HYDE PARK, NEW York, used to be famous as the hometown of U.S. president Franklin D. Roosevelt. To most food lovers today, it’s better known as the site of the Culinary Institute of America. The CIA, as it’s known, is America’s preeminent culinary school, the nursery that incubated countless top chefs.
For all its august stature now, the CIA had a fairly modest beginning. As the Second World War drew to a close, America faced a flood of newly demobilized young soldiers who needed jobs—and, in many cases, needed job skills, having spent the entirety of their brief adult lives in the military. The wife of Yale University’s former president figured some of those returning soldiers could find work as cooks, so she founded the New Haven Restaurant Institute to teach them how. The idea took off, and the little cooking school soon acquired grand ambitions and a grand name to match. By 1970, the CIA had outgrown its home near the Yale campus in Connecticut, and it moved into its present location on the Hudson River, an hour upstream from Manhattan. The big brick main building was once a Jesuit novitiate, a retreat for trainee priests—a nice parallel for today’s novices as they prepare for a lifetime’s dedicated service at the stove, rather than the altar.
Of all the CIA’s eminent instructors, the two best positioned to bridge the gap between the science of flavour and its application in the kitchen are Chef Jonathan Zearfoss and Dr. Chris Loss. Together, Zearfoss and Loss teach flavour science to the aspiring chefs. They’ve spent a lot of time pondering the science behind what tastes good, and both of them are comfortable in the lab as well as in the kitchen. I met the two over lunch at the CIA’s Italian restaurant, the Ristorante Caterina de’Medici. As we sip spritzers of grapefruit juice and mint, Zearfoss explains that much of what a chef does in designing a dish is to balance contrast and similarity of the ingredients. Foods go together, he says, either because the flavours of one echo those in another so that they blend well, or because their different flavours make one another stand out. Every chef navigates his or her own path between those two beacons. It’s trendy to serve Hendrick’s Gin with cucumber to underscore the cucumbery notes in the gin itself, for example. But Zearfoss—who says he’s more of a contrast guy—always asks for lime instead, because its angular acidity contrasts against the roundness of the gin’s cucumber notes.
Zearfoss and Loss are their own study in contrast and similarity. Zearfoss is a large, imposing man with a shaved, bullet-shaped head, small eyes, and regulation chef’s whites, and he speaks with the authority of someone who’s used to having his way in the kitchen. Loss is small, dark, and high-strung, with curly black hair and rapid speech. He’s in a suit, but tieless. It’s hard to give firm rules for managing contrast and similarity, Loss says. Almost everyone likes contrasting textures—a bit of crunch here, some creaminess there. Good chocolate brings its own internal textural contrast, as you snap off a bite only to have it melt in your mouth; ice cream gives a comparable textural treat. Often if chefs are using an unusual ingredient or presentation, they’ll pair this novelty—a contrast, in a way, with our expectations—with other, more familiar ingredients that help settle diners’ innate neophobia. But mostly, chefs just have to trust their instincts. “Sometimes it’s hard to identify what works,” he says. “It’s easier to pick out the flaws.”
One of their favorite lab exercises is to let students use the principles of similarity and contrast to pair wine with foods. The ideal wine for the job is sauvignon blanc. The similarity angle works because of sensory suppression and release, says Zearfoss. You take a sip of the wine and enjoy the balance of flavours. Then take a bite of green pepper. The pepper contains grassy-tasting methoxypyrazine, which makes your nose less responsive to the similar methoxypyrazine notes in the wine. As a result, the next time you sip the wine, you’ll probably notice one of its other flavours instead. The wine tastes different at each sip—a complexity that adds to its interest. Certain foods—pears, passion fruit, grapefruit, and others—will inhibit other parts of the wine’s aroma, and give different sensory experiences.
Still, a note of caution is in order. If anyone has a stake in the science of pairing food and wine, you’d think it would be Terry Acree. Acree is a flavour chemist at Cornell University with a wide-ranging intelligence that he loves to poke into the dusty cracks among scientific disciplines. Over the past few decades, Acree has put a huge amount of effort into cataloging the flavour molecules that we encounter in our food, and he’s published extensively on the flavour chemistry of wine. Here’s what he told me about the principles that determine whether a particular food and wine go together:
What does it mean to “go together”? My mother was an interior decorator, and when I was about five, I walked in and said to my mother, “My favorite color is red.” And she said, “No it isn’t, kid. That’s the stupidest thing I ever heard of. Nobody has a favorite color. Color has a place, and you have to find out where it belongs and where it doesn’t belong. It can only be your favorite if it’s in the right context.” So the first thing I’ve got to say about wine and food pairing is that it’s completely contextual, and almost entirely individual. It makes no sense to write a book on wine and food pairing, except to say there is such a thing as wine and food pairing, and go figure it out for yourself, because it’s your own pairing that counts.
As I’m talking wine with Zearfoss and Loss, the servers—CIA students getting some front-of-house practice, and clearly a bit nervous with the chef and professor at the table—bring our food. Zearfoss has ordered vitello tonnato, cold poached veal covered with a tuna sauce, while Loss is eating steak and fries. Loss pushes his fries into the center of the table for everyone to share. (Too many rich, delicious meals are clearly an occupational hazard to be resisted—both men ordered with restraint and ate abstemiously.) Zearfoss eats a fry and then gestures toward his vitello tonnato. “They should have served this with french fries. It’s the perfect combination.” He points at the uniform, soft beige of his dish. “There’s no brown, there’s no crunchy, there’s nothing here with a Maillard.” In short, not enough contrast for his taste.
The french fries come with a little crock of mayonnaise for dipping, which carries another lesson in balance. Without the mayo, the fries come across as too salty; with the mayo, they’re just right. “You don’t get the same salt impact with the mayo, because the fat coats the tongue,” says Zearfoss. “That’s the challenge of a chef. You’ve got the salt, the potato, the fat. Ultimately, what you’re trying to get in the customer’s mouth is a combination.”
