By DANIEL ENGBER New York Times
On a Sunday evening last September, stevia became famous. In the final episode of “Breaking Bad,” an image of the sweetener filled the TV screen. Lydia emptied the packet into her mug of camomile tea, not knowing that Walt, her former partner, had poisoned it. “How are you feeling?” he later asked. “Kind of under the weather, like you’ve got the flu? That would be the ricin I gave you. I slipped it into that stevia crap that you’re always putting in your tea.”
In an interview with The Guardian, published the next day, the actress who played Lydia laughed about the product anti-placement: “Sorry, stevia,” she said. “Oh, I suppose it feels a bit rubbish. Do you think anyone actually bought it anyway?”
Actually, yes. The natural, noncaloric sweetener, made from the leaves of a Paraguayan shrub, now sits in second place in the $400 million market for sugar-bowl sachets. (Sucralose hangs on at No. 1.) When Cargill introduced the leading brand of stevia, called Truvia, in 2008, the company heralded it as “a new category of sweet.” Sure enough, imitators followed. A few weeks later, Merisant put out PureVia — made from the same ingredient — and then the manufacturer of Sweet’N Low started filling light green packets with what it called Stevia in the Raw.
But the battle for the sugar-substitute market is not about packets on the table; the real money is in being the go-to additive for diet foods, especially diet drinks. When the F.D.A. approved a chemical called cyclamate in 1951, a brand-new industry emerged: No-Cal soda, Diet Rite and all the other sugar-free refreshments. Cyclamate was banned in 1969 for promoting bladder cancer in rats, but aspartame later took its place. In 1983, Coca-Cola put aspartame-based NutraSweet in Diet Coke, and sales soared. “The rule of thumb is that 60 percent of all high-potency sweetener sales is in beverages,” says John Fry, a food scientist and 40-year industry veteran who consults on sweeteners for Cargill. “You have to be in soft drinks, one way or another.”
As badly as stevia needs the soft-drink companies, the soft-drink companies may need stevia even more. While sweetened carbonated beverages still make up around one-fifth of all the liquids we consume, the volume sold has dropped, per capita, every year since 1998. We’re more afraid of sugar than we’ve ever been. What yesterday were seen as “empty calories” have today been designated “toxic.” Doctors warn that cans of soda put fat into your liver, weaken your response to insulin and increase your risk of heart disease and diabetes. The panic over sugar has grown so pervasive that other dietary boogeymen — salt and fat and gluten — seem like harmless flunkies in comparison. (In 2012, when the market-research firm Mintel asked consumers which ingredients or foods they were trying to avoid, sugar and added sugar topped the list, by a wide margin.)
The soda companies have tried to tack into the headwind: In 2010, PepsiCo promised to reduce the sugar in its products by 25 percent, and the following year Coca-Cola told the British government that it would cut the calories in soda. But consumers are not content to switch to artificial sweeteners. Sales volume of diet soda fell by 12 percent in the last six years. Far from serving as a life raft for the industry, that business is leaking dollars, too.
The problem is that for all the fear of fructose, consumers have grown just as wary of its beaker-born alternatives. To health fanatics, they seem noxious on their face: Sweet’N Low comes from a derivative of coal; Equal is made from methanol and converts to formaldehyde when digested; Splenda is a chlorinated sugar. Others worry over well-worn rumors of their ill effects — tumors, headaches and depression. More recent studies hint that diet drinks can cause the very problem they’re meant to solve and make us fat instead of thin. (Lab rats fed with noncaloric sweeteners sometimes start to overeat, as if the ersatz sugar primed their rodent tongues for other sweets.)
The science on these questions is inconclusive at best. There’s no clear evidence that artificial sweeteners cause cancer or obesity, at least in human beings. But the fear of artificial sweeteners was never quite a function of the scientific evidence — or never of just that. It stems as much from a sense that every pleasure has its consequences: that when we try to hack our taste buds in the lab — to wrench the thrill of sugar from its ill effects — we’re cheating at a game we’ll never win.
