. It makes for an entertaining parlor game to imagine the points of divergence for alternate histories that would have given us a different cheese industry. What if the railroad had come to a particular region earlier or later during the nineteenth century? What if the United States had had expensive land but cheap labor rather than the reverse? And what would that require? Some limit on westward expansion? A difference in patterns of immigration? Our own particular favorite is the question of what would have become of the British cheese industry if the Nawab of Bengal and his French allies had defeated the forces of the British East India Company, commanded by Robert Clive, at the Battle of Plassey in 1757. Victory at Plassey eventually allowed the British to leverage control of the Indian subcontinent, where, by the late nineteenth century, they had planted vast tea gardens.17 The tea from those plantations became the universal British drink, served with milk. Indeed, tea, milk, and sugar were, in the words of anthropologist Sidney Mintz, key nineteenth-century “proletarian hunger killers.”18 The ubiquity of tea drove the market for liquid milk, and the market for liquid milk made it commercially viable, even advantageous, to abandon farmhouse cheese production. Would defeat at Plassey have meant that farmhouse cheesemakers in the United Kingdom would be more than an endangered species today? Debating historical what-ifs is a fun game, but it only serves to underline how the selective environment has now shifted. Whereas in the mid-twentieth century, a British—or an American—dairy farmer could achieve a secure income simply by selling liquid milk, this is no longer the case. The mediumsized dairy farm is no longer viable. In the deregulated modern market, with a free-floating milk price, the profitability of farms is brutally bimodal: a dairy operation must either be small, highly extensive, and add value on the farm— ideally by making cheese—or it must be a megadairy pursuing the greatest operating efficiencies possible in a volatile commodity market.19 At the same time, our contemporary gastronomic culture venerates the farmer as never before. Within this new environment, the capacity of raw-milk farmhouse cheesemaking to allow farmers to discover and unlock the uniqueness of their milk and actively capture the flavors that come from admirable environmental stewardship suddenly becomes attractive. In ecological terms, the habitat is ready for these cheeses. It just needs to be populated. 24 4

•

REINVENTING THE WHEEL

As such, looking to the past to move forward is not mere antiquarianism. For people like the Hattans, the past is not just a source of recipes that provide a marketing edge or an aura of authenticity to their cheese. Rather, aware of the factors shaping the story as it has been lived, they are exploring divergent paths untaken and applying them to their present situation. Where subsistence for nineteenth-century Dales farmers meant preserving enough of their own produce to have enough to eat, now it means being able to sell their cheese at a price that guarantees the sustainability of their farms. The Hattans are the perfect illustration of the virtuous circle in practice: their highly extensive farming on marginal land is expensive, but as long as they can capture the unique properties of their milk in their cheese, they will be able to secure a price premium than can fund the farm without relying on state subsidy. To do so, they must resurrect dead ideas and structures, but they adopt and adapt these as need be, inventing their own traditions along the way. We began our discussion of Wensleydale with the claymation adventures of Wallace and Gromit, but it has become increasingly clear to us that there is an unheralded superior cinematic celebration of the process of raw-milk cheesemaking. The 1993 fi lm Jurassic Park, directed by Steven Spielberg and based on Michael Crichton’s 1990 novel of the same name, is the perfect allegory of modern cheese. A parable of the danger and futility of attempts to recreate complex ecosystems note by note, Jurassic Park depicts a situation not unlike that confronting the Hattans. Just like available information on Wensleydale production, the dinosaur DNA that has been harvested from mosquitoes preserved in amber is degraded and incomplete. And just as curious Anglo-Saxon cheesemakers have been forced to turn to French research to plug the gaps in their knowledge of traditional practice, the fictional geneticists of Jurassic Park use frog DNA to fi ll in the gaps of the dinosaur’s genetic code. It is this extra frog DNA that then allows the resulting dinosaurs chaotically to reproduce. The fi lm’s visceral thrills come from the peril that the dinosaurs pose to the human protagonists, but in our cheese world, the perspectives are reversed. It is true that if cheese competitions embraced the mortal peril of contestants getting eaten by velociraptors, more people would pay attention to them, but the issue of perspective is a wider one. The first faltering steps of communities of cheesemakers as they reemerge—and even thrive—in the marginal terrain of places like the Yorkshire Dales or the mountains of the Auvergne are an exercise in cultural conservation that is as ecologically REINVENTING THE WHEEL

•

245

important as work with any endangered species. Classic narratives of ancient and mysterious beasts present them as a threat to humanity, creatures to be beaten back and tamed. But as these Lost Worlds are repopulated with cheesefarmers, it is with pride and as a celebration of knowledge and human ingenuity that we are of the dinosaurs’ party. It is science that has brought them back to life.

24 6

•

REINVENTING THE WHEEL

ACKNOWLEDGMENTS

This is a book conceived in Burgundy, and it would not exist without the generous hospitality of the Seysses family. Diana, Jeremy, Rosalind, Jacques, and Alec shaped our understanding of what farming can be and how integrity is the route to success. Bronwen has been lucky to work with an amazing team, past and present, at Neal’s Yard Dairy, whose generosity with their knowledge and curiosity with their questions drove our understanding of cheese. Randolph Hodgson and Jemima Cordle are inspirations for anyone studying British territorial cheese. David Lockwood, Jason Hinds, and Clare Panjwani read drafts of this book and offered helpful advice, while Jennifer Kast was a valuable sounding board for our ideas. The world of wine has provided us with enormous support and encouragement. Neil Beckett first commissioned us to write together for The World of Fine Wine and is the greatest editor, mentor, and friend that any writer could want. Eric Asimov, Julia Harding, Andrew Jefford, Jancis Robinson, David Schildknecht, and Kelli White all shared their thoughts and ideas readily and were happy to consider familiar concepts in novel dairy situations. Jamie Goode gave us a template for how to engage people with technical details, while Peter Liem provided a model for reimagining an industry. Jon Bonné directly inspired us to write this book and then made the connections to make it happen. Jasmine Hirsch and Christian Holthausen believed in the project and taught us how to get our ideas out there. Vinny Eng, Mateja and Josko Gravner, Steve Lagier, Ted Lemon, Carole Meredith, Odessa Piper, Grant Reynolds, Matthew Rorick, Mitja Sirk, Terry Theise, and James Tidwell were all munificent with their expertise and enthusiasm. This book would not exist without many cheesemakers and farmers who were willing to give generously of their time. Guy Arpin, Will Atkinson, Dulcie Crickmore, Jonny Crickmore, Andy Hatch, Andy Kehler, Mateo Kehler, Graham Kirkham, Jamie Montgomery, Soyoung Scanlan, Albert Straus, and Rebecca

247

Williams all shared their considerable wisdom. Patrick and Rebecca Holden were open to putting theory into practice, and Nicholas Millard was our guide as we explored the mechanics of animal husbandry. We benefited hugely from our extensive discussions of dairy cow metabolism with Andrew Hattan, while Ivan Larcher helped us systematize our approach to the world of cheese. The advice, help, and assistance of the wider cheese community were also invaluable. Alison Lansley and Sue Sturman both offered valuable feedback on our manuscript. Caroline Hostettler and Jean-Pierre Missillier were generous with their time, knowledge, and connections in Switzerland and Savoie, respectively. Kate Arding, Omri Avraham, Elizabeth Chubbuck, Chris Dawson, Chris Dee, Laurent Dubois, Cathy Gaffney, Jane Hastings, Ursula Heinzelmann, Elaine Khosrova, Hervé Mons, Laurent Mons, Paul Neaves, Emily Shartin, Céline Spelle, Anthea Stolz, Cathy Strange, Fanny Thivoyon, Jon Thrupp, Sarah Weiner, Ari Weinzweig, Carlos Yescas, and Meg Zimbeck all provided commentary, ideas, and inspiration. Klara Halkjaer went out of her way to help arrange interviews. Our story is about science and scientists and would not have been possible without the warmth and openness of the scientific community. Benjamin Wolfe’s intellectual curiosity, mentorship, collaboration, and friendship have been second to none. He and Catherine Donnelly both commented in-depth on the full manuscript. Nathalie Desmasures, Bruno Martin, and Marie-Christine Montel offered insightful comments on draft chapters and were thoughtful and generous hosts. Rachel Dutton gave Bronwen the opportunity of a lifetime at her lab at Harvard. Dennis D’Amico, Marina Cretinet, Ulrike Eggert, Majid Ezzati, Arielle Johnson, Scott Jones, Harold McGee, Valérie Michel, and Elinor Thompson all gave advice and encouragement. This is a work that touches many disciplines. Kiera Butler offered advice on the manuscript, and Carrie Balkcom, Alison Martin, Tara Ramanathan, Vaughn Tan, and Harry West all reviewed chapters. Ophelia Deroy, Fuchsia Dunlop, James Hoffmann, Tomoko Iwaki, Sandor Katz, Jane Levi, Anette Moldvaer, Nicholas Morgan, Heather Paxson, Pat Roberts, Dan Saladino, Barry Smith, Charles Spence, and Jonathan Wilson all inspired us with their insights. Writing the history of your own family is a task fraught with potential pitfalls, and we thank Bronwen’s uncle Eddie Imsand and her parents, Eric and Pat Bromberger, for the good grace with which they helped us explore our own story. Nothing hones ideas more than the opportunity to talk about them in public. Dan Barber and Irene Hamburger at Blue Hill, together with Jill Isenbarger and Martha Hodgkins of the Stone Barns Center for Food & Agriculture, provided us with a forum for talking about raw milk microbiology. We were blessed with a brilliant editorial team. Our agent, Katherine Cowles, was involved from the very beginning and was instrumental in turning us toward writing a narrative. At the University of California Press, Kate Marshall immedi24 8

•

ACKNOWLEDGMENTS

ately understood what we were trying to achieve. Bradley Depew, Dore Brown, Alex Dahne, Peter Perez, Tom Sullivan, Genevieve Thurston, and the rest of the team were all vital in turning our vision into reality. We thank the Culinary Historians of New York, whose Scholar’s Grant funded the research that went into writing chapter 8. We also thank The World of Fine Wine for kind permission to reprint previously published material as part of chapters 7, 10, and 12. All errors that remain are entirely our own.

ACKNOWLEDGMENTS

•

249

This page intentionally left blank

APPENDIX H OW TO B U Y C H E E S E

Throughout this book, we have stressed the importance of an ecological approach to cheese. Cheese is not static. Rather, it is an exercise in evolution: the selective pressures applied to complex systems dictate everything from the shape of dairy animals to the contours of rural landscapes to the growth of microbes on a rind. If one understands the selective environment, one understands the cheese. If this all makes cheese sound very abstract, subject to grand and impersonal forces, it should not. Individual cheese lovers are not relegated to the role of spectator before the sweep of history. Their food choices and taste preferences, the cheeses they love and the cheeses they hate, are the forces that drive the evolution of cheese. Furthermore, the experience of eating cheese directly links the consumer with the landscape and practices they are sustaining: tasting it makes it real. We always think of the Thanksgiving dinner at which we served a piece of Salers. The cheese, presented without fanfare during the course of the meal, was a slow burner. It was strikingly different from the cheeses to which our extended family was accustomed: this was cheese from another planet. But from puzzlement came curiosity. Why did this cheese taste as it did? The mountain pastures of central France had never impinged on our family’s consciousness before, but now they were experiencing them firsthand as a sudden, visceral thrill. Cheese is an agricultural product inherently designed to reach distant markets: when made with sensitivity, it gives us the opportunity to taste places we have never been. So how can we encourage this sensitivity on the part of the cheesemaker? What drives the ecology? The ultimate selective pressure is how you as consumers choose to spend your money. Through our purchases, we shape the world according to our values, but that does not mean that our lives need to be dull but worthy. Cheeses made in ways that highlight admirable agriculture offer the lure of what chef Dan Barber calls “merchants of pleasure, not armies of virtue.” Eat these cheeses because they give you joy, and the wider questions look after themselves.

251

With that in mind, how should you buy cheese? Rather than a prescriptive list of producers, we present some simple guidelines: Buy where lots of people buy cheese.

The faster that a shop moves through its stock, the better the condition of its cheese will be. The moment that a cheese is cut, it begins to deteriorate: its surfaces start to dry and oxidize, and its flavor becomes less distinct. Even a great cheese will taste of little more than plastic and refrigerator when it has sat open on a shop counter for more than a week. You are far more likely to find great cheese at a shop that rockets through a large amount of a few well-chosen cheeses than one that chips away slowly at a selection of hundreds. Buy cheese made by farmers.

The pinnacle of what a cheese can achieve is dictated by the quality of the milk used. Cheesefarmers start their cheesemaking decisions in their pastures and their breeding programs, and they see the consequences of their actions directly in the vat. No milk contract, however specialized, can ever compete with this fine-grained control. The efforts of these individuals deserve our recognition. Buy unadulterated cheese.

Good milk is packed with inherent interest. If a cheesemaker hides behind added ingredients, whether smoke, added fruits or spices, or strong adjunct cultures, it is either a tragedy that they are masking the inherent potential of their milk or a sign that the milk was devoid of character in the first place. Buy raw-milk cheese.

Mature raw-milk cheeses are more complex and intense in flavor than their pasteurized counterparts. What’s more, heat treatments are a moral hazard if we care about the practice of agriculture: sloppy animal husbandry is readily concealed when the microbial slate is wiped clean by a kill step. Raw-milk cheese demands and rewards impeccable practice at every step. Buy complex cheese.

If ecology is the universal cheesemaking metaphor, biodiversity is the mark of excellence. And it is something that can be seen and tasted. The quickest way to evaluate a cheese counter is to look at the rinds: firstly, are there any rinds, or have the cheeses 2 52

•

APPENDIX

been matured under plastic or wax? Where rinds are present, are they monocultures, or do they wear their microbial diversity on their sleeves? The same applies when tasting: interesting cheeses embrace complexity. Buy from a cheesemonger.

Good cheesemongers are like museum curators. They will be eager to discuss their cheeses and offer you a taste. Cultivate this relationship, and ask them awkward, challenging questions. One of the great advantages of buying from cheesemongers is their ability to relay information back to the producer and be the conduit for an ongoing conversation. This is selective ecology in practice.

APPENDIX

•

253

This page intentionally left blank

GLOSSARY

For microbiological terms, refer to the sections on Dramatis Microbia in chapter 6, pages 100–105. affinage The practice of aging cheese, often carried out by independent cheese maturers, also known as affineurs. Also called maturation. artisan A term often used to describe specialty cheese. However, as it has no legal definition, it can be applied to cheese made on any scale. bacteria Microscopic organisms that are omnipresent in the environment and play a critical role in the cheesemaking and ripening process (not to be confused with yeasts or fungi). bloomy rind A category of soft cheeses whose rinds are formed by the growth of mold. While selected pure-white strains often dominate the rinds within this style, wild-type blue or green molds also exist. bulk A general description for milk that has been combined from multiple sources, whether from all of the individual animals of a herd (in a “bulk tank” on a farm) or from many different herds (whose milk is “bulked” en route to a processing plant). clothbound cheeses Cheeses that are wrapped in cotton or muslin at an early stage of their development to support the formation of a healthy rind and to help prevent cracking or slouching. A paste of flour and water or animal fat such as lard or butter is used to seal the cloth to the cheese. commodities Primary products that are functionally equivalent to others of their type. concentr ates High-energy supplemental feeds given to many dairy animals to satisfy their energy requirements for milk production. They can include various mixtures of grains, such as wheat, barley, and oats, and leguminous peas, beans, and soya. They may also contain agro-industrial by-products such as nut hulls, cotton seeds, and citrus peels, as well as vitamins and minerals.