Every creative chef approaches the challenge of balancing the flavours in a dish in his or her own way. Many think in terms of base notes, middle notes, and top notes, just like industrial flavourists do. A French onion soup, for example, might have base notes of the oniony flavours, middle notes of caramelized sugars from long cooking of the onions, and a top note of sherry vinegar to make the whole dish sing. Other chefs free-associate from one flavour to the next, building the finished dish in their imagination until it’s just right. There aren’t many common threads here, as a perusal of any collection of famous-chef cookbooks demonstrates clearly.
Where we can find commonalities, though, is in the chemistry of cooking. In a sense, a cook’s job is to collect and curate the right set of flavour molecules.
THE FIRST WAY cooks can intensify flavour is by extracting and concentrating aromatic molecules to deliver a more intense hit of flavour. Extraction is all about solubility: most flavour volatiles, such as the terpenes responsible for rosemary’s piney quality, are more soluble in oil than in water. As a result, if you toss a handful of rosemary into a stew, relatively few of the terpenes will end up in the liquid; instead, they vaporize into the air, making your kitchen smell delicious but doing nothing for the stew itself. Better to fry the rosemary in butter or oil first, along with onions and garlic, so that the terpenes extract into the oil and stay in the dish. Or buzz the herb in a blender with a little oil, then strain out the leafy bits, for an intensely flavoured rosemary oil to drizzle over the stew at the table.
On the other hand, sometimes you want to minimize extraction to keep as much flavour as possible in the food itself—especially if you’re going to discard the cooking liquid. Some of the key flavour molecules in asparagus, for example, are water soluble, so if you boil asparagus, they extract into the water and end up down the sink. Sauteeing the asparagus in butter or oil minimizes this loss and keeps more of the flavour in the vegetable.1 For the same reason, broccoli and beans—whose key odorants are oil soluble—retain their flavour better if steamed or boiled.
In high-end professional kitchens, chefs can concentrate flavour molecules extracted from herbs, spices, or almost anything else—including soil, seawater, and vegetation—using sophisticated (and expensive) distillation apparatus. Most of us lack the machinery to do that, but anyone can concentrate flavours via simple evaporation—for example, when we cook down a wine sauce to a syrupy consistency. Even the trainee chefs at the CIA quickly learn to have a cauldron of stock slowly reducing on the back of the stove. The process inevitably loses some of the flavour to the air, as a quick sniff will verify, but the reduced stock still packs a more intense flavour.
THE SECOND WAY cooks develop flavour in the kitchen is through cooking itself—the application of heat. The transformations that heat performs on flavour are largely a matter of breaking down big molecules such as fats and proteins into smaller, more volatile ones. This is most obvious with meat, so we’ll take that as an example. In the raw state, most meats have relatively little flavour. Anyone who’s eaten steak tartare or sushi knows how subtle, almost minimalist, the flavours are. In fact, you probably wouldn’t be able to tell much difference between cubes of raw beef, lamb, and pork. All have a mild flavour that’s often described as “bloodlike,” and a slight tang of iron. The vegetable world is diverse: sometimes we eat flower buds, sometimes leaves, sometimes roots, and sometimes fruits, and they carry a wide range of volatile molecules as attractants and as chemical defenses against marauding herbivores. By contrast, most of what we call “meat” is the muscle tissue of mammals or birds, and every one of those muscles is doing roughly the same thing with roughly the same set of biochemical tools. That’s why beef and lamb taste more like each other than beets and broccoli do.
The difference between one meat and the next is mostly a matter of the fat molecules they contain, with beef containing larger, less branched fat molecules and lamb, pork, and chicken increasingly more of the shorter, branched molecules. These fats—to be more precise, we should call them fatty acids—differ mildly in flavour on their own, but they break down into much different flavour molecules during aging and cooking. The fat that makes the most difference here, incidentally, isn’t the visible fat tissue on and between muscle fibers, the stuff you trim off if you’re meticulous about counting calories. Instead, most of the distinctive flavour of lamb, beef, and pork comes from the fat molecules known as phospholipids, which make up the membranes that enclose each cell. Researchers in England2 demonstrated this more than three decades ago by grinding up some lean beef, freeze-drying it, and extracting every last bit of intramuscular fat with petroleum solvent. After removing any traces of the solvent, they rehydrated the meat and cooked up the patties, boiling them in plastic bags for greater standardization. The aroma of the burgers—surprisingly, after all this chemistry—was indistinguishable from ordinary beef. The missing fat just didn’t matter. When they used chloroform and methanol to extract even the phospholipids, the resulting burgers had much less of the meaty aroma. So if you’re a carnivore, thank the cell membranes for the meaty flavour of your next stew or steak.
This mix of fats differs subtly depending on which part of the animal the meat comes from, the breed of animal involved, and diet. Grain-fed beef, for example, has more monounsaturated fatty acids, including the very tasty oleic acid. Animals that eat a pasture diet, by contrast, end up with more polyunsaturated fats, plus a few additional flavour compounds such as skatole,3 a molecule that adds a pleasant funk at the concentration it’s found at in meat, but which smells fecal at higher concentrations. But because cattle and sheep are ruminants—that is, they have complex stomachs where microbes break down their grassy diet, including the fats—the flavour of their meat doesn’t depend strongly on diet. In contrast, pigs and chickens have simple stomachs and the fats in their diets more often make it into the meat intact. That’s why you’ll see specialty pork producers proudly touting animals finished on chestnuts or acorns—like the prized jamón ibérico of Spain—but rarely see specialized diets used as a selling point for beef.