Because it came from a plant, stevia seemed to offer a way to sneak around the rules. “We’ve been consuming sweetened products for 8,000 years,” says Jim Kempland, a former sales manager for NutraSweet and later vice-president of marketing for Sweet Green Fields, a U.S.-based stevia producer. “The enormous rates of obesity both here and globally are not going to curtail our human cravings for sweet things. The alternative, or possibly the solution, would be to ask, How do we create those things naturally, so that they can fit into a lifestyle that allows us to have things we like?”
The beverage companies knew that sales growth in the natural category had almost tripled that of other foods and beverages, driven by the intuition that they’re better for our health. What if Diet Coke or Diet Pepsi could be made into a “natural” product — a sweet-tasting drink with zero calories and nothing artificial? Would a more wholesome substitute for sugar — one that comes from plants instead of factories — let us have our sweets and eat them too?
But the industry soon discovered that its salvation would have to be postponed. For all the hype, stevia had a fatal flaw: Its taste.
Deep inside Cargill’s corporate headquarters in Wayzata, Minn., where it runs its $136 billion business, a technician in a hairnet put out a bowl of strawberries. Melanie Goulson, a food scientist in the company’s corn-milling unit, had taken me to a laboratory kitchen outfitted with frying pans and cleavers and stir-plates spinning fluids with Teflon-coated bars. She waited as I dipped a berry in a sample of white granules and popped it in my mouth.
Truvia felt a lot like sugar on my tongue — much more so than the ***** brands — but there was something strange about its sweetness. The flavor dawdled and digressed, until it seemed as if I’d chewed a nub of licorice or soaked my gums in watered-down Campari. This has been stevia’s problem from the start: It has a bitter taste that lingers. The defect may be unobtrusive in small doses — the amount you sprinkle in your cappuccino — but it’s ruinous at the quantities it takes to make a diet soda. “Anybody who tasted stevia in 2008, when it was just about to be permitted in the United States,” Fry says, “would have been painfully aware that this was not an aspartame or a sucralose in terms of sweetness quality.”
Goulson and her team have tried to bolster stevia by blending it with other additives. “We’re trying to understand how sweet this product should be,” she said. “What features do people like? What don’t they like? How can we get the recipe just right?” In a perfect world, they’d find a way to sand down the jagged edges of its flavor, so stevia could match the taste of table sugar. (Sugar is “widely accepted as the gold standard for sweet taste,” Goulson told me.) At the very least they’re hoping to make stevia as appetizing as the chemicals in Diet Coke and Diet Pepsi. “Taste is king,” Goulson said. “I mean, the healthiest product in the world really isn’t relevant if people don’t enjoy the taste.”
A member of her team poured out several cups of orange liquid for me to try. It wasn’t orange juice but orange drink: The beverages had been watered down by more than half to cut the sugar content, then spiked with stevia and other natural flavors. I knew the first one they handed me would be the taste-test patsy, but even so, its awful flavor caught me by surprise: a hit of sour that slowly faded to a bitter, sticky sweet. Then I tried the second cup. Goulson had tweaked the recipe so that its sweet and sour flavors came and went in harmony. The drink contained the same amount of stevia but tasted more like orange juice. “It’s a science and an art,” she said. “Depending on the flavor, if you’re dealing with an apple-based or orange-based or berry-based system, we may change the ratio of the acids to help that flavor pop out more.”
Her artistry has limits, though. It’s possible to minimize the aftertaste of stevia, but only when there’s still some sugar to go with it. “Generally you’ve seen 30 percent, 35 percent, 50 percent [sugar] reductions on the market,” says Wade Schmelzer, one of Goulson’s lab lieutenants. The biggest soft-drink companies have shunted off the product to a little-known and little-loved class of carbonated drinks: the midcalorie soda. In parts of Europe, Coca-Cola now makes Sprite with a mix of stevia and sugar, to cut the calories by 30 percent; last June, it started selling Coca-Cola Life, another stevia-and-sugar drink, in Argentina. PepsiCo has also tried a stevia-sweetened, midcalorie version of its cola at markets in Australia. For now, their dreams of making fully natural, fully diet soda — the beverage market’s killer app — are on hold. “We know that in the end — in an ideal world,” Schmelzer said, “the customers would like to be at zero sugar.”