2 55

curd A milk gel created by the action of rennet, acidity, or a combination of the two. Forming a curd is a critical and universal step in cheesemaking, and the characteristics of the curd influence the progression of the later steps in the process. dual- (or triple-) purpose breeds Breeds adapted for diverse agricultural outputs, often a combination of milk, meat, or draft animals. Multipurpose breeds seldom perform as well in any single area as their more specialized counterparts, but their health and longevity, not to mention the economic value of their outputs, can stack up well against those of high-performance, single-purpose breeds in certain systems. direct-vat set cultures Direct-Vat Set (DVS) cultures, also known as Direct-Vat Inoculation (DVI) cultures, are highly concentrated freeze-dried mixtures of acidifying or ripening cultures that can be added directly to fresh milk at the start of the cheesemaking process. extensive farming A system that relies on relatively small inputs of fertilizer and labor to support a small number of animals on a given amount of land. Yields within these low-input systems are commensurately smaller than those of intensive systems. factor An archaic term for an agent who selected and bought cheeses from farmers and sold them on to merchants (e.g., Josiah Twamley was an eighteenth-century cheese factor). farmhouse cheese In the strictest sense, this is vertically integrated cheese; the milk and cheese are produced by the same business, and the cheese is made onsite. In French, this type of cheese is designated fermier. for age Plants eaten by ruminants, including grass, hay, and various types of silage. Most dairy animals’ diets consist of a combination of forage and concentrates in varying proportions, depending on the type of operation (intensive or extensive). fungi Organisms in the kingdom Fungi, which includes both yeasts and molds. While bacteria play the main role in cheesemaking, fungi are also important in the cheese-ripening process. ger le A type of wooden vat used in the production of Salers cheese in the Auvergne. Their porous surfaces are home to microbial communities that function as natural starter cultures. hay Dried grass that is often used as fodder. heifer A young cow that has not yet had a calf. inoculation The action of adding selected strains of bacteria or fungi to milk to drive the cheesemaking or ripening process. The rinds of fresh cheeses can also be inoculated directly, either through washing or spraying on cultures. intensive farming A system characterized by high inputs and high yields. Intensification of dairy farming during the late nineteenth and twentieth

256

•

GLOSSARY

centuries involved the adoption of specialized high-yielding breeds, the development of inorganic nitrogen fertilizers, a shift from hay to silage, and the use of concentrates in feed. lactic cheese A type of cheese for which the curd is coagulated primarily by the action of acid rather than rennet. These cheeses have a fairly friable texture and a sharp acid taste. make A cheesemaker’s term for the process used for making a given type of cheese (e.g., “The Cheddar make has become hotter and faster over the past hundred years”). marginal land Land that is unsuitable for growing crops (for whatever reason; e.g., it is too steep, too dry, too wet, or has poor soil) but that can be put to productive use by grazing animals. Animals subsisting on unimproved marginal land often eat a more diverse diet, making for more interesting milk and cheese. microbe (a.k.a. microorganism) Any microscopic life form. Microbes implicated in cheesemaking include bacteria, yeasts, and molds. molds Filamentous fungi that grow as a branching network (as opposed to singlecelled yeasts). Molds degrade organic matter, playing an important role in flavor and texture development during cheese ripening. natur al rind A rind that grows on a cheese that is ripened in the presence of oxygen, which allows yeasts, molds, and/or bacteria to grow on the surface. Cheeses can be produced without rinds by being dipped in wax or sealed in plastic pouches; this prevents both rind growth and moisture loss and creates a different set of flavors. pasteurization The process of heating milk to 161°F (72°C) for fi fteen seconds—or an equivalent time-temperature combination—which is required to destroy the most heat-resistant human pathogen potentially found in milk, Coxiella burnetii. While pasteurization is imperative where milk production is not under strict control, it also destroys a wide variety of microorganisms that play an important role in the cheesemaking process, necessitating their replacement by starter cultures. pathogen Any organism capable of causing disease. The line between pathogen and benign microbe is not always clear; some bacteria or fungi are normally harmless but can become pathogens if presented with the right opportunity. r aw milk Milk that has not been subjected to heat treatment. The use of raw milk for cheese demands extra control at the veterinary and milking stages to avoid the introduction of pathogens or spoilage microbes. rennet A cocktail of two enzymes (chymosin and pepsin) that has the capacity to turn liquid milk into a solid curd. Commercial rennet can take several forms. “Natural rennet,” or “animal rennet,” is extracted from calves’, kids’, or lambs’ stomachs on an industrial scale and is considered by many to be the premium option; it is also the most expensive. Vegetarian alternatives include “fermenta-

GLOSSARY

•

2 57

tion-produced chymosin,” which is synthesized by genetically modified organisms into which the gene for the enzyme chymosin has been cloned. A third type is “microbial coagulant,” which uses enzymes naturally synthesized by microbes that are similar enough to chymosin to clip the milk proteins in approximately the same way. However, their slightly different behavior can lead to bitterness during ripening. A major benefit of all commercial coagulants (as opposed to homemade extracts) is that they are diluted to a standard strength. ripening cultures Selected strains of bacteria or fungi that are added to milk or applied to cheese to impact its development during the maturation process. Many progressive cheesemakers believe that these are not necessary, and that given the right conditions, the naturally occurring microorganisms in the milk and the environment will perform the task even better. rumen The first chamber in a ruminant’s digestive tract, where the breakdown and microbial digestion of cellulose primarily occurs. If an animal is fed starchy feeds, these are also fermented within the rumen, but into different by-products, changing the rumen environment. silage Fermented feed for ruminants, which has a higher moisture content and acidity than hay. It is commonly made from grass and from some cereal crops, such as corn and barley. Silage has the potential to harbor certain pathogens and spoilage microorganisms, adding an extra degree of risk to the production of certain types of cheese. starter cultures Selected strains or communities of lactic acid bacteria that are added at the beginning of the cheesemaking process to acidify the milk. Before starters were developed, cheesemakers used techniques such as adding whey, using wooden utensils that inoculated the milk with acidifying bacteria, and incubating milk at warm temperatures to encourage the growth of sufficient levels of lactic acid bacteria to make cheese. territorial cheese Cheese linked with a particular place. Many British territorial cheeses have been defi ned by their counties of origin (Lancashire, Cheshire, and the like), but it would be equally appropriate to say that Salers from the Auvergne and Reblochon de Savoie are territorial cheeses. Many of these territorial associations form the basis of protected food name schemes, particularly in the European Union. terroir A French term that refers to the soil and climate of a particular place and the marks these local conditions leave on the foods that are grown and produced there. Many cheeses—as well as other foods—invoke the notion of terroir without engaging in a more nuanced discussion about its mechanisms. true cost accounting An economic principle designed to account for the external costs (e.g., social, public health, and environmental) that are not factored into the price of food. unpasteurized A term that may refer either to raw milk or to milk that has been subjected to a heat treatment less intense than pasteurization. Therefore,

258

•

GLOSSARY

raw-milk cheese is always unpasteurized, but not all unpasteurized cheeses are made from raw milk. vat A container in which milk is set into curd. washed rind Cheeses that are rubbed with a brine or water solution during the ripening process to discourage mold and encourage the growth of pigmented bacteria. It is a broad category that includes everything from lactic to hard cheeses. whey The milk serum that drains from the curd during the cheesemaking process. It contains soluble milk proteins, lactose, small amounts of fat, vitamins, and minerals. yeast Single-celled microorganisms that belong to the fungal kingdom and often play an important part in the early stages of cheese ripening.

GLOSSARY

•

2 59

This page intentionally left blank

NOTES

P R O LO G U E

1. R. Didienne, C. Defargues, C. Callon, T. Meylheuc, S. Hulin, and M. C. Montel, “Characteristics of Microbial Biofi lm on Wooden Vats (‘Gerles’) in PDO Salers Cheese,” International Journal of Food Microbiology 156 (2012): 91–101. 2. “Innovation Is the Answer, Say Acclaimed Dairy Producers,” Dupont Danisco, February 14, 2017, www.danisco.com/about-dupont/news/news-archive /2014/innovation-is-the-answer-say-acclaimed-dairy-producers.

1 . E CO LO G I E S

1. J. MacDonald, E. O’Donoghue, W. McBride, R. Nehring, C. Sandretto, and R. Mosheim, Profits, Costs, and the Changing Structure of Dairy Farming, USDAERS Economic Research Report no. 47 (September 1, 2007): 2–4. 2. C. Baker, Dairy Industry in the UK: Statistics, House of Commons Library, Social and General Statistics Section, Standard Note SN/SG/2721, last updated January 29, 2015, http://spotidoc.com/doc/732840/dairy-industry-uk-statistics. 3. R. Lovreglio, O. Meddour-Sahar, and V. Leone, “Goat Grazing as a Wildfire Prevention Tool: A Basic Review,” iForest 7 (2014): 260–268. 4. P. Smith, “For Gastronomists, a Go-To Microbiologist,” New York Times, September 19, 2012, D5. 5. Y. Uchida, H. Morita, S. Adachi, T. Asano, T. Taga, and N. Kondo, “Bacterial Meningitis and Septicemia of Neonate Due to Lactococcus lactis,” Pediatrics International 53, no. 1 (2011): 119–120; H. Kassamali, E. Anaissie, J. Ro, K. Rolston, H. Kantarjian, V. Fainstein, and G. Bodey, “Disseminated Geotrichum candidum Infection.” Journal of Clinical Microbiology 25, no. 9 (1987): 1782–1783. 6. J. van Rensburg et al., “The Human Skin Microbiome Associates with the Outcome of and Is Influenced by Bacterial Infection,” mBio 6, no. 5 (2015): e01315-15. 261

7. R. Dutton and P. Turnbaugh, “Taking a Metagenomic View of Human Nutrition,” Current Opinion in Clinical Nutrition and Metabolic Care 15 (2012): 448–454. 8. L. David et al., “Diet Rapidly and Reproducibly Alters the Human Gut Microbiome,” Nature 505 (2014): 559–563. 9. M. Blaser, Missing Microbes: How Killing Bacteria Creates Modern Plagues (London: Oneworld, 2014), 116. 10. P. VanRaden, “Impact of Genomics on Genetic Improvement” (presentation, Advancing Dairy Cattle Genetics: Genomics and Beyond, Phoenix, AZ, February 19, 2014). 11. J. Lundgren and S. Fausti, “As Biodiversity Declines on Corn Farms, Pest Problems Grow,” The Conversation, July 31, 2015, www.theconversation.com /as-biodiversity-declines-on-corn-farms-pest-problems-grow-45477.

2. REAL CHEESE

1. R. Bloomfield, The Farmer’s Boy; A Rural Poem (London: Vernor and Hood, 1800), 17–18. 2. J. Twamley, Dairying Exemplified, or The Business of Cheese-Making, 2nd ed. (Warwick: J. Sharp, 1787), 118. 3. Ibid., 73. 4. X. Willard, Address before the Cheese Makers Association (Albany: Van Benthuysen’s Steam Printing House, 1865), 3. Estimated 2015 value calculated using the indices at www.measuringworth.com. 5. P. Rance, The Great British Cheese Book (London: Macmillan, 1982), 135.

3. THE THIRD RAIL

1. A. Wilson, Forgotten Harvest: The Story of Cheesemaking in Wiltshire (Broughton Gifford: Cromwell Press, 1995), 179. 2. G. Edgar, Cheddar: A Journey to the Heart of America’s Most Iconic Cheese (White River Junction: Chelsea Green), 99.

4. BREED

1. R. Splan and D. Sponenberg, “Characterization and Conservation of the American Milking Devon,” Animal Genetic Resources Information 34 (2003): 11–16. 2. W. Scott, The Poetical Works of Walter Scott Esq., vol. 2 (New York: James Eastburn, 1818), 137.

2 62

•

NOTES TO PAGES 15 – 47

3. H. Ritvo, “Race, Breed, and Myths of Origin: Chillingham Cattle as Ancient Britons,” Representations 39 (1992): 1–22. 4. P. Visscher, D. Smith, S. Hall, and J. Williams, “A Viable Herd of Genetically Uniform Cattle,” Nature 409 (2001): 303. 5. Ibid. 6. Cattle: Their Breeds, Management, and Diseases; With an Index (London: Baldwin and Cradock, 1834), 242. 7. H. H. Dixon, “Rise and Progress of Shorthorns: Prize Essay,” Journal of the Royal Agricultural Society of England, 2nd ser., 1 (1865): 324. 8. J. Walton, “The Diff usion of the Improved Shorthorn Breed of Cattle in Britain during the Eighteenth and Nineteenth Centuries,” Transactions of the Institute of British Geographers 9, no.1 (1984): 22–36. 9. M. Derry, Bred for Perfection: Shorthorn Cattle, Collies, and Arabian Horses since 1800 (Baltimore: Johns Hopkins University Press, 2003), 22. 10. Cattle: Their Breeds, Management, and Diseases, 243. 11. For a description of the characteristics of the now-extinct local breeds of Somerset, see Cattle: Their Breeds, Management, and Diseases, 28. 12. Cattle: Their Breeds, Management, and Diseases, 234. 13. M. Quinn, “Corpulent Cattle and Milk Machines: Nature, Art and the Ideal Type,” Society and Animals 1, no. 2 (1992): 150. 14. Walton, “Diff usion,” 25. 15. Price adjusted by multiplying $40,600 by the percentage increase in the consumer price index between 1873 and 2015. 16. Derry, Bred for Perfection, 26. 17. J. Walton, “Pedigree and the National Cattle Herd circa 1750–1950,” Agricultural History Review 34, no. 2 (1986): 165. 18. D. Valenze, Milk: A Local and Global History (New Haven: Yale University Press, 2011), 89. 19. M. Derry, Masterminding Nature: The Breeding of Animals 1750–2010 (Toronto: University of Toronto Press, 2015), 116. 20. F. L. Houghton, Holstein-Friesian Cattle: A History of the Breed and Its Development in America (Brattleboro: Press of the Holstein Friesian Register, 1897), 17. 21. B. Theunissen, “Breeding without Mendelism: Theory and Practice of Dairy Cattle Breeding in the Netherlands 1900–1950,” Journal of the History of Biology 41, no. 4 (2008): 637–676. 22. B. Theunissen, “Breeding for Nobility or for Production? Cultures of Dairy Cattle Breeding in the Netherlands, 1945–1995,” Isis 103, no. 2 (2012): 278–309. 23. J. L. Lush, Animal Breeding Plans (Ames, IA: Collegiate Press, 1937). 24. Derry, Masterminding Nature, 119. 25. “Do You Know This about Holstein Cattle?” Holstein Association USA, accessed February 19, 2017, www.holsteinusa.com/pdf/fact_sheet_cattle.pdf. 26. B. Heins, L. Hansen, and A. Seykora, “Production of Pure Holsteins versus Crossbreds of Holstein with Normande, Montbéliarde, and Scandinavian Red,” Journal of Dairy Science 89 (2006): 2799–2804.