Most of the flavour of meat develops when we start to cook it, as the heat of cooking begins to break down fatty acids into smaller molecules, many of which carry strong flavours. (Dry aging of meat also breaks down fatty acids, so aged meat develops even more of these flavours.) The weak point in a fatty acid molecule is the carbon-carbon double bonds, the “unsaturated” parts of the molecule, so polyunsaturated fatty acids have more weak points, and break into smaller molecules, than monounsaturated or saturated fats do. These fatty acid degradation products account for most of the meaty aromas and flavours in cooked meat. They’re most obvious in meat that’s cooked at relatively low temperatures, such as by simmering or stewing. At higher temperatures, when meat begins to brown, another process dominates the flavour picture.
THE BROWNING REACTION—more formally known as the Maillard reaction after the early-twentieth-century French chemist who first described it—is responsible for a whole host of flavour changes that happen when foods cook. It’s the reason why bread is much tastier after it’s baked, why we roast our coffee beans, and why cauliflower roasted in the oven is more delicious than plain boiled cauliflower. It’s the reason we grill our steaks instead of poaching them, and why the best stews start by browning the meat in a little fat.
Even though it’s called the Maillard reaction, what we’re really dealing with here is a vast network of interconnecting chemical reactions, almost like a braided streambed. At the upstream end, the reaction begins when amino acids and sugars react with each other to form a series of unstable intermediate compounds. Those intermediates then react with one another, and sometimes with fatty acids and other molecules in the vicinity as the process quickly becomes too complex to keep track of fully. These products give the characteristic brown color, and many of them are also volatile flavour molecules. Each food has its own unique starting point into the stream as a result of its particular mix of amino acids and sugars, so the reaction proceeds differently. That’s why roast beef smells different from baking bread.
Lab chemists starting with pure amino acids and sugars have documented at least 621 different Maillard products;4 real foods, with their vastly greater diversity of chemical starting points, almost certainly produce even more products. We’ll leave their detailed identification to the flavour chemists. For now, suffice it to say that Maillard products are responsible for all the toasty, roasty flavours we get in baked goods, roasted and grilled meats, and anything else with a browned crust. Most of these Maillard products are present in only tiny quantities, but our senses are exquisitely sensitive at detecting them. On the downside, the Maillard reaction can also produce molecules like acrylamide and other carcinogens. Chemists are hard at work trying to find ways to guide the reaction into stream courses that enhance the beneficial flavours and avoid these unhealthy molecules.
For the cook, the most important thing to know about the Maillard reaction is that it requires high temperatures, typically well above the boiling point of water. That’s why fried and grilled foods brown, but stewed, steamed, and simmered foods don’t. It’s also why conscientious cooks dry the surface of their meat before searing—with less moisture to evaporate, the meat reaches Maillard temperatures more quickly so that more flavour develops. (Actually, the Maillard reaction does happen at lower temperatures, too—but so slowly that it rarely figures in cooking. Low-temperature Maillard reactions explain why powdered eggs sometimes turn brown after long storage, the impetus for some of the early research on the Maillard reaction. And black garlic, a cutting-edge ingredient these days, owes its complex, caramel-like flavour in part to Maillard reactions that take place over the span of a month at temperatures well below the boiling point.)
Because the Maillard reaction requires amino acids, or the proteins formed from them, it works most dramatically for protein-rich foods like meats, though most grains and vegetables also contain enough protein to generate some Maillard products. For some vegetables, especially sugar-rich ones like onions, a second browning reaction—caramelization—is also important. In caramelization, sugar molecules react with one another, rather than with amino acids, to form a similar cascade of flavourful products. Since sugars lack the nitrogen and sulfur atoms found in amino acids, however, caramelization produces a narrower range of flavour compounds, and less of the meaty, roasty flavours of Maillard products. From the cook’s point of view, though, both can be treated as a single, high-temperature browning process.
As complex as browning is, cooks can still steer the process to some extent. Meat that contains more fat will feed more fatty acid breakdown products into the reaction, producing more of the roasty furans that make a rib roast of beef or a leg of lamb so delicious—a big reason we like to roast these cuts without trimming off all the surface fat. Cooking temperature makes a big difference, too, by pushing the flow down one branch or another of the Maillard stream.
For any carnivore, of course, all this raises a practical question: What’s the best way to grill a steak? As it turns out, this has been the subject of sober scientific study by a meat scientist in (where else?) Texas named Chris Kerth. I phoned Kerth in his office at Texas A&M University, a hotbed of agricultural research, to get the scoop.
The hotter you cook a steak, the more you shift the balance from the beefy, brothy fatty acid breakdown products toward the roasty, nutty Maillard products. “You have a whole continuum that you can play with to customize the flavour,” says Kerth. “There’s a lot of restaurants where that’s their claim to fame, is cooking their steaks at 1,800 degrees—which is obviously going to be for a very, very short period of time. That’s what creates their signature flavour.” For thicker steaks, the outside would burn black before the middle cooks adequately, so these restaurants often sear their steaks to develop the right Maillard crust, then finish cooking in a gentler oven.
Most of us don’t have access to temperatures like that. To find out what works best at more typical cooking temperatures, Kerth embarked on the lab version of a cook-off.5 He bought whole beef strip loins and cut them into steaks either one-half, one, or one and one-half inches thick, then cooked the steaks to well done at one of three different temperatures: 350, 400, or 450 degrees Fahrenheit. It’s hard to get the temperature exact on a grill, so Kerth cooked his steaks in preheated cast-iron skillets instead. (Science demands some sacrifices. The bigger sacrifice here, actually, is that Kerth fed his cooked steaks not to hungry Aggie volunteers, but to a gas chromatograph, again in the interest of greater precision.) As you’d expect, the thinner steaks (and those in the hotter pans) cooked through more quickly, which left less time for roasty Maillard flavours to develop. In thin steaks, as a result, tallowy, fatty, green flavours tended to dominate, while thicker steaks were more roasty, nutty, and buttery—but also had more acrid flavours. Since then, Kerth has also cooked experimental steaks for actual people, and he reports that most of them preferred the flavour of thick steaks cooked at relatively low temperatures. “That’s been my recommendation, is to find a little bit lower temperature,” he told me. “There’s also an impact on tenderness—the lower, slower grilling results in more tender meat.”