Later in the day I met with Mark Brooks, Truvia’s business director. “There are those people who love artificial sweeteners,” he told me, “and there are those people for whom sugar is it; but the growing segment, and now the biggest segment (and we see this worldwide), is a group that we call the naturally splendids — those people who are really looking for that natural source.” Brooks’s department guessed that in 2010, 39 percent of household shoppers fit into this latter group, but that number may be rising.
Still, all the splendids in the world won’t help the bottom line if Truvia can’t appeal on flavor. It’s not enough to tell consumers that a product comes from farms in Paraguay. It’s not enough to let them think it’s better for their health. “At the end of the day,” Brooks said, “the consumer will buy you again because you taste great.”
Not every natural, noncaloric sweetener comes from stevia. As Cargill tries to grapple with its gremlins, other firms have done their best to find alternatives. In the summer of 2012, McNeil Nutritionals — the maker of Splenda — put out Nectresse, a product made from the Chinese mountain-orchard melon known as luo han guo, or monk fruit. Sold in tangerine-colored packets, Nectresse was supposed to be a more natural-seeming natural product. While few of us think of leafy plants as being sweet, monk fruit brings to mind a cantaloupe or a honeydew. “A fruit ingredient in your food or in your beverage is very intuitive for the customer,” says David Thorrold, the C.E.O. of BioVittoria, which now controls a major portion of the world’s monk-fruit supply.
Even Cargill is developing stevia alternatives. As Nectresse hit the market, the company received a bioprospecting permit from the South African government, giving it the right to exploit the molomo monate plants that grow on rocky slopes in northeast South Africa. At her lab in Minnesota, Goulson tested a sweet-tasting amino acid drawn from those plants and concluded that it would be enough to flavor diet beverages. John Fry called it one of the finest zero-calorie sweeteners he had ever tried.
Yet these ingredients have problems even worse than stevia’s. While monk-fruit extract isn’t quite as bitter, its flavor can be very slow to build, and it, too, lacks the oomph to sweeten diet soda. (It’s also more expensive.) The molomo extract tastes a lot like sugar, but when exposed to UV light, it undergoes a horrid transformation. In the early 2000s, scientists at Coca-Cola added the sweetener to bottles of Sprite, then left them on the roof over the weekend; by Monday morning, the soda had turned urine yellow and developed the smell of feces.
So the beverage-makers have returned to where they started. Coca-Cola set out to find what is often called the holy grail of diet beverages — a natural alternative to aspartame — about 10 years ago. If today we’re stuck with stevia, a curly-haired chemist named Grant DuBois may be as much to blame as anyone.
DuBois left Coca-Cola in 2011, after two decades of working for the company’s global research and development team. “I was feeding my ideas into this supersecret group responsible for flavor chemistry,” he said when we met for breakfast in the suburbs of Atlanta. Now 67, DuBois runs a sweetener consultancy from his Georgia home, and he seemed most at ease while making presentations. The one he’d prepped for me described how hard it is to find a sweetener that works and why the best natural product that we have is a compromise.
A new product should be considered for development, he explained, only if it meets nine specific requirements. It’s not enough to find a chemical that’s sweet; it also has to sweeten foods in the ways that matter most, to the right degree, without an aftertaste, without a linger and without diminishing effect. And even one that has the taste of granulated sugar would be deficient if it weren’t also stable in solution, resilient to hot and cold, safe to drink, cheap to make and amenable to patenting. An “all-natural” sweetener must meet these nine metrics for success, and then a 10th: It must be taken from a living thing.
That last requirement left DuBois with little room to work. In all, there are about a hundred natural, noncaloric chemicals that, by pure coincidence or some unlikely quirk of evolution, have found a way to hijack the mouth’s machinery for sensing sugar. One category comes from animals. Lysozyme, a chemical found in tears and spit, and also in the whites of eggs, can be very sweet. (In hens and turkeys, it’s sugary but pungent; in soft-shelled turtle eggs, it tastes more like licorice.) The rest derive from vegetation, including: one compound drawn from crushed hydrangea leaves, used in Japan for sweet tea; another from a Malaysian plant called lemba, with tiny yellow flowers and fruits that look like cloves of garlic; the seeds of a swollen caper berry in Yunnan, China, which locals chew as candy; and monk fruit, a cousin to the cucumber and the bitter melon, which grows in Guangxi.