NOTES TO PAGES 47–53

•

263

27. P. Sainsbury, Tradition to Technology: A History of the Dairy Industry in Devon (Tiverton: P. T. Sainsbury, 1991), 31. 28. “U.S. Registered Holsteins for Maximum Profit,” Holstein Association USA, accessed February 19, 2017, www.holsteinusa.com/pdf/print_material /USReg_Holsteins.pdf. 29. M. De Marchi, G. Bittante, R. Dal Zotto, C. Dalvit, and M. Cassandro, “Effect of Holstein Friesian and Brown Swiss Breeds on Quality of Milk and Cheese,” Journal of Dairy Science 91 (2008): 4092–4102. 30. P. Fricke, “14,000 Kg and Beyond: Current Benchmarks and Future Challenges for Dairy Cattle Reproduction,” Advances in Dairy Technology 16 (2004): 15. 31. At least in some cultures. Other premodern farmers rationalized that it was best to eat their healthiest animals and bred the next generation from the remaining stock that was too scrawny to be tempting for the pot. 32. R. H. Foote, “The History of Artificial Insemination: Selected Notes and Notables,” Journal of Animal Science 80 (2002): 1–10. 33. Ibid. 34. P. VanRaden, “Invited Review: Selection on Net Merit to Improve Lifetime Profit,” Journal of Dairy Science 87 (2004): 3125–3131. 35. “Montbéliarde: Bred for the French Cheese Industry,” Auzred Xb, accessed May 14, 2016, www.auzredxb.com.au/media/docs/MB_CheesemakersChoice .pdf. 36. Dr. Alison Martin, the executive director of the Livestock Conservancy, notes that this type of selective breeding is no less problematic from a biodiversity perspective than is the laser-like focus of the Holstein breeders on maximizing yields: “Kappa-casein B (and [a variant of milk protein called] A2 beta-casein) represent opportunities to use genomics. The problem we see in rare breeds is, while the tool itself is neutral, rapid screening for today’s market whim makes it possible to rapidly discard tremendous fractions of an already limited genetic pool.” 37. “Montbéliarde: Bred for the French Cheese Industry.” 38. The same dizzying statistics apply to the impact of individual animals. Born in 1962, the bull Pawnee Farm Arlinda Chief is the source of some 14 percent of the genes in the entire contemporary US dairy herd. He was also a carrier of a recessive genetic defect that causes midterm miscarriages in some of his offspring. In characteristic Holstein style, the cumulative economic impact of the genetic defects inherited from this single bull was calculated at $400 million. The impact of the enhanced milk yield from Pawnee Farm Arlinda Chief ’s superlative genes for milk production? $25 billion. In the “more milk equals more money” world of the Holstein, the genetic defect was absolutely worth it. P. Van Raden, “Impact of Genomics on Genetic Improvement” (presentation delivered at Advancing Dairy Cattle Genetics: Genomics and Beyond, Phoenix, AZ, February 19, 2014), slide 20. 39. X. P. Yue, C. Dechow, and W. S. Liu, “A Limited Number of Y Chromosome Lineages Is Present in North American Holsteins,” Journal of Dairy Science 98 (2015): 2738–2745. Semen from some of the extinct lineages has been preserved as part of the USDA’s National Animal Germplasm Program, which aims to preserve

264

•

NOTES TO PAGES 53– 61

and reintroduce genetic diversity back into the population and serve as an insurance policy against unknown and potentially detrimental effects of loss of genetic variability within populations. See C. Dechow, “Holstein Lineages Trace Back to Two Bulls,” Hoard’s Dairyman, March 25, 2015, 219. 40. L. Hansen, “Random Use of Holstein Semen on Holstein Females Is Becoming Risky,” University of Minnesota Extension, Dairy Extension, July 2015, www.extension.umn.edu/agriculture/dairy/reproduction-and-genetics/holsteinsemen-on-female-becoming-risky/index.html. 41. Ibid. 42. N. Zwald, “What Progressive Dairy Producers Need to Know about Inbreeding,” Alta, Dairy Basics, August 3, 2012, http://web.altagenetics.com/france /DairyBasics/Details/1060_What-Progressive-Dairy-Producers-Need-to-Knowabout-Inbreeding.html. 43. G. Pollott, A. Charlesworth, and D. Wathes, “Possibilities to Improve the Genetic Evaluation of a Rare Breed Using Limited Genomic Information and Multivariate BLUP,” Animal 8, no. 5 (2014): 685–694. 44. B. Cassell, “Dairy Crossbreeding: Why and How,” Virginia Cooperative Extension, publication 404-093, May 1, 2009, p. 1, https://www.pubs.ext.vt.edu /content/dam/pubs_ext_vt_edu/404/404-093/404-093_pdf.pdf. 45. W. Montgomerie, “Experiences with Dairy Cattle Crossbreeding in New Zealand” (paper prepared for the 53rd Annual Meeting of the European Association for Animal Production, Cairo, Egypt, September 1–4, 2002), 3. 46. P. VanRaden and A. Sanders, “Economic Merit of Crossbred and Purebred US Dairy Cattle,” Journal of Dairy Science, 86 (2003): 1036–1044. 47. Cattle: Their Breeds, Management, and Diseases, 22.

5. FEED

1. R. Dion, Histoire de la vigne et du vin en France des origines au XIXe siècle (Paris: CNRS Editions, 2011); A. Trubek, The Taste of Place: A Cultural Journey into Terroir (Berkeley: University of California Press, 2008), 18–53. 2. R. T. Conant, Challenges and Opportunities for Carbon Sequestration in Grassland Systems: A Technical Report on Grassland Management and Climate Change Mitigation, Plant Production and Protection Division, Food and Agriculture Organization of the United Nations, Integrated Crop Management, vol. 9 (Rome: Food and Agriculture Organization of the United Nations, 2010), 3. 3. “Policy Options to Support Transhumance and Biodiversity in European Meadows,” Mountain Research and Development 25, no. 1 (2005): 82. 4. N. Thompson, I. Barrie, and L. Harvey, “The Climatic Potential for Field Hay Drying in North-West Europe,” Journal of Agricultural Science 105 (1985): 167–181. 5. J. Twamley, Dairying Exemplified, or the Business of Cheese-Making: Laid Down from Approved Rules, Collected from the Most Experienced Dairy-Women, of Several Counties, 2nd ed., corrected & improved (Warwick: J. Sharp, 1787), 56. N OTE S TO PAG E S 62–70

•

2 65

6. J. Stadhouders, “The Manufacturing Method for Cheese and the Sensitivity to Butyric Acid Fermentation,” Bulletin of the IDF 251 (1990): 37–39; Stadhouders, “Prevention of Butyric Acid Fermentation by the Use of Nitrate,” Bulletin of the IDF 251 (1990): 40–45; R. Lodi, “The Use of Lysozyme to Control Butyric Acid Fermentation,” Bulletin of the IDF 251 (1990): 51–53. 7. B. Orland, “Alpine Milk: Dairy Farming as a Pre-Modern Strategy of Land Use,” Environment and History 10 (2004): 327–364. 8. P. McDonald, R. Edwards, C. Morgan, and J. Greenhalgh, Animal Nutrition (Noida: Pearson, 2002) 265; A. Hattan, e-mail message to authors, November 11, 2016. 9. A. Chamberlain and J. Wilkinson, Feeding the Dairy Cow (Lincoln: Chalcombe Publications, 1998), calculations based on table 18.12, p. 173. 10. M. Galyean and J. Rivera, “Nutritionally Related Disorders Affecting Feedlot Cattle,” Canadian Journal of Animal Science 83, no. 1 (2003): 13–20. 11. Chamberlain and Wilkinson, Feeding the Dairy Cow, 138; A. Hattan, “Reviving the Northern Dairy Shorthorn: A Model for Using Low-Yielding Breeds within Extensive Dairy Systems” (presentation, Science of Artisan Cheese Conference, Somerset, UK, August 23, 2016). 12. With its high metabolic toll, a high-concentrate approach to rearing highyielding dairy cattle has its downsides. But some simple calculations show why the dairy industry chose this path. The energy requirements for body maintenance (before milk production) of a cow is based on her weight. Take a hypothetical cow that weighs just over 550 kilograms (around 1,200 pounds). Her energy requirement for basic bovine activities—such as breathing and maintaining body temperature— is about sixty megajoules per day (just over 14,000 calories). If she were a Salers producing ten liters of milk per day, she would need about 12,000 more calories to supply that energy. However, if she pumped out thirty-four liters of milk a day (like an average Holstein cow, according to Holstein USA), she would require over 40,000 calories above her base requirements to stoke the furnace. But although the Holstein would eat more, all of that extra energy would go directly into making milk. If we do the math, the Holstein consumes around 1,600 calories for every liter of milk she produces, while the Salers requires 2,600 calories per liter, or almost 40 percent more energy. Higher-yielding cows are simply more efficient: given the same feed price, farmers can feed fewer bodies and convert more of that feed into milk. (Calculations based on table 18.12 in Chamberlain and Wilkinson, Feeding the Dairy Cow, 173). 13. Chamberlain and Wilkinson, Feeding the Dairy Cow, 140. 14. Orland, “Alpine Milk,” 338. 15. J. Milner, “On Hay-Making in General, and Particularly in Wet Weather,” Belfast Monthly Magazine 7, no. 37 (1811): 140. 16. W. Baron and A. Bridges, “Making Hay in Northern New England: Maine as a Case Study, 1800–1850,” Agricultural History 57, no. 2 (1983): 171. 17. D. Gamble and T. St. Pierre, eds., Hay Time in the Yorkshire Dales: The Natural, Cultural, and Land Management History of Hay Meadows (Lancaster: Scotforth Books, 2010), 33; Baron and Bridges, “Making Hay,” 173.

266

•

N OTE S TO PAG E S 7 1 –76

18. Gamble and St. Pierre, Hay Time, 152. 19. V. Smil, “Detonator of the Population Explosion,” Nature 400 (1999): 415. 20. V. Smil, “Nitrogen Cycle and World Food Production,” World Agriculture 2, no. 1 (2011): 9–13. 21. Smil, “Detonator”; Smil, Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation of World Food Production (Cambridge: Massachusetts Institute of Technology Press, 2001). 22. C. Chesterton, Environmental Impacts of Land Management, Natural England Research Report 30 (Sheffield: Natural England, 2009), 65. 23. N. Fish, “The History of the Silo in Wisconsin,” Wisconsin Magazine of History 8, no. 2 (1924): 166. 24. P. Brassley, “Silage in Britain, 1880–1990: The Delayed Adoption of an Innovation,” Agricultural History Review 44, no. 1 (1996): 74. 25. Gamble and St. Pierre, Hay Time, 122. 26. Brassley, “Silage,” 77, 87. 27. J. Murdoch, Making and Feeding Silage (Ipswich: Farming Press, 1961), 22–23. 28. A. Farruggia, D. Pomiès, M. Coppa, A. Ferlay, I. Verdier-Metz, A. Le Morvan, A. Bethier, F. Pompanon, O. Troquier, and B. Martin, “Animal Performances, Pasture Biodiversity and Dairy Product Quality: How It Works in Contrasted Mountain Grazing Systems,” Agriculture, Ecosystems and Environment 185 (2014): 235. 29. Gamble and St. Pierre, Hay Time, 125. 30. Ibid., 122. 31. Ibid., 128; M. Kamm, State of the Birds: Massachusetts Breeding Birds—A Closer Look (Mass Audubon Bird Conservation Programs, 2013), 22. 32. Brassley, “Silage,” 80. 33. B. Martin, I. Verdier-Metz, S. Buchin, C. Hurtaud, and J. Coulon, “How Do the Nature of Forages and Pasture Diversity Influence the Sensory Quality of Dairy Livestock Products?” Animal Science 81 (2005): 205–212. 34. P. Kindstedt, American Farmstead Cheese (White River Junction: Chelsea Green, 2005), 44. 35. M. Coppa, I. Verdier-Metz, A. Ferlay, and B. Martin, “Effect of Different Grazing Systems on Upland Pastures Compared with Hay Diet on Cheese Sensory Properties Evaluated at Different Ripening Times,” International Dairy Journal 21 (2011): 815–822. 36. Martin et al., “Nature of Forages and Pasture Diversity,” 210. 37. A. Cornu, A. Farruggia, E. Leppik, C. Pinier, F. Fournier, D. Genoud, B. Frérot, “Trapping the Pasture Odorscape Using Open-Air Solid-Phase Micro Extraction, a Tool to Assess Grassland Value,” PLOS ONE 10, no. 11 (2015): e0140600, doi:10.1371/journal.pone.0140600. 38. J. Ruechel, Grass-Fed Cattle (North Adams: Storey, 2006), 4. 39. D. Pallett, “Changes in Plant Species Richness and Productivity in Response to Decreased Nitrogen Inputs in Grassland in Southern England,” Ecological Indicators 68 (2016): 73–81.

NOTES TO PAGES 77–88

•

2 67

40. Gamble and St. Pierre, Hay Time, 176–177. 41. Ibid., 160. 42. Although the Holdens do not practice transhumance, the name of their cheese is an old Welsh word for “summer dwelling,” which harks back to the days when British cows were moved to high-altitude pastures for summer grazing, just as in much of continental Europe. 43. G. Monbiot, Feral: Rewilding the Land, Sea and Human Life (London: Penguin, 2013), 153–166. 44. Martin, “Nature of Forages and Pasture Diversity,” 205. 45. H. Buller and C. Morris, “Eating Biodiversity: An Investigation of the Links between Quality Food Production and Biodiversity Protection: Full Research Report,” ESRC End of Award Report, RES-224–25–0041 (Swindon: ESRC, 2007), 32.

6. MICROBES

1. L. Clavreul, “L’avenir du Camembert au lait cru,” Le Monde, March 14, 2007. 2. “Fromages AOP: Les Camemberts de Normandie attaquent les industriels en justice,” Lafranceagricole.fr, November 18, 2011, www.lafranceagricole.fr/actualites /fromages-aop-les-camemberts-de-normandie-attaquent-les-industriels-enjustice-1,0,84185870.html; A. Conté, “Retour à la Case Départ pour l’AOP Camembert de Normandie?” Réussir Lait, June 25, 2015, http://lait.reussir.fr/actualites /retour-a-la-case-depart-pour-l-aop-camembert-de-normandie:I091KYK9.html. 3. M. Steinberger, Au Revoir to All That: Food, Wine, and the End of France (New York: Bloomsbury, 2009), 122–123. 4. J. Richard, “Nature de la flore microbienne dominante et sous-dominante des laits crus très pollués,” Le Lait 63 (1983): 168. 5. R. Mills, “BactoScan™ FC: Beating Bacteria in Milk,” In Focus 30, no. 2 (2006): 8–9. 6. Ibid., 8. 7. D. D’Amico, “Microbiological Quality and Safety Issues in Cheesemaking,” in Cheese and Microbes, ed. Catherine Donnelly (Washington, DC: ASM Press, 2014), 260. 8. D. Corroler, I. Mangin, N. Desmasures, and M. Guéguen, “An Ecological Study of Lactococci Isolated from Raw Milk in the Camembert Cheese Registered Designation of Origin Area,” Applied and Environmental Microbiology 64, no. 12 (1998): 4729–4735. 9. N. Desmasures, F. Bazin, and M. Guéguen, “Microbiological Composition of Raw Milk from Selected Farms in the Camembert Region of Normandy,” Journal of Applied Microbiology 83 (1997): 53–58. 10. Y. Bouton, “Faut-il des laits de plus en plus propres? Incidences du niveau de flore d’un lait de fabrication sur la qualité d’un fromage à pâte pressée cuite,” Les Nouvelles du Comté, no. 30 (2000): 1–4.

268

•

NOTES TO PAGES 88–99

11. W. Hammes and C. Hertel, “The Genera Lactobacillus and Carnobacterium,” in Prokaryotes, ed. Martin Dworkin (New York: Springer-Verlag, 2006), 4:335. 12. Ibid., 369. 13. M. Teuber and A. Geis, “The Genus Lactococcus,” in Prokaryotes, ed. Martin Dworkin (New York: Springer-Verlag, 2006), 4:205–228. 14. J. Alomar, P. Loubière, C. Delbès, S. Nouaille, and M. C. Montel, “Effect of Lactococcus garvieae, Lactococcus lactis and Enterococcus faecalis on the Behaviour of Staphylococcus aureus in Microfi ltered Milk,” Food Microbiology 25, no. 3 (2008): 502–508. 15. G. B. Mahajan, “Antibacterial Agents from Actinomycetes—A Review,” Frontiers in Bioscience 4 (2012): 240–253. 16. The legal standard for pasteurization temperature was raised slightly to target another pathogen, Coxiella burnetii, in the late 1950s, but many Actinobacteria are heat-insensitive enough to survive pasteurization. See O. Cerf and R. Condron, “Coxiella burnetii and Milk Pasteurization: An Early Application of the Precautionary Principle?” Epidemiology & Infection 134 (2006): 946–951. 17. D’Amico, “Microbiological Quality,” 256. 18. B. Wolfe, J. Button, M. Santarelli, and R. Dutton, “Cheese Rind Communities Provide Tractable Systems for In Situ and In Vitro Studies of Microbial Diversity,” Cell 158 (2014): 422–433. 19. M. C. Montel, S. Buchin, A. Mallet, C. Delbès-Paus, D. Vuitton, N. Desmasures, and F. Berthier, “Traditional Cheeses: Rich and Diverse Microbiota with Associated Benefits,” International Journal of Food Microbiology 2, no. 177 (2014): 136–154. 20. P. Deetae, J. Mounier, P. Bonnarme, H. Spinnler, F. Irlinger, and S. Helinck, “Effects of Proteus vulgaris Growth on the Establishment of a Cheese Microbial Community and on the Production of Volatile Aroma Compounds in a Model Cheese,” Journal of Applied Microbiology 107 (2009): 1404–1413. 21. C. Callon, C. Arliguie, and M. C. Montel, “Control of Shigatoxin-Producing Escherichia coli in Cheese by Dairy Bacterial Strains,” Journal of Food Microbiology 53, part B (2016): 63–70. 22. M. Gray, N. Freitag, and K. Boor, “How the Bacterial Pathogen Listeria monocytogenes Mediates the Switch from Environmental Dr. Jekyll to Pathogenic Mr. Hyde,” Infection and Immunity 74, no. 5 (2006): 2505–2512. 23. H. Hilbi, S. Weber, C. Ragaz, Y. Nyfeler, and S. Urwyler, “Environmental Predators as Models for Bacterial Pathogenesis,” Environmental Microbiology 9, no. 3 (2007): 563–575. 24. F. Monsallier, I. Verdier-Metz, C. Agabriel, B. Martin, and M. C. Montel, “Variability of Microbial Teat Skin Flora in Relation to Farming Practices and Individual Dairy Cow Characteristics,” Dairy Science and Technology 92, no. 3 (2012): 275. 25. V. Michel, A. Hauway, and J. Chamba, “La flore microbienne de laits crus de vache: Diversité et influence des conditions de production,” Le Lait 81 (2001): 590.