THE THIRD MAIN way that cooks can create flavour in the kitchen is through fermentation, the process that creates such diverse flavour wonders as cheese, bread, soy sauce, kimchi, and beer and wine. Actually, fermentation is probably better described as a form of herding than cooking, because what we’re really doing is managing the microbes that are doing the hard work of breaking down sugars and other molecules in the food, releasing volatile flavour molecules as they go. Often, a whole ecosystem of microbes—bacteria, yeasts, and other fungi—is involved in a fermentation. As we saw for wine in the previous chapter, the outcome of fermentation depends on exactly which microbes are involved.
That’s easiest to see in the case of cheese. Several species of lactic acid bacteria attack the lactose in milk, generating sour-tasting lactic acid as a waste product. As the milk acidifies, its proteins curdle into a semisolid mass that the cheesemaker strains and presses to form the basic starting point for the cheese. Now things get far more diverse, as cheese makers can encourage different sets of microbes6 to take over the job. If a fungus called Penicillium camemberti settles in, its microscopic filaments form a whitish rind on the outside of the cheese and secrete enzymes that break down the casein protein, gradually liquifying the center of the cheese and generating the sharp, ammonia aromas of degraded proteins that mark a ripe Camembert. On the other hand, the related Penicillium roqueforti favors a different set of enzymes that break down the milk fats in the cheese, yielding sharp-flavoured fatty acids and 2-heptanone, the signature flavour compound of blue cheeses such as Roquefort. Bacteria in Swiss cheese produce propionic acid, which contributes the nutty flavour of that cheese. The reddish rind of Limburger cheese is rich in the bacterium Brevibacterium linens, which produces sulfury by-products that give the cheese its stinky, body-odor quality (an apt analogy, since a related species lives in human armpits). Many other microbes contribute minor notes to the flavour of cheeses—indeed, the complexity these minor microbes add is the main reason for the deeper, more complex flavours of raw-milk cheeses. (These complex microbial ecosystems are also what makes sourdough bread more flavourful than bread risen from plain old, cultured baker’s yeast.)
ONE OF THE big questions both professional chefs and amateur home cooks want to know about flavour is which ingredients go well together. Until recently, however, every cook—from the most primitive tribeswoman to world-famous chefs with three Michelin stars—has worked entirely by trial and error. We learn what goes well together by combining ingredients and seeing if the result tastes good. (Actually, most people cook what their culture has always cooked: Vietnamese savor fermented fish sauce, hot chilis, and lime; those from southern India favor mustard seed, coconut, and tamarind; southern Italians mix tomato, garlic, and basil. But this merely pushes the trial and error into the distant past.) The approach has obviously worked well, as any stroll through the restaurant districts of New York or San Francisco would reveal. But it’s hard to get very far off the beaten path using trial and error. To really explore the far reaches of possibility, it would help a lot if we could find some general principles that underlie and guide our choice of flavour combinations that work.
You’ll sometimes hear chefs cite the principle, “What grows together, goes together,” as a basis for their flavour pairings. It’s not hard to come up with some outstanding examples: morels with asparagus, lamb with thyme and rosemary from the Mediterranean hillsides where it once grazed, venison with cranberries and wild forest mushrooms. Zearfoss particularly likes the combination of apricots with the chanterelle mushrooms that grow in apricot orchards. But is there any scientific reason this principle should hold true?
At one level, yes—because it directs your attention to ingredients that are local and in season, and therefore most likely to be at their peak of flavour. Why would you not pair morels and asparagus in the springtime, when they’re both at their best? At a slightly deeper level, the grows-together/goes-together principle can be seen as an endorsement of traditional flavour pairings. After all, for most of the history of civilization, cooks had no choice but to combine foods that grew together—local, seasonal food was all that was available (especially if you think of winter storage foods as another type of seasonal ingredient). Over the course of generations, cooks learned which combinations were most pleasant, and these became fixed by tradition. Meanwhile, no one much notices the pairings that didn’t make the cut. (Spinach grows well in the springtime, too, but no one makes a big deal about pairing spinach with morels.) The upshot is that the grows-together/goes-together pairs that come to mind are largely the ones that have passed our ancestors’ taste tests. Following one of those pairings is likely to yield a better result than you’d get with a random pairing of two ingredients—chili peppers and turnips, say—that aren’t naturally found together and therefore haven’t been vetted by tradition.
On the other hand, there’s probably no basis in the science of flavour chemistry for expecting that ingredients from the same place would combine particularly well. As we’ve seen, the molecules that give fruits and vegetables their flavour do not come from the soil directly but are made by the plants themselves. That means there’s no reason why two plants that grow together should be more likely to make similar flavour molecules, or molecules that are compatible in some other way.
We can push this a little further, though. Because our expectations and our previous experience play a big role in our perception of flavour, and especially in our flavour preferences, we might predict that familiar, traditional combinations of ingredients—based, of necessity, on foods that grow together—would strike us as more pleasant, on the whole, than novel combinations. What grows together, goes together not necessarily because it’s intrinsically better, but because we’ve tried it before and we expect to like it.
THERE’S ANOTHER PROBLEM inherent in puzzling out pairs of ingredients that go well together: it can quickly become overwhelming, because the number of possible combinations explodes far too fast to evaluate all of them. Consider the pizza problem: If I have twenty-five different toppings, I can make twenty-five different one-topping pizzas, so it’s reasonable to ask you which one you like best. But if I’m making a two-topping pizza, you’ve got six hundred different combinations to evaluate (that’s 25 × 24, for the math geeks—we won’t consider pepperoni and pepperoni as a two-topping option), and only an obsessive would work through the entire list to pick the best pair. And if you’d like a three-topping pie, you’ve got nearly fourteen thousand combinations to choose from. It’s little wonder most pizzas use the same standard set of toppings, over and over and over again.