Most of these, DuBois knew, would never stand a chance. He pushed aside his knife and fork and tilted the laptop’s screen to show me why. For the natural-sweetener project, he started with the first and most important of his tests, shown on his display as “Taste Quality Metric 1: Maximal Response.” Were any of the natural compounds strong enough to sweeten Diet Coke? At first this seemed an easy obstacle to overcome: Lysozymes are at least 10 times as sweet as sugar; monk-fruit extract is 20 times as strong as lysozyme; and the sweetener drawn from the lemba plant is 10 times more intense than monk fruit. But those numbers correspond only to tiny doses of the chemicals, enough to match the strength of, say, a teaspoon of sugar mixed into a cup of water — a 2 percent solution. To make a cup of Coca-Cola, you’d need much more: about six teaspoons’ worth of sugar, for a 10.4 percent solution. (Pepsi is a little sweeter, at 11 percent. Root beers and some fruit-flavored sodas can be 12 percent or more.) That’s where many substitutes fall short.
DuBois had a set of graphs tracking how the power of a sweetener changes with its concentration. He included curves for six different compounds, from saccharin to stevia, but they all looked very much the same. Each curve rose steeply, gaining sweetness with every increment in milligrams per liter, then appeared to hit a ceiling, a point at which the sweetness flattened out. Once you reach that threshold concentration, a compound loses its effect: No matter how much more of it you pump into a beverage, you’ll never get a sweeter taste. The ceilings for some chemicals are high enough to flavor carbonated drinks. Aspartame, for example, can match the taste of sugar in a 16 percent solution. But others reach their limit much too soon. That’s why today you’ll never find a Diet Coke that’s made with saccharin. At best it would match the sweetness of a sugar drink at 10.1 percent.
He gestured at the curve for stevia, which didn’t seem much better than the one for saccharin. “That looks like a death sentence,” he said. But when I asked him to name his favorite noncaloric sweetener, DuBois demurred and cleared his throat. “None of them,” he said, “tastes like sugar.”
Blending stevia extract and dextrose for Stevia in the Raw. Floto + Warner for The New York Times
We clicked ahead to Metric 2 — “Flavor Profile.” Most products have at least a hint of bitterness or licorice; some have a metallic note or even menthol cooling. Metric 3 considers how the compound’s taste develops on the tongue. Sip a soda made with sugar, and the taste should reach its peak of sweetness in four seconds, then fall off 10 seconds later. Zero-calorie substitutes tend to lag behind: They come on too slowly, and then — much worse — they stay too long, clinging to your mouth in a disconcerting glaze. Even aspartame takes an extra second to hit its sweetness high and hangs around an extra four before it goes away.
It’s hard enough to handle these requirements, DuBois said, when you’re making chemicals from scratch. Never mind if you have to meet them using something natural. As he searched for a plant-based substitute to put in diet beverages, DuBois remembered that he played around with a compound drawn from stevia while working for a start-up in the 1970s. At the time, he figured out a way to make it work about as well as aspartame. “It still had a lingering sweetness, but no bitterness,” he said. But his lab work changed the compound’s shape. It wasn’t natural anymore.
Could the chemical in stevia — called rebaudioside-A — work in an unaltered state? It seemed more promising than any other option on his list. Stevia was commercialized in Japan more than 30 years ago and then bred to make as much Reb-A as possible. While the sweetener has some bitterness and licorice, its warts are modest next to those of monk fruit or lemba or any of the other plant derivatives. Also a company in Kuala Lumpur called Stevian Biotechnology (later renamed PureCircle) said it could produce Reb-A on the cheap. In 2003, DuBois passed along his findings: Coca-Cola should try to make a sweetener from stevia. Others at the company agreed and made a deal with Cargill to get the plant-based ingredient approved. Coke would do the basic scientific work on stevia; Cargill would work on the supply chain. “There was no alternative,” DuBois told me, looking very serious. “I decided that of all these 100 or so compounds — all the natural, noncaloric sweeteners that are known — the one that meets the metrics best — and it’s not perfect — is Reb-A.”