NOTES TO PAGES 100 –108

•

2 69

26. M. C. Montel, “Effect of Post Dipping Treatment after Milking on Teat and Milk Microbiota” (presentation at the Conference on the Science of Artisan Cheese, Somerset, United Kingdom, August 19, 2014). 27. J. Sanders, G. Venema, and J. Kok, “Environmental Stress Responses in Lactococcus lactis,” FEMS Microbiology Reviews 23 (1999): 483–501. 28. V. Lafarge, J. C. Ogier, V. Girard, V. Maladen, J. Y. Leveau, A. Gruss, and A. Delacroix-Buchet, “Raw Cow Milk Bacterial Population Shifts Attributable to Refrigeration,” Applied and Environmental Microbiology 70, no. 9 (2004): 5644–5650. 29. P. Boisard, Camembert: A National Myth (Berkeley: University of California Press, 2003), 7–10. 30. Ibid., 39. 31. Ibid., 42–44. 32. J. Long and J. Benson, Cheese and Cheese-Making, Butter and Milk, with Special Reference to Continental Fancy Cheeses (London: Chapman and Hall, 1896), 35. 33. Boisard, Camembert, 72. 34. Ibid., 77.

7. R I S K

1. One of the most enlightened aspects of the French system is that cheeses (as well as meat and other animal foods) are regulated by the same authority that is responsible for overseeing animal health. 2. D. Valenze, Milk: A Local and Global History (New Haven: Yale University Press, 2011), 210. 3. P. Atkins, Liquid Materialities: A History of Milk, Science, and the Law (Farnham: Ashgate, 2010), 232. 4. R. Koch, “Die Ätiologie der Tuberkulose,” Berliner Klinische Wochenschrift 15 (1882): 221–230. 5. R. Nimmo, Milk, Modernity, and the Making of the Human: Purifying the Social (Abingdon: Routledge, 2010), 103. 6. Ibid., 111. 7. Ibid., 264. 8. Valenze, Milk, 213. 9. H. Kay, “The National Institute for Research in Dairying, Shinfield, Reading,” Proceedings of the Royal Society of London, Series B, Biological Sciences, 138, no. 890 (1951): 23. 10. A. Langer, T. Ayers, J. Grass, M. Lynch, F. Angulom, and B. Mahon, “Nonpasteurized Dairy Products, Disease Outbreaks, and State Laws—United States, 1993–2006,” Emerging Infectious Diseases 18, no. 3 (2012): 387. 11. A. Eck and J. C. Gillis Cheesemaking: From Science to Quality Assurance (Paris: Lavoisier, 2000), 696–697. European Food Safety Authority data show that the vast

270

•

NOTES TO PAGES 109 –124

majority of food-associated Campylobacter outbreaks are caused by poultry; the convention is to report illness due to milk and cheese separately, and the report for 2013 shows three outbreaks associated with milk but none with cheese. See European Food Safety Authority, European Centre for Disease Prevention and Control, “The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-Borne Outbreaks in 2013,” EFSA Journal 13, no. 1 (2015): 3991. Within the United States, however, several Campylobacter outbreaks have been associated with cheese. Yet the defining feature of the moral panic over the risks of these fresh cheeses is that they are produced outside of a formal quality system. Risk factors include production within the home, use of illegally acquired raw milk, and illegal importation. See L. Gould, “Outbreaks Attributed to Cheese: Differences between Outbreaks Caused by Unpasteurized and Pasteurized Dairy Products, United States, 1998–2011,” Foodborne Pathogens and Disease 11, no. 7 (2014): 545–551. 12. B. Silk, B. Mahon, P. Griffi n, L. Gould, R. Tauxe, S. Crim, K. Jackson, P. Gerner-Smidt, K. Herman, and O. Henao, “Vital Signs: Listeria Illnesses, Deaths, and Outbreaks—United States, 2009–2011,” Morbidity and Mortality Weekly Report 62, no. 22 (2013): 448–452; C. Farrokh et al., “Review of Shiga-ToxinProducing Escherichia coli (STEC) and Their Significance in Dairy Production,” International Journal of Food Microbiology 162 (2013): 190–212. 13. C. Little, J. Rhoades, S. Sagoo, M. Greenwood, V. Mithani, K. Grant, J. McLauchlin, and the Food, Water and Environmental Surveillance Network, “Microbiological Examination of Cheeses Made from Raw or Thermised Milk from Production Establishments and Retail Premises in the United Kingdom,” European Commission Co-ordinated Programme for the Official Control of Foodstuffs (2004); C. Little, J. Harris, S. Sagoo, M. Greenwood, V. Mithani, K. Grant, J. McLauchlin, and the Food, Water and Environmental Surveillance Network, “Microbiological Examination of Cheeses Made from Pasteurised Milk from Production Establishments and Retail Premises in the United Kingdom,” European Commission Coordinated Programme for the Official Control of Foodstuffs (2005). 14. “Food for Space Flight,” NASA, last modified April 7, 2002, www .spaceflight.nasa.gov/shuttle/reference/factsheets/food.html. 15. J. Ross-Nazzal, “ ‘From Farm to Fork’: How Space Food Standards Impacted the Food Industry,” in The Societal Impact of Spaceflight, ed. R. Lanius and S. Dick (Washington, DC: National Aeronautics and Space Administration Office of External Relations, 2007), 219–236. 16. “Regulation (EC) No 852/2004 of the European Parliament and of the Council of 29 April 2004 on the Hygiene of Foodstuffs,” Official Journal of the European Communities, April 30, 2004, http://eur-lex.europa.eu/legal-content /EN/TXT/?uri=CELEX%3A32004R0852. 17. C. Donnelly, “Achieving Control over Listeria during Artisan Cheese Production” (presentation, Science of Artisan Cheese Conference, Somerset, UK, August 28, 2012). 18. J. Higginson, R. Walters, and N. Fulop, “Mortality and Morbidity Meetings: An Untapped Resource for Improving the Governance of Patient Safety?” BMJ

NOTES TO PAGES 125 –129

•

27 1

Quality and Safety 21 (2012): 576–585, doi:10.1136/bmjqs-2011–000603. The Agency for Healthcare Research and Quality of the US Department of Health and Human Services runs a dedicated website, Patient Safety Network (www.psnet.ahrq.gov), which publishes reports of what went wrong and helps physicians improve the structures in which they operate. 19. C. Mariani, N. Oulahal, J. Chamba, F. Dubois-Brissonnet, E. Notz, and R. Briandet, “Inhibition of Listeria monocytogenes by Resident Biofi lms Present on Wooden Shelves Used for Cheese Ripening,” Food Control 22 (2011): 1357–1362. 20. The full text of the Food Safety Modernization Act can be found at www .fda.gov/Food/GuidanceRegulation/FSMA/ucm247548.htm#SEC103. 21. M. Batz, M. Doyle, J. Morris Jr., J. Painter, R. Singh, R. Tauxe, M. Taylor, and D. Lo Fo Wong, “Attributing Illness to Food,” Emerging Infectious Diseases 11, no. 7 (2005): 993–999. 22. E. Scallan, T. Jones, A. Cronquist, S. Thomas, P. Frenzen, D. Hoefer, C. Medus, and F. Angulo, “Factors Associated with Seeking Medical Care and Submitting a Stool Sample in Estimating the Burden of Foodborne Illness,” Foodborne Pathogens and Disease 3, no. 4 (2006): 432–438. 23. Ibid., 435. 24. “APHIS Factsheet, Veterinary Services: Questions and Answers; Bovine Tuberculosis,” United States Department of Agriculture, Animal and Plant Health Inspection Service, March 2014, www.aphis.usda.gov/publications/animal_health /content/printable_version/faq_bovine_tb_.pdf; Policy Paper: 2010 to 2015 Government Policy; Bovine Tuberculosis (Department for Environment, Food and Rural Affairs, updated May 8, 2015), www.gov.uk/government/publications/2010-to-2015government-policy-bovine-tuberculosis-bovine-tb/2010-to-2015-government-policy-bovine-tuberculosis-bovine-tb. 25. “Food Safety and Raw Milk,” Centers for Disease Control and Prevention, last modified March 10, 2015, www.cdc.gov/foodsafety/rawmilk/raw-milk-index.html. 26. Langer et al., “Nonpasteurized Dairy Products,” 389. 27. P. Slovic, “Perception of Risk: Reflections on the Psychometric Paradigm,” in Social Theories of Risk, ed. Sheldon Krimsky and Dominic Golding (Westport, CT: Praeger, 1992), 117–152. 28. “Fatality Analysis Reporting System (FARS) Encyclopedia,” National Highway Traffic Safety Administration, accessed August 3, 2016, www-fars.nhtsa.dot .gov/Main/index.aspx. 29. “Dairy Products: Per Capita Consumption, United States, 1975–2014 (In Pounds per Person),” Dairy Data, USDA Economic Research Service, accessed August 3, 2016, www.ers.usda.gov/data-products/dairy-data.aspx. 30. Joint FDA/Health Canada Quantitative Assessment of the Risk of Listeriosis from Soft-Ripened Cheese Consumption in the United States and Canada: Interpretive Summary (Food Directorate, Health Canada and Center for Food Safety and Applied Nutrition, Food and Drug Administration, U.S. Department of Health and Human Services, July 2015), www.fda.gov/downloads/Food/FoodScience Research/RiskSafetyAssessment/UCM429420.pdf.

27 2

•

NOTES TO PAGES 131–136

31. “Camembert de Normandie,” Fromages AOP de Normandie, accessed August 2, 2016, www.fromage-normandie.com/fr/anglais/camembert-normandie .html. 32. Interpretive Summary: Quantitative Assessment of the Risk to Public Health from Foodborne Listeria monocytogenes among Selected Categories of Ready-to-Eat Foods (Center for Food Safety and Applied Nutrition, Food and Drug Administration, US Department of Health and Human Services, September 2003), www.fda .gov/downloads/Food/FoodScienceResearch/UCM197329.pdf. 33. W. Schlech, P. Lavigne, R. Bortolussi, A. Allen, E. Haldane, A. Wort, A. Hightower, S. Johnson, S. King, E. Nicholls, and C. Broome, “Epidemic Listeriosis: Evidence for Transmission by Food,” New England Journal of Medicine 308, no. 4 (1983): 203–206. 34. Ibid., 203. 35. C. Sunstein, Laws of Fear (Cambridge: Cambridge University Press, 2005), 13. 36. Ibid., 1–9. 37. Ibid., 129–148. 38. Committee on the Review of the Use of Scientific Criteria and Performance Standards for Safe Food, National Research Council, Scientific Criteria to Ensure Safe Food (Washington: National Academies Press, 2003), 245. 39. “Mortality Risk Valuation” Environmental Economics, US Environmental Protection Agency, accessed July 29, 2016, www.epa.gov/environmental-economics /mortality-risk-valuation. 40. J. Hirsch, “A Review of the Literature on Rural Suicide: Risk and Protective Factors, Incidence, and Prevention,” Crisis 27, no. 4 (2006): 189–199. 41. W. McIntosh, E. Spies, D. Stone, C. Lokey, A. Trudeau, and B. Bartholow, “Suicide Rates by Occupational Group—17 States, 2012,” Morbidity and Mortality Weekly Report 65 (2016): 641–645. 42. Sunstein, Laws of Fear, 126. 43. Assured Code of Practice (London: Specialist Cheesemakers Association, 2015), www.specialistcheesemakers.co.uk/assured-guide.aspx. 44. M. Nestle, Safe Food: Bacteria, Biotechnology, and Bioterrorism (Berkeley: University of California Press, 2003), 194–219. 45. S. Strom and K. Severson, “F.D.A. Rule May Alter Cheese-Aging Process,” New York Times, June 10, 2014, www.nytimes.com/2014/06/11/business/us-rulemay-alter-cheese-aging-process.html. 46. H. Paxson, The Life of Cheese: Crafting Food and Value in America (Berkeley: University of California Press, 2013), 158–186. 47. D. Kahan, “Fixing the Communications Failure,” Nature 463 (2010): 296–297. 48. D. Kahan, “The Politically Motivated Reasoning Paradigm, Part 1: What Politically Motivated Reasoning Is and How to Measure It,” Emerging Trends in Social and Behavioral Sciences, November 29, 2016, 1–16. 49. Paxson, Life of Cheese, 165.

NOTES TO PAGES 136 –143

•

273

50. We should also stress that ours is a purely theoretical assignment of values to players within the industry based on their public pronouncements; we would actively welcome the empirical and experimental testing of our schema. 51. Oldways Cheese Coalition, www.oldwayscheese.org. 52. Raw Milk Freedom Riders, www.rawmilkfreedomriders.wordpress.com. 53. J. Gusfield, “On Legislating Morals: The Symbolic Process of Designating Deviance,” California Law Review 56, no. 1 (1968): http://scholarship.law.berkeley .edu/cgi/viewcontent.cgi?article=2827&context=californialawreview. D. Kahan, “Fear of Democracy: A Cultural Evaluation of Sunstein on Risk,” Yale Law School Faculty Scholarship Series, paper 104 (2006): http://digitalcommons.law.yale.edu /fss_papers/104. 54. L. Reiley, “Farm to Fable: At Tampa Bay Restaurants, You’re Being Fed Fiction,” Tampa Bay Times, April 13, 2016, www.tampabay.com/projects/2016/food /farm-to-fable/restaurants.

8 . C U LT U R E S

1. J. Twamley, Dairying Exemplified, or the Business of Cheese-Making, 2nd ed., (Warwick: Sharp, 1787); J. Twamley, Essays on the Management of the Dairy (London: J. Harding, 1816). 2. D. Valenze, Milk: A Local and Global History (New Haven: Yale University Press, 2011), 129–133. 3. Twamley, Management of the Dairy, 15–29. 4. P. Kindstedt, Cheese and Culture: A History of Cheese and Its Place in Western Civilization (White River Junction: Chelsea Green, 2012). 5. P. Kindstedt, “Starter Culture: The Heart of Cheesemaking,” in American Farmstead Cheese: The Complete Guide to Making and Selling Artisan Cheese, ed. Paul S. Kindstedt (White River Junction: Chelsea Green, 2005), 57. 6. R. Richardson, Lancashire Cheese-Making (Including Small Cheese) (Preston: Geo. Toulmin and Sons, 1921), 10. 7. The lur is a Bronze Age musical instrument and a national symbol for Denmark. Lurmark was first registered on October 23, 1901, and became Lurpak in 1957. “The History of Lurpak,®” Arla Foods UK, accessed August 30, 2016, www .arlafoods.co.uk/brands/lurpak/the-history-of-lurpak1. 8. W. Lynch, Scientifi c Butter-Making (Toronto: C. Blackett Robinson, 1883), 13. 9. K. O’Rourke, “Late 19th Century Denmark in an Irish Mirror: Towards a Comparative History” (Department of Economics and IIIS, Trinity College Dublin, IRCHSS Government of Ireland Senior Fellow, 2004), 1. 10. Ibid., 7. 11. S. Keillor, “Agricultural Change and Crosscultural Exchange: Danes, Americans, and Dairying, 1880–1930,” Agricultural History 67, no. 4 (1993): 61.