A few years ago, Michael Nestrud, a sensory scientist then at Cornell University who is also a CIA-trained chef, realized that an arcane branch of mathematics called graph theory might provide fresh solutions to the pizza problem by helping identify appealing food combinations more quickly. Despite its name, graph theory has nothing to do with the bar charts and zigzag lines most of us call graphs. Instead, it’s all about groups of connected objects—in this case, foods that go together. Nestrud’s insight was that you should be able to recognize a good three-topping pizza by seeing that each topping pairs well with the other two. Mathematically, this is identical to picking out “cliques” of Facebook friends where every member of the clique is friends with every other member.
So Nestrud made a list of pairs of possible pizza toppings and asked several hundred university students to give a thumbs-up or thumbs-down to each suggested pairing. From their answers, he compiled a list of “good” topping pairs, such as pepperoni and mushroom. Then he used the mathematics of graph theory to pull out sets of three or more toppings for which all the pairs were on the “good” list. These three-topping pizzas also turned out to be more popular than you’d expect7 by chance.
Of course, you don’t need advanced mathematics to top a pizza. However, Nestrud’s approach attracted serious interest from the U.S. Army, which desperately wanted to make tastier field rations. Soldiers in combat situations need food that is light, nutritious, and—above all—durable, and for decades, that has meant the dreaded MRE (the acronym stands for “meal, ready-to-eat”). MREs are precooked meals sealed into foil pouches. From the Army’s point of view, MREs are great. They’ll last for years, and soldiers can just grab them and go. The problem is, soldiers quickly get bored with the meals. It’s already hard to get soldiers to eat enough when they could be shot or blown up at any moment, and boring food doesn’t help matters. So the Army puts a lot of effort into making MREs as appealing as possible under the circumstances.
Each MRE contains an entree, side dishes, fruit, dessert, snacks, condiments, candy, and beverages, each chosen from up to thirty-two different options. These components could conceivably be mixed and matched in any which way—more than twenty-two billion different combinations in total. Which ones would the soldiers like? The Army hired Nestrud—fresh out of graduate school with his PhD in pizza toppings—to figure it out.
Using the same approach he took with the pizza, Nestrud designed a questionnaire that listed possible pairs of items and asked soldiers whether they’d want to eat them at the same meal: beef roast with vegetable couscous, meatballs and gravy with barbecue sauce, beef taco filling with jalapeño cheese spread, chicken fajita with bacon cheese spread, and so on. Using the pairs most commonly approved by the soldiers, Nestrud could then assemble entire MRE menus that he predicted would prove popular (top of the list: chili with beans, Mexican mac and cheese, herb-citrus seasoning, crackers with chunky peanut butter, fruit, cookies, and cheese pretzels), and others that he predicted the soldiers would hate. When he showed those menus to actual soldiers and asked them to rate their compatibility, the soldiers’ ratings matched his predicted ones almost perfectly—real-world validation that you really can use Nestrud’s approach to predict flavour pairings.
Nestrud’s next move was to a consulting company, where he used his graph-theory technique to help identify which snack foods people like to buy at the same time. Grocery stores and fast-food restaurants could then display these “go togethers” next to each other in the store so that consumers who bought one might happen to buy the other, as well—seemingly on a whim, but really the result of careful thought and planning by the seller. (Nestrud has no idea whether his clients actually put his recommendations into practice.)
In his current job as sensory scientist with Ocean Spray, America’s largest cranberry company, Nestrud is trying a different twist to this flavour-association game. Every day during the winter of 2015–2016, he searched through Twitter’s daily archives and gathered every tweet that mentioned certain flavour-related key words. (The details are secret, of course, and Nestrud was careful not to mention the word “cranberry” when I spoke to him, but it’s a pretty safe bet that it was one of his key words.) Cleaning up the data took a lot of work. He had to toss out key word hits that didn’t really refer to flavour, such as references to cranberry-colored paint for the bathroom or mentions of orange that referred to the color of the uniforms of the Denver Broncos football team. And conversely, he had to ensure that “cranberry,” “cran-apple,” and “cran-raspberry” grouped together as similar flavours, and likewise “orange,” “mandarin,” and “tangerine.”
In the first four months, Nestrud accumulated nearly twelve thousand relevant tweets—enough to get a good sense of what other flavours the Twitterverse thought of when it thought of cranberry. Better yet, his sample included both Thanksgiving and Christmas, as well as the postholiday lull, so he could see how flavour pairings changed through the seasons. The results might not lead directly to new products, but they’re the first step in a long creative process. “The ultimate goal is not to make any final decisions,” says Nestrud. “It’s to generate hypotheses about products we wouldn’t have thought of on our own that we can then go out and validate with real consumer testing.”
PROFESSIONAL CHEFS LIKE to push the limits of tradition, and so do adventurous eaters. One of the great joys of eating is to discover a novel mix of ingredients that works unexpectedly well together, breaking us out of the comfortable confines of tradition and into a new world of possibility. We can find these new combinations through trial and error, of course, or we can rely on the intuition of gifted cooks, which is essentially trial and error inside the cook’s imagination. But perhaps our search for delicious novelty can also get some guidance from what we know about the science of flavour.
One of the first really promising steps in this direction came a decade ago, when world-famous chef Heston Blumenthal of the Fat Duck restaurant was experimenting with salty ingredients8 in desserts, and discovered that white chocolate and caviar make a great flavour combination. The pairing was so bizarre—yet so delicious—that Blumenthal mentioned it to a colleague at an industrial flavour company. A little work soon revealed that both members of this unlikely pairing are rich in a compound called trimethylamine, which has a fishy flavour.