At a greenhouse in East Lansing, Mich., a scientist named Ryan Warner plucked a pair of shriveled-looking leaves from a specimen of stevia. They looked like wrinkled basil, but when I put one in my mouth, it had the taste of Sen-Sen and the feel of leaded paint. Warner laughed as I did my best to gulp it down. “Did you see me spit mine out?” he said.
Since 2010, Warner, an associate professor of horticulture at Michigan State University, has led a team of scientists funded by PureCircle, the top stevia producer, to engineer a better version of the crop. Rather than alter the chemical, scientists are trying to alter the plant. They’d like to find a novel strain that makes sweet-tasting compounds with less bitterness and aftertaste — a kind of superstevia. Every year, he breeds a set of cultivars and tests the composition of their leaves. It’s a funny job for someone who has so little interest in sweet taste: “I’m a fat, oil and beer kind of person,” Warner told me. “The best dessert is a cheese plate.”
What I tasted in his greenhouse might be called Stevia 1.0: It’s not so far off from the crude leaf extract that has been sold in health-food stores in the United States since the early 1980s. The composition of these extracts — and thus their taste — varies with the plants from which they’re made, which may be one reason that they’ve never been cleared as a food additive by the F.D.A. (It’s possible that chewing on raw stevia is dangerous.)
It’s Warner’s task to tweak the makeup of the organism, using only natural means. Every stevia plant carries a certain mix of metabolites, called steviol glycosides, that define its sweetness signature. The most abundant glycoside in wild stevia has an overly bitter taste that wouldn’t fly in foods or beverages. Enzymes in the plant convert that glycoside into several dozen others, including rebaudioside-A — the substance that DuBois recommended as a natural sweetener. But for any given plant, it’s hard to know how much of each you’ll get. “There’s still not a great understanding of why there’s so much variability,” Warner said.
Until recently, PureCircle wanted plants that make as much Reb-A as possible. (That was Stevia 2.0.) Now PureCircle hopes that Warner and his team can start this process over and optimize the plant for a different set of glycosides with still fewer imperfections in their flavor. The project could take many years of crossing plants. In the wild, the shrub grows at the borders of grasslands and marsh. When cultivated by farmers — in Paraguay, Kenya, China and California’s Central Valley — the shrub demands enormous care: Crops are often raised from cuttings, dug into the ground by hand. The work seemed just as hard in Michigan. A few miles from the greenhouse, Warner’s team had planted rows along a highway and left them under strips of plastic for weed control. When I visited last spring, Warner’s seedlings had just gone in that morning. He told me that his team would harvest the foliage and grind it to a powder in the Red Devil — a souped-up paint-shaker on which a member of the lab had drawn a demon-horned Bart Simpson spearing a leaf of stevia with a pitchfork.
Warner tests the light green powders for their glycosides and then selects the ones with better profiles for another round of breeding. The details of his work are secret, but it seems as if he’s searching for a pair of chemicals — rebaudioside-D and rebaudioside-X — that tend to show up in very small amounts. If the lab can find a way to make these more abundant, the rewards would be enormous. “These next-generation stevia sweeteners hold a lot of potential for unlocking new growth within the stevia industry,” says Robert Brooke, C.E.O. of Stevia First, an agricultural and biotech firm. “Reb-A has done quite well. . . . Reb-D and Reb-X can take it to a whole new level.”
The beverage giants feel the same. In a patent filed last September, PepsiCo scientists put out a set of spider plots to show that Reb-D is sweeter than Reb-A, with far less bitterness and licorice. The new ingredient would allow for the creation of a “natural carbonated cola beverage product” sweetened entirely with stevia and without the need for “taste-masking agents.” Coca-Cola’s researchers followed with a similar patent at the end of May, this one with graphs to show that Reb-X tastes even better than Reb-D and would make an even more delicious can of diet cola.