274

•

NOTES TO PAGES 143–150

12. J. Leisner, “Weeds, Heat and Pure Cultures: On the Differential Success of New Technologies in the Danish and American Creamery Industries in the 1890s,” Food Control 30 (2005): 423. 13. I. Henrikson, “ ‘To the Taste and Fancy . . .’: Quality Control in the Value Chain” (paper presented at the Economic History Society Annual Conference, University of York, York, UK, April 5–7, 2013), 14. 14. Leisner, “Weeds, Heat and Pure Cultures,” 421. 15. Quoted in S. Knudsen, “Starters,” Journal of Dairy Research 2 (1931): 137–163. 16. K. Abildgren, “A Chronology of Denmark’s Exchange-Rate Policy 1875– 2003,” Working Papers of Danmarks Nationalbank 12 (2004): 9. Calculation made using the index at www.measuringworth.com. 17. Knudsen, “Starters.” 18. Ibid. 19. Leisner, “Weeds, Heat and Pure Cultures,” 424. 20. Ibid. 21. “Scientific Methods of Butter-Making,” Brisbane Courier, December 18, 1894, 6. 22. Lynch, Scientific Butter-Making, 105. 23. Leisner, “Weeds, Heat and Pure Cultures,” 425. 24. Ibid., 426. 25. Lynch, Scientific Butter-Making, 105. 26. Leisner, “Weeds, Heat and Pure Cultures,” 427. 27. The “home made rennet” to which he refers was made by infusing strips of dried calves’ stomachs in whey, the same technique Alpine cheesemakers continue to use as a combined starter-rennet preparation today. 28. G. Marty, “Brick Cheese Making,” in Fifteenth Annual Meeting of the Wisconsin Cheese Maker’s Association, ed. U. Baer (Madison: Democrat Printing Company, 1907): 66–73. 29. W. Sandine, “Commercial Production of Dairy Starter Cultures,” in Dairy Starter Cultures, ed. T. Cogan and J. P. Accolas (New York: VCH, 1995), 192. 30. D. Saker, Practical Cheddar Cheese-Making (St. Albans: Camfield, 1917), 40–41. 31. R. Leitch, Cheddar Cheese-Making: Faults in Cheese (Glasgow: Scottish Agricultural Publishing Company, 1932), 25. 32. V. Cheke, The Story of Cheese-Making in Britain (London: Routledge and Kegan Paul, 1959), 236–237. 33. A. Johnson and L. Williams, A Field Guide to Fermentation (self-published, 2016), 83. 34. Equipment and Supplies for Farm Dairy, commercial catalog (Villefranche sur Saône: Ets Coquard, 2013/2014), 9. 35. “The Results from Our Cultures Trial Project Are In!” Australian Specialist Cheesemakers’ Association, November 2, 2015, www.australiancheese.org/news /?id=155.

NOTES TO PAGES 151–167

•

275

36. R. Cornish, “Cheese Revolution as Milk Bacteria Undergo DNA Sequencing,” Good Food, October 27, 2014, www.goodfood.com.au/archived/cheese-revolutionas-milk-bacteria-undergo-dna-sequencing-20141023-11akkz. 37. Regional Fermentation Cultures for Victorian Cheesemakers: Final Technical Report on Culture Preparation (Dairy Innovation Australia, November, 2015). 38. Sandine, “Commercial Production,” 195–196. 39. “Create Your Own Cheese Sensory Signature,” Chr. Hansen, accessed August 31, 2016, www.chr-hansen.es/noticias/singlenoticias/create-your-owncheese-sensory-signature.html. 40. “CHOOZIT MM100,” Bob-White Systems, accessed August 31, 2016, www.bobwhitesystems.com/products/choozit-mm100-50dcu. 41. J. Cole, “Comfortable Cows, Delectable Cheese,” myFarmLife.com, accessed August 30, 2016, www.myfarmlife.com/features/comfortable-cows-delectablecheese. 42. “Products,” Orchard Valley Dairy Supplies, accessed August 30, 2016, www .orchard-dairy.co.uk/pages/BUY-SUPPLIES-ONLINE.htm 43. B. Wolfe, J. Button, M. Santarelli, and R. Dutton, “Cheese Rind Communities Provide Tractable Systems for In Situ and In Vitro Studies of Microbial Diversity,” Cell 158 (2014): 422–433. 44. N. Brennan, A. Ward, T. Beresford, P. Fox, M. Goodfellow, and T. Cogan, “Biodiversity of the Bacterial Flora on the Surface of a Smear Cheese,” Applied and Environmental Microbiology 68, no. 2 (2002): 820–830. 45. N. Marcellino and D. Benson, “The Good, the Bad, and the Ugly: Tales of Mold-Ripened Cheese,” in Cheese and Microbes, ed. C. Donnelly (Herndon: ASM, 2014), 108. 46. E. Hastings, A. Evans, and E. Hart, The Bacteriology of Cheddar Cheese (Washington, DC: Government Printing Office, 1912). 47. Ibid., 9. 48. The greatest advances in the microbial understanding of fermentation were a product of the nineteenth century. See, for example, L. Pasteur, “Mémoire sur la fermentation appelée lactique,” Comptes Rendus Chimie 45 (1857): 913–916; and J. Lister, “On the Lactic Fermentation and Its Bearings on Pathology,” Transactions of the Pathological Society of London 29 (1878): 425–467. 49. Twamley, Dairying Exemplified, 55. 50. N. Klijn, F. Nieuwenhof, J. Hoolwerf, C. van der Waals, and A. Weerkamp, “Identification of Clostridium tyrobutyricum as the Causative Agent of Late Blowing in Cheese by Species-Specific PCR Amplification,” Applied and Environmental Microbiology 61, no. 8 (1995): 2919–2924. 51. Twamley, Dairying Exemplified, 56. 52. Klijn et al., “Identification of Clostridium tyrobutyricum”; J. Stadhouders, “The Use of Lysozyme to Control Butyric Acid Fermentation,” Bulletin of the International Dairy Federation 251 (1990): 51–53.

276

•

N OTE S TO PAG E S 1 67–174

9 . FA M I L I E S A N D FAC TO R I E S

1. M. Winstanley, “Industrialization and the Small Farm: Family and Household Economy in Nineteenth-Century Lancashire,” Past and Present 152 (1996): 171. 2. Ibid., 160 3. Ibid., 171 4. J. Binns, Notes on the Agriculture of Lancashire, with Suggestions for Its Improvement (Preston: Dobson and Son, 1851), 16. 5. Ibid. 6. H. Jenkins, “Report on the Cheese-Factory System and Its Adaptability to English Dairy Districts,” Journal of the Royal Agricultural Society of England, 2nd ser., 6, no. 11 (1870): 181. 7. Ibid., 193. 8. X. Willard, Address before the Cheese Makers’ Association (Albany: Van Benthuysen’s Steam Printing House, 1865), 13. 9. Ibid., 3. 10. “American Cheese,” Penny Magazine of the Society for the Diffusion of Useful Knowledge 11, no. 638 (March 12, 1842): 98–99. 11. Willard, Address, 5. 12. Ibid., 7. 13. Ibid. 14. X. A. Willard, Practical Dairy Husbandry (New York: D. D. T. Moore, 1872), 276–277. 15. J. Harding, “Recent Improvements in Dairy Practice,” Journal of the Royal Agricultural Society of England 21 (1860): 86. 16. Ibid. 17. Ibid., 89. 18. Ibid., 90. 19. Ibid., 91. 20. Willard, Practical Dairy Husbandry, 271. 21. Ibid. 22. Ibid., 262. 23. Ibid., 275. 24. “Artisanal Somerset Cheddar,” Slow Food Foundation for Biodiversity, accessed March 1, 2016, www.fondazioneslowfood.com/en/slow-food-presidia /artisan-somerset-cheddar. 25. Willard, Address, 19. 26. Ibid., 1–2. 27. E. Thompson, “Time, Work-Discipline, and Industrial Capitalism,” Past and Present 38 (1967): 60. 28. X. Willard, quoted in Jenkins, “Report on the Cheese-Factory System,” 198. 29. Ibid.

NOTES TO PAGES 177–185

•

27 7

30. Willard, Practical Dairy Husbandry, 272. 31. Jenkins, “Report on the Cheese-Factory System,” 199. 32. Ibid., 202. 33. Ibid., 199. 34. R. Blundel and A. Tregear, “From Artisans to ‘Factories’: The Interpenetration of Craft and Industry in English Cheese-Making, c1650–1950,” Enterprise and Society 7, no. 4 (2006): 705–739. 35. D. Saker, Practical Cheddar Cheese-Making (St. Albans: Camfield, 1917), 40. 36. L. van Slyke, The Science and Practice of Cheese-Making (New York: Orange Judd, 1910), 55–65. 37. S. Babcock, The Cold Curing of Cheese, US Department of Agriculture, Bureau of Animal Industry, Bulletin no. 49 (Washington, DC: Government Printing Office, 1903), 53. 38. V. Cheke, The Story of Cheese-Making in Britain (London: Routledge and Kegan Paul, 1959), 273. 39. Ibid. 40. P. Rance, The Great British Cheese Book (London: Macmillan, 1982), 69. 41. Thompson, “Time, Work-Discipline, and Industrial Capitalism,” 91.

10. EXPERTISE

1. H. West, “Thinking Like a Cheese: Towards an Ecological Understanding of the Reproduction of Knowledge in Contemporary Artisan Cheesemaking,” in Understanding Cultural Transmission in Anthropology: A Critical Synthesis, ed. R. Ellen, S. J. Lycett, and S. E. Johns (Oxford: Berghahn Books, 2013), 320–345. 2. F. J. Lloyd, Observations on Cheddar Cheese-Making (London: William Clowes and Sons, 1892), 4. 3. Ibid., 42. 4. Ibid., 26. 5. Ibid., 44. 6. Ibid., 75. 7. Ibid., 76–77. 8. F. J. Lloyd, Observations on Cheddar Cheese-Making (Bath: Bath and West and Southern Counties Society, 1895), 12. 9. In the 1908 Census of Production, there were 362,000 hundredweight (16,000 tons) of English farmhouse cheese sold, compared to just 53,000 hundredweight (2,300 tons) of cheese from English factories. By 1935, the respective figures had become 130,000 hundredweight (5,600 tons) from farms and 817,000 hundredweight (35,500 tons) from domestic factory production. 10. D. Taylor, “Growth and Structural Change in the English Dairy Industry, c1860–1930,” Agricultural History Review 35, no. 1 (1987): 49.

278

•

NOTES TO PAGES 185 –200

11. N. Verdon, “Women and the Dairy Industry in England, c. 1800–1939,” (paper presented at the 14th International Economic History Congress, Helsinki, Finland, August 21, 2006), 1. 12. H. D. Kay, “The National Institute for Research in Dairying, Shinfield, Reading,” Proceedings of the Royal Society of London, Series B, Biological Sciences 138, no. 890 (1951): 17. 13. Ibid., 20. 14. Ibid., 23. 15. Ibid., 25–26. 16. V. Cheke, The Story of Cheese-Making in Britain (London: Routledge and Kegan Paul, 1959), 248. 17. S. McMurray, “Women’s Work in Agriculture: Divergent Trends in England and America, 1800 to 1930,” Comparative Studies in Society and History 34 (1992): 258. 18. D. Fink, “ ‘Not to Intrude’: A Danish Perspective on Gender and Class in Nineteenth-Century Dairying,” Agricultural History 83, no. 4 (2009): 446–476; Verdon, “Women and the Dairy Industry,” 4. 19. D. Valenze, Milk: A Local and Global History (New Haven: Yale University Press, 2011), 118–133. 20. J. Twamley, Dairying Exemplified, or the Business of Cheese-Making: Laid Down from Approved Rules, Collected from the Most Experienced Dairy-Women, of Several Counties, 2nd ed. (Warwick: J. Sharp, 1787), v. 21. Verdon, “Women and the Dairy Industry,” 12. 22. Ibid., 13. 23. Ibid., 14. 24. Cheke, Cheese-Making in Britain, 214. 25. D. Saker, Practical Cheddar Cheese-Making (St. Albans: Camfield, 1917), 33. 26. R. Leitch, Cheddar Cheese-Making: Faults in Cheese (Glasgow: Scottish Agricultural Publishing Company, 1932), 40–41. 27. “A Promising Career,” Farmer and Stockbreeder, August 9, 1926, p. 1654. 28. Saker, Practical Cheddar Cheese-Making, 34. 29. Taylor, “Growth and Structural Change,” 52. 30. Ibid., 63. 31. Kay, “National Institute for Research in Dairying,” 23. 32. Cheke, Cheese-Making in Britain, 272. 33. Twamley, Dairying Exemplified, 20. 34. F. J. Lloyd, Observations on Cheddar Cheese-Making (Bath: Bath and West and Southern Counties Society, 1894), 34. 35. Ibid., 38. 36. C. B. Donnelly, J. Leslie, and L. Black, “Production of Enterotoxin A in Milk,” Applied Microbiology 16, no. 6 (1968): 917–924. 37. P. Rance, The Great British Cheese Book (London: Macmillan, 1982), 8. 38. M. Porter, The Competitive Advantage of Nations (Basingstoke: Macmillan, 1990), 156.

NOTES TO PAGES 200 –214

•

279

39. Ibid., 151. 40. Committee on the Future of the Colleges of Agriculture in the Land Grant System, National Research Council, “Colleges of Agriculture at the Land Grant Universities: A Profi le” (Washington, DC: National Academies Press, 1995). 41. M. Hall, “Ryegrass,” Penn State University Agronomy Facts 19 (1992): 2. 42. C. Lee, S. Smith, G. Lacefield, and J. Herbek, “Grain and Forage Crop Guide for Kentucky, AGR-18,” University of Kentucky College of Agriculture Cooperative Extension Service, 2007. 43. S. Charters and V. Michaux, “Strategies for Wine Territories and Clusters: Why Focus on Territorial Governance and Territorial Branding?” Journal of Wine Research 25, no. 1 (2014): 1–4.

11. MARKETS

1. “Cheese Day: Un festival de fromages, ” Gourmets & Co, accessed February 19, 2017, www.gourmetsandco.com/actus/15103-cheese-day-un-festival-de-fromages. 2. M. Winstanley, The Shopkeeper’s World, 1830–1914 (Manchester: Manchester University Press, 1983), 129.

12 . REINVENTING THE WHEEL

1. O. Burdett, A Little Book of Cheese (London: G. Howe, 1935), reprinted in A Guide to English Cheeses: A Collection of Historical Articles on Varieties of English Cheese (Redditch: Read Books, 2011), 16. 2. J. Prescott, Taste Matters: Why We Like the Foods We Do (London: Reaktion Books, 2012), 62–81; M. Moss, Salt Sugar Fat: How the Food Giants Hooked Us (New York: Random House, 2013). 3. H. Lawless and H. Heymann, Sensory Evaluation of Food: Principles and Practices. (New York: Springer-Verlag, 2010), 4–8. 4. V. Agnew, “History’s Affective Turn: Historical Reenactment and Its Work in the Present,” Rethinking History 11, no. 3 (2007): 299–312. 5. C. Allsop et al., “75 Best Cheeses of 2015,” Culture: The Word on Cheese 7, no. 6 (2015): 25. 6. “The Valley Fromage Blanc,” Culture, accessed August 19, 2016, www .culturecheesemag.com/cheese-library/The-Valley-Fromage-Blanc. 7. F. Percival, “Changing Tastes,” World of Fine Wine 33 (2011): 109. 8. Moss, Salt Sugar Fat. 9. P. Rance, The Great British Cheese Book (London: Macmillan, 1982), 75–76; K. Calvert, Wensleydale Cheese, (Clapham: The Dales, 1946), 13–17; D. Font-Réaulx, Painting and Photography: 1839–1914 (Paris: Flammarion, 2012), 32–64.