This got Blumenthal thinking. If a shared flavour molecule accounts for the success of this odd pairing, maybe similar “molecular rhymes” might point us to other surprising flavour combinations. The idea makes some intuitive sense. As we’ve seen, chefs often balance similarity and contrast in building their dishes—and since flavour is all about molecules, then similar flavours should share flavour molecules. As Blumenthal pursued these molecular similarities, he came up with a whole kitchenful of brilliant, unexpected matches: liver and jasmine, which share sulfur compounds; carrots and violets, which share a molecule called ionone; pineapple and blue cheese; snails and beets.
In the years that followed, Blumenthal’s insight sparked a whole gastronomic movement. Going by the name of “food pairing,” it makes these molecular rhymes the centerpiece of combining foods. There’s even a commercial service (foodpairing.com) that, for the price of a monthly subscription, will let professional chefs and enthusiastic amateurs start with any ingredient and follow a web of molecular similarities to find other foods with supposedly complementary flavours.
A Canadian sommelier named François Chartier has begun to investigate the pairing of wine with food along the same lines, based on chemical similarities between ingredients in the food and aroma compounds in the wine. For instance, Chartier suggests pairing a rosemary-scented lamb stew with a dry Riesling wine, to take advantage of the citrusy, floral-smelling molecules in that wine, which echo those in the rosemary. This “molecular sommellerie” was novel enough to earn Chartier’s book9 on the subject, Taste Buds and Molecules, a prize as the “best innovative food book in the world” at the 2010 Gourmand World Cookbook Awards.
You’d think, with all this excitement, that food scientists would be eager to sink their teeth into molecular food pairing to see if it really works. But very few have actually done so—and even fewer have published their results in the scientific literature. (Foodpairing, Inc., the company selling food-pairing ideas to chefs, hasn’t released any evidence to back up its approach.)
The obvious test would be to have people rate how well pairs of ingredients go together, and see whether those that share more flavour molecules receive higher ratings. Wender Bredie, a Danish food scientist at the University of Copenhagen, did exactly that a few years ago, using fifty-three different pairs of ingredients ranging from cinnamon and apple to cinnamon and garlic, malt and cocoa to malt and blue cheese. They found that the number of flavour molecules in common made absolutely no difference to the rated pleasantness. “I have never seen an experimental study with a correlation that was so low,” recalls Bredie. (It’s worth noting that Bredie’s study—like an earlier study by a different group that reached much the same result—was only presented at a scientific conference, not published in a scientific journal. That means the research has not been vetted by other experts, so the conclusions should be regarded as preliminary.)
Bredie’s group did find one interesting result, though. Pairs with fewer molecules in common tended to be perceived as more novel than those that shared more molecules—a feature at least a few high-end chefs might like to make use of, though novelty is not the same as pleasantness. “With high-end restaurants, you want something unique and surprising for your customers,” says Bredie. “And it doesn’t have to be nice. If you go to Noma”—the New Nordic restaurant in Copenhagen that was rated number one in the world for several years—“the foods aren’t very nice. You go there, and you have a fantastic experience. But if you ask the customers, ‘Is this something you’d really like to eat more frequently?’ they’d probably say no.” (When Bredie says that Noma’s dishes “aren’t very nice,” he means that the ingredients and techniques are often unusual and challenging, such as the moss cooked in chocolate that is featured on the menu as I write this.)
A second approach would be to look at the flavour combinations that people actually use, and count the molecules their ingredients have in common. If these real combinations are more likely to share flavour molecules than random sets of ingredients are, that would be evidence that molecular rhymes really do make combinations taste better. The data are out there: the Internet age has provided a huge treasure trove of real flavour combinations, in the form of online recipes, and anyone with a few hundred dollars to spare can subscribe to a database that lists all the flavour compounds in any given food ingredient. The big challenge is making sense of the tangled web of recipes, ingredients, and flavour molecules.
Enter Sebastian Ahnert. By day a theoretical physicist at the University of Cambridge and by night an enthusiastic amateur cook, Ahnert has exactly the skill set necessary to pick apart the problem. A few years ago, he and his colleagues downloaded more than fifty-six thousand recipes from three online recipe archives (Epicurious, Allrecipes, and a Korean database called Menupan) and studied their molecular overlaps.10 Real recipes, they found, had a slight tendency to share more flavour molecules in common than random collections of ingredients—but only for North American, Western European, and Latin American cuisines, where common ingredients like milk, eggs, butter, and wheat share overlapping flavour profiles. Asian recipes actually shared fewer molecules than random ingredients, because their most common ingredients, such as soy sauce, scallion, ginger, and rice, have largely nonoverlapping flavours. When Ahnert left those most common ingredients out of his analysis, he found no evidence at all to support the food-pairing hypothesis.
Ahnert’s analysis made a big splash when it was published in a topnotch scientific journal, but he wasn’t satisfied. Recipes aren’t an ideal starting point, because some ingredients—flour and eggs come to mind—are often included more for structural reasons than because they contribute important flavours. So Ahnert went back to the drawing board. This time, instead of recipes, he used ingredient pairings recommended by well-known chefs, which he found in a best-selling book called The Flavour Bible11 by Karen Page and Andrew Dornenburg. He found that chef-recommended pairs share more flavour molecules than randomly paired ingredients—and the pattern gets stronger when he counts only the most abundant flavour molecules, or those with the most food-related odors.
So maybe there’s something to food-pairing theory after all. Not everyone is convinced yet, particularly because as I write this, Ahnert has not yet published his most recent reanalysis. But even if foods that go well together do tend to share flavour molecules, that’s not the same as saying that foods that share flavour molecules necessarily go well together. This molecular-rhyming approach is probably an idea generator at best.