All these efforts raise a basic question, though. As Grant DuBois discovered in the 1970s, stevia can taste as good as NutraSweet, if you’re willing to apply the tools of synthetic chemistry. But even if you avoid these sorts of interventions, what about the science in between — from crossbred crops to lab-grown glycosides and fancy formulations? At what point in the process of refinement does a natural product lose its link to nature?
At the kitchen lab in Minnesota, Goulson offered a taste of what might be next. Her team set out a Sprite-type drink, fizzy lemon-lime, in which the sugar had been cut by 75 percent — more than has been feasible with natural sweeteners so far. “This is outside of the space where anybody has been operating,” Goulson said. She’d used an experimental extract that might soon be on the market, and while the soda had a bitter aftertaste, its defects were subdued. Goulson wouldn’t tell me much about the mystery ingredient, only that it’s “in the world of stevia.”
Other signs suggest that Cargill is about to make the jump to next-gen natural sweeteners. On March 6, a multimillion-dollar deal with the Swiss biotech firm Evolva was announced; with Cargill they would develop a yeast-fermented version of stevia. Instead of raising stevia from cuttings of the plant, then steeping the leaves and sucking out the glycosides, Evolva makes them by using microbes to perform the needed feats of biochemistry.
When Evolva gets its system going — which should happen in the next few years — the cost of stevia-based sweeteners could plummet. Cargill wouldn’t be beholden to rain and sun and weeds, nor to the natural limits of plant biology. Fermentation would allow the company to make whichever glycoside it wanted, at whatever quantity and purity it needed.
The Evolva process, like another that’s in development at Stevia First, makes use of homemade microorganisms, with foreign genes stuffed into their nuclei. That’s one way to pump out Reb-D or Reb-X, but what about the consumers who are drawn to “natural” claims — will they still go for stevia when it flows from a vat of G.M.O.’s? And will regulators object?
Later that afternoon, I put these questions to David Henstrom, Cargill’s global business director for health ingredients and the man now in charge of selling Truvia. “Country by country they have different ways that they describe what you can say is natural and what you can’t,” he told me. In the United States, food-and-beverage companies get to make the judgment for themselves. “There might be some products that aren’t trying to make that hard ‘all-natural’ claim,” he continued. “Some people are claiming naturally sourced. Some people are claiming nature-identical. It comes down to the product and what the product is trying to say and deliver to the customer.”
Natural zero-calorie sweeteners have so far been caught between two imperatives: What they want to say and what they can deliver. It used to be that natural sweeteners weren’t sweet enough; now they have an added problem: They aren’t fully natural.
“‘Natural’ would mean that I picked it from the ground,” said Donna LiVolsi, the director of operations at Cumberland Packing Corporation, which invented Sweet’N Low, the first artificial sweetener sachet, in 1957. I met her near the Navy Yards of Brooklyn, where Cumberland still makes Sweet’N Low, along with value brands of aspartame and sucralose and a couple of natural-sugar substitutes — Stevia in the Raw and Monk Fruit in the Raw. When I asked LiVolsi if she thought these latter two were “natural,” she said she couldn’t answer, because each consumer has a sense of what the word means to them.
It’s a question that has bedeviled beverage-makers, too. In the fall of 2012, a German food company surveyed 4,000 people in eight European countries, to find out how they understood the “natural” claim. Almost three-quarters said they thought that natural products were more healthful and that they’d pay a premium to get them. More than half argued that natural products have a better taste. But the respondents weren’t sure what degree or form of processing would be enough to strip a product of its natural status. Some drew a line between sea salt (natural) and table salt (artificial). Others did the same for dried pasta and powdered milk, though both are made by dehydration.
My visit to the Navy Yards showed how confusing this can get. LiVolsi took me through the loading bays, where drums of ingredients arrive in giant bags from overseas, and then into a noisy room where mostly women sit in front of large machines, sweeping packets into pink boxes. A set of tubes suspended from the ceiling ferried blends of saccharin or stevia from an upper floor, and wayward dust dissolved into the atmosphere. With every breath, I could feel a sweetness tingling my tongue like tiny snowflakes.