280

•

NOTES TO PAGES 214–241

10. P. McCouat, “Early Influences of Photography on Art,” Journal of Art in Society, accessed August 17, 2016, www.artinsociety.com/early-influences-ofphotography.html. 11. Ibid. 12. Ibid. 13. Ibid. 14. M. Hartley and J. Ingilby, Dales Memories (Lancaster: The Dalesman, 1986), 118; Rance, Great British Cheese, 78. 15. Rance, Great British Cheese, 132. 16. M. Shaffer, “Minimum Population Sizes for Species Conservation,” BioScience 31, no. 2 (1981): 131–134. 17. J. Fromer, “ ‘Deeply Indebted to the Tea-Plant’: Representations of English National Identity in Victorian Histories of Tea,” Victorian Literature and Culture 36, no. 2 (2008): 531–547. 18. S. Mintz, “Time, Sugar, and Sweetness,” in Food and Culture, ed. C. Counihan and P. Van Esterik (New York: Routledge, 2008), 100–101. 19. P. Tozer and R. Huffaker, “Dairy Deregulation and Low-Input Dairy Production: A Bioeconomic Evaluation,” Journal of Agricultural and Resource Economics 24, no. 1 (1999): 155–172.

NOTES TO PAGES 241–244

•

281

This page intentionally left blank

INDEX

Appleby’s Cheshire, 37 fig. 4, 203–204 arable land, 87; biodiversity on, 16 Arapawa Island goats, 43 Arkwright, Richard, 176–178 Arthrobacter, 103 artificial insemination, 53–56 artisanal, 29 artisan cheese: defined, 29, 255; vs. factory cheese, 183 artisanship, 195–197, 206 Artisan Somerset Cheddar, 183 associated dairies, 179, 185 Aubrac, 45–46, 82 Aubree, David, 93–94, 114 Australian cultures trial, 166–168 Australian Specialist Cheesemakers’ Association, 166–168 authoritarian values, 144 Auvergne, ix–xiii, 9, 12, 22, 82, 85–86, 149, 195, 219, 245

abomasum, 33 Abondance, 120, 219 acid development: effect of, on curd texture, 34–38; in rumen, 73–74. See also lactic acid bacteria acidimeter, 198–199, 200, 201, 207, 211 acidosis, 73–75 ACTALIA, 146 actinobacteria, 102–104 additives: to cheese, fruit and spices, 252; to feed, 74; lysozyme, 71; potassium nitrate, 70–71 affinage, 17–18, 40 alleles, 62 almond hulls, 4, 240 Alouette, 219 alpine cheeses, 35, 112, 173 alternate histories, 244 American Cheese Society, 1, 146, 158, 231–232 American Milking Devons, 43–46, 67 anachronistic cheesemaking, 239 anaphylaxis, 137 Andante Dairy, 193–195 annatto, 83 antibiotics, 15, 56, 74, 102 anti-flavor elite, 240 antimicrobial compounds, 100, 108–109. See also terpenes Appelation d’Origine Protegée regulations, 9, 94, 113, 116, 144 fig. 7, 190, 220, 224–228

bacteria, 255; community formation and interactions, xii, 13–15, 160–161; conflation with disease, 20; in fermented foods, 30–31; and flavor development, 39; purified strains, 97, 148–149, 151, 167, 169–170; in raw milk, 20, 95–112, 149; thermoduric, 103, 110; total bacterial count, 96–99, 107–108, 110, 149. See also lactic acid bacteria; ripening bacteria; spoilage organisms; starter cultures; names of individual bacteria

283

19; cultured, 149–153, 202; Danish butter industry, 149–152; and starter culture origin, 152–153; texture of, related to animal feed, 83 buttermilk, 150, 152 buttery flavors, 159, 166 buying guide for cheese, 251–253

bacteriocins, 100 bacteriophage, 156–157 Bactoscan test, 96 Bates, Thomas, 50–51 Bath and West and Southern Counties Society, 196, 203 Battle of Plassey, 244 Bauman, Howard, 126 beta-carotene, 83, 85, 219 “Betty Crocker cheeses,” 170 biodiversity, 16, 252; cheese biodiversity chain, 90, 91 fig. 6; and cheese safety, 97–100; decrease of, in raw milk, 98–99; and economic value, 16, 25–26, 89–90; of fields, 80, 86, 88, 236; and flavor, 5, 82, 82, 84–85, 90, 117, 214–215; of livestock breeds, 15–16, 264n36, 264n38, 264n39; microbial, 98, 117; multilevel, 16, 25; and novel drug development, 102; and restoration of fields, 86–88. See also inbreeding biofi lms, 5, 16, 130–131 Blaser, Martin, 15 Bleu de Termignon, 164–165, 173–174 bloat, feedlot, 74 blown cheese, 70–71, 155; and Clostridium tyrobutyricum, 174 Boisard, Pierre, 115 Brà, 13, 215 breeding: crossbreeding, 63–64; inbreeding, 48, 50, 61–65, 264n36, 264n38, 264n39; modern, 49–51; “true breeding,” 62; in US, 52–55, 63. See also genomics breeds: biodiversity of, 15–16, 264n36, 264n38, 264n39; defined, 48; dualpurpose, 113, 235, 256; rare, 16, 44, 49, 61, 118, 236, 264n36. See also names of individual breeds Brevibacterium, 103, 171–172 Brexit, 23 Brick cheese, 154–155 British cheese. See United Kingdom brucellosis, 7, 124 Buller, Henry, 89–91, 91 fig. 6 Burgundy, 11–12 butter: American market for, 153; British market for, 19, 152; and cheese quality,

284

calcium phosphate, 33–37, 206 California, 2–5, 110, 193–195, 215–217 calorie intake, 73, 266n12 Cal Poly San Luis Obispo, 215–217 calving, ease of, 54, 55, 60, 63 Cambozola, 164–165, 173 Camembert: Champsecret, 113–118; de Normandie AOP, 28, 93–99, 113, 190, 220–221, 229; Gillot, 190; Graindorge, 93, 220; Le Rustique, 219; rind of, 115–116; safety of, 125, 136–139; stabilized, 28–29, 37 fig. 5 Campaign for Real Ale, 25 Campylobacter, 124, 270n11 Candy Method, 197, 222 Cannon, Edith, 197–204, 208, 210–212, 222, 232 Cannon Method, 197 Cantal, xi–xii, 9, 59 cattle: health metrics, 62; herd size, 4, 81, 207; imported to US, 49, 51; wild, 46–49. See also breeding; breeds; names of breeds CCP (Certified Cheese Professional), 231–232 CDC. See Centers for Disease Control and Prevention Cellars at Jasper Hill, 1, 36 fig. 3, 80–81, 143–144, 144 fig. 7 cellulose, 72–74 Centers for Disease Control and Prevention, 124, 133–136, 139, 142 Certified Cheese Professional, 231–232 Chambon, Guy, ix–xi, 4, 14, 16–17, 20, 221, 237 chaparral, 6 Chassard, Patrice, 85–87 Chaumes, 219 Cheddar, 38 fig. 5, 209 fig. 8; American, 40–41, 178–181; cloth binding of,

•

INDEX

rumen, 72–75; of soil, 87–88; synthetic, 165–168, 221 components, of milk, 60, 63, 221 Comté, 35, 36 fig. 2, 38 fig. 5, 59, 70, 99, 108, 130, 163, 173, 229 concentrates, 54, 74, 84, 87, 255, 266n12 conception rate, 54 Conference on the Science of Artisan Cheese, 12 conjugated linoleic acid, 83 consistency, 159, 216, 231; of commercial rennet, 34, 158; of factory cheese, 186–187; and pasteurization, 151–152 consolidation of production, 3–5, 110, 220–222 construction of value/taste, 50–51, 238–242 consumption (per capita): of cheese, 135–136; of liquid milk, 122, 207 contamination: of biofi lms, xii, 131; of liquid milk, 123–124, 134; by Listeria, 104, 134; and off-flavors, 199, 210; post-pasteurization, 125; resistance to, xii, 14; vs. competitive advantage, 39–40, 173; with wild-type molds, 115–116 cooperative, 29 cooperative dairying, 150–152, 179 corn silage, 78, 83, 258 corynebacteria, 103, 172 cost, of cheese, 229; and AOP regulations, 224–225, 227, 229; cost-benefit analysis of, 138–140; of factory cheeses, 159, 180, 186, 188, 221; premium, xii, 21–22, 38; and sustainability, 245; unpriced externalities, 22, 258 cost, of food, 22, 89; and true cost accounting, 89, 258 cost-benefit analysis, 138–140 Coxiella burnetii, 257, 269n16 creamy Lancashire, 188–189 Cretinet, Marina, 93–95, 113, 117 Crickmore, Jonny, 58–61 crossbreeding systems, 63–64 crumbly Lancashire, 175, 188–189 cultural cognition model, 142–146, 144 fig. 7 culture-based methods, 96, 107

181; farmhouse vs. factory, 184–191, 221; flavor of, 35; flavor of, in Brick cheese, 154–155; masculinization of, 208–211; Montgomery’s, 204, 208, 211; number of producers, 25, 212, 243; reductionist approaches to, 195–201, 206; Somerset, 50, 181–184; sweet, 40–41, 72, 163, 169, 240; Victorian, 53, 182–184; Westcombe, 211; wholesale price of, in 1865, 22. See also Cannon, Edith; Harding, Joseph; Lloyd, Frederick Cheddarmaster, 190 cheese biodiversity chain, 90, 91 fig. 6 cheese by the clock, 176–177, 182, 185–186, 189–190 cheesefarmer, 239, 242, 246, 252 cheesemaking process, 29–38 cheesemongers, 24–25, 176, 228–233 Cheese of Choice Coalition, 143–144, 144 fig. 7 Cheke, Valerie, 187, 201, 204, 212 Cheshire, 37 fig. 4, 38 fig. 5, 183, 208, 243 chèvre, 192. See also goat cheese Chillingham herd, 46–49 Chr. Hansen, 157–158, 160–165, 215 chymosin, 33, 34, 204 citrus peel, 4, 240, 255 classification: of cheese, 29, 37–38, 38 fig. 5; of microbes, 102 closed systems, 69, 237–238 Clostridium tyrobutyricum, 174. See also blown cheese cloth binding, 181, 183–184, 224, 255 clusters/clustering, 212–217 Codex Alimentarius, 127 cold-ripening, 187 Colling, Charles and Robert, 50 color, link to feed, 83–85, 236 Comet, 50 commercial cultures, xii, 102, 104, 149–155, 158–173 commodity: by-products, in feed, 4, 8, 240; cheese as, 67; markets, 4, 24, 26, 87, 139–140, 221, 244 community: assembly, 15, 170–172; of gut, 14–16; microbial, xii, 13–16, 20, 30, 94, 97–101, 104–109, 141, 155, 210; of

INDEX

•

285

Dutton, Rachel, 12–15, 104, 170–173 DVI/DVS (Direct-Vat Inoculation / Direct-Vat Set) cultures, 169, 256

cultured butter, 149–153, 202 culture houses, 157–162, 169, 215 cultures, commercial, xii, 102, 104, 149– 155, 158–173. See also starter cultures curd mill, 37, 182, 208–209, 209 fig. 8, 211 customer engagement, 221, 226–229, 232–233, 251–253 cut-to-order, 221, 231

“eating biodiversity chain,” 90–91, 91 fig. 6 ecological niche, 20, 98 ecology: as cheesemaking metaphor, 195, 252; of knowledge, 195–196, 201, 207, 212; and microbial community, xii, 13–16, 20, 30, 94, 97–101, 104–109, 141, 155, 172, 210. See also biodiversity; evolution; selective environment Edgar, Gordon, 41 education: Certified Cheese Professional program, 230–233; Opus Caseus Concept, 230–231. See also dairy schools élevage, 17. See also maturing elitism, 22 embryo transfer, 56 Emmenthal, 38 fig. 5, 223 emperor’s new clothes, 163, 242 empiricism, xii, 82, 141–143, 182 endangered species: breeds, 44, 49; cheesemakers as, 221, 244; cheeses as, 223, 246 energy, 69, 72–75, 78, 266n12 Enlightenment principles, 21, 49, 147, 203 environmental determinism, 68–72 environmental impact, 79–80, 87–90. See also hay meadows; true cost accounting epidemiology, 123–125, 133–140 Epoisses, 38 fig. 5, 221 Escherichia coli, 13, 104–105, 134; highly pathogenic serotypes of, 125, 127. See also Gram-negative bacteria evolution: of cheese styles, 69–72, 174; commercial selective pressure and, 158–159, 243–244, 251–253; microbial, 15, 98. See also selective environment export: of American cheese to Europe, 178–183, 186; of Shorthorns to North America, 49; subsidies, 223 extension programs, 4, 215–216 extensive farming systems, 256; environmental benefits of, 70, 88–91, 91 fig. 6; impact of, on flavor of cheese, 81–85; profitability and, 87, 244–245 extra-farm inputs, 237–238

dairy farm, products of, 69, 237 Dairy Innovation Australia Limited, 166–168 dairymaids, 21, 70, 148, 174, 185, 203, 242 dairy schools, 154–155, 183, 196–201, 215–217 Dales cheese, 234–236, 239–243, 245 D’Amico, Dennis, 140 Danisco, xiii, 169 Danish Agricultural Research Laboratory, 151–152 Danish Heater, 153 Daviet, Madeleine, 120–122 Dawson, Trish, 158–159, 164, 215 “dead milk,” 98, 106 defined starters, 156–157 demineralization, 32–34. See also calcium phosphate Denmark, 149–153, 157–158 depletion, of microbial diversity in raw milk, 98–99 Desmasures, Nathalie, 95, 98–99, 107, 113, 117 dicotyledons. See terpenes; wildflowers Direct-Vat Set/Direct-Vat Inoculation cultures, 169, 256 disinfectant. See sanitizers diversity, microbial, 98, 117. See also biodiversity Domaine de Saint Loup, 93–94, 113, 117 Donnelly, Catherine, 129, 133 draft animals, 43, 49. See also dual-purpose breeds drainage, 34–38 driving, risk of vs. eating raw-milk cheese, 135–136 dual-purpose breeds, 113, 235, 256 Dubois, Laurent, 228–231, 233 Durham Ox, 51

286

•

INDEX

xi–xii, 120, 130–132; research on cheese, xi–xii, 12, 14, 66, 82–85, 109, 212, 215– 216; specialist retailers, 226–231, 233. See also Auvergne; Institut national de la recherche agronomique; Loire Valley; Normandy fraud, 145 freeze-dried culture. See Direct-Vat Set/ Direct-Vat Inoculation cultures freezing milk, 8, 240 Friesian, 52–53, 59, 63 Fromagerie Gillot, 190 Fromagerie Graindorge, 93, 220 fruit and spices, added to cheese, 252 FSMA (Food Safety Modernization Act), 128, 232 functional redundancy, 98 fungi, 105, 256, 257

factory production, 25, 86, 114–115, 150– 154, 156, 158–162, 176–191, 202, 220– 222; associated dairies, 179, 185 farmhouse cheesemaking, 256. See also fermier farmhouse cheesemaking, decline of: in Britain, 186, 200, 206–207, 231, 244, 278n9; in France, 220–222; in US, 179–180 farming: specialization of, 3–4, 71, 239; subsistence, 76, 235, 245. See also extensive farming systems; intensive farming systems farm-specific cultures, 165–168 fatty acid composition, 83–84 faults: avoidance of, 174, 239; and starter cultures, 155–156 FDA (US Food and Drug Administration), 101, 127, 136–137, 141, 144, 146 feed, impact of on characteristics of cheese, 81–85, 236 Fen Farm Dairy, 58–61 Ferme du Bois Joli, 85–87 Ferme du Champsecret, 113–114, 117 fermentation, 30–31, 34, 156–157, 206; abnormal, 187; during rumination, 72; in silage, 78–79; spontaneous, 149–151 fermier, 29, 113, 219, 256 fertility: of livestock, 54, 62–63; of soil, 88–89 fertilizers, 11, 25, 77–78, 80, 242 Fjord, N.J., 151, 153 flavor: and biodiversity, 5, 82, 82, 84–85, 90, 117, 214–215; development of, 38–40; effect of feed on, 84–85; impact of pasteurization on, 90, 165; from starter and adjunct cultures, 41, 163, 169, 240; moral dimension of, 242 Flora Danica, 166 fondue, 222–224 Food Safety Modernization Act, 128, 232 food-safety systems, 125–133 forage, 71, 74, 236, 256 4-H club, 6–7, 9 France: gastronomic heritage, 94, 218; legal classification of cheese, 29, 82; market for cheese, 218–222; native breeds, x, 5, 59–60, 94, 113, 120; regulation of cheese,