TO EXPLORE A really high-tech approach to finding unusual, exciting flavour combinations, I visited IBM’s Thomas J. Watson Research Center in Yorktown Heights, New York. The company has a long history of taking on the biggest challenges in artificial intelligence, and the Watson Research Center is where it all happens. IBM’s Deep Blue supercomputer made headlines back in 1997 when it beat world chess champion Garry Kasparov in a six-game match. Then, in 2011, Deep Blue’s successor, Watson (named, like the research center itself, after the company’s longtime president from the first half of the twentieth century), beat two human champions at the quiz game Jeopardy. With Watson’s win under their belts, IBM’s researchers started looking for new applications for their expertise. Why not, they asked, turn Watson’s immense powers to the kitchen? After all, cooking is both highly creative and a familiar, everyday activity that millions of people do regularly. And Watson’s computing prowess should pay off big time, because the computer could learn more recipes, more ingredients, more techniques than any human ever could, just like it had for Jeopardy trivia. Let’s do it, the Watson team decided.
The Thomas J. Watson Research Center sits in forested hills just off the Taconic State Parkway, less than an hour’s drive north of midtown Manhattan. The main building—a vast, three-story curving facade, designed by renowned architect Eero Saarinen—looms over visitor parking, and the main entrance, sheltered by a thrusting, flaring overhang, leads into a 1960s-futuristic foyer. It’s all very high concept, very expensive looking, and very formal. Just what you’d expect from IBM, the corporation long notorious for its severely conservative dress code.
In such a setting, Florian Pinel comes as a distinct shock. What you first notice about the French-born software engineer is not his broad face, blue eyes, or unruly mop of stringy brown hair. It’s the four stainless-steel studs that pierce the corners of his mouth and lower lip, and the longer, blade-shaped fin that emerges from the center of his lower lip, just above his chin. Dressed in jeans and a ratty shirt, instead of IBM’s traditional white shirt and tie, Pinel looks more at home in a restaurant kitchen than in an IBM conference room.
Appearances aren’t deceiving in this case—Pinel is indeed comfortable in the kitchen. While working at IBM, he spent his weekends studying at New York’s respected Institute of Culinary Education, earning a chef’s ticket in 2005. For a while after that, he worked Saturday nights as a line cook in a Manhattan restaurant, just for the thrill of it. “That was a big rush,” he recalls. Eventually, though, he gave that up and focused on cooking at home in his newfound leisure time. When Watson came along, he was ready.
How do you teach a computer how to cook? Not the way you’d teach your kid, by having him or her stand by your elbow and watch. And not the way Pinel learned, in culinary school. Instead, you feed the computer data. Lots and lots of data. Food chemists have identified the key flavour chemicals in most ingredients, and psychologists have measured how pleasing we find each of them. Cyberspace is full of recipes that show how people all over the world cook: which ingredients they use, and how they combine them. Pinel and his team input all this information into Chef Watson’s memory bank. From this mass of data, the computer chef extracted patterns: the sets of ingredients that were likely to go well together, and the sequence of steps to use in combining those ingredients. (A big help for the latter was getting access to Bon Appétit magazine’s archive of nine-thousand-plus recipes, all tested and carefully edited into a standard format.)
Anyone with a computer can consult Chef Watson for free (at least as of this writing) at ibmchefwatson.com. You simply type in an ingredient or two, and Chef Watson suggests a few other ingredients you might try. Once you’ve settled on a core set of four ingredients—and, optionally, specified a style such as French, summer, or vegetarian—you can choose from a list of suggested recipes, complete with measurements and cooking techniques. It’s that simple.
Behind the scenes, though, a lot is going on. To come up with its recommendations for ingredients that work together, Chef Watson looks for ingredients that are already used together in existing recipes somewhere in the world, or sets of ingredients that share several flavour chemicals like Heston Blumenthal’s white chocolate and caviar. But for Pinel, finding these sets of ingredients alone isn’t enough to make Watson truly creative. “We think there are two things that make something creative,” he says. “It has to be novel, and it has to be valuable.” For a recipe, “value” equals deliciousness—something that Chef Watson could estimate from its knowledge of which flavour chemicals people like best, and from its calculations of chemical overlaps. And novelty was an easy one—Watson just calculated how similar a recipe’s ingredients are to those used in other recipes. Tomatoes, garlic, oregano? Not very novel. Asparagus, pig’s feet, and Indian spices? You bet. For each ingredient combination, Chef Watson gives you a “synergy” score—essentially a composite of compatibility, pleasantness, and surprise. A high synergy score means Watson is confident of its choice of ingredients, says Pinel. “This is going to work well, and it’s not going to be trivial, either.”
For a foodie, software like this is endlessly intriguing, and it’s easy to get sucked down the rabbit hole, browsing one idea after another. But it doesn’t take much experimentation to realize that for all of Chef Watson’s knowledge and computing power, it lacks the finely honed kitchen instincts of a Michelin-starred chef. Instead, it’s more like your brilliant-but-loopy buddy who blurts out whatever thought happens to pass through his mind, no matter how bizarre. I tested Chef Watson in late January, right around Robbie Burns Day, the great Scottish haggis-and-whisky fest. Since the traditional accompaniment, “neeps and tatties” (turnips and potatoes, for the uninitiated), is traditionally boring, I thought I’d see whether Chef Watson had any better ideas. I started with the key ingredient, turnips, and specified that I wanted a Scottish recipe. Then—to continue the Scottish theme—I added another of that country’s favorite foods: beer.
The chef’s suggestion: “Scottish Turnip Meatball,” a veal/turkey meatball served in a sauce of chili powder, the Indian spice mix called garam masala, turnip, avocado, and clam juice. It sounds like a bizarre jumble, and I almost didn’t try it. But in the spirit of research, I finally subjected my family to it one night—and, to our surprise, it worked pretty well. The creamy unctuousness of the avocado made just the right counterpoint to the turnip’s bitterness, and the garam masala and clam juice added a subtle depth to the flavour profile. In fact, it was good enough that we served it to dinner guests a few weeks later. Maybe there really is something to this whole molecular food-pairing thing.
Want something to cook for your Super Bowl party? Just specify “Super Bowl” in the “Pick a Style” field, and Chef Watson suggests some ingredients to start with: raisins, garlic, chocolate chips, and endive. The synergy score is off the charts, well over 90 percent, so for some reason Chef thinks this is a great combination. I think I’ll try another spin of the wheel.