Every sweetener at Cumberland goes through the same routine, from a blending tub upstairs into a giant hopper and then down across the belts and wheels below. LiVolsi pointed to what looked like a whipping Catherine wheel, with spikes that spurted powder into paper packets, 50 blurring past in every second. It doesn’t matter if it’s made from Chinese orchard fruit or a derivative of coal, the stuff ends up in this contraption, strobing pink for saccharin, blue for aspartame, gold for sucralose, pale green for stevia. “We get the ingredient, we get the bulking agent, we blend it together, we make the batch, we test the batch, we put it in the packets and life moves on,” LiVolsi said. They’re different flavors to us, she said, “that’s how we look at it.”
Some consumers also wonder if the natural sweeteners aren’t simply different flavors of the products they’ve been trying to avoid. At the beginning of July, just as PepsiCo got approval for Reb-D and Coca-Cola said it would be working on Reb-X, a 58-year-old woman living in Hawaii filed suit against Big Stevia. In March she bought a box of Truvia at Walmart because she thought it was a natural product. Now she’s convinced it’s no such thing. Her complaint declared that “Reb-A is not the natural crude preparation of stevia,” and that its manufacture is not “similar to making tea,” as Cargill’s packaging suggests. Rather, it’s “a highly chemically processed and purified form of stevia-leaf extract.”
Hers was not the only attack on Cargill’s natural sweetener. In ongoing negotiations to settle a similar suit, Cargill has offered to remove the phrase “similar to making tea” from the packaging and/or add an asterisk to the product’s tagline, “Nature’s Calorie-Free Sweetener,” directing people to a website F.A.Q. That page would explain that Truvia contains very little stevia, by weight, and that its main ingredient — erythritol — comes from yeast that may be fed with genetically modified corn sugar. “As with almost all finished food products,” the F.A.Q. would say, “the journey from field to table involves some processing.”
“Five years ago, these lawsuits were unusual,” says Rebecca Cross, an attorney with BraunHagey & Borden in San Francisco who has helped a dozen food-product companies defend themselves against charges of health-claim mischief and natural-product fraud. “Now in California, you have four or five every week. Before 2008, I don’t know if you’d get four or five every year.”
So far, none of the natural-product cases have gone to trial. They have either been settled out of court or dismissed by a judge. But many more are in the pipeline, and the F.D.A. has done little to clear up the regulations. The agency began to draft a formal definition of the term “natural” in the early 1990s but never finished. It fell back instead on an advisory opinion, hinting only that a natural food is one without artificial or synthetic additives. Judge William Orrick of the U.S. District Court, meanwhile, ruled in October that natural-products lawsuits can’t be put off until the feds produce a more specific rule. “The court is skeptical that the F.D.A. will develop a policy regarding the term ‘natural’ anytime soon,” he wrote, “especially since it has considered the matter for over two decades but still has not provided further guidance.”
In the face of this uncertainty, manufacturers have chosen to be pragmatic: They won’t say that stevia is more healthful than aspartame or sucralose just because of how it’s made. Nor will they acknowledge that one form of stevia might be more or less natural than another. “I think the important thing is that there’s a choice for everyone, and you can pick what’s right for you and what you need,” Goulson told me on our tour of tasting labs at Cargill. “What’s good for one person may not be good for another.”
Earlier, she showed me where her colleague Nese Yurttas feeds Truvia to employees from the Cargill campus or kids from local schools, in order to study their responses. Yurttas earned her masters at the University of Minnesota, working on cheddar-cheese aromas. Now she runs panel tests to see if average consumers like the taste of stevia. In 2011, Yurttas asked a group of children to test diet chocolate milk, with the sugar reduced by 30 percent. They gave it some mixed reviews. “Tasted so good it blowed my head off,” one kid told Yurttas. “I think it should be sweeter,” another said. On the whole, however, when asked to rate its flavor on a “nine-point hedonic scale,” the children gave the drink an average score of 7.9, the same as standard chocolate milk. They liked the taste, all right, but would their parents say it’s natural? There is no nine-point scale to measure that.