INDEX

game theory, 226 gas production: in cheese (“heaving”), 70, 122, 155–156, 169, 174, 187; in cows (“feedlot bloat”), 74 gender roles, 184–186, 202–208 “generally regarded as safe” (GRAS), 101 genetic/genetics: bottlenecks, 62–63; and Chillingham herd, 46–49; improvement, 49–64; variation, importance of, 16, 264n36, 264n39. See also breeds; breeding genomics, 56–57, 62–64 geographical indications, 9, 19 Geotrichum candidum, 14, 130, 170, 172, 192 gerles, x–xii, 1, 5, 12, 14, 16, 149, 173, 256 Gloucester, 49, 63 goat cheese, 173, 192–194, 229; chèvre, 192 goats: Arapawa Island, 43; genomics of, 57 Gouda, 51, 71, 169, 174 grain, 73–74, 83. See also concentrates Gram-negative bacteria, 102, 104, 107, 112 Gram-positive bacteria, 102–104, 107, 111–112 grass: adapted to silage making, 80; effect of nitrogen fertilizers on, 80, 87–88; impact on milk and cheese characteristics, 81–87; native varieties, 216, 236; and rumination, 72–75. See also hay; silage

•

2 87

hygiene, 95, 96, 116; dairy, 21; milking, 95–97, 104, 108–110; regulations, 122, 130–133 hygiene through sterility. See Pasteurian perspective

grass-based systems, cattle breeds suited to, 45, 64, 66, 120, 235 grazing. See grass; overgrazing greenhouse gas, from cows, 82 grey market (unbranded) cheese, 223, 225 Gruyère, 23, 35, 70, 163, 173, 222–226 Gubbeen, 171–173 gut microbiota: of humans, 14–16, 98, 101; of ruminants, 72–75

Illumina, 57 immunocompromised persons, 101, 126 inauthenticity, 25, 28, 40–41, 173, 220 inbreeding, 48, 50, 61–65, 264n36, 264n38, 264n39 industrial mentality, xiii, 20, 22, 24, 37, 54, 77, 88–89, 113, 117, 130, 141, 152–153, 178–180, 183, 190–19; at homestead scale, 5–9 Industrial Revolution, 115, 176–177 industriel, 29, 113, 117 ingredients, 31, 115, 163, 166, 252 inoculation, 16, 164, 173, 242, 256. See also ripening cultures; starter cultures INRA (Institut national de la recherche agronomique), xii, 12, 82, 85, 95, 212, 215–216 insects, 16, 70, 80 inspections, 123, 127–128, 140, 194 institutional infrastructure, 196, 212, 231. See also dairy schools; ecology: of knowledge; education Institut national de la recherche agronomique (INRA), xii, 12, 82, 85, 95, 212, 215–216 intensive farming systems, 8–9, 28, 66, 74, 77, 84, 89, 256 Isigny Sainte-Mère, 94 ISO unit, 205

Haber-Bosch process, 77 HACCP (Hazard Analysis and Critical Control Point) system, 126–129, 132– 133, 140 Hafnia alvei, 104 Hansen, Christian, 157 Harding, Joseph, 181–186, 197 HARPC (Hazard Analysis and RiskBased Preventive Controls) system, 128–129, 133, 140 Hatch, Andy, 90–92 Hattan, Andrew, 235–245 hay, 256; dryers, 81, 114; effect of, on cheese, 83–85, 91–92, 120; making, 70, 75–76; and plant biodiversity, 80. See also grass; silage; wildflowers haycocks, 76 hay meadows, 77, 80, 87–88, 235–236 health metrics, cattle, 62 Helicobacter pylori, 15 herd size, 4, 81, 207 heterosis, 63 high-energy rations. See concentrates high-risk food, 120, 125, 133, 138–139 high-throughput methods, 95, 107, 159 Holden, Patrick, 88–89 holistic system, 69–72, 237–238 Holstein, 51–54; health concerns, 54, 60; inbreeding of, 62; yield, 44, 59 60–64, 66 homozygosity, 61–62 Horner, Marc-Henri, 222–226 hot iron test, 197, 201, 205–207 hoven cheese. See blown cheese human microbiome, 15, 101 hybrid vigor. See heterosis

288

Jameson effect (competitive inhibition of pathogens), 131 Jasper Hill, 1, 36 fig. 3, 80, 143, 144 fig. 7 Jensen, C. O., 151 Johansen, Eric, 158–161 Johnson, Arielle, 161 Jurassic Park, 245 Kahan, Dan, 142–145 kappa-casein B, 59, 264n36 Kehler, Mateo, 1, 80–81, 143 “kill step,” 127–128, 138, 184, 252

•

INDEX

Lloyd, Frederick, 196–204, 206–208, 210–212, 214 Loire Valley, 34, 173, 192–193, 195, 197 Long, James, 115 Low Riggs Farm, 235–236, 239, 242 Lurpak butter, 150, 274n7

Kirkham, Graham. See Kirkham’s Lancashire Kirkham’s Lancashire, 175–176, 188–191 Kraft, James, 178 Lactalis, 94, 220 lactic acid bacteria, 100–102; first isolation of, 151; and gas production, 156; in raw milk, 95–96, 98–102, 107–108, 110–111; role of, in fermentation, 30–31, 33, 34, 38, 148–152; in rumen, 73–74. See also acid development; bacteria; starter cultures lactic cheeses, 34–35, 35 fig. 1, 192–194, 228, 257 lactobacilli, 101 Lactobacillus helveticus, 41, 163, 240 Lactococcus garviae, 101–102 Lactococcus lactis, 14, 98, 100–101. See also lactic acid bacteria; starter cultures lactofermentation, 30 lactose, 30, 32, 137 lactose intolerance, 137 Laguiole, 46 Lancashire, 37, 38 fig. 5, 175–178, 183, 188–191, 210, 234, 243, 258; creamy, 188–189; crumbly, 175, 188–189 land grant colleges, 215–216 Lansley, Alison, 168 Le Cheese Day, 218–220 legumes, 77, 91 Leitch, Renwick, 205 leuconostocs. See buttery flavors; Flora Danica Lewandrowski, Lorraine, 23–24 libertarianism, 143–145 line breeding, 62. See also inbreeding liquid milk. See milk Listeria monocytogenes: competitive inhibition of, 130–131; discovery of, as foodborne pathogen, 137; environmental sources of, 134; importance of testing for, 129, 146; mechanism of pathogenicity, 106; post-pasteurization contamination, 103–104, 125; and pregnancy, 136–137; in raw milk, 98 Listeria spp., 81, 107; cold-adaptation, 112 Livarot, 104, 114, 220 lizard brain, 240–241

INDEX

Maine, 45, 76 Maison MonS, 226–228, 230–231 manure, 52, 69, 76–78, 88, 110 marginal land: biodiversity of, 79–80, 90; dairy farming as use of, 6, 16, 71, 89, 177–178, 245, 257. See also extensive farming systems markets, 219–233; cheese as way to access distant, 23, 180, 251. See also merit indexes; supermarkets Martin, Bruno, 82–86, 88–90 mastitis, 57, 59–61, 101, 103, 109, 124 Mathieu, Bruno, 130–133 matriarchs, 203–204. See also gender roles maturing, 17, 31, 40. See also affinage Meadowbrook Dairy, 2–5, 8 meadows, hay, 77, 80, 87–88, 235–236 megadairies, 3–4 Meilleur Ouvrier de France, 228, 230 Mercier, Patrick, 113–114, 116–118, 221 merit indexes, 58, 65 Messier, Norah, 42–43 methane, 82 Metz, Monica, 141, 144 micelle, casein, 32–35 Michel, Valérie, 146 Microbacterium, 103 Microbacterium gubbeenense, 172 microbial diversity, 98, 117. See also biodiversity microbial terroir, 167, 171. See also biodiversity microfi ltration, 90, 99, 101 milk: adulteration of, 18–20; components of, 60, 63, 221; digestion of, 33–34; flavor of, 18, 20; freezing, 8, 240; Holsteins, optimization for, 52–54; market for, 2–3, 200, 206–207, 241, 244; microbiology, 20, 95–105, 108–113, 123, 195; optimization of, for fresh milk processing, 97, 110–111; pasteurization of, 103; price

•

289

National Institute for Research in Dairying, 200–201, 203, 207, 210–211 native starters, 164–168, 172 Neal’s Yard Dairy, 2, 9, 120, 243 Netherlands, 51, 52, 70 New York, 23–24, 51, 55, 123, 126, 177–181, 187, 194 NIRD (National Institute for Research in Dairying), 200–201, 203, 207, 210–211 nitrogen fertilizers, 77–78, 80, 87–88, 239, 242, 257 noble grapes, 65–66 noma, 161–162 Normande, 94, 113 Normandy, 93–95, 98, 112–117, 242 North Devon cows, 43, 50 Northern Dairy Shorthorn, 235–236. See also Shorthorn numbers game, 232

milk (continued) of, 4, 23–24, 110, 207, 221, 244–245; quality metrics, 18–19, 60–66, 67, 96–97, 99, 107, 123, 239; recording, 57; safety factors, vs. cheese, 122–125; shelf life of, 99, 103, 124; skimmed, 150–152; urban vs. rural, 23–24, 123; yield, xi, 2, 4, 5, 43, 52–54, 63, 66, 67, 73–74, 227, 235, 237. See also raw milk milk fat globules, 32, 33, 67 milk fi lter testing, 81 milking: hygiene, 95–97, 104, 108–110; mobile units, 86; monitoring of data, 52–53; for raw-milk cheesemaking vs. for pasteurized liquid milk, 110; robotic milking systems, 57–58; seasonal, 236 Milk Marketing Board, 207 milk-quality payments, 96–97 millionaire bulls, 56. See also inbreeding minimum viable population, 243 Missillier, Jean-Pierre, 119–122, 130 misuse, of starter, 155–156 mites, 181, 187 molds, 39, 93, 105, 107, 131, 257; in silage, 78; wild-type, 115–116. See also fungi; ripening cultures Monbiot, George, 89 monoculture, xiii, 116, 157, 173, 253. See also biodiversity Mons, Hervé, 226–227, 230 Mons, Laurent, 226, 230 Montbéliarde, 59–61, 64–66, 84–85, 109, 227 Mont d’Or, 31, 35, 38 fig. 5, 59 Montel, Marie-Christine, x, xii–xiii, 1–2, 5, 12, 14, 20, 66, 82, 109, 145 Montgomery’s Cheddar, 204, 208, 211 moral dimension of flavor, 242 moral panic, 140, 142, 270n11 muck. See manure mycobacteria, 13, 103 Mycobacterium tuberculosis, 103, 106 Mycobacterium bovis, 124

off-flavors, 122, 165, 199, 210 Oldways Cheese Coalition, 143, 144 fig. 7 opportunistic pathogens, 14, 101, 103 Opus Caseus, 230–231 organic: certification, 19; conversion, 88, 110 Orland, Barbara, 71 Ossau Iraty, 38 fig. 5 outbreaks, of illness, 127, 133–137, 146 overgrazing, 75, 89 painting and photography, 241–242 panda bears, 243 Parisian World Expo, 122 Pasteurian perspective, 141, 143–145 Pasteur Institute, 116 pasteurization, 257; economic implications of, for raw-milk cheesemakers, 138– 140; link with starter cultures, 148– 152, 173; of Camembert, tension over, 94; impact of, on cheese flavor, 90, 165; of liquid milk, 103; post-pasteurization contamination, 125; reasons for, 96, 124, 134; as route to consistency, 23, 164–165. See also Danish Heater; microfi ltration; raw milk pastoralism, 69–71, 115, 118

NASA, 125–126 National Animal Germplasm Program, 264n39

290

•

INDEX

proteins, 31–33, 39, 77, 159 protozoa, 72, 74, 106 Pseudomonads, 30, 106, 112 Pseudomonas aeruginosa, 106 public health impacts, 89, 124, 133–140, 142

pathogens, xii, 14, 20, 97–106, 124–134, 137, 141, 168, 210, 257; opportunistic, 14, 101, 103 Paxson, Heather, 141, 143–144 PDO (Protected Designation of Origin), 9, 227–228 pedigree, 50–52, 60, 63–64, 66. See also breeding; breeds Peltier, Vincent, 192–193, 195–196, 212, 228 Penicillium spp., 102, 105, Penicillium candidum, 115–116, 164 Penicillium roqueforti, 38, 164 pH. See acid development; pH meter phage. See bacteriophage phenotypic data, 57 pH meter: limitations of, 194, 205. See also acidimeter; hot iron test; rennet test Pillsbury, 126 plastic coating, 184, 221, 253. See also cold-ripening Pleasant Ridge Reserve, 90–92, 216 Plimoth Plantation, 42–43 population-level analysis of microbial communities, 107. See also highthroughput methods porous materials: legality of, for use with cheese, 132, 145, 214–215; as reservoir for microbial communities, x–xiii, 14, 121, 130–132, 149. See also biofi lms; gerles; Jameson effect; wooden boards Porter, Michael, 214 potassium nitrate, 70–71, 174 Pouligny-Saint-Pierre, 227 Powell, Ian, 167–168 precautionary principle, 137–140 pregnancy, as risk factor, 125, 136–137, 146 preripening, 112, 149 preservation, fermentation as method of, 30 price. See cost, of cheese; cost, of food probiotics, 101, 157 producers’ groups, 216, 220, 221, 223 professionalization, 230–232 profitability: beyond yield, 87; of Holstein cows, 54; of Montbéliarde cows, 59, 61 Protected Designation of Origin, 9, 227–228 protected names, 19, 220, 227–228

INDEX

quality: milk-quality payments, 96–97; social construction of, 238–242 quotas, for milk and cheese, 223, 225 railroads, 123, 181–182, 244 Rance, Patrick, 25 rare breeds, 16, 44, 49, 61, 118, 236, 264n36 RA series starter cultures, 159 raw milk: consumed at a distance, 122–123; importance for gerle system, xii; optimization of, for cheese, 18, 81, 83–85, 95–100, 111–112, 237; safety issues, 123–124, 150–151 raw-milk cheese: attempts to simulate, 165–168; complementary starter cultures, 160–161; cost-benefit analysis, 138–140; cultural cognition theory applied to, 142–146; factory production of, 221; flavor of, 90, 99; in name only, 139; quality systems, 127–129, 132; safety of, 120, 125, 134–137 Raw Milk Freedom Riders, 143 real cheese, 25–26 Reblochon, 37, 38 fig. 5, 59, 82, 119–122, 130–132, 139, 214–215, 237 reference libraries, microbial, 98 refrigeration, 111–112, 122, 147, 149, 179 regulations: AOP, 9, 82, 86–87, 113, 190, 221, 223–228; and hygiene, 122, 130–133; impact of values on enforcement of, 137–146 relative index test, 107 relative risk, 134–136. See also risk rennet, 34–35, 37, 70, 237, 257; advent of commercial, 157–158; role in ripening, 187; whey-based, 149, 154, 275n27 rennet cup. See rennet test rennet test, 197, 201, 204–205, 207 reservoirs, microbial, 20, 98, 106, 108, 130. See also biodiversity