Watson’s next suggestion: pork belly, shallot, ginger, and white pepper. Ah, more promising. One of the suggested recipes for that combination is Superbowl Pork Belly Bolognese, which sounds plausible. But the recipe itself is over the top, calling for ground chicken breast, ground pork belly, and ground chicken wings (bones in, I wonder?), as well as chorizo sausage and a quarter cup of horseradish.
To drink? Why not a Cauliflower Bloody Mary, made with Pernod and ouzo, not vodka or gin, and swapping out the seasoned tomato juice for a puree of cauliflower, shiitake mushrooms, and onion. Garnish with grape wedges, Chef suggests—a peculiarly fussy take on the usual lime wedge, and certainly one I wouldn’t have thought of on my own. (Weird recipe steps like that show up frequently, because Watson bases its procedures on existing Bon Appétit recipes, substituting similar ingredients as needed. In this case, probably, Watson decided to substitute grapes for lime because both are fruits with plenty of acid, and it just borrowed the technique verbatim. That likely also explains the ground chicken wings in the pork belly bolognese. You’ll soon find your own humorous combinations.)
I’m poking fun here, because Chef Watson leaves some low-hanging fruit. But even its goofiest ideas often have a kernel of real inspiration. When I started with Italian sausage and broccoli, it suggested a recipe adapted from a braised brisket dish: rub the sausage with a seasoning mix, “working the paste into all the cracks,” then put the sausage in a casserole “fat side up.” Clearly Watson doesn’t understand the difference between a brisket and a sausage. But that night as I lay in bed, I realized that a spice rub might add a nice touch to something as prosaic as a grilled bratwurst, or even a hot dog. Good idea, Chef. And maybe, with the right drink, a garnish of grape wedges wouldn’t be such a bad idea after all.
The jury’s still out on whether Chef Watson is a major step forward in culinary creativity or just an amusing sideline. So far, the app is generating about fifty thousand ingredient pairings a month, says Pinel—which sounds like a lot until I realize that I’ve probably done fifty today, all by myself. Some users just look for suggestions of ingredient pairs, and build or adapt their own recipes; others click on complete recipes. The next step, says Pinel, is to add nutritional information to the mix, so that Chef Watson can double as Dietician Watson.
WITH MENU CHOSEN and kitchen chemistry properly deployed to marshal just the right flavour molecules, there’s one more step that a savvy cook can take to boost the flavour of a meal: serve it properly.
We’ve seen that presentation can legitimately be considered as part of the flavour experience: Changing the color, shape, or weight of the plate or bowl can make the food taste sweeter or more bitter. Charles Spence, the psychologist behind that study, took that notion still further in another experiment. Working with Charles Michel, a professional chef, Spence presented volunteers with a salad containing identical ingredients plated in one of three ways. Some diners got an ordinary tossed salad, some got a salad with each ingredient stacked neatly in separate piles, and some got a salad that was dramatically plated in a splash of colors and shapes that resembled a Wassily Kandinsky painting.12 The diners who got the Kandinsky salad found it both more pleasing aesthetically and also tastier than those who got the boring versions. For any cook, at home or in a restaurant, an attractive presentation is more than just window dressing—it makes the food itself more flavourful.
We can apply the same principle to wine: drinking it from an elegant glass should make it taste better. There are functional reasons for this as well as the psychological ones. A large, tulip-shaped glass that tapers inward toward the lip offers more space for volatile molecules to gather above the liquid, enhancing the aroma. Studies have verified that the same wine does indeed taste more pleasant from a glass like this than from a straight-sided water glass. On the other hand, there’s little hard evidence that you get any extra flavour boost from having a different glass shape for Bordeaux-style wines than for Burgundies, as some high-end crystal makers suggest. I recommend spending your money on the wine, instead. (I asked one expert in the oenology department at the University of California, Davis, who’s done some of the wine glass studies, what kind of wine glasses she uses. “Whatever I get free from the winery,” she replied, with a laugh.) Many wine lovers like to decant their wine before serving, especially for reds. Besides the obvious benefit of eliminating the sediment that some wines accumulate in the bottle, decanting is supposed to “let the wine breathe” and improve its flavour. In molecular terms, decanting allows the escape of some of the off flavours that may have developed in the bottle, and it may also allow oxygen—nearly absent in the bottle—to react with the wine and create some new flavour compounds. Whatever the reason, it does seem to work.
But if a little decanting is good, would more be better? Nathan Myhrvold thinks so. Myhrvold, the former chief technology officer at Microsoft (and before that a physicist who studied under Stephen Hawking) has, in recent years, taken an engineer’s approach to high-end cuisine, where he clearly enjoys finding iconoclastic approaches to familiar kitchen tasks. For wine, Myhrvold recommends “hyperdecanting” by pouring the wine into a blender and buzzing it on high for thirty to sixty seconds. “Even legendary wines, like the 1982 Chateau Margaux, benefit from a quick run through the blender,”13 he writes in his six-volume cookbook Modernist Cuisine.
Of course I had to try it (though not with the Chateau Margaux, which is well beyond my means). I decanted one-third of a bottle of wine the usual way, by pouring it into a decanter, and put a third into the blender, keeping the last third undecanted in the bottle. Then I had my son, who was underage at the time, pour all three into numbered glasses so that the rest of the dinner party could taste them without knowing which was which. Myhrvold turned out to be right—sort of. The wine from the blender practically leaped out of the glass, with a huge and vivid aroma. It was much better than the other two, which we all thought indistinguishable.
But when I returned to the glass five or ten minutes later, the blenderized wine was lifeless and spent—it appeared to have no flavour left. If I’m pouring six or eight small glasses from a bottle to be consumed immediately, I’d definitely go the blender route. If I’m going to share a bottle with my wife over a leisurely dinner, though, I’ll keep the blender in the cupboard and pour the normal way.