•

291

Scotch Method, 197, 222 seasonal production, 70, 75, 91–92, 236, 268n42 selective environment, impact of: on breed, 48, 62–63; on cheese characteristics, 211, 243–244, 251–253; on cheese retail, 231; on microbes, 15, 97, 172–174; technology and, 157–161, 214. See also evolution sensitivity, of measurement, 197, 204–205, 213 separators, 150, 182 septicemia, 101 sex-selected semen, 56 sheep, 57, 73, 83 shelf life of liquid milk, factors influencing, 99, 103, 124 shift structure. See time discipline Shiga-toxin producing E. coli, 125. See also Escherichia coli Shorthorn, 49–51, 53. See also Northern Dairy Shorthorn silage, 77–81, 83, 86, 103, 258 silo, 78–79 Singleton’s Dairy, 188 skimmed milk, 150–152 skimmed-milk cheese, 19 Slovic, Paul, 135 Slow Food, 13, 215 Slow Food Presidium, and Artisan Somerset Cheddar, 183 slow vat, 210–211 smallholders, 177–178 smoked cheese, 252 soil: compaction, 75; fertility, 88–89, 235, 257; microbes, 11, 70, 87, 98, 101–103, 106, 172; minerals, 199; seed bank, 80; terroir, 68, 258 somatic cell count, 55, 109 Somerset, 181–183, 187, 197–198, 203–204, 208, 222, 232 souring. See acid development; fermentation space race, 125–126 specialization: of farming systems, 3–4, 71, 239; of Holstein breed, 54, 64; of knowledge, 206–207, 216; of microbial communities, 15, 100

retail models. See cheesemongers; supermarkets Richard, Jean, 95–97 ripening bacteria, 39, 103, 107–108, 121, 164; biofi lms as reservoir of, 130, 132; impact of, on shelf life of liquid milk, 110–111; impact of refrigeration on, 112 ripening cultures, 105, 258; competition from native strains, 170–172; as convergence factor, 90, 115, 139, 169, 173 risk, 119–146; factors affecting susceptibility to, 101–102, 136–137; relative, 134–136. See also cost-benefit analysis; pathogens risk reduction. See HACCP (Hazard Analysis and Critical Control Point) system; HARPC (Hazard Analysis and Risk-Based Preventive Controls) system robotic milking systems, 57–58 Roger, Georges, 116 Roquefort, 38 fig. 5, 219 rumen, 72–75 rumen stasis. See acidosis rumination, 72–73 rural-urban divide, 23–24 Rush Creek, 91–92 safety, of cheese. See risk Saint-Agur, 219, 221 Sainte-Maure de Touraine, 192, 227–228 Saint-Marcellin, 38 fig. 5 Saint-Nectaire, 59, 85–86, 195, 214, 219, 227 Saker, Dora, 204–206, 208, 209 fig. 8, 211–212 Salers: breed, x, 17, 53, 66–67, 74, 85, 227; Tradition, ix, xi–xiii, 4–5, 9, 14, 16, 22, 37, 145, 173, 219, 237, 256. See also Cantal; gerles; Montel, Marie-Christine Salmonella, 104 salt, 34, 37, 39, 124, 125, 170–171, 240 saltpeter. See potassium nitrate sanitizers, 141, 108–109, 121. See also Pasteurian perspective sauerkraut, 30, 73, 78, 100 Savencia, 219–220 scalding, 182, 208, 224 Scanlan, Soyoung, 193–195, 215 schools. See dairy schools; education Science of Artisan Cheese conference, 12

2 92

•

INDEX

taste: preferences, 152, 163, 187, 251; and social construction of quality, 238–242 teat microbial communities, 108–109 temperature: effect of, on curd drainage, 35–36, 153, 174, 224; effect of, on microbes, 39, 100, 103, 111–112, 124; measuring, 149, 197–198. See also pasteurization terpenes, 84–86 territorial cheeses, 25, 38, 38 fig. 5, 71, 201, 203, 206, 212, 243, 258 terroir, 68–69, 167, 171, 212, 215, 258; microbial, 167, 171 testing, microbiological, 96, 107, 113, 122, 126, 128–129, 133, 201, 210 texture: determinants of, 28–29, 31, 33–38, 35 fig. 1, 36 fig. 2, 36 fig. 3, 37 fig. 4, 38 fig. 5, 159, 176, 180, 187, 189, 210; relationship of, to feed, 82–84 thermoduric bacteria, 103, 110 Th ivoyon, Fanny, 226 Thompson, E. P., 188 time discipline, 177–178, 181–182, 185, 190, 210; and farmhouse vs. factory Cheddar, 184–191, 221. See also factory production; Thompson, E. P. titratable acidity, 198, 200, 204–206, 211. See also hot iron test; rennet test tomme de montagne, 86 Tong, Phil, 215–216 total bacterial count, 96–99, 107–108, 110, 149 total mixed ration, 4, 74 trade over distance, 153, 180–181 tradition: as historical practice, 21, 130, 236; as marketing term, 29, 213, 218; in cultural cognition theory, 143–146; mutability of, 79, 184, 211, 245 transatlantic trade. See trade over distance transhumance, 70 transport, 115, 122, 181–182, 185, 207 “true breeding,” 62 true cost accounting, 89, 258 tuberculosis, 7, 13, 52, 103, 123–124, 134, 151. See also Mycobacterium tuberculosis; Mycobacterium bovis Twamley, Josiah, v, 21–22, 147–148, 174, 203, 208, 256

spoilage organisms, 30, 40, 70, 78, 95–99, 108, 112, 131, 151, 181 stabilized cheeses, 28–29, 37, 38 fig. 5, 169 standardization: of approaches to food safety, 125–129; of cheese, xii, 71, 114, 223; of measurement systems, 200, 205 standard plate count. See total bacterial count staphylococci, 103, 172 Staphylococcus aureus, 102–103, 111, 124, 210–211 starch, effect of on rumination, 73–74, 258 starter cultures, 16, 30, 34, 131, 148–149, 152–162, 169–170, 172–174; defined, 156–157; and microbial community, xii, 13–16, 20, 30, 94, 97–101, 104–109, 141, 155, 210; misuse of, 155–156; native, 164–168, 172; RA series, 159; requirement for pasteurized cheese, 97; undefined, 156, 166, 169. See also acid development; ripening cultures; whey starters starter flavor, 41, 90, 154–156, 163–166 Stilton, 38 fig. 5, 71, 180, 235 stochastic variation, 243 Storch, Vilhelm, 151–152, 156, 166 Straus Family Creamery, 110–111 Streptomyces, 102 subsidies, 223, 236 subsistence farming, 76, 235, 245 suicide rates, in farming, forestry, and fishing, 139–140 Sunstein, Cass, 137–140 supermarkets, 3, 24–25, 117, 162–163, 213, 218, 220–221, 231 sustainability: economic, 21–22, 25, 87, 236, 245; environmental, 88–90, 117; through greater efficiency, 5. See also biodiversity sweetness. See Lactobacillus helveticus; ripening cultures “swill milk,” 123 Swiss Cheese Union, 223 Switzerland, 70, 222–226 taints. See off-flavors task-oriented time, 177, 184

INDEX

•

2 93

Van Raden, Paul, 15–16 vegetarian coagulant, 257–258 Vindeløv, Jannik, 158–161 Vinny cheese, 38 volatile sulfur compounds, 39, 83, 104, 171

undefined starters, 156, 166, 169 UNESCO, 11, 218 uniqueness, xii, 5, 16, 21, 66, 163, 168, 171, 244 United Kingdom: adoption of silage in, 79; dairy schools in, 196–201; industrialization of cheese production in, 25, 177; maritime climate, effect of on agriculture, 71, 79, 180; market for imported American Cheddar in, 180–182; native breeds in, 43, 49, 63, 235; politics in, 23; price of cheese in Victorian-era, 22; technical network in, 140. See also Cheddar; Cheshire; Lancashire; Stilton; Wensleydale United States: adoption of silage in, 78; approach to selective breeding in, 52–55, 63; associated dairies in, 179, 185; cattle importation to, 49, 51; food safety regulations in, 127–129, 132; hay-making in marginal areas, 76, 81; politics in, 23, 27; price of land in, 186, 244; trends in dairy farming, 3–4, 202. See also California; New York; Wisconsin unpriced externalities. See true cost accounting Uplands Cheese Company, 90–91 urban vs. rural milk, 23–24, 123 US Centers for Disease Control and Prevention. See Centers for Disease Control and Prevention US Department of Agriculture, 3, 16, 58, 172, 187 US Food and Drug Administration, 101, 127, 136–137, 141, 144, 146

Wallace and Gromit, 234–235, 241, 245 wax coating, 184. See also cold-ripening Wensleydale, 38 fig. 5, 75, 234–236, 239– 241, 243, 245 West, Harry, 195 Westcombe Dairy, 211 whey, 34–35, 37, 93, 114, 121, 159, 182, 185, 198, 206–208, 211, 216, 237, 259 whey starters, 149, 154, 173, 192, 226, 258, 275n27 whisky, 212–214 wildflowers, 76, 80, 85, 91, 236 Willard, Xerxes, 178–187 Williams, Jesse, 179 wine, 11–12, 17, 65–66, 113, 217, 218, 223, 229, 238–240 Winnimere, 35 fig. 3 Wisconsin, 56, 78, 90, 154–155, 165, 179, 187, 216 Wisconsin Cheese Makers Association, 154–155 Wolfe, Benjamin, 13–15, 170–171 women’s work. See gender roles wooden boards, 121, 130–131, 141, 145–146, 214–215. See also biofi lms; community: microbial wood wool, 108–109 yeast, 20, 105, 107, 255, 256, 257 yield: of cheese, 59, 61, 159, 182, 187, 198, 216; of crops, 42, 71, 76–79; economic, 61, 87, 89; of milk, xi, 2, 4, 5, 43, 52–54, 63, 66, 67, 73–74, 227, 235, 237. See also extensive farming systems; intensive farming systems yogurt, 18, 23, 31, 35 fig. 1, 38, 41, 110, 207

Vacherin Fribourgeois, 223–226 Valençay, 38 fig. 5, 227 Valenze, Deborah, 203 value/taste, social construction of, 238–242 van Gogh, Vincent, 241–242

294

•

INDEX

C a l i for n i a S t u di e s i n Food a n d C u lt u r e Darra Goldstein, Editor 1. Dangerous Tastes: The Story of Spices, by Andrew Dalby 2. Eating Right in the Renaissance, by Ken Albala 3. Food Politics: How the Food Industry Influences Nutrition and Health, by Marion Nestle 4. Camembert: A National Myth, by Pierre Boisard 5. Safe Food: The Politics of Food Safety, by Marion Nestle 6. Eating Apes, by Dale Peterson 7. Revolution at the Table: The Transformation of the American Diet, by Harvey Levenstein 8. Paradox of Plenty: A Social History of Eating in Modern America, by Harvey Levenstein 9. Encarnación’s Kitchen: Mexican Recipes from NineteenthCentury California: Selections from Encarnación Pinedo’s El cocinero español, by Encarnación Pinedo, edited and translated by Dan Strehl, with an essay by Victor Valle 10. Zinfandel: A History of a Grape and Its Wine, by Charles L. Sullivan, with a foreword by Paul Draper 11. Tsukiji: The Fish Market at the Center of the World, by Theodore C. Bestor 12. Born Again Bodies: Flesh and Spirit in American Christianity, by R. Marie Griffith 13. Our Overweight Children: What Parents, Schools, and Communities Can Do to Control the Fatness Epidemic, by Sharron Dalton 14. The Art of Cooking: The First Modern Cookery Book, by the Eminent Maestro Martino of Como, edited and with an introduction by Luigi Ballerini, translated and annotated by Jeremy Parzen, and with fi ft y modernized recipes by Stefania Barzini 15. The Queen of Fats: Why Omega-3s Were Removed from the Western Diet and What We Can Do to Replace Them, by Susan Allport 16. Meals to Come: A History of the Future of Food, by Warren Belasco 17. The Spice Route: A History, by John Keay

18. Medieval Cuisine of the Islamic World: A Concise History with 174 Recipes, by Lilia Zaouali, translated by M. B. DeBevoise, with a foreword by Charles Perry 19. Arranging the Meal: A History of Table Service in France, by Jean-Louis Flandrin, translated by Julie E. Johnson, with Sylvie and Antonio Roder; with a foreword to the Englishlanguage edition by Beatrice Fink 20. The Taste of Place: A Cultural Journey into Terroir, by Amy B. Trubek 21. Food: The History of Taste, edited by Paul Freedman 22. M. F. K. Fisher among the Pots and Pans: Celebrating Her Kitchens, by Joan Reardon, with a foreword by Amanda Hesser 23. Cooking: The Quintessential Art, by Hervé This and Pierre Gagnaire, translated by M. B. DeBevoise 24. Perfection Salad: Women and Cooking at the Turn of the Century, by Laura Shapiro 25. Of Sugar and Snow: A History of Ice Cream Making, by Jeri Quinzio 26. Encyclopedia of Pasta, by Oretta Zanini De Vita, translated by Maureen B. Fant, with a foreword by Carol Field 27. Tastes and Temptations: Food and Art in Renaissance Italy, by John Varriano 28. Free for All: Fixing School Food in America, by Janet Poppendieck 29. Breaking Bread: Recipes and Stories from Immigrant Kitchens, by Lynne Christy Anderson, with a foreword by Corby Kummer 30. Culinary Ephemera: An Illustrated History, by William Woys Weaver 31. Eating Mud Crabs in Kandahar: Stories of Food during Wartime by the World’s Leading Correspondents, edited by Matt McAllester 32. Weighing In: Obesity, Food Justice, and the Limits of Capitalism, by Julie Guthman 33. Why Calories Count: From Science to Politics, by Marion Nestle and Malden Nesheim 34. Curried Cultures: Globalization, Food, and South Asia, edited by Krishnendu Ray and Tulasi Srinivas

35. The Cookbook Library: Four Centuries of the Cooks, Writers, and Recipes That Made the Modern Cookbook, by Anne Willan, with Mark Cherniavsky and Kyri Claflin 36. Coffee Life in Japan, by Merry White 37. American Tuna: The Rise and Fall of an Improbable Food, by Andrew F. Smith 38. A Feast of Weeds: A Literary Guide to Foraging and Cooking Wild Edible Plants, by Luigi Ballerini, translated by Gianpiero W. Doebler, with recipes by Ada De Santis and illustrations by Giuliano Della Casa 39. The Philosophy of Food, by David M. Kaplan 40. Beyond Hummus and Falafel: Social and Political Aspects of Palestinian Food in Israel, by Liora Gvion, translated by David Wesley and Elana Wesley 41. The Life of Cheese: Crafting Food and Value in America, by Heather Paxson 42. Popes, Peasants, and Shepherds: Recipes and Lore from Rome and Lazio, by Oretta Zanini De Vita, translated by Maureen B. Fant, foreword by Ernesto Di Renzo 43. Cuisine and Empire: Cooking in World History, by Rachel Laudan 44. Inside the California Food Revolution: Thirty Years That Changed Our Culinary Consciousness, by Joyce Goldstein, with Dore Brown 45. Cumin, Camels, and Caravans: A Spice Odyssey, by Gary Paul Nabhan 46. Balancing on a Planet: The Future of Food and Agriculture, by David A. Cleveland 47. The Darjeeling Distinction: Labor and Justice on Fair-Trade Tea Plantations in India, by Sarah Besky 48. How the Other Half Ate: A History of Working-Class Meals at the Turn of the Century, by Katherine Leonard Turner 49. The Untold History of Ramen: How Political Crisis in Japan Spawned a Global Food Craze, by George Solt 50. Word of Mouth: What We Talk About When We Talk About Food, by Priscilla Parkhurst Ferguson 51. Inventing Baby Food: Taste, Health, and the Industrialization of the American Diet, by Amy Bentley

52. Secrets from the Greek Kitchen: Cooking, Skill, and Everyday Life on an Aegean Island, by David E. Sutton 53. Breadlines Knee-Deep in Wheat: Food Assistance in the Great Depression, by Janet Poppendieck 54. Tasting French Terroir: The History of an Idea, by Thomas Parker 55. Becoming Salmon: Aquaculture and the Domestication of a Fish, by Marianne Elisabeth Lien 56. Divided Spirits: Tequila, Mezcal, and the Politics of Production, by Sarah Bowen 57. The Weight of Obesity: Hunger and Global Health in Postwar Guatemala, by Emily Yates-Doerr 58. Dangerous Digestion: The Politics of American Dietary Advice, by E. Melanie duPuis 59. A Taste of Power: Food and American Identities, by Katharina Vester 60. More Than Just Food: Food Justice and Community Change, by Garrett M. Broad 61. Hoptopia: A World of Agriculture and Beer in Oregon’s Willamette Valley, by Peter A. Kopp 62. A Geography of Digestion: Biotechnology and the Kellogg Cereal Enterprise, by Nicholas Bauch 63. Bitter and Sweet: Food, Meaning, and Modernity in Rural China, by Ellen Oxfeld 64. A History of Cookbooks: From Kitchen to Page over Seven Centuries, by Henry Notaker 65. Reinventing the Wheel: Milk, Microbes, and the Fight for Real Cheese, by Bronwen Percival and Francis Percival